Principles of Corrosion Engineering and Corrosion Control, Elsevier (2006), 0750659246
660 Pages

Principles of Corrosion Engineering and Corrosion Control, Elsevier (2006), 0750659246

Course Number: ME ME78212, Spring 2010

College/University: Institut Teknologi Bandung

Word Count: 244776


Document Preview

• ISBN: 0750659246 • Pub. Date: September 2006 • Publisher: Elsevier Science & Technology Books PREFACE The phenomenon of corrosion is as old as the history of metals and it has been looked on as a menace which destroys metals and structures and turns beauty into a beast. Our human civilization cannot exist without metals and yet corrosion is their Achilles heel. Although familiarity with corrosion is...

Unformatted Document Excerpt
Coursehero >> Indonesia >> Institut Teknologi Bandung >> ME ME78212

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

ISBN: • 0750659246 • Pub. Date: September 2006 • Publisher: Elsevier Science & Technology Books PREFACE The phenomenon of corrosion is as old as the history of metals and it has been looked on as a menace which destroys metals and structures and turns beauty into a beast. Our human civilization cannot exist without metals and yet corrosion is their Achilles heel. Although familiarity with corrosion is ancient, it has been taken very passively by scientists and engineers in the past. Surprisingly, it is only during the last six decades that corrosion science has gradually evolved to a well-defined discipline. Corrosion Science and Engineering is now an integral part of engineering curriculum in leading universities throughout the world. With the rapid advances in materials in the new millennium, the demand for corrosion engineering courses has dramatically increased. This has necessitated the need for the publication of new books. Professor U. R. Evans, Prof. H. H. Uhlig and Prof. M. Fontana wrote a classical generation of basic text books covering the fundamentals of corrosion science and engineering. These books served as texts for decades and some of them are still being used. Several new books in corrosion have been published in recent years to cater to the needs of corrosion science and engineering students. As a teacher of corrosion engineering for the last twenty-five years, I found the material to be deficient in corrosion engineering content. However, sufficient coverage was given to the understanding of corrosion science. In this book, chapters on cathodic protection, materials selection, concrete corrosion and coatings have been written to cater to the needs of corrosion engineering students as well as corrosion engineers. These chapters contain simple and sufficient information to enable students to design corrosion preventive measures. A large number of illustrative problems are given in the chapter on cathodic protection to show how simple cathodic protective systems may be designed. The chapter on material selection is devoted to an understanding of the art of selection of materials for corrosive environment and applying the knowledge of corrosion prevention - the objective of corrosion engineering students. Concrete corrosion is a global problem and of particular interest to civil, chemical and mechanical engineering students. The chapter on boiler corrosion would be of specific interest to corrosion engineering students and corrosion engineers who desire to refresh their knowledge of the fundamentals of boiler corrosion and water treatment. The chapter on concrete corrosion has been added in view of the global interest in concrete corrosion. It presents the mechanism of rebar corrosion, preventive measures and evaluation methods in a simplified form with eye-catching illustrations. And the unique feature of the book is the follow-up of each chapter by keywords, definitions, multiple-choice questions, conceptual questions and review questions. A solution manual will soon be available to students containing solutions of problems and answers to multiple-choice questions. These are intended to test the readers comprehension of the principles covered in the text. I have put all my lifetime teaching experience into writing this book for corrosion engineering students in the sophomore or junior year. Graduate students lacking background in corrosion will also benefit from the book. It is expected that the students would be able to understand the principles of corrosion science and engineering in a simple and logical manner and apply them for solutions to corrosion engineering problems. This book is written with a new approach and new philosophy and it is hoped that it will fulfill their aspirations. While writing this book, I passed through the most turbulent period of my life with the loss of my most beloved son Intekhab Ahmad who passed away suddenly on April 20, 2004 leaving Preface XV a sea of unending tears and sadness in my life. It was followed by my own sickness, operation and desertions of some of my closest ones. I am grateful to Almighty Allah that I passed through this traumatic period and am able to complete the book. The success of my efforts will depend on how well this book is received by the students and the corrosion community. This book will not only be found very useful by corrosion engineering students but also by corrosion scientists and engineers in their problems in their professional capacity and those interested in corrosion. Table of Contents 1. Introduction 2. Basic concepts 3. Corrosion kinetics 4. Types of corrosion materials and environments 5. Cathodic protection 6. Corrosion control by inhibition 7. Coatings 8. Corrosion prevention by design 9. Selection of materials for corrosive environments 10. Atmospheric corrosion 11. Boiler corrosion 12. Concrete corrosion Index I N T R O D U C T I O N TO CORROSION Thenard (1819) suggested that corrosion is an electrochemical phenomenon. BACKGROUND • Hall (1829) established that iron does not rust in 'Lay not up for yourselves treasures upon earth where the absence of oxygen. • moth and rust doth corrupt and where thieves break- Davy (1824) proposed a method for sacrificial protection of iron by zinc. through and steal' (Mathew6:14) • De la Rive (1830) suggested the existence of microcells on the surface of zinc. he word corrosion is as old as the earth, but it has been known by different names. The most important contributions were later Corrosion is known commonly as rust, an unde- made by Faraday (1791-1867) [1] who estabsirable phenomena which destroys the luster and lished a quantitative relationship between chembeauty of objects and shortens their life. A Roman ical action and electric current. Faraday's first philosopher, Pliny (AD 23-79) wrote about the and second laws are the basis for calculation of destruction of iron in his essay 'Ferrum Cor- corrosion rates of metals. Ideas on corrosion rumpitar.> Corrosion since ancient times has control started to be generated at the beginaffected not only the quality of daily lives of ning of nineteenth century. Whitney (1903) people, but also their technical progress. There provided a scientific basis for corrosion control is a historical record of observation of corrosion based on electrochemical observation. As early by several writers, philosophers and scientists, but as in eighteenth century it was observed that there was little curiosity regarding the causes and iron corrodes rapidly in dilute nitric acid but mechanism of corrosion until Robert Boyle wrote remains unattacked in concentrated nitric acid. his 'Mechanical Origin of Corrosiveness.' Schonbein in 1836 showed that iron could be Philosophers, writers and scientists observed made passive [2]. It was left to U. R. Evans to procorrosion and mentioned it in their writings: vide a modern understanding of the causes and control of corrosion based on his classical electro• Pliny the elder (AD 23-79) wrote about chemical theory in 1923. Considerable progress towards the modern understanding of corrosion spoiled iron. • Herodotus (fifth century BC) suggested the was made by the contributions of Evans [3], Uhlig [4] and Fontana [5]. The above pioneers use of tin for protection of iron. of modern corrosion have been identified with • Lomonosov (1743-1756). • Austin (1788) noticed that neutral water their well known books in the references given at the end of the chapter. Corrosion laboratories becomes alkaline when it acts on iron. • 1.1 HISTORICAL T 2 Principles of Corrosion Engineering and Corrosion Control (C) Corrosion is an aspect of the decay of materials by chemical or biological agents. (D) Corrosion is an extractive metallurgy in reverse. For instance, iron is made from hematite by heating with carbon. Iron corrodes and reverts to rust, thus completing its life cycle. The hematite and rust have the same composition (Fig. 1.1). (E) Corrosion is the deterioration of materials as a result of reaction with its environment (Fontana). (F) Corrosion is the destructive attack of a metal by chemical or electrochemical reaction with the environment (Uhlig). established in M.I.T., USA and University of Cambridge, UK, contributed significantly to the growth and development of corrosion science and technology as a multi disciplinary subject. In recent years, corrosion science and engineering has become an integral part of engineering education globally. 1.2 DEFINITIONS Corrosion is a natural and costly process of destruction like earthquakes, tornados, floods and volcanic eruptions, with one major difference. Whereas we can be only a silent spectator to the above processes of destruction, corrosion can be prevented or at least controlled. Several definitions of corrosion have been given and some of them are reproduced below: Despite different definitions, it can be observed that corrosion is basically the result of interaction between materials and their environment. Up to the 1960s, the term corrosion was restricted only to metals and their alloys and it did not incorporate ceramics, polymers, composites and semiconductors in its regime. The term cor(A) Corrosion is the surface wastage that occurs rosion now encompasses all types of natural and when metals are exposed to reactive envi- man-made materials including biomaterials and nanomaterials, and it is not confined to metals ronments. (B) Corrosion is the result of interaction and alloys alone. The scope of corrosion is consisbetween a metal and environments which tent with the revolutionary changes in materials development witnessed in recent years. results in its gradual destruction. Alloying Pmamf Sted ft * C (Si, etc) Corrosion Process Pe PV*] Added (Heat) 3 ftiOs (Hematite) tan Ore <^ Released Figure 1.1 Refining-corrosion cycle Introduction to corrosion 3 i .3 CORROSIVE • ENVIRONMENT Corrosion cannot be defined without a reference to environment. All environments are corrosive to some degree. Following is the list of typical corrosive environments: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) dyes, packaged goods, etc. with dire consequences to the consumers. Nuclear hazards. The Chernobyl disaster is a continuing example of transport of radioactive corrosion products in water, fatal to human, animal and biological life. The magnitude of corrosion would depend upon the sensitivity of a particular metal or Air and humidity. alloy to a specific environment. For instance, Fresh, distilled, salt and marine water. Natural, urban, marine and industrial copper corrodes rapidly in the presence of ammonia and it is a serious problem in agriatmospheres. cultural areas. Many historical statues made Steam and gases, like chlorine. from bronze have been destroyed by ammonia Ammonia. released from fertilizers. Environmental condiHydrogen sulfide. tioning offers one method of controlling corSulfur dioxide and oxides of nitrogen. rosion, such as the use of inhibitors and oil Fuel gases. Acids. transmission pipelines. Alkalies. Soils. It may, therefore, be observed that corrosion 1 .5 C O S T OF CORROSION is a potent force which destroys economy, depletes resources and causes costly and untimely failures In a study of corrosion cost conducted jointly of plants, equipment and components. by C. C. Technologies Inc., USA [6], Federal Highway Agencies (FHWA), USA [7] and National Association of Corrosion Engineers [8], the direct corrosion cost was estimated to be 1 .4 C O N S E Q U E N C E S OF around 276 billion US dollars, approximately 3.1% of the national gross domestic product. CORROSION Based on an extensive survey conducted by Some important consequences of corrosion are Battelle Columbus Laboratories, Columbus, summarized below: Ohio, USA and National Institute of Standards and Technology (NIST), in 1975, the cost was esti• Plant shutdowns. Shutdown of nuclear mated to be 82 billion US dollars, which would plants, process plants, power plants and have exceeded 350 billion US dollars in view of refineries may cause severe problems to price inflation over the last twenty-five years. industry and consumers. Because of the long time involved in conduct• Loss of products, leaking containers, storage ing cost structure, it is not possible to update tanks, water and oil transportation lines and the information every year. However, both studfuel tanks cause significant loss of product and ies show that corrosion costs are staggering and may generate severe accidents and hazards. a figure of about 350 billion US dollars appears It is well-known that at least 25% of water is to be a reasonable estimate for another two to three years. At least 35% of the above amount lost by leakage. • Loss of efficiency. Insulation of heat could have been saved by taking appropriate corexchanger tubings and pipelines by corrosion rosion control measures. In UK, the corrosion products reduces heat transfer and piping cost is estimated to be 4-5% of the GNP [4]. In Japan, the cost of corrosion is estimated to be capacity. • Contamination. Corrosion products may 5258 trillion Yen per year. For most industrialized contaminate chemicals, pharmaceuticals, nations, the average corrosion cost is 3.5-4.5% 4 Principles of Corrosion Engineering and Corrosion Control and petrochemical industries are the largest contributor to corrosion expenditure. The highway sector in USA alone includes 4 000 000 miles of highways, 583 000 bridges, which need corrosion remediation maintenance [8]. The annual direct corrosion cost estimated to be 8.3 billion US dollars. The direct corrosion of transportation sector is estimated to be 29.7 billion US dollars. It includes the corrosion cost of aircraft, hazardous materials transport, motor vehicles, railroad car and ships [8]. In the oil sector, drilling poses severe hazards to equipment in the form of stress corrosion cracking, hydrogen induced cracking and hydrogen sulfide cracking [6]. In USA alone this sector costs more than 1.2 billion US dollars. The cost is very staggering in major oil producing countries, like Saudi Arabia, Iran, Iraq and Kuwait. The direct cost of corrosion to aircraft industry exceeds 2.2 billion US dollars [8]. Corrosion has a serious impact on defense equipment. In the Gulf War, a serious problem of rotor blade damage in helicopter was caused by the desert sand. The thickness of the blade was reduced to 2-3 mm in some instances. The desert erosion-corrosion offered a new challenge to corrosion scientists and engineers. The storage of defense equipment is a serious matter for countries with corrosive environments, such as Saudi Arabia, Malaysia and Southeast Asia. Humidity is the biggest killer of defense hardware. Storage of defense equipment demands minimum humidity, scanty rainfall, alkaline soil, no dust storms, no marine environment and minimal dust particles. From the above summary, it is observed that corrosion exists everywhere and there is no industry or house where it does not penetrate and it demands a state of readiness for engineers and scientists to combat this problem. of the GNP. Below are some startling figures of corrosion losses: • The corrosion cost of gas and liquid transmission pipelines in USA exceeds seven billion US dollars. The figure for the major oil producing countries in the Gulf region are not known, however the cost expected to be very high because of highly corrosive environment in the region [8]. The corrosion-free life of automobiles in the coastal regions of Arabian Gulf is about six months only [9]. Nearly 95% of concrete damage in the Arabian Gulf coastal region is caused by reinforcement corrosion and consequent spalling of concrete [10]. It is estimated that 10% of all aircraft maintenance in USA is spent on corrosion remediation [11]. Major annual corrosion losses to the tune of £350 million in transport, £280 million in marine, £250 million in buildings and construction and £180 million in oil and chemical industries, have been reported in UK [12]. These are uncorrected 1971 figures. About $120 billion is spent on maintenance of aging and deteriorating infrastructures in USA [13]. Automotive corrosion costs 23.4 billion US dollars annually in USA [8]. Every new born baby in the world now has an annual corrosion debt of $40. • • • • • • • 1 .6 B R E A K D O W N OF S P E N D I N G ON C O R R O S I O N The petroleum, chemical, petrochemical, construction, manufacturing, pulp and paper and transportation (railroad, automotive and 1 .7 CORROSION SCIENCE aerospace) industries are the largest contributors AND CORROSION to corrosion expenditure. The cost of corrosion differs from country to country. For instance E N G I N E E R I N G in USA, the transportation sector is the largest sector contributing to corrosion after public util- The term science covers theories, laws and ities, whereas in the oil producing countries, explanation of phenomena confirmed by intersuch as the Arabian Gulf countries, petroleum subjective observation or experiments. For Introduction to corrosion instance, the explanation of different forms of corrosion, rates of corrosion and mechanism of corrosion is provided by corrosion science. Corrosion science is a l o w i n g why' of corrosion. The term engineering, contrary to science, is directed towards an action for a particular purpose under a set of directions and rules for action and in a well-known phrase it is 'knowing how.' Corrosion engineering is the application of the principles evolved from corrosion science to minimize or prevent corrosion. Corrosion engineering involves designing of corrosion prevention schemes and implementation of specific codes and practices. Corrosion prevention measures, like cathodic protection, designing to prevent corrosion and coating of structures fall within the regime of corrosion engineering. However, corrosion science and engineering go hand-in-hand and they cannot be separated: it is a permanent marriage to produce new and better methods of protection from time to time. 5 European Federation of Corrosion, Japan Society of Corrosion Engineers and others are playing leading role in the development of corrosion engineering education. Detailed information on corrosion education, training centers, opportunities in corrosion can be found in various handbooks and websites. Some sources of information are listed in the bibliography. As a consequence of cumulative efforts of corrosion scientists and engineers, corrosion engineering has made quantum leaps and it is actively contributing to technological advancement ranging from building structures to aerospace vehicles. 1.10 FUNCTIONAL A S P E C T S OF CORROSION Corrosion may severely affect the following functions of metals, plant and equipment: (1) Impermeability: Environmental constituents must not be allowed to enter pipes, process equipment, food containers, tanks, etc. to minimize the possibility of corrosion. (2) Mechanical strength: Corrosion should not affect the capability to withstand specified loads, and its strength should not be undermined by corrosion. (3) Dimensional integrity: Maintaining dimensions is critical to engineering designs and they should not be affected by corrosion. (4) Physical properties: For efficient operation, the physical properties of plants, equipment and materials, such as thermal conductivity and electrical properties, should not be allowed to be adversely affected by corrosion. (5) Contamination: Corrosion, if allowed to build up, can contaminate processing equipment, food products, drugs and pharmaceutical products and endanger health and environmental safety. (6) Damage to equipment: Equipment adjacent to one which has suffered corrosion failure, may be damaged. Realizing that corrosion effectively blocks or impairs the functions of metals, plants 1.8 INTER-DISCIPLINARY N A T U R E OF CORROSION The subject of corrosion is inter-disciplinary and it involves all basic sciences, such as physics, chemistry, biology and all disciplines of engineering, such as civil, mechanical, electrical and metallurgical engineering. 1 .9 CORROSION EDUCATION The subject of corrosion has undergone an irreversible transformation from a state of isolated and obscurity to a recognized discipline of engineering. From the three universities in USA which offered courses in corrosion in 1946, corrosion courses are now offered by almost all major technical universities and institutions in USA, UK, Europe, Southeast Asia, Africa and Japan. Corrosion is now considered as an essential component of design. Learned societies like National Association of Corrosion Engineers, 6 Principles of Corrosion Engineering and Corrosion Control the product image which is a valuable asset to a corporation. Surface finishing processes, such as electroplating, anodizing, mechanical polishing, electro polishing, painting, coating, etching and texturing all lead to the dual purpose of enhancement of aesthetic 1 .10.1 H E A L T H , S A F E T Y , value and surface integrity of the product. ENVIRONMENTAL AND (5) Product life: Corrosion seriously shortens PRODUCT LIFE the predicted design life, a time span after which replacement is anticipated. Cars have, in general, a design life of twelve years, but These can involve the following: several brands survive much longer. A DC-3 aircraft has a design life of twenty years but (1) Safety: Sudden failure can cause exploafter sixty years they are still flying. The sions and fire, release of toxic products and Eiffel Tower had a design life of two years collapse of structures. Several incidents of fire only, and even after one hundred years it is have been reported due to corrosion causing still a grand symbol of Paris. The reason for leakage of gas and oil pipelines. Corrosion their survival is that the engineers made use adversely affects the structural integrity of of imaginative designs, environmental resiscomponents and makes them susceptible tant materials and induction of corrosionto failure and accident. More deaths are free maintenance measures. Distinguished caused by accidents in old cars because of and evocative designs always survive whereas the weakening of components by corrosion designs of a transitory nature deteriorate damage. Corrosion has also been a significant to extinction with time. Design life is a factor in several accidents involving civil and process of imagination, material selection military aircraft and other transportation and corrosion-free maintenance. vehicles. Corrosion failure involving bridges, ships, airports, stadiums are too numerous (6) Restoration of corroded objects: Objects of to be mentioned in detail in this chapter outstanding significance to natural history and recorded in the catalog of engineering need to be preserved. Many historical strucdisasters [11]. tures have been lost through the ravages of (2) Health: Adverse effects on health may be corrosion. One recent example is the call for caused by corroding structures, such as a help to restore the revolutionary iron-hulled plumbing system affecting the quality of steamships Great Britain built in 1843. It has water and escaping of products into the been described as mother of all modern ships, environment from the corroded structures. measuring 3000 feet in length and weighing (3) Depletion of resources: Corrosion puts 1930 tons. A plea for £100 000 has been made a heavy constraint on natural resources for its restoration. of a country because of their wastage by corrosion. The process of depletion outweighs the discovery of new resources which may lead to a future metal crisis similar to the l.ll FIVE GOOD R E A S O N S past oil shortage. (4) Appearance and cleanliness: Whereas anes- T O S T U D Y C O R R O S I O N thetics numb the senses, aesthetics arouse interest, stimulate and appeal to the senses, (1) Materials are precious resources of a country. particularly the sense of beauty. A product Our material resources of iron, aluminum, designed to function properly must have copper, chromium, manganese, titanium, an aesthetic appeal. Corrosion behaves like etc. are dwindling fast. Some day there a beast to a beauty. It destroys the aeswill be an acute shortage of these materials. thetic appeal of the product and damages An impending metal crisis does not seem and equipment, appropriate measures must be adopted to minimize loss or efficiency of function. Introduction to corrosion 7 anywhere to be a remote possibility but a 3. Which is the most common cause of corrosion damage, corrosion fatigue, stress reality. There is bound to be a metal crisis corrosion cracking or pitting corrosion? and we are getting the signals. To preserve these valuable resources, we need to under- 4. Describe with an example how corroded structures can lead to environmenstand how these resources are destroyed by tal pollution. corrosion and how they must be preserved by 5. Does corrosion affect humans? If so, applying corrosion protection technology. explain how. (2) Engineering knowledge is incomplete without an understanding of corrosion. 6. Describe two engineering disasters in which corrosion played a leading role. Aeroplanes, ships, automobiles and other transport carriers cannot be designed with- 7. State two important corrosion websites. out any recourse to the corrosion behavior of 8. How can corroded structures be injurious to materials used in these structures. human health? (3) Several engineering disasters, such as crash- 9. Name three cities in Southeast Asia and the Middle East which have the most corrosive ing of civil and military aircraft, naval and environment. passenger ships, explosion of oil pipelines and oil storage tanks, collapse of bridges and 10. What is the best way to minimize the corrodecks and failure of drilling platforms and sion of defense equipment during storage? tanker trucks have been witnessed in recent 11. What is the relationship between depletion of years. Corrosion has been a very important natural resources and corrosion? factor in these disasters. Applying the knowledge of corrosion protection can minimize such disasters. In USA, two million miles of pipe need to be corrosion-protected for R EFERENCES safety. (4) The designing of artificial implants for the [1] Walsh, F. (1991). Faraday and his laws of electrolysis. Bulletin of Electrochem, 7, 11,481-489. human body requires a complete under[2] Schonbein, C. (1936). Pogg. Ann., 37, 390. standing of the corrosion science and [3] Evans, U.R. (1972). An Introduction to Metallic engineering. Surgical implants must be Corrosion, 2nd ed. London: Arnold. very corrosion-resistant because of corrosive [4] Uhlig, H.H. (1985). Corrosion and Corrosion Control, 3rd ed. New York: John Wiley and nature of human blood. Sons. (5) Corrosion is a threat to the environment. For [5] Fontana, M.G. (1986). Corrosion Engineerinstance, water can become contaminated ing, 3rd ed. New York: McGraw-Hill Book by corrosion products and unsuitable for Company. consumption. Corrosion prevention is inte[6] C. C. Technologies Laboratories, Inc. (2001). Cost of corrosion and prevention strategies in gral to stop contamination of air, water and the United States, Ohio: Dublin, USA. soil. The American Water Works Association [7] Federal Highway Administration (FHWA), needs US$ 325 billion in the next twenty years Office of the Infrastructure and Development to upgrade the water distribution system. (2001). Report FHWA-RD-01-156. [8] National Association of Corrosion Engineers (NACE) (2002). Materials Performance, Special Issue, Houston, Texas, USA, July. Jointly with C. C. Technologies and FHWA. [9] Ahmad, Z. (1996). Corrosion phenomena in QUESTIONS coastal area of Arabian Gulf. British ^Corrosion Journal 31, (2), 191-197. [10] Rashid-uz-Zafar, S., Al-Sulaiman, G.J. and C ONCEPTUAL Q U E S T I O N S Al-Gahtani, A.S. (1992). Symp. Corrosion and the Control, Riyadh, Saudi Arabia, May, 110. 1. Explain how corrosion can be considered as [11] Tullmin, M.A.A., Roberge, P.R., Grenier, L. and extractive metallurgy in reverse. Little, M.A. (1990). Canadian Aeronautics and Space Journal, 42, (2), 272-275. 2. Listfiveimportant consequences of corrosion. 8 Principles of Corrosion Engineering and Corrosion Control WEBSITES [20] [21] [22] [23] [12] Hoar,T.P. (1971). Report of the Committee on Corrosion and Protection, London: HMSO. [13] Latanisian, R.M., Leslie, G.G., McBrine, N.J., Eselman, T. etal. (1999). Application of practical ageing management concepts to corrosion engineering, Keynote Address. 14th ICC, Capetown, South Africa, 26 Sep-10 Oct. GENERAL REFERENCES [14] Hackerman,N. (1993). A view of the history of corrosion and its control. In: Gundry, R.D. ed. Corrosion 93 Planery and Keynote Lectures, Texas: NACE, Houston, 1-5. [15] Pliny (1938). Natural History of the World. London: Heinemann. [16] Hoare, T.P. (1971). Report of the Committee on Corrosion and Protection. London: HMSO. [17] Uhlig, H.H. (1949). Chemical and Engineering News, 97,27'64. [18] N. B. S. (1978). Corrosion in United States, Standard Special Publication. [19] Bennett, L.H. (1978). Economic Effects of Metallic Corrosion in USA, Special Publication 511-1, Washington, DC. SOFTWARE [24] NACE: Basic Corrosion (National Association of Corrosion Engineers, Houston, Texas), Course on CD-Rom, 2002. [25] Corrosion Survey Data Base (COR.SUR), access via NACE website, NACE, Houston, Texas, 2003. [26] Rover Electronic Data Books™, William Andrew, Inc., New York, USA, 2002. [27] Peabody, A.W. Control of Pipeline Corrosion, 2nd Edition, Ed. R. Bianchette, NACE, Houston, Texas, 2003. [28] Corrosion Damage: A Practical Approach, NACE, Houston, Texas, 2003. BASIC CONCEPTS IN CORROSION or corrosion to take place, the formation of a corrosion cell is essential. A corrosion cell is essentially comprised of the following four components (Fig. 2.1). • • • • Anode Cathode Electrolyte Metallic path. F Anode: One of the two dissimilar metal electrodes in an electrolytic cell, represented as the negative terminal of the cell. Electrons are released at the anode, which is the more reactive metal. Electrons are insoluble in aqueous solutions and they only move, through the wire connection into the cathode. For example, in a battery, zinc casing acts as the anode. Also in a Daniel cell, zinc is the anode as oxidation occurs on it and electrons are released (Fig. 2.2). Corrosion nomenclature is the opposite of electroplating nomenclature, where an anode is positive and the cathode is negative. Cathode: One of the two electrodes in an electrolytic cell represented as a positive terminal of a cell. Reduction takes place at the cathode and electrons are consumed. Example, carbon electrode in a battery, copper electrode in a Daniel cell. Figure 2.3 shows the reduction of hydrogen ion. The electron is always a reducing agent. Electrolyte: It is the electrically conductive solution (e.g. salt solution) that must be present for corrosion to occur. Note that pure water is a bad conductor of electricity. Positive electricity passes from anode to cathode through the electrolyte as cations, e.g. Z n + + ions dissolve from a zinc anode and thus carry positive current away from it, through the aqueous electrolyte. Metallic Path: The two electrodes are connected externally by a metallic conductor. In the metallic conductor, 'conventional' current flows from (+) to (—) which is really electronsflowingfrom (—) to (+). Metals provide a path for the flow of conventional current which is actually passage of electrons in the opposite direction. Current Flow: Conventional current flows from anode (—) to cathode (+) as Z n + + ions through the solution. The current is carried by these positive charged ions. The circuit is completed by passage of electrons from the anode (—) to the cathode (+) through the external metallic wire circuit (outer current). Electron Flow: H + + e -+ H, 2H -* H 2 f Although the anode (e.g. Fe or Zn) is the most negative of the two metals in the cell, this reaction does not occur there because its surface is emanating Fe + + ions which repel H + ions from discharging there. The circuit is completed by negative ions (—) which migrate from the cathode (+), through the electrolyte, towards the anode (—). They form 10 Principles of Corrosion Engineering and Corrosion Control Current flow in an electrochemical cell is shown in Fig. 2.4. Current carrying electrolyte -A-v 2 .1 Non-corroding area Corroding area ANODIC AND CATHODIC REACTIONS The anode is the area where metal is lost. At the anode, the reactions which take place are oxidation reactions. It represents the entry of metal ion into the solution, by dissolution, hydration or by complex formation. It also includes precipitation of metal ions at the metal surface. For example F e 2+ + 2 0H~ -» Fe(OH)2. Ferrous hydroxide or rust formation on steel surface is a common example. Some more examples are: Figure 2.1 Corrosion cell in action Fe(OH)2 when they enter the cloud of Fe + + ions coming from the anode. Anions'. Migrate towards the anode (OH~) but precipitate as Fe(OH)2 before reaching it. (a) 2A1 + 6HC1 » 2A1C13 + 3H2 t 2+ (b) Fe + 2HC1- FeCl2 + H 2 t Cations'. Migrate towards the cathode (Fe ). Anode T h ~ (Znrod) I U -ju+ | + ^ Cathode (Cured) In I Z n 2 * S04*tact> II Cu2* S & t a q ) ! Cu Chemical reaction — * > Figure 2.2 A galvanic cell (Daniel cell) Basic concepts i n c orrosion 11 involve oxidation to a higher valence state. Reactions (a-d) can be written in terms of electron transfer as below: Anode — r 4 " Cathode (a) (b) (c) (d) Al-> Al 3+ + 3e Fe-+ F e 2+ + 2e Z n - y Z n 2 + + 2e Z n - > Zn2+ + 2e Anodic reaction in terms of electron transfer is written as M -> M n + + ne 2.2 Figure 2.3 Figure showing the reduction of hydrogen in an acid electrolyte ANODIC REACTIONS CHARACTERISTICS (1) Oxidation of m etal t o an ion w ith a c harge. (2) Release of e lectrons. (3) Shift t o a h igher valence state. The process of o xidation i n m ost metals a nd alloys represents corrosion. Hence, if o xidation is stopped, corrosion is s topped. (c) Z n + H 2 S 0 4 - + Z n S 0 4 + H 2 t (d) Zn + HC1 -> ZnCl2 + H 2 t T he reactions ( a-d) i nvolve t he release of h ydrogen gas. All the r eactions shown above Cathode ||2e* Cu 2e art" Fe Anode Hydrogen picks up an v Q electron and tabbies V ^ off at cathode m !2e1 . Current (Fe** tons) leaves surface of iron causing corrosion Negative current enters surface of copper and protects copper m €) © % tI © I t!f© conventional current flow ——-# electron flow Fe(OH)a Iron atom loses two electrons and combines with hydroxyl ions to " form ferrous hydroxide precipitate Figure 2.4 Current flow in an electrochemical cell 12 Principles of Corrosion Engineering and Corrosion Control 2.3 CATHODIC REACTIONS CHARACTERISTICS Cathodic reactions are reduction reactions which occur at the cathode. Electrons released by the anodic reactions are consumed at the cathode surface. Unlike an anodic reaction, there is a decrease in the valence state. The most common cathodic reactions in terms of electrons transfer are given below: > (a) 2H + + 2e —• H2 f ( m ac id solution) (b) 0 2 + 4H+4e -> 2H 2 0 (in acid solution) (c) 2H 2 0 + 0 2 + 4e -+ 4 0 H " (in neutral and alkaline solutions) (d) Fe 3+ + e - » Fe +2 (metal ion reduction in ferric salt solutions) (e) Metal deposition: M 2 + + 2e -> M N i + + + 2e -> Ni Cu2+ + 2e -> Cu (f) Bacterial reduction of sulfate: SO^ - + 8H+ + 8e -> S" + 4H 2 0 2.4 T Y P E S OF CORROSION CELLS There are several types of corrosion cells: (1) (2) (3) (4) Galvanic cells Concentration cells Electrolytic cell Differential temperature cells. 2 .4.1 GALVANIC C E L L S The galvanic cell may have an anode or cathode of dissimilar metals in an electrolyte or the same metal in dissimilar conditions in a common electrolyte. For example, steel and copper electrodes immersed in an electrolyte (Fig. 2.5), represents a galvanic cell. The more noble metal copper acts as the cathode and the more active iron acts as an anode. Current flows from iron anode to copper cathode in the electrolyte. Electricity Hows from cathode to the anode through the metallic conductor. Current The metal having Ilia lower energy level (in this case Cu), no corrosion occurs at the cathode. »Cathode Anode*— Cu The metal having the higher level of energy (In this case Fe), corrosion occurs at the anode, 41 I I Electrolyte - * conventional current flow ~* electron flow Water and dissolved salts conduct the flw of electricity from steel anode to the copper cathode. Figure 2.5 Typical galvanic cell Basic concepts in corrosion 13 2 .4.2 CONCENTRATION CELLS This is similar to galvanic cells except with an anode and cathode of the same metal in a heterogeneous electrolyte. Consider the corrosion of a pipe in the soil. Concentration cells may be set up by: (a) Variation in the amount of oxygen in soils. (b) Differences in moisture content of soils. (c) Differences in compositions of the soil. Concentration cells are commonly observed in underground corroding structures, such as buried pipes or tanks (Fig. 2.6). The inequality of dissolved chemicals causes a potential difference which establishes anode in the more concentrated region and cathode in the less concentrated region. 2 .4.3 ELECTROLYTIC C E L L External power source ^JEfectolyle - * conventional current flow ~* electron flow Figure 2.7 Electrolytic cell. The cathode and anode This type of cell is formed when an external cur- can be any metal rent is introduced into the system. It may consist of all the basic components of galvanic cells and concentration cells plus an external source of electrical energy. 2 . 5 MECHANISM OF Notice that anode has a (+) polarity and cathode has (—) polarity in an electrolytic cell, C O R R O S I O N O F IRON where external current is applied. This is the type of cell set up for electrically protecting the struc- Consider a piece of iron exposed to humid air tures by cathodic protection. The polarity of an which acts as an electrolyte. Fe 2+ ions are released electrolytic cell is opposite to that in a galvanic from the anode by oxidation and OH~ ions from (corrosion) cell (Fig. 2.7). the cathode by reduction on the metal surface. Air day (low oxygen) :^yy^yM>:^^y. V^4L Loam (High oxygen) Yf///^ Anodic area ^ ^ ^ C a t h o d l c a o ^ W Figure 2.6 Concentration cell formation in an underground pipeline 14 Principles of Corrosion Engineering and Corrosion Control where aA = activity of the substance, T = absolute temperature, R = gas constant. From the Gibbs-Helmholtz equation: The negative and positive ions combine Fe + + + 2 0 H " ->Fe(OH) 2 (white green precipitate) Fe(OH)2 is insoluble in water and separates from the electrolyte. A more familiar name of Fe(OH)2 is rust. Details of reactions involved in the corrosion of iron-based materials is given below: (1) Fe + H 2 0 U FeO + 2H+ + 2e~ Forms monolayer of FeO islands (2) Fe + 2H 2 0 - • Fe(OH)2 + 2H+ + 2e (3) 3FeO + H 2 0 -> Fe 3 0 4 + 2H+ + 2e (Black) (Magnetite) (Brown) and G= H-TS (2.2) Substituting H = U + PV from the First Law of Thermodynamics F = PV - TS Differentiating: dG = dU + PdV + VdP - TdS - SdT But for a reversible process, dU = qTeY — PdV at constant pressure and grev = TdS at constant temperature So, dG = TdS - PdV + PdV + VdP (2.5) (2.4) (2.3) (4) 2 F e 3 0 4 + H 2 0 - > 3 ( y - F e 2 0 3 ) + 2 H + + 2 e (5) 2[y - F e 2 0 3 ] + 3HzO -* 6(y - FeOOH) (Brown) (Yellow hydrated oxide) The formation of rust, Fe(OH)2, is shown in Fig. 2.8. 2 . 6 C O N C E P T OF F R E E ENERGY - TdS - SdT i.e. dG = VdP — SdT, for a reversible process at constant pressure and temperature. (2.6) In this section the relationship between free energy and equilibrium constant is shown. The contribution made by one mole of any conAt constant pressure, dP vanishes and this stituent, A, to the total free energy, G, of a mixture becomes: is GA> which may be represented by dG GA = Gl + RT In aA (2.1) Oj+2H20+4e" Figure 2.8 Formation of rust in seawater Basic concepts in corrosion and at constant temperature dT vanishes and: dG=VdP 15 So, for 1 mole of forward reaction A + RTln { = C AG = AG° ^-Ar20^T>T^( £=% a a However, the ideal gas equation for 1 mole is PV = RT V= RT P , /RT\ , >dG=\— \dP, since dG=VdP \P ) (2.7) (Van't Hoff reaction isotherm) (2.11) At equilibrium AG=0 and I acrai\ K where Kis the e uilibrium Integration between the limits G'4, GA and p, A p ives: ° \vjr ' ^ / -/(f)* G'A P'A GA PA , x \aAaB/ constant. 0 = AG° + RT\nK AG° = -RTlnK (2.8) (v AG° is always = -RT In K) (2.12) GA-G'=RT -m = RTIn PA 2.6.1 FREE ENERGY (THE If GA is taken to refer for standard conditions, GA-G'A D RIVING F O R C E OF A CHEMICAL REACTION) ^ chemical reaction at constant temperature and pressure will only occur if there is an overall . ,_ . decrease in the free energy of the system during If activities are used instead of pressures, then as t h e r e a c t i o n > Consider the following two reactions in equation (2.1) for example: or GA = G°A+RT\naA Applying AG for the reaction aA+bB**cC+dD gives In example (b), both AH° and AS° work in AG = (cGc + dGd)- (aGA + bGB) (2.10) °PP o s i t e s i 8 n s t 0 e a c h o t h e r i n t e r m s of direction of energy change, however, both of them proceed spontaneously as indicated. It is important tQ k n o w w h k h o n e w o u l d d e d d e t h e d i r e c t i o n G = G* = # T In P (2.9) (a) ZnO(s) + C(s)^Zn(s)+CO(g) at 1373K A5° = +285JK- 1 mor 1 AH° = +349.9kJ• mol - 1 (b) Fe(s) + 2± 0 2 ( s ) ^ F e O ( s ) y AS° = - 7 1 J K ~ 1 m o r 1 FeO AH 0 = +265.5 kj.mol" 1 FeO Using expression (2.1) for GA, GB) Gc and GD: AG = c[G£+jRTln ac\ + d[Gp + RT In ap] -a[G°A+RTIn aA]-b[G0B+RTIn aB] of the reaction (AH° or A 5°). We must, therefore, introduce a single function expressing the combined effect of change of both AH and A5. o f the total enthalpy of the system only a part 16 Principles of Corrosion Engineering and Corrosion Control is converted to useful work, which is called free Illustrative Example 2.1 energy (AG). It can be defined in terms of AH, A 5 and T. The relationship between the three is Calculate A G° for the following reaction at 500 K. deduced as CuO(s)+H 2 (g)->Cu(s)+H 2 0 AH°-AG° = TAS° (2.13) which is a statement of the Second Law of Thermodynamics and may be rearranged to: AG = AH-TAS (2.14) Solution: AH5°00 = - 8 7 k J - m o r 1 AS£00 = + 47JK- 1 mor 1 AG = A f f ° - r A 5 ° = -87000-(500x47) J-mol" 1 AG is thus a driving force for a reaction to occur. Reactions at constant temperature and = 1 10.5kJ-mor 1 pressure proceed in a direction which tends to cause a decrease in free energy. AG can be used to predict the feasibility of a reaction as shown AG° is negative (—ve), hence the reaction is feasible. below. Ever if AG° were positive, a reaction producFor a chemical reaction where the reactants ing gas pressures of products of less than unit (A+B) react to yield the product (C+D) activity (=lbar, for gases) is possible. The sign according to of AG° only predicts spontaneity if all activities A+B-^C+D (2.15) are unity. the free energy change of the reaction can be expressed as AG=GP-GR AG = (Hp-HR)-T(SP-SR) AG = A H - T A S where P = products R = reactants. (a) If the value of free energy change is negative, the reactions will be spontaneous and take place from left to right (A+B^C+D). (b) If the value of the free energy change is zero, the reaction is at equilibrium (A+B<=± C+D). (c) If the value of the free energy change is positive, the reaction will proceed in the reverse direction (A+B<^C+D). It is not possible to obtain the absolute value of free energy for a reaction. However, the change in free energy can be measured. (2.16) (2.17) (2.18) Electrochemical cells generate electrical energy due to electrochemical reactions. The electrical energy available is Electrical energy=volts x current x time = volts x coulombs = EQ where Q=nF n is the number of electrons involved in the chemical reaction Fis Faraday's constant = 96 500 C (g e quiv.) -1 The gram equivalent is the number of moles divided by the number of electrons involved in the reaction £is electromotive force (emf) of the cell (volts). Any work performed can only be done through a decrease in free energy of the cell 2 .6.2 CELL POTENTIALS A ND E M F Basic concepts in corrosion reaction, hence, AG = -nFE where AG = free energy change (joules) E = standard potential of the reaction (volts). Illustrative Example 2.2 (1) Calculate the standard enthalpy change for the following reaction at 298 K: 2PbS(s)+30 2 (g)-+2PbO(s)+2S0 2 (g) Solution: Given that the standard enthalpies of formation are: AH298PbS = -94.5kJ-mor 1 AH298PbO(s) = -220.5 k j-mol - 1 A H^ 98 S0 2 (s) = -298kJ-moi _ 1 Rewrite the equation: 2PbS(s)+30 2 (g)^2PbO(s)+2S0 2 (£) 4 4 Dr Hr 17 (b) Deduct Kp for the reaction using the relationship AG° = A H ° - T A 5 ° for the reaction. Given enthalpies and entropies of reaction: AH30o = +65kJmor 1 A53oo = +13.7JK" 1 mor 1 A Hi 200 = + 1 80.9kJ-mor 1 ASi20o = +288.6 J K ^ m o r 1 Solution: (a) For reaction at 300 K: (1) AG° = 65000-(300xl3.7) = 60890J-mor 1 For reaction at 1200 K: (2) AG° = 180900-(1200x288.6) = -165420J-mor 1 As A G° is negative for the reaction at 1200 K, the reaction proceeds from left to right in the forward direction at 1200 K. The forward reaction is not feasible at 300 K to produce lbar of C 0 2 from 1 bar of CO: but it would form much lower C0 2 pressures spontaneously. Such low C0 2 pressures would not be sufficient to 'push back the atmosphere' and so effectively no Zn would be produced. (b) Calculation of Kp. The equilibrium constant (K) is given by 2 (-94.5) 3(0) 2(-220.5) 2(-298.0) The standard enthalpy of formation of an element is taken as zero at 298 K. Using equation: AH = H2 — H\, where H2 = total enthalpy of formation of products and Hi is for the reactants AH = [2(-220.5)+2(-298)] - [2(-94.5)+0] = -848.0 kj (2) For the equilibrium Zn(s)+CO(g) ZnO(s)+C(s)<£» ( a%4) a c a d\ -^-r (2.19) (a) Deduce the direction in which the reaction would be feasible at 300 and 1200 K, when all reactants and products are in their standard states (i.e. unit activities). but the term in parenthesis is not K, but is any arbitrary non-equilibrium activity for which A G is the free energy change if such a reaction was carried out. 18 Principles of Corrosion Engineering and Corrosion Control At equilibrium AG = 0, and then interface with a potential difference between the a^a^/a^ag — K equilibrium constant = metal and electrolyte. The atoms of the metal, M, ionize producing aquo-ions, M n+ (ag), and AG° +RT In K. electrons, ne, according to So,AG° = -RT\nK (2.20) Equation (2.19) can be rewritten as: M = Mn+(aq) + ne (2.21) where M represents metal atoms. The metal is left with a negative charge and its positively \PAPBJ charged metal ions, Mn+(aq), in the electrolyte where P is partial pressure (actual) are attracted back towards the metal surface. assuming A, B, C and D are gases. The Thus, a potential difference and dynamic equiabove equation connecting AG and Kis librium between the metal and the solution a form of van't Hoff isotherm. is established. The atoms of the metal conNow calculate Kp tinue to ionize until the displacement of electrical charges produced balances the tendency AG° of the metallic atoms to ionize into the elecp RT trolyte. Consider, for example, a piece of zinc metal in water. Zinc dissolves producing positively A U200K, charged zinc ions (cations): -165420 610 p 2.303x8.314x1200 (2.22) Zn-*Zn2++2e = 7.197 The zinc ions in the solution remain very close to the metal surface. The zinc metal becomes negatively charged as the positive ions leave its surface. The excess electrons on the zinc surface orient themselves opposite to a layer of zinc ions of equal and opposite charge on the water side of the zinc/water interface. Such a process leads to the formation of an electrical double layer of about 1 nm (10~7 cm) thickness along the metal/solution interface (Fig. 2.9). Figure 2.9A shows the metal/solution interface at the moment of immersion and the formation of double layer is shown in Figs 2.9B and C. The double layer shown in Figs 2.9B and C is formed as a result of attraction between the negative ions (anions) and positive ions (cations) on one hand, and repulsion between similarly charged ion, anions or cations on the other hand. As a result of the above interactions, cations diffuse amongst the anions until an equi2 .7 R E V E R S I B L E librium is established between the metal and solution, and between the bulk of the soluE LECTRODE P OTENTIAL tion and the layer adjacent to the metal surface. When a metal is immersed in an electrolyte, The plane passing through the ions absorbed on a dynamic equilibrium is established across the the metal surface is called the Inner Helmholtz So, Kp = 1.574 xlO 7 . Similarly at 300 K, Kp = 2.512 x 1 0" 11 . The value of Kp at 1200 shows that the reaction proceeds fast (left to right) compared to the reaction at 300 K, which proceeds at a negligible rate. At 1200 K, bright red hot, reaction kinetics are fast and it is safe to interpret a large K value as a meaning that reaction is fast. But the reaction 2 H 2 + 0 2 = 2 H 2 0 has a very large K value at room temperature, yet is negligibly slow (unless sparked), due to the very slow kinetics at room temperature. K is a thermodynamic factor, but kinetics may allow only a negligible rate. Basic concepts in corrosion 19 Hetal Electrons on metal surface Netal ions adjacent to surface ^•A*****^-*^-!-* «.'*» *»Mfr Layer of positive ions immediately adjacent to surface Diffused layer of ions Solution (A) At the instant of immersion. (B) First stage of formation of electrical double layer (C) Complete double layer Figure 2.9 Representation of electrical double layer at a metal/solution interface 1.0 mol of zinc salt, as for example ZnS04 per liter M^M + ne ( 2.23) MCu2+ = 1.0 mol of CUSO4 per liter. r When writing down a cell, it is common pracbecomes equal to the rate of discharge across the tice to put the anode on the left-hand side and the double layer: cathode on the right-hand side. The anode is the electrode where oxidation reaction takes place, ( 2.24) Mn^ + ne-^M and the cathode is the electrode where reduction reaction takes place. Accordingly, the reactions in In other words, the rate of forward reac- a Daniel cell can be written as: tion (equation (2.23)) becomes equal to the rate of backward reaction (equation (2.24)) on the Zn -> Zn 2 + +2e (Oxidation at anode) (2.26) metal electrode. This is described as an exchange current. Cu 2 + +2e -> Cu (Reduction at cathode) (2.27) The double layer described above is responsible for the establishment of potential differ- An emf of 1.10 V is produced in a Daniel cell. This ence between the metal and solution, and this potential is equal to the difference in potentials potential difference is referred to as 'Absolute between the anode and the cathode. It is called the Electrode Potential' More details will be given in 'Reversible Cell Potential* (£rev) or 'Equilibrium Potential' (Eeq). Chapter 3. n+ plane, whereas the plane passing through the center of solvated cation is called the Outer Helmholtz plane, which also marks the beginning of a diffuse layer when an excess of charges is neutralized (Fig. 2.9C). At this stage, it is sufficient to understand the meaning of double layer to know the electrode potential. Returning to equation (2.21), it should be noted that under the equilibrium conditions in the system, the rate of ionization on the metal surface: 2 .7.1 EXAMPLES OF REVERSIBLE CELLS Figure 2.2 shows a Daniel cell which is a good example of a reversible cell. A Daniel cell can be presented as follows: Znmetal/MZn2+//MCu2+/Cumetai where: MZn2 + : (2.25) 20 2 .7.2 Principles of Corrosion Engineering and Corrosion Control DIFFERENCE BETWEEN AND REVERSIBLE STANDARD POTENTIAL Electrode reaction Zr + 4 +4e = Zr T i + + + 2 e = Ti A l +3 + 3e = Al Hf + 4 +4e = Hf U+3+3e = U Be + + +2e = Be M g + + + 2e = Mg Na + + e = Na Ca+++2e = Ca K+ + e = K Li+ + e = Li FeO~~ + 8 H + +3e = Fe+ 3 +4H 2 0 . . . Co+3 + e = C o + + P b0 2 + S O ^ " + 4 H + + 2 e = PbS0 4 +2H 2 0 Ni0 2 +4H++2e = N i + + + 2 H 2 0 . . . . Mn + 3 + e = M n + + Pb02+4H++2e = P b + + + 2 H 2 0 . . . . Cl2 + 2e = 2C1" C r 2 07~ + 14H+ +6e = 2Cr +3 + 7 H 2 0 0 2 + 4 H + + 4 e = 2H 2 0 Br 2 (I)+2e = 2Br~ Fe+3 + e = F e ++ 02+2H++2e = H202 I 2 +2e = 2 r 0 2 + 2 H 2 0+4e = 4 0 H " Hg2Cl2 + 2e = 2 H g + 2 C r AgCl+e = Ag + C r S O~~+4H + + 2 e==H 2 S0 3 +H 2 0.... C u + + + e ^ Cu+ Sn+ 4 +2e = Sn + + AgBr+e = Ag+Br~ Cu(NH3)+ + e = C u+2NH 3 Ag(CN)" + e = A g+2CN PbS0 4 +2e = Pb+S0 4 HPbO~+H20+e = Pb+30H2H 2 0+2e = H 2 + 2 0 H Zn(NH 3 )+ + +2e = Zn+4NH 3 E° (V) -1.53 - 1.63 - 1.66 -1.70 - 1.80 - 1.85 - 2.37 -2.71 -2.87 -2.93 -3.05 1.9 1.82 1.685 1.68 1.51 1.455 1.3595 1.33 1.229 1.0652 o.771 0.682 0.5355 0.401 0.2676 0.222 0.17 0.153 0.15 0.095 - 0.12 -0.31 - 0.356 -0.54 -0.828 - 1.03 POTENTIAL If a metal is immersed in a solution of its own ions, such as zinc in ZnS04 solution, or copper in CUSO4 solution, the potential obtained is called the reversible potential (£rev)« If, on the other hand, the substances taking part in the process are in their standard states, such that the activities of the metallic ions are equal to unity or gases are at 1 bar pressure, the potentials obtained are called 'Standard Electrode Potentials (Table 2.1). A standard potential refers to the potential of pure metal measured with reference to a hydrogen reference electrode. The details of reference electrodes are provided later in this chapter. Table 2.1 Standard reduction potentials, 25°C (Modified from Uhlig) Electrode reaction Au +3e = Au P t + + + 2 e = Pt Hg++ + 2e = Hg P d + + + 2e = Pd A g + + e = Ag H g+ + +2e = 2Hg Cu+ + e = Cu C u + + + 2e = Cu 2 H++2e = H 2 P b + + + 2 e = Pb S n + + + 2 e = Sn M o + 3 + 3e = Mo N i + + + 2 e = Ni Co+++2e = Co T l + + e = Tl In+ 3 + 3e = In C d + + + 2 e = Cd F e + + + 2 e = Fe Ga + + + 3e = Ga Cr + 3 + 3e = Cr Z n + + + 2 e = Zn Nb +3e = Nb Mn+++2e = Mn +3 +3 E° (V) 1.50 ca 1.2 0.854 0.987 0.800 0.789 0.521 0.337 0.000 -0.126 -0.136 ca -0.2 -0.250 -0.277 -0.336 -0.342 -0.403 - 0.440 -0.53 -0.74 -0.763 2 .7.3 HALF CELLS ca -1.1 -1.18 The term half cell is frequently used in electrochemistry. To illustrate the formation of Basic concepts in corrosion half cells, consider a Daniel cell, shown in Fig. 2.2. In the cell shown, zinc rod is immersed in one molar solution of zinc sulfate, and copper rod in one molar solution of copper sulfate. The two solutions, zinc sulfate and copper sulfate, are separated by a porous pot, which prevents them from mixing but allows electrical contact. As shown in the figure, Zn in ZnSC^ (1.0M) forms one half cell and Cu in Q1SO4 (1.0M) forms another half cell. Each half cell has its own potential. 21 The above cell (Daniel cell) can also be represented as: Zn(s)/ZnSO4(1.0M)//CuSO4(1.0M)/Cu(s) (2.32) 2.7.5 S T A N D A R D H Y D R O G E N POTENTIAL ( S H E ) A N D M E A S U R E M E N T OF SINGLE E L E C T R O D E POTENTIALS Absolute single electrode potential is a characteristic property of a metal. All metals have characteristic electrode potentials. Absolute single electrode potentials cannot be measured directly. They can be measured with respect to a standard electrode such as standard hydrogen electrode (SHE). The standard hydrogen electrode is a widely used standard electrode. It is arbitrarily assumed to have zero potential at all temperatures by definition. Thus, it provides a zero reference point for electrode potentials. The construction of the standard hydrogen electrode is shown in Fig. 2.10. As shown in the figure, 2 .7.4 REACTIONS IN THE CELL In Daniel cell, zinc dissolves and a potential is set up between the zinc ions and zinc metal in the zinc half cell. In the copper half cell, copper ions are deposited on the copper metal and a potential is set up. Once the half cells are in equilibrium, no further deposition or dissolution would occur. If, now, the two half cells are joined by a conductor, as shown in Fig. 2.2, electrons would flow from zinc (anode) to copper (cathode). As a result of electronflow,the equilibrium in the half cells is distributed, and, therefore zinc will dissolve further according to the anodic reaction (oxidation), Zn^Zn2++2e (2.28) and copper ions will discharge from copper sulfate solution in order to remove the excess electrons according to the cathodic reaction (reduction of C u + + ions to Cu metal): Cu2++2e^Cu (2.29) The overall reaction of the cell is the sum of the above two half cell reactions: Zn+Cu 2+ ^Zn 2+ Piatinizad Platinum +Cu (2.30) Electrolyte Electrolyte Bridge The emf of the cell is the algebraic sum of the potentials of the half cells or 'Standard Electrode Potential:' £ceu - £Zn/Zn2+ + £Cu2+/Cu (2.31) F i g u r e 2l0 Hydrogen electrode 22 Principles of Corrosion Engineering and Corrosion Control one of the electrodes is SHE and the other is the metal electrode of which we want to measure the single electrode potential. The emf of the cell expressed in equation (2.34) is given by: Ecell = EMn+/M — £^+/H2 (2.38) a platinum electrode is dipped in a solution containing hydrogen ions at unit activity. To obtain hydrogen ions at unit activity, a solution of one mol per liter HC1 can be used. To obtain unit activity of hydrogen gas, hydrogen is passed at one atmospheric pressure over platinum plate in the solution. The SHE is expressed as H2-*2H++2e (2.33) The most acceptable method of obtaining standard electrode potentials is by comparing the electrode potential of metals with the standard £cell = £M«+/M (2.39) hydrogen electrode. Since the SHE has zero electrode potential at all temperatures by definition, the electrode potential of a metal is numerically 2 .7.6 THE POTENTIAL (EMF) equal to the emf of the cell formed by SHE and the metal electrode. In other words, the emf of the cell O F A C E L L represents the electrode potential of the half cell formed by the metal with respect to the standard The emf of any cell is, therefore given by hydrogen electrode. In such a cell, reaction on the hydrogen electrode is oxidation and reaction on £cell = [% s ht-£°(SHE)] R i g h t the other electrode is reduction. Such a cell can be - [£ L eft-£°(SHE)] Left (2.40) expressed as: Pt(s)/H 2 (latm)//H + , (aH+ = 1)//M/M"+ (2.34) The activity or concentration of the M n + ion solution needs to be specified, as cell potential will vary with it. The electrode reaction for such a cell can be written as (considering reduction reactions as convention): M n + +rce= M for metal electrodes, and nH+ + ne=-H2 (2.36) (2.35) and £cell = ^Right ~ ^Left (2.41) where £ ^+/H (SHE), the electrode potential of standard hydrogen electrode and is assumed equal to zero at all temperatures by convention. Therefore, equation (2.38) becomes: As the potential of SHE cancels, no matter what value it is, the emf of the cell, £cen, is not affected. The cell potential (emf) is positive if the left-hand electrode is negative and the righthand electrode, positive. Consequently, oxidation would occur on the left-hand electrode, and reduction on the right-hand electrode. 2.8 CONCENTRATION for the standard hydrogen electrode (SHE). The cell reaction is obtained by taking the difference of equations (2.35) and (2.36) above: CELL In a concentration cell, the emf results from transportation of an electrolyte (anion, cation or both) from the more concentrated solution (2.37) to the more dilute solution. Here the metal is M" + + - H 2 = M +rcH + in contact with two half cells having the same Reaction (2.37) is the complete reaction for electrolyte at two different concentrations. Such the cell expressed in equation (2.34) above, where a cell is represented, for example, by two zinc Basic concepts in corrosion electrodes immersed in zinc sulfate solutions at different concentrations: Zn(s)/Zn 2+ (0.1 M )//Zn 2+ (1.0M)/Zn(s) (2.42) Another example of a concentration cell is two standard hydrogen electrodes immersed into two HC1 solutions of different concentrations: Pt(s)H 2 (g, latm)/HCl(a 2 )// HCl(ai)/H 2 (g, latm)/Pt(s) (2.43) 23 In the above cell, HC1 is in two different concentrations. The activity (molality x activity coefficient) a\ is greater than activity a2; a\ > a2. Several types of concentration cells are encountered in corrosion. For example, a concentration cell is formed if one end of a pipe is exposed to soil and the other end to air. The end of the pipe in air is exposed to a high concentration of oxygen than the end of the pipe in the soil. The formation of a concentration cell leads to differential aeration corrosion in buried structures in the soil. 2.9 LIQUID JUNCTION POTENTIAL For accurate measurement of emf, two conditions must be satisfied: (a) The cell reaction must be reversible and (b) no current must be drawn from the cell. The boundary between two electrolytic solutions with different concentrations is a source of irreversibility in measuring the emf of the cells. The potential difference developed at the boundary is called'Liquid Junction Potential' The potential difference is caused by the migration of ions from one electrolyte to another electrolyte. Let us take the case of concentrated HC1 forming a junction with dilute HC1. Both H + and Cl~ ions diffuse from concentrated HC1 to dilute HC1. The hydrogen ion moves faster and, therefore, the dilute solution becomes positively charged due to ingress of H + ions from the stronger acid solution and the more concentrated solution thus acquires a relative negative charge. The consequent potential difference across the double layer, which is a positively charged layer in the dilute solution side and a negatively charged layer in the concentrated solution, may effect the emf of the cell seriously in certain cases. It is generally believed that the difference in potential is caused by the difference in the rate of diffusion of oppositely charged ions. In the above example, the charge acquired by a dilute solution is that of the faster moving ion (H + ); chloride ions (Cl~) diffuse much slower. The magnitude of the liquid junction potential may affect the reversible potential positively or negatively depending on the mobilities of the ions. The most common method of removing liquid junction potential is to insert a KC1 salt bridge between the electrolytes. Porous barriers are also used. The introduction of salt bridge minimizes the liquid junction potential. The KC1 contained in the bridge, allows the migration of K + and Cl~ ions which carry most of the current and migrate with the same mobility. The potential of zinc is more negative, hence it is considered as an anode (Zn—>-Zn2++2e) and the potential of nickel is less negative than zinc, hence it is considered as a cathode (Ni 2+ + 2e -> Ni). The sign of the potential is changed for zinc as it oxidizes. From the potential obtained (+0.51 V), it is clear that the reaction will proceed spontaneously. 2.lO APPLICATION OF F R E E E N E R G Y TO CORROSION C E L L S It is known that corrosion reactions produce electrical energy. The amount of work done by a cell is equal to the quantity of electrical energy which it generates under constant pressure, temperature and concentration of the reaction. In an electrochemical reaction, electrical energy available is equal to the product of potential of the cell and the quantity of electricity involved (volts x amperes x time). That is, Electrical energy = ExQ. This is equal to the net work done by the cell. From Faraday's law, Q is 24 Principles of Corrosion Engineering and Corrosion Control one Faraday (F) for each gram equivalent of the Hence, reactants. For n gram equivalent of the reactants, Q is equal to nF. Work can only be performed if R eaction = - 0 - 0 0 2 V the free energy of the cell is decreased, as shown in Section 2.6.2: Reaction does not take place simultaneously. AG = -nFE where A G = change in Gibbs free energy of a cell in joules per mole n = number of electrons involved in the reaction E = emf of the cell in volts. If all substances are in their standard states (i.e. at unit activity) the expression (2.44) becomes: AG° = -nFE° (2.45) (2.44) 2 .1 1 NERNST EQUATION The Nernst equation expresses the emf of a cell in terms of activities of products and reactants taking place in the cell reaction. Consider a general cell reaction: Mi +M£ + < * M2 + M* + = (2.46) Illustrative Example 2.3 F orthecell[Cu/Cu 2+ (1.0M)]//[Zn 2+ (1.0M)/Zn], and the cell reaction Cu+Zn 2 + ->• C u 2+ + Zn: If and AG298oK = - 147.5 AG298oK = 63.35 mol Solution of the Problem: AGLaction = Sn,AG(0-E«;AG(i) where i = products j — reactants n = number of moles •••AG£eaction = 2 [63.35+0-(-147.5+0)] So, kT -^ f orZn 2+ mol —r for Cu + Mi and M2 represent metal electrodes, such as Cu and Zn, in a cell, and the above reaction (2.46) can be written as: Cu+Zn 2 + +± Zn + Cu 2+ (2.47) The free energy change (AG) of a reaction is given by the difference in the molar free energy of products and reactants as shown in Section 2.6. AG = TlniAG(i)-i:njAG(j) (2.48) Therefore, the free energy change for the reaction (2.46) can be expressed as: AG° = ( G ^ + G ^ + ) - ( G ^ + G ^ „ + ) (2.49) If the substances are in their standard states the above expression becomes: AG0 = ( G^ 2 + G ; „ + ) - ( G ^ + G ; „ + ) (2.50) A G ^ =425.1 J l Since, AG Free energies of any metal, such as Mi at its standard state and any arbitrary state are related through the following equation: Qvii - Gjtf! = RT In atAi (2.51) Reaction = -^ £ = -2x96500x£ where ay[x is the activity of metal Mi, R is the universal gas constant, and equal to 8.314 joules/degree/mole, and T is absolute Basic concepts in corrosion temperature in degrees Kelvin, °K, which is (273.16+f°C). Subtracting equation (2.50) from equation (2.49), we get: A G - A G 0 = (GM2-GM2) + ^GM«+-G^„+J 25 At25°C 0.059 £M: zE M~ logK+] - (GMl - G^J - [GMn+ - G^n+J Substituting from equation (2.51), we obtain: A G - AG° = RTIn aM2+RTIn aMl Equation (2.53) is a general form of the Nernst equation. It expresses the emf involved in an electrochemical reaction, such as equation (2.46) in terms of R, T, F, n activities of products and activities of reactants. However, the IUPAC sign convention requires the above minus sign to be replaced by a plus sign. The Nernst equation is re-written to comply with the IUPAC sign convention, as: RT, oxidized state activities product rt E = E° + — I n — *—: nF reduced state activities product — RT In ayix — RT In aMn+, AG-AG° = RTln (2.52) Mi%»+ a which is written regardless of which way round where the V values are any arbitrary activity the redox reaction is written, Hence, for ++ ++ values, for which AG is the corresponding free Zn = Z n + 2 e ~ , and for Z n + 2 e ~ = Zn, energy of that reaction. A G° is the free energy change for the reaction involving standard states E = E°-\ In a7n++. (unit activities). Zn nF As AG = -nFEi and also AG° = -nFE°y from equations (2.44) and (2.45), respectively, the 2 .11.1 A P P L I C A T I O N O F above equation (2.52) becomes: N E R N S T E Q U A T I O N TO A E-E° = — In nF whence (2.53) a CORROSION REACTION A corrosion reaction can be considered as composed of two half cell reactions. One of the half cell reactions corresponds to 'oxidation reaction taking place on the 'anode,' and the other half cell reaction corresponds to 'reduction reaction' taking place on the 'cathode of the cell. The contribution of each half cell reaction to the Nernst expression can be derived as follows: M i - » M j + + ne oxidation reaction at anode, and M^ + ne^M2 (II) (2.55) (I) (2.54) M 2 Mj+ a nF |_ areactants J Suppose M->Mz++ze RT EM = E^- — \n[a^xzaze] because au and ae — \. reduction reaction at cathode. 26 Principles of Corrosion Engineering and Corrosion Control Equation (2.60) is more commonly used in the form of E quation (2.53) can also be re-arranged into expressions to correspond to reactions (I) and (II): E = E° + E = EX+E2 E= (2.56) x log 0.05915 n product of activities of reactants product of activities of products (2.63) « s-& „«+N + (2.57) ^ + 3 + nFl n ( 3 k j \a Ml The expression (2.61) can also be written with the activities inverted, as E = E° n w here E° = standard potential, and E° = £ ° + £ £ (2.58) Mi . RT ^ p roduct of activities of products In-— fnF p roduct of activities of reactants (2.64) nF £ 2 = £2° + ^ l n nF ' an+' flM2 (2.59) and E=ET+—in nF (2.60) %!flM2 Equations (2.58) and (2.59) are Nernst equations for the half cell reactions and equation (2.60) is the Nernst equation for the complete cell reaction. Equation (2.60) can be put into a mathematical form: RT, activities of reactants rt E = E° + — In nF activities of products But this equation is incomplete without a sign convention. Quite separate from the above considerations, standard Redox potentials are tabulated with reactive metals like Zn negative instead of positive, which accords with the experimental fact that Zn will actuallybecome n egative because when Z n + + i ons leave it, then two electrons will appear in the Zn metal, to maintain a charge balance. Such reduction potentials are given in Table 2.1. By convention the cell reaction is written as if the oxidation process is occurring at the electrode on the left and reduction on the right. This practice is in accordance with the International Union of Pure and Applied Chemistry (IUPAC) Convention held in 1953 and modified in 1960. 2.12 SIGN CONVENTION (2.61) But, in addition to this statement (2.61), a sign convention is necessary as already mentioned and this will be discussed in the next section. Since RT _ 2.303 x 8.314 x (273+25) 96500 (2.62) The older American Convention has been to use oxidation potentials rather than reduction potentials. The present practice is to use reduction potentials instead of oxidation potentials in accordance with the recommendation of IUPAC (International Union of Pure and Applied Chemistry) and existing European practice - the general convention is that the cell emf is given by: ^cell — ^ Right' ~nF n ~ •K .eft = 0.05915 at 25°C T he standard hydrogen electrode (SHE) is the reference point. The construction of hydrogen Basic concepts in corrosion electrode is shown in Fig. 2.10. By definition, SHE has zero potential at 25°C. It is assumed that oxidation occurs at the hydrogen electrode so that it is placed at the left-hand side of a cell diagram. If a cell is composed of SHE and some other electrode, then the measured voltage of the cell is the electrode potential of the second electrode: Ece\\ = Ex-0 ^cell = Ex 27 2 .12.1 C E L L E M F A N D T H E D IRECTION O F S P O N T A N E O U S R EACTION The following is a step-by-step procedure to apply the above sign convention to a cell reaction in order to determine cell emf and direction of spontaneous reaction correctly. (1) Write the right-hand electrode reaction. The right-hand electrode reaction is always reduction and the electron is always a reducing agent: M++ e~-+M R For example, in a galvanic cell made of a hydrogen electrode and a zinc electrode [Zn/Zn 2+ (1.0 M) ], the measured voltage is E = —0.763 volts. 170 770 r-o And reduction electrode potential is ££. (2) Write the left-hand electrode reaction. The £c°ell = - 0.763 V left-hand electrode reaction is oxidation. However, we write it as reduction reaction In the older textbooks, hydrogen electrode was because of the convention we are using: placed on the right, and the cell potential becomes: ^cell-^Zn-^ ECe\\ = En2—Ex And the reduction electrode potential is E[. The reaction was made oxidation and the sign was (3) Subtract the left-hand electrode reaction and potential from those of right-hand changed. Thus the electrode potential E° ++ electrode to obtain cell reaction and cell is +0.763 volts according to old convention. potential: £c°ell = -EH2-(£zn)> £ B| n = +0.763 V cell • • ^Zn - F° M++M L = M R +M+ £c°ell = 0.763 volts For example, calculate E° for the Daniel cell. which is opposite to that shown above. This From the table of reduction potentials, write the should clarify the confusion caused in student's corresponding reactions: minds by different conventions. The simplest procedure avoids specifying 'right and 'left by Cu 2 + +2e->Cu E° = 0.337 (R) just writing: Zn 2+ + 2e -> Zn RT, oxidized state activities product E = E° + — I n — nF reduced state activities product 0 E° = - 0.763 V(L) Ece\\ = E^ — E[ = 0 .337-(-0.763V) which is written regardless of which way round the chemical equation is written. = 1.100V 28 Principles of Corrosion Engineering and Corrosion Control Therefore, the tendency of the right-hand electrode is reduction (acceptance of electron; M R E + £ = M R, if £ > 0 ) . At the same time, 17° ro r?o positive emf signifies an excess of electrons on ^cell-^R-^L the left-hand electrode. Therefore, the electrode reaction is oxidation ( ML = M ^ + e) on that electrode. Thus, if E > 0, the cell reaction has a R ->L tendency to proceed from left to right: that is, electrons are produced by the left-hand electrode = -0.126-(-0.136) and consumed by the right-hand electrode. It also = 0.10V indicates that the reaction is spontaneous. Conversely, if E < 0, the reaction at the rightM++e^M hand electrode is oxidation and the reaction at the left-hand electrode reduction. The tendency Zn-^Zn+++2e of the reaction would, therefore, be to proceed from right to left and the reaction would not be Cu+++2e^Cu spontaneous. Figure 2.11 explains theflowof electrons in the two situations explained above. The same convention can also be understood in terms 2 . 1 2 . 2 POSITIVE AND of flow of positive current (positive charged ions). It is to be noted that according to the convention, NEGATIVE E M F if a cell is short circuited, and the positive current A positive emf (E > 0) signifies a deficiency of flows through the electrolyte from left to right, electrons on the right-hand electrode because of the emf is positive. In this case the left-hand electhe extraction of electrons from that electrode. trode is the anode and the right-hand electrode A nother example: Using a cell of P b 2 + / P b and Sn 2 + /Sn, calculate JB°ell: Tendency of electron flow Tendency of electron flow + © A © © © © © © © © ® © © -' - I R © 0 0 0 r ^ © © © © © © r 0 0 0 ©: 0 £>G (ER E<0 (&<EL) > Ei) Figure 2.11 Implications of the sign convention Basic concepts in corrosion Table 2.2 emf Current flow conventions Current flow L -*R R^L Anode L R Cathode R L 29 + the cathode. If the current flows from right to left, the emf is negative. Table 2.2 summarizes the current flow convention in a cell. From knowledge of cell potential it is thus possible to predict whether a reaction would proceed spontaneously or not. From knowledge of potential of half cells, and the cell potentials, it is possible to predict whether a metal is an anode or Z n 2 + + 2 e ^ Z n , ££n = - 0.763 V cathode or whether a corrosion reaction is likely to take place or not. An understanding of elecCu 2+ +2e +± Cu, ££u = +0.34 V trode potential is of fundamental importance to the understanding of corrosion mechanism. This Subtracting anodic reaction from cathodic reacis illustrated in typical problems given below. tion to obtain the cell reaction Illustrative Problem 2.4 Two half cell reactions are given below: Cu2++2e^Cu, £° = 0.34(V) Z n 2 + + C u ^ Z n + C u 2 + and £c°ell = - 1.10 V From the table of standard reduction potentials (Table 2.1), the reduction potential of Cu is +0.34 volts and that of Zn is —0.76 volts. The emf of the cell is positive. Hence the reduction is spontaneous and it should proceed from left to right. If the position of the electrodes are now interchanged such that the copper electrode is placed on the left and the zinc electrode on the right, the sign of the cell emf will change. Here, zinc electrode being on the right is treated as cathode and the copper on the left as anode according to the convention discussed earlier in the chapter. Therefore: The emf obtained is now negative which indicates that the reaction is not spontaneous and the current flows from right to left. The leftZ n 2 + + 2 e ^ Z n , E° = - 0.763 (V) hand electrode where the current originates Their reduction potentials are given opposite to is, therefore, the anode, and the right-hand electrode, the cathode. The polarity of the cell is, each reaction. Calculate: therefore, clearly established. (a) the emf of the cell and Illustrative Problem 2.5 (b) show the spontaneous cell reaction. Calculate the reversible potential for a zinc electrode in contact with ZnCl2 when the activity of Solution: 3 In Fig. 2.2, the left-hand electrode (Zn) is the zinc is aZn2+ = 10~ , Use IUPAC convention. anode and the right-hand electrode (Cu) is the cathode according to the convention we studied Solution: earlier on. Therefore, zinc will undergo oxidation We write the half cell reaction and copper reduction as shown below. Hence: Z n 2 + + 2 e ^ Z n , £° n = - 0.76 V (a) £c°eU = £ ° - El (according to IUPAC Apply Nernst equation: convention) = 0.337-(-0.763) £c°ell = +1.100volts (b) Spontaneous cell reaction is given by: cell reaction = cathodic reaction = anodic reaction Cu 2 + +Zn = Zn 2 + +Cu E Zn = E ^ — ln[aZn2+] 0.0591 r -. £zn = -0.76+ - y - log [aZn2+ J EZn = - 0.76+0.03 log [10 - 3 ] EZn = - 0.85 V 30 Principles of Corrosion Engineering and Corrosion Control Illustrative Problem 2.6 t h e i o n s i n t h e s o i u t i 0 n, and a{ is the activity Show that for the reduction reaction given below: coefficient. The activity coefficient a{ for a given concentration can be found from a table of activ2 H + + 2 e ^ H 2 , £ H = - 0.0591 pH ity coefficients given at the end of the chapter. In dilute solutions, the activity coefficient is taken Solution'. t 0 be unity. if w e t a k e a0l =PQ2/P° = 1, when P0l = The reaction is 2 H + + 2 e ^ H 2 . Using Nernst equation we obtain for the above reaction: po = x b a r > s t a n dard state, and R = 8.314 J/mol-K, F = 96 485C(g equiv.) -1 , T = 298°K, and the [reaCtantS] £ ( H + / H 2 ) = £° + * I In term 2303RTIF = 0.0591, and substitute in the 2 nF [products] above expression, the oxygen electrode potential becomes: RT [oxidized state] 0 or % + /H 2 ) = £ (H+/H2 ) + -^ ^ [ r e d u c e d s t a t e ] % ) 2 / O H - ) = ^ 0 2 / O H - ) - ° - 0 5 9 1 l o 8 a (OH-) ( IUPAC-Nernst) Changing to 2.303 log and substituting for £ = 8.314J/(mol-K), T = 298°K, F = 96490C, tion f o r 2.303^7/^ = 0.0591V. It can similarly be shown that in acid solu» r e a c t i° n > ° 2 + 2 H 2 0 + 4 c -> 2H 2 0, the oxygen electrode potential becomes: , ,4 DT n E(o-\ = £/°n ^ H In • ( U2J «H20 4rcF The Nernst expression above can, therefore, be Substituting the values for the constants as written as: above, we get: 17 CO inA r fll rrr+T E (02) = E (02) { 2) + 0 ' 0592 lo g fl H+ %+/H2) = £(H+/H2)+°-0591OB[H 1 or since pH = — log <?H+ > Now substituting pH = —log(H+) in the above expression we get: £ (H+/H 2 ) £(02) = £(°o2)— 0.0592 pH Illustrative Problem 2.8 Calculate the potential of oxygen electrode at pH = 14.0. = - 0.0591 pH Illustrative Problem 2.7 Show that for the reaction: 0 2 +2H 2 0+4e<=^ 4 (OH _ ), Solution: Oxygen electrode reaction: in basic and neutral £ ( O 2 /OH-) = £(°O 2 /OH-) - 0.05911ogaoHenvironment is expressed as: Solution: The reaction is 0 2 + 2 H 2 0 + 4 e ^ 4 ( O H ) (in a basic solution), and E,r* , ™- ^ = E° M02/OH ) _x: Q 2+2H 2 0 + 4e-2H 2 0 a t 2 5 o C is g i v e n b y Illustrative Electrode potential equation for the electrode p ro blem 2.7: Eo, = En, ~ 0.0592 log a(OH- ^ U2 4- — In in a2 ° ( 0 2 / O H - ) ^ np / x4 °2 & (OH > where Note that 0 2 is the oxidized state and OH~ is the reduced state, for the above reaction (from IUPAC-Nernst expression). The activity (a;) of an ion is defined by a ^ M i t f , where Mi = molality (mol/kg) of lo Bfl(OH-)= _ 1 4 - l o g fl(H+) _ — F (from the knowledge of pH) Basic concepts in corrosion Therefore, £o 2 =£° 2 -0.0592(-14+pH) Substitute ££2 = 0.401 (from Table 2.1). Therefore, £b 2 =0.401 + 1 4x0.0592-0.0592pH £b2 = 0.401+0.828-0.0592 pH and so a t p H = 14, Eo2 = 0.401+0.828-0.828 and Eo2 = 0.401 log aFe2+"| 0.301 xnF flSn2+J~ 2.303£T where: asn = 1 and ape = 1. £cell = - 0.301 + £cell = 0, Therefore, 0.301 = 2.303£T, aVei+ —-log^nF aSni+ 2303RT, a Fe 2+ —— log nF ' aSni+ 31 At the turn of polarity. Assuming the reaction is taking place at t — 25°C,rc = 2 ,T = 2980K,£ = 8.315 J/mol and F = Illustrative Problem 2.9 96500 C(gequiv.) _1 In the cell reaction given below, what is the ratio Then of the activities of ionic species required to make the polarity reverse? :e2+l_ 0.301 X2X 96500 0 ^2+ log ~ 314x298 S ] ^+J~~ 2.303x8.31 F e 2+ + S n->Sn 2 + +Fe = 6.4xlO~ n at25°C Solution: Cell reaction: At20°C = 5 x l O - n a Fe2++Sn^Sn2++Fe Ratio of activities to make the polarity reverse at 25°C. Standard cell potential: 17 o po 2 Illustrative Problem 2.10 The emf of a cell made of Zn (anode) and H2 electrode (cathode) immersed in 0.7 M ZnCk is +0.690 volts. What is the pH of the solution? Solution: Given: [Zn 2+ ] = 0.7M, ForZn 2 + , activity coefficient, y = 0.6133 .-. a Zn 2+= 0.7x0.6133 = 0.424 (Right) - > 2 H + + 2 e - ^ H 2 (Left)-»Zn2++2e~^Zn £ ° n = - 0.762 V po ^ cell - ^ Fe +/Fe ^Sn +/Sn 2 = - 0.441-(-0.140) = -0.441+0.140 = - 0.301 volts The reaction in this case is non-spontaneous. That is, Fe2+/Fe electrode will act as 'anode' and S n 2+ /Sn electrode will act as 'cathode,' the reverse case of what was assumed in the problem. To reverse the polarity: Overall cell potential is given by: TIO , - ^ 1 t a Reactant] fccell = £,.11+ - 7 In 7 £ C ell "J [ ^Product] = % 2" H ~^Zn 0.0592 [log(pH2)-log (<£+)] = - 0.301 + ^ ^ nF lo,^2+aSn aSni+ flFe 0.0592 N[flZnH°g[>Zn2+]] 32 Principles of Corrosion Engineering and Corrosion Control 0.0592 = - 0.111 volts £cell = —0.111 volts Illustrative Problem 2.12 Calculate the pressure of H2 required to stop corrosion of iron immersed in 0.1 M FeCk, pH = 4. Solution: (R)^2H++2e"->H2 (L)^Fe2++2e~^Fe £°e = +0.44 [a Fe2+ ] = yx[Fe 2 + ] £Cell = [2 pH]+0.762 0.0592 , , + —£—[-log(0.42)] 0.690= -0.0296[2pH]+0.762 + 0.0296(0.376) 0.0296[2pH] = 0.762-0.690+0.011 =0.083 0.083 2v H = p =>2pH = 2.8 F 0.0296 => pH=1.4 Illustrative Problem 2.11 Calculate the theoretical tendency of nickel to corrode in deaerated water of pH = 8. Assume the corrosion products are H2 and Ni(OH)2 and the solubility product is 1.6x 10~16. Solution'. Given: Ksp = 1.6 x 10 _16 ,pH = 8, £ ^ = - 0 . 2 5 V Since, pH+pOH=14=>pOH = 14-8 = 6 = 0.1x0.75 ECZ]1 = ER-EL £cell = £ H 2 -0-0296[log(pH 2 )-log(a H+ ) 2 ] - [£F°e - 0.0296 (log(aFe) -log(a Fe2+ ))] Given: pH = 4 =» [aH2] = 10~4 ^ £ = £ Q - 0 . 0 2 9 6 [ 1 O g ( p H+2 i [ ^ 2 + ] l J L [H ] 2 O = 0.44 - 0.0296 [log(7.5 x 106) p H 2 ] = 0.44 , , = log[7.5 x 106]Fp H. 8 2 0.0296 L J [ OH"] + 2 1 0" 12 (R)->2H +2e"->H2 (L)->Ni +2e"->Ni 2+ = (7.5 x 106) p H 2 = 7.32 x 1014 =» ^cell = ^H2 - ~Y~ = - \ E; P°g(PH2) ~ 2 l °g (flH+ )] [log(flNi) p H 2 = 9 .7x 107atm Illustrative Problem 2.13 Calculate if silver would corrode when immersed in 0.5 M CuCl2 to form solid AgCl. What is the corrosion tendency? Solution: 0.0592 -log(aNi2+)]j| > £cell = -0.0296[2 pH]+0.25 +0.0296[-log(1.6xl0" )] • £cell = -0.0296[2 x 8] + 0.25 +0.0296[4-log(1.6)] 4 E°Cu = 0.337 V, £ ° gC1 = 0.220 yCuCi2=0.7 aCu2+= 0.47x0.5 = 0.233 "Cu = 0 .47x2x0.5 = 0.47 Basic concepts in corrosion 3 3 (R)->Cu2++2e~^Cu (L) -> 2AgCl + 2e~ -> 2Ag+ + 2C1" 0.0592 r ^cell = £ C u ~ £AgCl — L[log(flCu) The electrode potential for hydrogen £ H 2 c a n be determined as follows: 2H+ + 2 e ^ H 2 17 17° ^ R T (2.65) \ (fl - log(a C u 2 + )]-[log(Cr) 2 - log(AgCl)]] =» £ceii = 0.337 - 0.220 - 0.0296 [ - log( aCu2+) - 2log(a cl 2+)] = 0.117-0.0296[0.632-2(-0.327)] = 0.079 volts £cell = 0.079 volts H+)2 , £H+/H2 = V / H 2 + -£ loS " ^—j" <2-66) 0 ,,v where a^+ is activity of hydrogen ions, andpH2 *s hydrogen partial pressure. > At one atmosphere pressure (pH2)> 0H2 = 1 and E^+/H = 0 by definition. Therefore, £ H + / H 2 = 0.059 log ( « H + ) (2.67) or in terms of pH, £ H +/H 2 = - 0.059 pH (2.68) 2-13 REFERENCE 2 .13.2 ELECTRODE SILVER-SILVER ELECTRODES 2 .13.1 HYDROGEN C HLORIDE E L E C T R O D E This electrode is composed of a silver wire coated with silver chloride and immersed in a solution of chloride ions (Fig. 2.12). The chloride equilibrium is given by: AgCl^Ag++Cr (2.69) The hydrogen electrode is used as a reference for electrode potential measurements. Theoretically, it is the most important electrode for use in aqueous solutions. The reversible hydrogen electrode in a solution of hydrogen ions at unit activity exhibits a potential, which is assumed to be zero at all temperatures. The electrode consists of a platinum wire immersed in a solution (Fig. 2.10) containing hydrogen ions and saturated with hydrogen gas. Platinum is immersed completely in aqueousarsenic free hydrochloric acid, and hydrogen gas free from oxygen and carbon monoxide is bubbled to the platinum surface. Slowly, the air is displaced by hydrogen and the reversible potential is achieved. Unfortunately, this electrode has some drawbacks. First, the reversibility of hydrogen electrode cannot be maintained in oxidizing media. Second, if a current is withdrawn from the electrode, the electrode acts as an anode because of the ionization of gas molecules. Also, the electrode is fragile and delicate to handle. Two other reactions involve a dynamic equilibrium between deposition and dissolution of silver together with solubility equilibrium between silver chloride and its ions. The metallic silver reaches equilibrium with silver ions according to the following reaction: Ag+ + e ^ A g (2.70) The overall electrode reaction is, therefore, given by: AgCl+e^Ag+Cr (^gCi/Ag = ° - 0 2 4 V ) (2.71) 34 Principles of Corrosion Engineering and Corrosion Control The equation holds at 25°C. It can also be written in the following form: ^Ag/AgCl = 0.224 - 0.0592 log acr (2.75) Plastic tubing At low concentration log a cl - can be replaced by pH as Cl~ is provided by HC1 acid, [Cl~] = [H + ] and hence — log a cl - can be replaced by — l og[H + ]. Therefore, - l o g [ H + ] = pH Hence, 0mmm Electrical wire Injected sealant Soldered joint u , <,~y>^ J* ,, ^Ag/AgCl = 0.224 - 0.0592 pH (2.76) Sliver wire Silver The following are the values of ^Ag/AgCl f° r different HC1 concentrations: Concentration Electrode potential (volts) (M) 0.1 0.28 0.34 0.01 0.40 0.001 2 .13.3 T H E C A L O M E L ELECTRODE chloride Silver-Silver chloride reference electrode It is the most commonly used reference electrode. It has a constant and reproducible potential. The electrode basically consists of a platinum wire dipped into pure mercury which rests in a paste of mercurous chloride and mercury. The paste is in contact with a solution of potassium chloride The electrode potential, ^Ag/AgCb is given by: which acts as a salt bridge to the other half of the cell (Fig. 2.13). RT 0AgCl : :£ ln The most commonly used concentrations of (2.72) ^Ag/Cl !g/AgCl + - ^ aAg x aaKC1 are 0.1 N, 1.0 N and 3.5 N and saturated KC1. The saturated calomel is used when the liquid flAg=l junction potential is to be kept low. The potential «AgCl = 1 of electrode at 25°C is 0.241 V in saturated KC1 solution. Therefore, Mercurous chloride is slightly soluble, and it is in equilibrium with mercurous ions accord/ 2.3031*1 V ing to: ^Ag/ci = £Ag/Agci ~ ^ — p — ) l o g acr (2.73) Figure 2.12 Hg+ + e = Hg or £Ag/cl = ^g/AgCl "0.0592log acr (2.74) The overall equilibrium is expressed by: Hg2Cl2+2e^2Hg+2Cr Basic concepts in corrosion 3 5 2 .13.4 Electrical connection Bung COPPER-COPPER SULFATE ELECTRODE This is a reference electrode which is easy, robust and stable. It is used mainly in cathodic protection measurements, such as the measurement of pipe-to-soil potential. It has a lower accuracy than other electrodes used for laboratory work. It consists of copper metal placed in a solution containing copper sulfate and copper sulfate crystals placed in a non-conducting holder with a porous plug (Fig. 2.14). The copper sulfate crystals maintain the solution at a fixed ion concentration. Necessary contact with the earth is made through the porous plug. It is easily recharged when it becomes contaminated. The equation for the copper-copper sulfate electrode potential is given by: £ Glass tube Platinum wire Glass tube Mercury Mercury (1) Chloride Saturated aqueous potassium chloride Porous plug Porous plug Figure 2.13 A saturated calomel reference electrode Cu-CuS04 = 0.316 + 0.0009(25°C)volts (2.77) The reaction of the CU-CUSO4 half cell is The mercurous chloride and mercury are at unit activity. Therefore, the electrode potential can be written as: RT 2P a Cu2++2e^Cu and the electrode potential: 0.0592, £ (2.78) ^Calomel — -c H" Cu 2 +/Cu = £ Cu 2 +/Cu + — — log fl (Cu2+) ( 2 ' 7 9 ) where 0.0592 or ^Calomel — ^ l og*, tr (Cu2+) = activity of copper which is unity, ££u = 0.34 V at 25°C. The value of E° for the half cell reaction of A saturated solution of 1.47 M CuS0 4 at 25°C is calomel electrode is 0.267 V. Thus, the electrode used. potential becomes: aCu2+ = [Molarity (M) x Activity coefficient y] aCui+ = 1.47 x 0.037 (y is found from the table 0.0592, , , of activity coefficients - Table 2.3) ^Calomel = 0 .267 l og ( O Q - ) So, aCu2+=0.051 Substituting in the above equation we obtain £ Electrode Potentials of Calomel Electrode (Standard hydrogen electrode taken as reference) Electrode Potential (volts) Hg/Hg2Cl2/KCl (Sat.) 0.2444 0.289 Hg/Hg2Cl2/KCl (1.0 N) Hg/Hg2Cl2/KCl 0.3356 (0.1 N)/Salt Bridge 0.0592 Cu2+/Cu = 0 . 3 4 + — — log(0.051) Cu2+/Cu = ° - 3 0 v o l t s (2.80) (2.81) £ £Cu2+/Cu = 0.316+0.009 (T°C) volts, as seen earlier in this section (equation (2.77)). 36 Principles of Corrosion Engineering and Corrosion Control o 1H O O O • < — II I III II N O oo r^ O N LO \ N o H oo en | CN r l ON CN l > CN LO > I I D ^O N I I <NI I (N <' — .876 .036 i-H T-H 603 o CO r-H i-H dq i-H CN CO 00 1 q 1 1c 1 o 00 ON T-H oo CN o ON 00 00 CN r-H d o NO ON i-H m odd 1 1 11 m ^ 1 t^ 1 1 i-H i-H d 11 i 5N i-H dq d m i °° o CO ^ i-H q CN q i-H i-H r—1 1 CO LO m en ON i-H 1o in dddddd om NO i d e n i £ NO NO NO 1 oooo 1 r^ 00 00 TH m ^inVOr-HNO .COOOt-H 0 0 ddd NO o ON in CN 00 'd R d dd CN o in NO NO 1 .809 ON ON CN CO 00 NO o i-H NO CO t^ c O o oo Ln Ln^LnONCOr-H^ONCNOOONNOOCNNOON oo m ^ t^ co r^i^r-Hvot^Ln^HcoLncoooNo^OLn r^ oo O N NO «tf O CO r>* i>* C N t ^ O OOO ^^O^OOMinHQNNH CN ON 0 0 ^ mCNrHr-HCNNOr-HNOinOOO r^ r ^ N O N o r ^ q l N N oo o' CD cS CD d> odd oo d CN d NO m m ON 0 0 NO \^F_(^(TJH^Ir-Hi-HNOTf'-H Lnooor^r^rvT-Hcor^i^oo ododooodooddddddddddd m O N [^ O m m(N^OOcOin\ONO^fOOOHH ^ inT^mooor^i^oor-H^t^oooo oooooooooooooooooooooo N NM I . OCNOONT—I 0 0 r-H f v 00t^CN^m^CN'-H m o d CN O . I o co o ^ m CN CN CO CN "^ CN CN ON 00 00 CO 00 00 N . ^ ) N ^ 0 0 H O N 0 0 H ( S i n 0 0 H ( N ^ O \ ( N M ^ . l cS <5 <6 <6 cS d> I i n i n i n M ( N N i n ( N \ o o o o o o o o x H i n o o o o <6<6<6<5<5S<6<6<6<6<6c5cD<6d>c5cSd> m T-H r o t^ t^ m . NO N O 00 0 0 ^ I 00 00 CN .NO^OO'-HNONOT-HOONCN INONOCOCOOONOCOt^OON CN . I>> OO . loOOO I CN Tt< v o 0 0 ^O 0 0 0 0 . I d odd o ON O m I N ON ON ^ do <^ CDCSCD'CDCDCDCSCDCDC^ in CN i—i r>> O O NN^ ^ t ON do CN T co m H TH m - * co o o o ^ N H ON ON ON ON odd r-H N ON ON O 0 0 ON . I q d o o^ CN N CN .o oooooooooooooo .927 .927 .927 .785 in o o d i 1 ^ t^ CN CN ON O N CN 1 ON ON dd o dd CN r H in in m ON O N d o dd LO m in ON LO 00 ddddd 00 o CN 00 CO o 1 ON dooo m 1 00 CN CN C 1 °° N ON 1 d^ o rd 1-H o o 11 CN 1 1 1 00 00 NO NO 1 00 NO 1 ON mmm ON ON 00 11 ON ddd o o NO m o NO NO CO ON ON 00 dddd 11 1 1 00 00 00 oooo ON d 00 NO CO d 1 00 o ON ON ON dddd dd 00 m m in NO NO 1 1 ON ON ON co NO NO 1 00 ON ON 1 OOS ^ CN O / A* CM crj 2 u ^ ^ UC H SH ££2 <<PHUUUUUUU£ .932 ON 1 .639 oo r^ i-H Basic concepts in corrosion 3 7 00 I I I I i I I I I CO i—I od o NO T-H CN i s oo d I I I I II I I 00 ^ o ON o O (N O I-H © I I m I I I I till I II o dd NO I NO t ^ O T t N © © © © m I I I IS I I I I I I I co m oo ^ mm^ . oo q q q I Io © «-H © © m T-H .064 i-H .686 NO CN NO q 00 m ts in , I H m ^ I I ^H CO H 00 M CD © ts in NO NO 1 NO is o in in o 00 CN i n m co q m rjn CD CD d dddd d dddd odd . I Nm N ^H NO m m 00 m CN 0 0 t s CN CN NO CN CN 0 0 ON 0 0 CN i s NO CN i n © o NO NO m CN NO NO NO ooooooooooooooo NO odd m . CO ON 0 0 NO I m m co o dodo m o o NOCOCOCOONOONOCNONCO^^HONOCN NinHH^^^fO^NtNOOtSfCits s, © © © © © © © © <im — © in oo i n <H CDCDCDCDCDCDCDCDCDCDCDCDCDCD'CD 00 O O ONNooooo^cocoomooocor^Lnr^. NiftHHNtsN^NNoOOON^N CD'CDCDCD'CDCDCD'CDCDCDCDCDCDCDCD rjn CO 00 d ,CN ICN d I .NO 1^ d I CO ^ .ONOOOOOOT-HCN^NOCNCOCN I NNN^OOOOOOOOOOl^OO CD'CDCDCDCDCDCDCDCD'CDCD IS N C O C OO O^ is C t s t s t s IS O CDCDCDCDCDCDCDCD .743 om m T-H IS is ON O i s oo NO . m I dodo oo IS oo ,CN I co . m is t• — TtfO^O\NNOOO\N^NOO loqoqooLnoooqoooqoqNoooLnNq I I I I ISI O O ON d . I O ^ d O ©©©©©©0 000©©© i-H ON I S CO I I I I .OOISOOISOO'-HT-HO'—lOr-^i^O loOOOOONOONONONONONC^ONNONOON CDCDCDCD'CD'CD'CDCDCDCDCDCDCDCD O CO ON . I . I d f-H,—,^^ ONTji 00 .T-HOr^cNcOCNcO'^cOt^cOONOcO lONONONlSONONONONONlSONt^lSON CDCDCDCD'CD'CDCDCDCDCDCDCDCDCD I II I IS IS 0 0 ^ dd co m co i s LnmNOLn^Lno^ ONONONONOOONOOOO CDCDCDCDCDCDCDCD CO ON I , I ^ ON CN « * — ^ ^ ^ (N ON ON ON 00 I III dodo NO ON I , I T-H NO ON CD Tt NO ON CD ON rt m is ON 00 CD CD I0 N O 0 OO NO NO NO IS NONOtSNOOOlSNOOO ONONONONOOONOOOO CDCDCDCDCDCDCDCD Tf r H - . I I I I I 00 00 O IS CO <* - *§ CO < CO O O CN o o u o o U « 2 P H u PQ M o Z ^ .0 2 oo 0 0 CO £ £ £ 03 C3 c3 (3 « c3 c3 z 3 2 2 2222Z 2 2 ^ 2 2 2 2££ O oo ^§§3 NN 38 Principles of Corrosion Engineering and Corrosion Control Plastic insulated flexible cable Cable seal watertight if electrode Is to be fully immersed to retain "" bung Sam cap Joint between copper rod md cable sealed in to mbber bung Copper rod Plastic case CyS04 solution Excess CuSO* crystals Plug of ash (wood) or porous unglazed porcelain (shaped according to method of making contact) Figure 2.14 Reference copper-copper sulfate electrode The following is the conversion table for important reference electrodes (Table 2.4): Conversion of Cu/CuSC>4 electrode potential into Ag/AgCl and hydrogen electrode potentials is expressed by the following equations: (a) emf (vs Cu/CuS0 4 ) = emf (vs Ag/AgCl) 50 mV (b) emf (vs Cu/CuS0 4 ) = emf (vs H 2 /H + ) 316 mV The conversion table for reference electrodes is shown in Table 2.4. Illustrative Problem 2.14 Convert -0.900 V (on SCE, Sat.) to the SHE scale. Solution: - 0.900V (on SCE, Sat.) = -0.900+0.242V (on SHE) = -0.658 V (on SHE) Illustrative Problem 2.15 Convert -0.916 V (on SCE. 1.0 N) to the SHE scale. Table 2.4 Conversion table for important reference electrodes Potential vs Hydrogen (mV) -400 -425 -450 -470 -500 -525 -550 -575 -600 -625 -650 Copper-copper sulfate (mV) -84 -109 -134 -159 -184 -209 -234 -259 -284 -309 -334 Silver chloride (mV) -170 -203 -228 -253 -278 -303 -320 -353 -370 -403 -428 Basic concepts in corrosion Solution: - 0.916V (on SCE, 1.0 N) = - 0.916+0.280 V (on SHE) = -0.636 V (on SHE) Illustrative Problem 2.16 What is the potential on SHE scale, for an electrode which is at a potential of —0.920 V relative to a Ag/AgCl reference in 0.01 N KC1 at 25° C. Solution: Cell reaction: AgCl+e—>-Ag+Cl- from equation (2.63) Cell potential: £ Ag+/Ag = 0.22+0.059 log or c 39 Solution: For reaction (a): 0.0592 r i £o2/OH- = £02/OH- + " " J " Ll0gP02 - loga£ H _] 0.0592, £o2/OH~ = 0.401 + - ^ — l ogP 02 - 0.0592 log OQHReplacing log OOH- = pH—14 in the above equation: £ O 2 /OH- = 0.401-0.0592[pH-14] +0.0592logPO2 £Ag+/Ag = - 0.222-0.059log [ a c r ] a c r = [ y ± ] [ c r ] = (0.901)(0.01) where y± y± Cl~ Cl~ acr £ 02/OH" = 1.23- 0.0592 pH + 0.059, log P Q 2 For reaction (b): £ O 2 /H 2 O = £ S 2 /H 2 O + ^ — [logPo 2 +loga^ + ] = activity coefficient for chloride ions = 0.901 = concentration of chloride ions = 0.01N = 9 x 1 0" 3 0.0592, £b2/H2o = 1.23+0.0592logaH+ + — — logPo2 0.0592, £b2/H2o = 1 .23- 0.0592 pH + — — log PQ2 Ag+/Ag = - 0.222-0.059 log [9 x 1 0 - 3 ] £Ag+/Ag = 0.343 V with respect to SHE. Therefore, -0.920 V (on Ag/AgCl) = ( -0.920+0.343) V 2 .13.S E M F A N D S ERIES GALVANIC Table 2.1 gives the standard electrode potentials of metals with reference to standard hydrogen = -0.577 V electrode (SHE) which is arbitrarily defined as (with respect to SHE) zero. Potentials between metals are determined by taking the absolute difference between their standard potentials. The determination of standard Illustrative Problem 2.17 electrode potential is shown Fig. 2.15. The following reduction reactions are given: As shown above, the electrode potential of two (a) 0 2 + 2 H 2 0 + 4 e = 4 0 H different metals in an electrode can be compared. (b) 0 2 + 4 H + + 4 e = 2 H 2 0 Each metal in contact with an electrolyte of its ion forms a half cell. The most practical method Show that the single electrode potential for of obtaining reliable and consistent value of relaeach reaction at 25°C has the same potential tive electrode potential is to compare the value of dependence. each half cell with a common reference electrode. 40 Principles of Corrosion Engineering and Corrosion Control (£°e = - 0.443 V) and zinc (£° n = - 0.763 V); zinc is more active than iron. Between copper ( £° u = +0.334 V) and silver (+0.80 V), Ag is more noble than copper. If the electrode reactions occurring on different metals at room temperature are arranged in accordance with the value of standard potentials, an electrochemical series at 25°C is obtained (Table 2.1). 2 .13.6 S U M M A R Y O F C HARACTERISTICS O F S E R I E S ( TABLE 2 .1) EMF Figure 2.15 Method of determination of standard electrode potential This common electrode is the standard hydrogen electrode which consists of a platinum wire platinized by electrolysis surrounded by a solution having a H + ion activity of (aH+ = 1) of one and enveloped in a hydrogen gas at 1 atmosphere pressure. The potential of the hydrogen electrode is given by: Pt(s)|H 2 (latm)|H + (a H + = l),£ = 0V The electrode potential of all metals is compared with the standard hydrogen electrode and it is called the standard electrode potential (E°). Between two metals, such as zinc and aluminum, aluminum is more active than zinc [£°! = - 1.66 V, £° n = - 0.763 V]. A metal with a more negative potential has a higher tendency to corrode (dissolve) than a metal with a less negative potential, although kinetic factors may intervene. If the potential of a metal is less than hydrogen potential, reduction rather than oxidation takes place (electrons are gained), M + + e —• M. > Metals which correspond to relatively lower standard potentials E° are called active metals and metals which corresponds to relatively higher standard potential or less negative potentials are called noble metals. Cu, Ag, Au are examples of noble metals, whereas K, Li and Mg are between iron (1) Metals with large positive potentials are called 'noble' metals because they do not dissolve easily. Examples are copper, silver, gold, etc. The potential of a noble metal is preceded by a positive (+) sign. (2) The electrode potentials are thermodynamic quantities and have little relevance to potential of metals in solution encountered in service. (3) The emf series lists only the electrode potentials of metals and not alloys. Alloys are not considered in the emf series. (4) The emf series is based on pure metals. The more active metals, such as Na, Mg, Al, Zn are called 'active metals' (5) Alloys are not considered in the emf series. (6) From the reversible electrode potential in the standard emf series of metals, it is possible to predict whether a particular metal will spontaneously dissolve. (7) It gives an indication of how active the metal is but does not necessarily predict corrosion accurately for reason to be explained later in the next chapter. 2 .13.7 A P P L I C A T I O N O F E M F SERIES Following are useful applications of emf series: (1) A less electropositive metal would displace a more electropositive metal from one of its salts in aqueous solution. For instance, if a rod of zinc is placed in a solution of copper Basic concepts in corrosion sulfate, zinc would dissolve in the solution and copper would be discharged: Zn+CUSO4 —> ZnSC>4+Cu Zn -* Z n + + +2e (oxidation) C u + + + 2 e -> Cu (reduction) 41 hydrogen which may enter the metal and cause its embrittlement. The emf series is also useful for the electrolytic refining of the metals. or (2) (3) (4) (5) Consequently, copper will deposit on the zinc rod. Electrode potentials indicate also the tendency of cations in aqueous solutions to be reduced at a cathode. For example, silver ions are reduced more readily than cupric ions, because silver is more electropositive. Metal ions above hydrogen are more readily reduced than the hydrogen ions with 100% efficiency. The metals in the series with high positive potentials are recognized as metals with good corrosion resistance. They show a little tendency to pass from a metallic state to an ionic state. Conversely, metals with high negative potentials show a tendency to corrode, but whether they corrode or not would depend on other factors also. For instance, iron has a potential of —0.440 V and indicates tendency to corrode, but if it develops a film of oxide it would not remain active, and hence, it would not corrode. This effect is used in formulating stainless steels, which are covered by an invisible oxide film. The metal with a more negative potential is generally the anode, and the metal with a less negative potential, the cathode. If zinc and aluminum are coupled, aluminum would become the anode (E^ = — 1.67V) and zinc (££n = -0.763V) the cathode. It may be expected that in the presence of metals which are more negative than hydrogen in the emf series, hydrogen reduction would be the preferred process. That is, however, not always the case. Metals as negative as zinc (E^n = —0.763 V) can be plated from an acid solution without liberation of hydrogen. In the case of aluminum, however, hydrogen evolution would be the preferred process, and, therefore, aluminum is deposited by electrolysis of a non-aqueous melt in order not to give any chance for the liberation of 2 .13.8 L I M I T A T I O N S SERIES OF EMF The following are the obvious limitations of the EMF series: (1) The emf series lists only pure metals which have only a limited use in engineering applications. Alloys are of major interest to engineers rather than pure metals. (2) The electrode potential has little relevance to potentials of metals in solutions, in which the potential of interest is the corrosion potential and not the electrode potential of the metal. (3) The position of metals in the emf series is determined by the equilibrium potential of the metal with the concentration of ions at unit activity. Prediction about galvanic coupling can only be made when the two metals forming the galvanic couple have their ionic activities at unity. If the above conditions are not met, accurate prediction of galvanic conversion by emf series would not be possible. The activity of ions in equilibrium with a given metal vary with environment and, therefore, accurate predictions of galvanic corrosion are not generally possible. (4) The emf series predicts the tendency to corrode but it cannot predict whether corrosion would actually take place. For instance, on the basis of some negative potential, iron shows tendency to corrode, however, if it develops a passive film in some environment it would not corrode. (5) The emf series cannot always predict the effect of environment. For instance, in food cans, the polarity of tin may be critical. The potential of iron (££e = —0.44) is more negative than the potential of tin (££n = - 0.1369). However, in the presence of certain type of foods in food cans, tin can become active to iron. Such a change cannot be predicted by emf series. 42 Principles of Corrosion Engineering and Corrosion Control The effect of film formation on the tendency of the metal to passivate solution cannot be predicted by emf series. For instance, titanium and aluminum are more negative than iron. However, in certain environments they form a film which makes their potential less active than iron. The effect of film formation on the tendency of the metals to corrode is kinetic and cannot be predicted by the thermodynamic emf series. -600 i soo UJ X i L. t Vo 1 a a 1 -1000 I - JL._ JL- I l l 2 2 .13.9 IR DROP (OHMIC DROP) 4 6 Distance (m) The IR drop is the ohmic voltage that results from the electric current flow in ionic electrolytes, such as dilute acids, salt water, etc. When the reference electrode for measurement of potential is placed in the electrolyte, an electrolytic resistance exists along the line between the test and the reference electrode. Because the current flows through an electrolyte resistance an ohmic voltage drop is also automatically included in the measurement of potential. h = litest -Preference Figure 2.16 Extrapolation method possible in the electrolyte. The electrode is moved at definite intervals of distance and potential (£) is measured. Emf is plotted as a function of distance and the curve is extrapolated to zero distance: Fig. 2.16. This method is applicable in electrolytes of high resistance, such as soils. b. Current Interruption Method When there is no current flow there cannot be any IR drop. However, when a current is flowing IR drop is included in the measurement. At a certain time, t> the current is interrupted so that 1=0, hence IR is also zero. The test electrode, therefore, shows a potential free from IR drop. Hence current interruption provides a good method for measurement of IR free cell potential. Unfortunately, depolarization may also occur when the current is interrupted. If the system is depolarized it is too negative at the anode area and too positive at the cathode. Various types of commercial electronic interrupters are available. The current can be interrupted by as much as 10 milliseconds of each half second. The potential must be read quickly, at the instant of interruption, by a fast response voltmeter such as an oscilloscope. IR drop (ohmic drop) is an unwanted quality which must be eliminated to obtain an accurate potential measurement. By Ohm's Law, E = IR. The IR drop in a flash light battery, for instance, contribute to the internal resistance. For any measurement, the reference electrode is placed closed to the working electrode, if this is not done the measurement would be inaccurate as it would include an IR drop. This is because the potential which is measured will always include the potential difference across the electrolyte which is between the reference and working electrode. As E is measured as IRy the potential is called IR drop. The IR contribution must be subtracted to obtain the current emf of the cell. Methods to Remove IR Drop a. Extrapolation Method 2.13.10 GALVANIC SERIES The emf is measured with the reference elec- Having examined the limitations of the emf series, trode placed as close to the working electrode as the question arises, is there a better way to predict Basic concepts in corrosion c orrosion of metals and alloys than predicted by emf series. Fortunately, the limitations imposed by emf series are overcome by another series called 'galvanic series' w hich will be now discussed (Table 2.5). 43 Table 2.5 G alvanic series in seawater at room temperature 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1 1. 12. 13. 14. 15. 16. 17. 18. 19. 2 0. 2 1. 2 2. 2 3. 2 4. 2 5. 2 6. 2 7. 2 8. 2 9. 30. 3 1. 32. 3 3. 34. 3 5. 36. 37. M agnesium M g alloy AZ-3 IB M galloyHK-31A M g alloy (hot-dip, die cast or plated) Beryllium (hot pressed) Al 7072 clad on 7075 A 12014-T3 A11160-H14 Al 7079-T6 C admium (plated) U ranium Al 218 (die cast) Al 5052-0 Al 5052-H12 Al 5456-0, H353 Al 5052-H32 Al 1100-0 Al 3003-H25 A 16061-T6 Al A360 (die cast) A 17075-T6 Al 6061-0 I ndium Al 2014-0 Al 2024-T4 A 15052-H16 T in (plated) Stainless steel 430 (active) Lead Steel 1010 I ron (cast) Stainless steel 410 (active) C opper (plated, cast or wrou ght) N ickel (plated) C hromium (plated) T antalum A M350 (active) 3 8. 3 9. 4 0. 4 1. 4 2. 4 3. 4 4. 4 5. 4 6. 4 7. 4 8. 4 9. 50. 5 1. 5 2. 5 3. 54. 5 5. 5 6. 5 7. 5 8. 59. 6 0. 6 1. 6 2. 6 3. 6 4. 6 5. 6 6. 6 7. 6 8. 6 9. 70. 7 1. 7 2. 7 3. 74. 7 5. 16. 11. 7 8. 7 9. 80. 8 1. 8 2. 8 3. 84. Stainless steel 310 (active) Stainless steel 301 (active) Stainless steel 304 (active) Stainless steel 430 (active) Stainless steel 410 (active) Stainless steel 17-7PH (active) Tungsten Niobium (columbium) 1% Zr Brass, yellow, 268 Uranium 8% Mo Brass, n aval, 464 Yellow brass Muntz metal 280 Brass (plated) Nickel-silver (18% Ni) Stainless steel 316L (active) Bronze 220 Copper 110 Red brass Stainless steel 347 (active) Molybdenum, commercial pure Copper-nickel 715 Admiralty brass Stainless steel 202 (active) Bronze, phosphor 534 (B-l) Monel 400 Stainless steel 201 (active) Carpenter 20 (active) Stainless steel 321 (active) Stainless steel 316 (active) Stainless steel 309 (active) Stainless steel 17-7PH (passive) Silicone bronze 655 Stainless steel 304 (passive) Stainless steel 301 (passive) Stainless steel 321 (passive) Stainless steel 201 (passive) Stainless steel 286 (passive) Stainless steel 316L (passive) AM355 (active) Stainless steel 202 (passive) Carpenter 20 (passive) AM355 (passive) A286 (passive) Titanium 5A1, 2.5 Sn Titanium 13 V, 11 Cr, 3 Al (annealed) Titanium 6 Al, 4 V (solution treated and aged) (Contd) 44 Principles of Corrosion Engineering and Corrosion Control (Contd) are expressed with reference to this electrode (Fig. 2.17). Table 2.5 85. 86. 87. 88. 89. 90. 91. 92. (1) In the galvanic series, instead of potentials the relative positions of the metals and alloys are indicated. (2) The series is based on practical measurement of corrosion potential at equilibrium. The potential of a corroding metal in a given medium can be obtained by connecting the metal or alloy to the negative terminal of a voltmeter and the positive terminal to a The galvanic series is an arrangement of reference electrode, generally, the calomel metals and alloys in order of their corrosion electrode. potentials in the environment. The potentials of the metals and alloys are measured in the (3) The galvanic series indicates that alloys can desired environments, with the most noble (posbe coupled without being corroded. For itive) at the top and the most active at the instance, alloys close to each other in the bottom. Table 2.5 shows a galvanic series of some series can be safely coupled. As shown in commercial metals and alloys in seawater. The the table, monel can be coupled to copper, or bronze, without any risk of galvanic corpotentials are measured in seawater by means of rosion. However, brass cannot be coupled a saturated calomel electrode and all potentials Titanium 6 Al, 4 V (annealed) Titanium 8 Mn Titanium 13 V, 11 Cr, 3 Al (solution heat treated and aged) Titanium 75 A AM350 (passive) Silver Gold Graphite 2 .13.11 CHARACTERISTICS OF G ALVANIC S E R I E S ( TABLE 2 .S) Potentiometer or high-resistance voltmeter Insulated lead Salt bridge <y r O0 C T3 0 r II ¥Z I Electrolyte I \J I Electrolyte I ee Figure 2.17 Laboratory technique for measuring the single electrode (corrosion) potential £Corr of metals and alloys in aqueous environments. Use of salt bridge Basic concepts in corrosion with tin, because the two are far away in the table and coupling them may cause serious galvanic corrosion. (4) Some alloys exist in two places in the table. For instance, steel (18/8) exists in the passive state as well as in the active state. Steel in the passive state means steel with a film of oxide, which shifts the potential from active to passive. Joining of the two types (active and passive) may lead to serious corrosion. An example of stainless steel in the active state is if it is being continuously scraped. (5) Metals and alloys in brackets can be conveniently joined with one another without any risk of corrosion. Although galvanic series is widely used by designers, nevertheless, it suffers also from certain limitations, such as: (1) Each environment requires a different galvanic series, for example, a galvanic series in static seawater cannot be used to predict galvanic corrosion in turbulently flowing seawater. (2) Galvanic corrosion also depends on the extent of polarization of the metals in alloys and not only on how close they are in the galvanic series. Predictions based on their position in the galvanic series may not provide enough information on galvanic corrosion. For example, in galvanic corrosion the process of cathodic polarization predominates. For instance, titanium has a tendency to polarize cathodically in seawater. Any less resistant metal attached to titanium will not undergo corrosion as would be expected because of the cathodic polarization of titanium. In engineering applications, alloys rather than pure metals are used, however, the emf series has only a limited value. This series is of extreme importance for design engineers. In spite of the limitations mentioned above, the galvanic series provides valuable information to engineers and scientists on the galvanic corrosion of metals and alloys in different environments. 45 2.14 POURBAIX DIAGRAMS (STABILITY DIAGRAMS) 2.14.1 INTRODUCTION Potential-pH diagrams are also called Pourbaix diagrams after the name of their originator, Pourbaix (1963), a Belgium electrochemist and corrosion scientist. These diagrams represent the stability of a metal as a function of potential and pH. They are analogues to phase equilibrium diagrams, where the stability of various phases is shown as a function of temperature and percentage composition of the metal. At a particular temperature and composition a stable phase can be easily determined. Similarly, at a particular combination of pH and potential, a stable phase can be determined from the Pourbaix diagram. In such diagrams, the redox potential of the corroding system is plotted on a vertical axis and the pH on a horizontal axis. These diagrams are constructed from calculations based on Nernst equations and solubility data for metal and its species, such as Fe, Fe2C>3, Fe(OH)2, Fe304, etc. in equilibrium. A typical diagram for Fe-F^O system is shown in Fig. 2.18. 2 .14.2 CHARACTERISTICS OF A DIAGRAM POURBAIX (1) pH is plotted on the horizontal axis and redox potential E vs SHE on the horizontal axis. (2) The horizontal lines represent electron transfer reactions. They are pH -independent, but potential-dependent. These lines separate the regions of stability, e.g. Fe and Fe 2+ in a potential-pH diagram for Fe-F^O system. Variation of concentration of F e 2+ (10~6 - 1) leads to several parallel lines. (3) The vertical lines are potential-independent but pH-dependent and not accompanied by any electron transfer, e.g. lines corresponding to the following reactions: Fe2++2H2O^Fe(OH)2+2H+ Fe3 + 3 H 2 0 ^ Fe(OH)3 + 3H+ 46 Principles of Corrosion Engineering and Corrosion Control 1.6 Fe042* CORROSION 12 m 04 0 Redox Potential (V) -04 «&8 IMMUNITY 12 ~1S 4 6 pH Fe(OH)2 HFeO, CORROSION i a i n i 12 14 Figure 2.18 Potential-pH diagram for iron (4) The sloping, straight lines give the redox potentials of a solution in equilibrium with hydrogen and oxygen, respectively. These equilibria indicate electron transfer as well as pH, e.g. Fe203+6H-+2e^2Fe2++3H20 The above reaction indicates both electron transfer and pH change. (5) The concentration of all metal ions is assumed to be 10 - 6 mol per liter of solution. At lower concentration, corrosion should not occur. (6) The diagram is computed for the equilibrium conditions at 25°C. (7) The upper end of the redox potential axis is the noble end and the lower end, the active end, meaning that the oxidizing power increase with increasing potential. Basic concepts in corrosion (8) The hydrogen and oxygen lines are indicated in Pourbaix diagrams by dotted line. It can be noted that Pourbaix diagrams may but be constructed for all elements. The diagrams subdivide the potential-pH plots into regions of immunity, corrosion or passivation. These are very useful in prediction of tendency of metals to corrode. These diagrams, however, has several limitations which will be summarized at the end. £ Fe2+/Fe = -0.44+0.0301og [aFe2+] because a(Fe) = 1 47 a=10~ 6 mole/liter £Fe2+/Fe = - 0.44+0.59[-6] £ = 0.62 2.14.3 C A L C U L A T I O N S I N V O L V E D IN C O N S T R U C T I O N OF POURBAIX DIAGRAMS This value is marked on the potential axis for line 2. Line 3 The equilibrium reaction is Fe+2H 2 0 «=• Fe(OH)2 + 2H+ Numerous examples of applying Nernst equation to determine the potential and pH are given in this chapter. These equations are essential tools to calculate the redox potentials in the Pourbaix diagrams. For instance, consider Fig. 2.18, line 1 and let us determine the potential corresponding to reaction occurring in line 1. Here, Fe 3+ is in equilibrium with Fe 2+ . Applying Nernst equation, we obtain the reaction Line 1 Fe3 + e ^ F e 2 + E = KFe3+/Fe: There is no charge transfer in the above equilibrium system, or Fe2+20H"^Fe(OH)2 Here the precipitation of Fe(OH)2 is shown, and solubility product comes into play ( XSP), which can be described by KsP = [ a F e^][«OH-] 2 = 10-14-71 (Check table of solubility products) f [%e2+] .-. l og[a Fe 2+]+2log[a OH -] = "14.71 but log [a O H -] = - 2 pH+13.29 Substituting for a= 1 0 - 6 - 6 = -2pH+13.29 pH = 9.65 (one electron transfer) £ = 0.77+0.059log ?l£L\ L%e2+J where a =10 6 Take a = 0.6 moles. The log term drops out here as the term a becomes The corresponding pH value is indicated on the unity. Hence E = 0.77 V, and we indicate this pH axis in the figure. value as the potential axis. Line 4 The equilibrium reaction between Fe 2+ and Line 2 Fe203 is The equilibrium reaction is Fe 2 + +2e=^Fe E F e 2 0 3 + 6 H + + 2 e = 2Fe 2 + +3H 2 0 Note that there is some degree of potentialdependence shown by electron transfer as well = Fe +/Fe + ~^~ £ 2 l0g ^ 2+ J 48 Principles of Corrosion Engineering and Corrosion Control Other lines in the diagram: • 0.0591 [aFe2o3] E = EFe203/Fe2+ + — - —log-. ^ „ /T._2 [ a Fe 2+] 0.0592 Reduced some degree of pH-dependence, and the slope is finite as shown by the line. The expression is Hydrogen line (a) H+ + e =-H 2 +- log[aH+]b H20+e=-H2+OHor Oxygen line (b) 2H 2 0 = 0 2 + 4 H + + O H 40H-^02+2H20+4e Multiplying [ ^ ] l o g [ a H + + ] 6 by 3, putting 0Fe2o3 = l> moving aFe++ to numerator, and changing sign in the log term, we get £ = 0.73-0.059log[a Fe 2+]-0.177pH The above expression shows the relativity between the pH and potential. Hence, the slope is now 0.777 which is indicated in the figure. Line 5 It has also a finite scope and involves equilibrium between Fe 2+ ions and Fe3C>4. The equation for the reaction is Fe 3 0 4 +8H+2e = 3Fe 2 + +4H 2 0 £ or Below hydrogen line (a) hydrogen is produced by a reduction of H + and H 2 0 , and above oxygen line (b), 0 2 is produced by oxidation of H 2 0 and OH~. Between the lines a and b water is stable. Parallel lines identified by exponent of the activity of Fe 2+ ions in solution (aFe2+ = 10°, 10 - 2 , 10~4, 10~6). For instance line 8 corresponds to the formation of Fe 2 0 3 from solution of aFe3+ > aFe2+. Curves identified as 0, —2, —4, —6, corresponds to 10°, 10~2, 10" 4 , 10~6, respectively (Fig. 2.19). F°e3O 4 /Fe 2 + - =0 98 Putting £° = 0.98 for Fe 3 0 4 /Fe 2+ in the Nernst equation for the above reaction and solving E = 0.98 - 0.088 log (aVei+) - 0.236 a slope of —0.236 is obtained, which is indicated in the figure. The slope shows a pH and potential dependence. Similarly, we can calculate the potentials for other equilibrium reactions involving Fe 2 0 3 and Fe3C>4. The reaction is given by 3 Fe 2 0 3 + 2H+ +2e <± 2 F e 3 0 4 + H 2 0 £ ° = + 0.221VSHE 2 .14.4 R E G I O N S OF IMMUNITY, C O R R O S I O N A N D P A S S I V I T Y IN POURBAIX DIAGRAMS Figure 2.20 shows the regions of immunity, corrosion and passivity which are described below. 0.0551 = 0.221 - 0.059pH 9 £ = 0.221 + — — log ( OH+) 2 Immunity Region The region of immunity shown in the Pourbaix diagram for Fe-H 2 0 indicates that corrosion canSuch calculations can be carried out for differ- not occur in this region. For instance, at a point ent metal-water systems to construct Pourbaix X in the diagram as the activity of Fe 2+ would be diagrams. very low (~10~ 10 ). Basic concepts in corrosion 49 E(SHE) -0.2 Figure 2.19 Simplified Pourbaix diagram for iron/water system 50 Principles of Corrosion Engineering and Corrosion Control EN SHE CORROSION PASSIVITY IMMUNITY PH Figure 2.20 Pourbaix diagram for the iron-water system at 25° C showing nominal zones of immunity Corrosion Region As iron is transformed to soluble species, it is expected that iron would corrode. Region of Passivation An oxide species in contact with an aqueous solution along a boundary would not allow corrosion to proceed if it is impervious and highly adherAI2O3+H2O- 2A10~+2H+ ent. The thin layer of oxide on the metal surface, such as Fe3C>2 or Fe3C>4 is highly protective under In neutral solutions (4-8 pH), the hydroxide is the above condition. Metals like aluminum and insoluble which makes aluminum surface passive. steel are known to resist corrosion because of Aluminum dissolves both in acids and bases. development of oxide films in the air. (x-axis) from acidic at low pH to caustic at high pH. Line a is the hydrogen line below which water is no longer stable and decomposes into hydrogen and OH~ (alkalization). Line b is the oxygen line above which water decomposes into hydrogen, oxygen and H + (acidification). Water is stable between regions (a) and (b). In acidic conditions Al dissolves as Al +3 . In alkaline conditions aluminum dissolves as AIO2. 2 .14.S SOME EXAMPLES OF POURBAIX DIAGRAMS Pourbaix Diagram for Copper A Pourbaix diagram for copper/water system is shown in Fig. 2.22. Copper (£° = 0.337 V) is more noble than iron (£pe = —0.444 V), however, it is more stable in water (SHE) than iron. A Pourbaix diagram for aluminum and H2O Copper is not passive in acid electrolytes. The system is shown in Fig. 2.21. The pH varies oxide of copper, Cu + and Cu 2+ are only Pourbaix Diagram for Aluminum Basic concepts in corrosion 51 ^ ^ ^ ^ <b) —r— ^ ^^ * "^ ^ (2L E/V SHE ( 1 A l* A1203 AJO " Dissolution Passivation Dissolution ~2 M 12 18 pH Figure 2.21 Pourbaix diagram for the aluminum-water system at 25° C protective in weakly acidic and alkaline electrolytes. The region of immunity extends above the line (a) in the diagram which represents oxygen evolution. If the potential of copper is made more noble, it would corrode under mildly acidic and strongly acidic conditions. It would also corrode under strongly alkaline conditions in higher oxidizing potentials. The scope of the chapter does not permit a CuO broader treatment of Pourbaix diagrams. By iden"*"~*» ~~ ^^ *~ tifying the pH and potential ranges, it is possible ( Br — Cu* "* "**— . " *•* to control the corrosion of metals in aqueous E/V _HL^ ^ """"-^ environment. Aluminum is highly active but it SHE ^-^_ cino^^J can be used with a minimum risk of corrosion because of its tendency to form protective oxide (a) — - ^ T ^ *" *" films which prevent the metal from corrosion. Cxi Although titanium is highly active with a very narrow immunity zone towards the bottom of the diagram, it has an excellent tendency for passivation as shown by a highly extended passive PH zone which extends over the entire range of pH. Titanium, however, corrodes under reducing and highly oxidizing conditions as shown in the lower Figure 2.22 Pourbaix diagram for the copper-water system at 25°C left region of the pH-potential diagram. ""*"""" *"*"*"•' •—.. 52 Principles of Corrosion Engineering and Corrosion Control OF POURBAIX 2 .14.6 B E N E F I T S DIAGRAMS Although the above disadvantages appear to be substantial, the advantages offered by the Pourbaix diagrams far outweigh their limitations. Pourbaix diagrams offer a large volume of thermodynamic information in a very efficient and compact format. The information in the diagrams can be beneficially used to control corrosion of Q U E S T I O N S pure metals in the aqueous environment. By altering the pH and potential to the regions A. M U L T I P L E C H O I C E of immunity and passivation, corrosion can be controlled. For example, on increasing the pH Q U E S T I O N S of environment in moving to slightly alkaline Select one best answer: regions, the corrosion of iron can be controlled. This can be achieved by water treatment. Similarly, changing the potential of iron to more 1. The electrode potential of a metal is: negative values eliminate corrosion, this technique is called cathodic protection. Also, raising [ ] The potential which exists at the interthe potentials to more positive values reduces the face between the metal and the eleccorrosion by formation of stable films of oxides trolyte on the surface of transition metals. Steel in rein[ ] The potential between the anodic and forced concrete does not corrode if an alkaline cathodic areas of the metal environment is maintained. On the contrary, an [ ] The potential between two metals alkaline environment for aluminum is a disaster immersed completely in the same elecif the pH exceeds 8.0. The above example clearly trolyte demonstrate the merits of Pourbaix diagrams in [ ] The potential of a metal with respect prediction and control of corrosion. However, to another metal, which is at a higher there are several limitations of these diagrams, concentration which are summarized below: 1. These diagrams are purely based on thermodynamic data and do not provide any information on the reactions. The thermodynamic stability may not be achieved to the kinetics of the reaction. No information is provided in the rates of reaction. 2. Consideration is given only to equilibrium conditions in specified environment and factors, such as temperature and velocity are not considered which may seriously affect the corrosion rate. 3. The activity of species is arbitrarily selected as 10~6 g mol which is not realistic. 4. Pourbaix diagrams deal with pure metals which are not of much interest to the engineers. 5. All insoluble products are assumed to be protective which is not true, as porosity, thickness, and adherence to substrate are important factors, which control the protective ability of insoluble corrosion products. 2. The double layer is formed as a result of [ ] attractive forces between negative charged metal surface and positive ions only [ ] repulsive forces between like positive ions only [ ] both attractive and repulsive forces between ions [ ] None of the above 3. The most acceptable method of obtaining standard electrode potential is by [ ] comparing the electrode potentials of a metal half cell with a hydrogen half cell [ ] comparing the electrode potential of a metal immersed in a solution of its ions at any concentration, with the hydrogen half cell Basic concepts in corrosion [ ] comparing the electrode potential of a metal with any standard electrode, such as Ag-AgCl or Calomel electrode [ ] None of the above 4. A galvanic cell is formed [ ] does not proceed [ ] None of the above 53 9. The most common electrodes used for measurement of corrosion potentials are: (Mark two correct answers) I [ [ [ ] ] ] ] Ag-AgCl Hg-Hg2Cl2 C u-CuS0 4 Hydrogen electrode [ ] when two metals are immersed in solutions differing in concentration [ ] when two different metals are immersed in one electrolyte [ ] when two different metals are exposed to air 10. [ ] when two metals are brought close together and electrically insulated from one another 5. In concentration cells [ ] the metal is in contact with two half cells having the same electrolyte but at different concentrations [ ] the metal is in contact with two half cells, having the same electrolytes with the same concentration [ ] no liquid junction is present [ ] there is no migration of ions from one electrolyte to another electrolyte 6. If the free energy of a reversible process is negative, it implies that [ ] the cell reaction is spontaneous [ ] the cell reaction is not spontaneous [ ] the cell reaction proceeds from right to left [ ] no reaction takes place at all 7. The value of (2303RT)/F at 25°C varies with [ [ [ [ ] ] ] ] temperature the metal being considered the melting temperature of the metal None of the above A galvanic series is [ ] a list of alloys arranged according to their corrosion potentials in a given environment [ ] a list of metals and alloys according to their corrosion potentials in a given environment [ ] a list of standard electrode potentials of alloys or metals arranged in order of their values [ ] a grouping of metals and alloys based on their ability to get oxidized in a stated environment B. R EVIEW Q U E S T I O N S 1. Distinguish between: a) Metallic conduction and electrolytic conduction. b) Standard electrode potential and corrosion potential. c) Anode and cathode. d) Electronic conduction and ionic conduction. 2. In the sign convention adopted by IUPAC: a) What does the right-hand electrode indicate? b) What does the left-hand electrode indicate? c) How the cell potential, £ceu, is obtained? d) What does the positive sign of the cell indicate? 8. If E > 0, the cell reaction [ ] proceeds in forward direction [ ] proceeds in reverse direction 54 Principles of Corrosion Engineering and Corrosion Control are 1 and 0.01, respectively. The activity coefficient for Q1SO4 is 0.047 and for Z nS0 4 is 0.70. 5. Calculate E for the half cell in which the reaction Cu + + (0.1 m) + 2e~ = Cu(s) takes place at 25°C. 6. Calculate the potential for each half cell and the total emf of the cell (£ceii) at 25°C: Pb |Pb 2+ (0.0010 M)/Pt,Cl 2 (l arm)/ Cr(0.10M) £°Pb = Pb 2+ /Pb° = - 0 . 1 3 V E° - (Cl 2 - CI) = 1.358 V 7. Calculate the emf and the free energy of the cell given below: Fe 2 + -FeandNi 2 + -Ni 3. If E > 0, in which direction will the cell reaction proceed, and conversely if E < 0, in which direction the reaction would proceed? 4. State which of the following statements are true: a) When two metals, e.g. Zn and Cd, are connected and placed in a solution containing both metal ions, the metal with the lower standard potential would corrode. b) Conversely, the metal with the higher potential would be deposited. c) The cell and cell reaction are written in opposite orders, for instance, for the cell Fe/Fe 2+ (ag)/Cu 2+ (ag)/Cu, the reaction is F e + C u->Cu +Fe d) The cell potential is obtained by subtracting the electrode potential of the right-hand electrode from the left-hand electrode. 2+ 2+ (Obtain data from the literature.) 8. A piece of copper is immersed in an aqueous solution of KC1 with a concentration of 1 kmol/m3. The solubility product of CuCl at 25°C is (1.7 x 10~6). Calculate the potential 5. State the limitations of the emf series and of the copper electrode. the advantages of galvanic series for an 9. A test piece of cadmium is placed (a) in flowengineer. ing seawater, and (b) in stagnant seawater. Predict under which conditions cadmium would corrode. The following information C. P R O B L E M S is provided: pH = 7.0 1. Devise electrochemical cells in which the Concentration of cadmium ions: following overall reactions can occur: Cd(OH)2 = 1.2 x 1 0" 14 kmol/m3 at25°C a) Zn(s)+Cu 2 + (ag)^Cu(s)+Zn 2 + (ag) in stagnant condition. Concentration of cadmium ions in flowing b) Ce + 4 (a^)+Fe 2 + (a^)->Ce 3 + (a^) conditions: +¥e3+(aq) = 10~6 kmol/m3 c) Ag+(flq)+Cr(^)->AgCl(s) E°cd = - 0.42 V d) Zn(s)+2Cl 2 (g)->ZnCl 2 (a^) 10. Aluminum samples are exposed in Arabian Gulf water. It has been found that aluminum 2. What is the mole fraction of NaCl in a solucorrodes either as Al(OH)3 or AICI3 in seation containing 1.00 mole of solute in 1.00 kg water. Show by calculations in which form ofH 2 0? aluminum corrodes or does it corrode in 3. What is the molarity of a solution in which both forms. The concentration of A l + + + 1.00 x 102 g of NaOH is dissolved in 0.250 kg in AICI3 = 10 - 4 . The solubility product of Al(OH)2 is 3.7 x 10" 15 . ofH 2 0? 4. What is the voltage (Ece\\) of a cell com- 11. A piece of zinc measuring 2" x 1.5" is placed in a 0.002 molar solution of ZnCl2. Show prising a zinc half cell (zinc in ZnS04) and whether zinc would corrode in the given a copper half cell (Cu in CUSO4)? The medium. metal concentrations of ZnSC>4 and CUSO4 Basic concepts in corrosion 55 12. A p iece of nickel measuring 6" x 4" is immersed in deaerated water having a pH of 8.0. The solubility product Ni(OH) 2 is 1.6 x 1 0" 1 6 . D etermine the potential of nickel in the given conditions and state whether nickel would corrode or not. 13. C alculate the emf of the following cell at 2 5°C: F e 3 + (a = 0.01), Fe 2 + (a = 0.0001), C u + (a = 0.01)Cu. Given: £ ° , F e 3 + / F e 2 + = - 0 . 7 7 1 V and E°,Cu+/Cu = 0.5211. 14. F ind the potential of a cell where the reaction N i + S n 2 + - > N i 2 + + S n , proceeds. The concentration of N i 2 + is 1.3 and the concentration of Sn 2 + is 1.0 x 10~ 4 . Predict if the reaction would proceed from right to left or left to right. 15. W rite a balanced cell reaction and calculate the emf of the following cell: Pt/Sn 2 + (a = 0.10), Sn 4 + (a = 0.10), Fe 3 + /Fe (a = 0.200). 16. T he following is the reaction when iron corrodes in an acid: Fe+2HCl(a<2) + - 0 2 ^ F e C l 2 ( a < ? ) + H 2 0 a) In what direction shall the reaction proceed if the activity of Fe 2 + is 1, and a(H+) = 1? b) If the temperature is maintained at 25° C, what activity of iron (aFe2+) would be required to stop the corrosion in acid? 17. C alculate the voltage of the following cell: Z n + C d 2 + (a cd 2+ =0.2) = Z n + + (aZn2+ = 0.0004)+Cd. 18. T he potential of an electrode is measured as —0.840 V relative to a 1N Calomel electrode. What is the electrode potential on a standard hydrogen scale? 19. W hat pressure would be required to stop corrosion of zinc in deaerated water at 25°C? The major corrosion produced is Zn(OH) 2 a nd the solubility product is 1.8 x 10~ 1 4 a t 20°C(TakepH = 7). 2 0. C alculate the theoretical tendency of cadmium to corrode when it is immersed in a solution of 0.001 M C dCl 2 ( pH = 2.0). 2 1. F or the cell H 2 (1 atm)-HCl-HgCl-Hg, E° = 0.2220 V at 298 K. If pH = 1.47, determine the emf of the cell. 2 2. C opper corrodes in an acid solution of pH = 2. Hydrogen is bubbled continuously in the solution. Calculate the maxim u m concentration of C u 2 + i ons that would result. 2 3. A cell is composed of a pure copper and pure lead electrode immersed in solutions of their bivalent ions. For a 0.3 molar concentration of C u 2 + , the lead electrode is oxidized and shows a potential of 0.507 V. What would be the concentration of Pb ion at 25°C? S U G G E S T E D BOOKS FOR READING [1] [2] [3] [4] [5] Davis, J.P. ed. (2000). Corrosion: Understanding the Basics. 2nd ed. Ohio: Metals Park, USA. Revier, R.W. ed. (2000). Uhlig's Corrosion Handbook. New York: John Wiley. ASM Handbook (1987). Corrosion. Vol. 13, ASM, Ohio: Metals Park, USA. Stansbury, E.E. and Buchanan, R.A. (2000). Fundamentals of Electrochemical Corrosion. Ohio: ASM, USA. Piron, D.L. (1991). The Electrochemistry of Corrosion, Texas: NACE, USA. KEYWORDS Anode The region of the electrical cell where positive current flows into the electrolyte. Anode is the site where oxidation occurs. In a corrosion cell anode is the region which is dissolving. Cathode is the region of an electrical cell where positive electric current enters from the electrolyte. In a corrosion cell reduction reaction takes place at the cathode. Electrochemical cell An electrochemical system comprising of an anode and a cathode in a metallic contact and immersed in an electrolyte. Electrode potential It is the potential of an electrode in an electrolyte as measured against a reference electrode. Electrolyte It is electrically conductive. It is usually a liquid containing ions that migrate in an electric field. 56 Principles of Corrosion Engineering and Corrosion Control Reference electrode A stable electrode with a known and highly reproducible potential. Reversible potentiel(Equilibrium potential) It is the potential of an electrode where the forward rate of reaction equals to the reverse rate of reaction. Standard electrode potential (E°) It is reversible potential of an electrode measured against a standard hydrogen electrode (SHE) consisting of a platinum specimen immersed in a unit activity acid solution through which H2 gas at 1 atm pressure is bubbled. The potential of the hydrogen electrode (half cell) is taken to be zero. Free energy From the total heat content of a system (enthalpy), only a part can be converted to useful work. This part of total enthalpy is called free energy, G. Absolute value of free energy (G) cannot be measured, and only changes in free energy (AG) are measured. Galvanic series A table of metals and alloys arranged according to their relative corrosion potential in a given environment. Half cell An electrode immersed in a suitable electrolyte for measurement of potential. IR drop Voltage drop caused by current flow in a resistor. CORROSION KINETICS FARADAY'S LAWS OF ELECTROLYSIS AND I TS A P P L I C A T I O N IN DETERMINING THE CORROSION RATE The masses of different primary products formed by equal amounts of electricity are proportional to the ratio of molar mass to the number of electrons involved with a particular reaction: Mi (3.2) 3 .1.2 THE SECOND LAW T 3.1 3 .1.1 he classical electrochemical work conducted by Michael Faraday in the nineteenth century produced two laws published in 1833 and 1834 named after him. The two laws can be summarized below. where m\,mi M i, M2 n\, H 2 Z\, Z2 LAW m\ c< — a Z\ n\ mi oc — oc Z2 n2 (3.3) THE LAWS THE FIRST = = = = masses of primary product in grams molar masses (g.mol -1 ) number of electrons electrochemical equivalent. The mass of primary products formed at an electrode by electrolysis is directly proportional to the quantity of electricity passed. Thus: m oc It or m = ZIt where J t m Z = = = = current in amperes time in seconds mass of the primary product in grams constant of proportionality (electrochemical equivalent). It is the mass of a substance liberated by 1 ampere-second of a current (1 coulomb). (3,1) Combining the first law and the second law, as in equation (3.1) m = ZIt Substituting for Z, from equation (3.2) into (3.1) M m = k—It n or 1M -.-It Fn (3.5) (3.4) where F = Faraday's constant. It is the quantity of electricity required to deposit the ratio of mass 58 Principles of Corrosion Engineering and Corrosion Control Below a re s ome examples showing h ow F araday's laws a re u sed t o d etermine t he c orrosion rate. t o t he v alency of a ny s ubstance a nd expressed i n c oulombs p er m ole ( C (g e quiv.) - 1 ). It h as a v alue of 96 485 coulombs p er g ram equivalent. This is s ometimes written as 96 485 coulombs p er m ole of electrons. 3.1.4 3.1.3 APPLICATIONS OF ILLUSTRATIVE EXAMPLES F A R A D A Y ' S L A W S IN DETERMINATION OF CORROSION R A T E S OF METALS AND ALLOYS C orrosion rate h as d imensions of m ass x r eciprocal of t ime: (g-y 1 Example 1 Steel corrodes in an aqueous solution, the corrosion current is measured as 0.1 A • c m - 2 . Calculate the rate of weight loss per unit area in units ofmdd. Solution: F orFe —>• F e 2 + 2e w Mi nF orkg-s l ) where Tt" M = 55.9g.mol" 1 i = 0.1 A-cm" 2 n =2 (3.9) I n terms of loss of w eight of a m etal with time, from equation (3.5), w e get dw MI - £- = — dt nF U = current) (3.6) Substituting the values in equation (3.9), we obtain w ~At Now converting g t o mg ( x 10 ) , we get (2.897 x H T ^ g c n r V " 1 ) = 2.897 x 1 0~°gcm~2z .s- 1 T he rate of c orrosion is p roportional t o the c urrent passed a nd to the m olar mass. Dividing equation (3.5) by the e xposed area of the m etal in t he alloy, w e get w At MI ~n~FA (3.7) = 2.897 x 10~5mg cm~ 2 s _ 1 and converting from mg c m w e get (3.8) 2 But, — = current density (i). Then: A w Mi —=— At nF s l to mg d m 2 s l, (i = current density) (2.897 x 1 0 - : > m g c m " V ' X l x 1 0 z c m z / d m z ) For converting t he a bove expression t o m illigrams p er d ecimeter square p er day (mg d m - 2 d a y - 1 ) , we m ultiply b y 24 h/day x 3600 s/h, a nd o btain t he c orrosion rate i n the d esired units: (2.897 x 1 0- 5 mgcm~ 2 s" 1 )(24)(3600) = 2.503 m g d m - 2 d a y " 1 T he above equation h as b een successfully used t o d etermine t he r ates of c orrosion. A very useful practical unit for r epresenting the corrosion rate is m illigrams p er d ecimeter square p er day ( _ 2 .day _ 1 ) o r m dd. O ther practical units a re m illimeter p er year ( m m y - 1 ) and mils p er y ear (mpy). Corrosion kinetics Example 2 Iron is corroding in seawater at a current density of 1.69 x 10 - 4 A/cm2. Determine the corrosion rate in (a) mdd (milligrams per decimeter2 day) (b) ipy (inches per year) Solution: (a) Apply Faraday's law as before mdd = 1.69 x 10""4 A/cm2 x 3600 s/h x 24h/day x 100cm2/dm2 55.85 59 Example 4 Penetration unit time can be obtained by dividing equation (3.8) by density of the alloy. The following equation can be used conveniently: Corrosion rate, r = C where p i M n C = = = = = density (g/cm3) current density (A/cm2) atomic weight (g • m ol - 1 ) number of electrons involved constant which includes F and any other conversion factor for units, for instance, C = 0.129 when corrosion rate is in mpy, 3.27 when in mm/year and 0.00327 when units are in mm3/year. Mi np (3.10) g/mol x 103 mg/g For instant, the above relationship can be used to establish the equivalent of corrosion current of = 422.8 mg d m d ay |xA/cm2 with the rate of corrosion for iron in mpy (b) Converting 422.8 mg d m - 2 d ay - 1 to inches as shown below per year (ipy) with the conversion factor r(55.8)(l)l ? [mdd x 0.00144/p], p = density 1 |xA/cm2 = 0.129 = 0.46 mpy w -2 -1 1 A • s • mole 96 495 L 2 7.86 J = 422.8 x Thus, for iron 1 |xA/cm2 of current is equivalent to a corrosion rate of 0.46 mpy for iron. = 0.077 ipy The equation can be extended to establish the or 77 mpy, because 1 mil = 1/1000 of an inch. equivalent in other units also. The above example shows the correspondence Example 3 between penetration rate and current density for A sample of zinc anode corrodes uniformly with a metal. A similar correspondence between the a current density of 4.27 x 10 - 7 A/cm2 in an penetration rate and current density for alloys aqueous solution. What is the corrosion rate of can be established. However, it would require the determination of equivalent weight (M/n) for zinc in mdd? the alloy. Solution: Following is the relationship which is used to +2 determine the equivalent weight of an alloy: Zn -> Zn + 2e CR = (4.27 x 10~7 A/cm2) (l x 102 c m 2 /dm 2 ) / 65.38 \ g/mol \ (24 x 3600 s • h/h • day) I x 0.00144 7.86 (3.10a) Equivalent weight = ] (3.10b) ["i/^ij where j\ — mean fraction of an element present in the alloy n\ = electron exchanged Mi = atomic mass. x 103mg/g I A • s • mol | 5& V 9 6 495 / = 1.25 mdd 60 Principles of Corrosion Engineering and Corrosion Control 3600s/lhxl000mg/lg 2 x 96500coulombs/mol 65.38 g/mol 2 x 96 500 coulombs/mol =» Corrosion rate of Z n = 12.29 mg/dm2/day (b) We can also use the relationship given below to determine the rate of corrosion in mm/year or other units by changing the constants. The constants for mm/year is 0.00327. Corrosion rate, r = C - — np where p is the density in g/cm3, i is the current density in (xA/cm2, and C is the constant = 0.00327 for mm/year. Corrosion rate = 0.00327 x 65.38 g/mol x 4.2 2x7.13 Example 5 Determine the corrosion rate of AISI 316 steel corresponding to 1 |xA/cm2 of current. Following is the composition of alloys: Cr = 18% Ni = 10% Mo = 3% Mn = 2% Fe = balance You mayfindthe composition of alloys from ASM Handbook. Solution: Using equation (3.10a): 1 uA/cm2 = 0 •mmm\ T 52.3 1 , 54.94 ", 54.94 I + 0.128 ["— 0.02 .(2)(7.45)_J L ( 2)(7.45) T 95.95 I + 0.128 — -0.03 L 2 10.1 J [" 55.65 " I + 0.128 L(2)(7.86)J 0, 07 mpy = 0.0629 mm/year Corrosion rate = 0.0629 mm/year = 0.514587 mpy Example 6 A sample of zinc corrodes uniformly with a current density of 4.2 x 10 - 6 A/cm2 in an aqueous solution. (a) What is the corrosion rate of zinc in mg/dm2/day? (b) What is the corrosion rate of zinc in mm/year? Solution: (a) Given current density, i = 4.2 x 10~6 A/cm2 = (4.2|xA/cm2), we know that for zinc atomic weight, M = 65.38 g/mol, density, p = 7.1 g/cm3, n = 2, F = 96 500 coulombs/mole. From the formula: 4.2 x l O - 6 A/cm2 Corrosion rate = 2 x 96 500 coulombs/mol 100cm 2 /ldm 2 x24h/lday 2x96 500 coulombs/mol The proportionality constant is 0.129 for mils per year (mpy). We can also convert 12.29 mg/dm2/day to mm/year as below: =» 12.29 mg/dm2/day 1 365 days 100 dm2 = 12.29x-x— -x 1 year 1 m2 p x 1000 mm 1kg x6 lm 10 mg 1 = 12.29 x 7130kg/nr> 365x100x1000 1000000 12.29x365 71300 = 0.0629 mm/year =>> Corrosion rate = 0.0629 mm/year Corrosion kinetics Example 7 AISI 316 steel has the following nominal composition: Cr = 18% Ni = 8% Mo = 3% Fe = 70% n= 1 n= 2 n= 1 n= 2 p = 7.1 g/cm2 At. wt. = 52.01 g/mol p = 8.9 g/cm2 At. wt. = 58.68 g/mol p = 10.2 g/cm2 At. wt. = 95.95 g/mol p = 7.86 g/cm2 At. wt. = 55.85 g/mol Solution: For a penetration rate equivalent to 1 uA/cm2 Mi Corrosion rate = C • — np Given: M= Equivalent weight x 3 = 8.99 x 3 = 26.97 Density = p = 2.71 g/cm3 n=3 i = 1 |xA/cm2 = > Corrosion rate = > 0.128 x 26.97 x 1 3 x 2.71 61 Find the equivalence between the current density of 1 |xA/cm2 and the corrosion rate (mpy). Solution: Find the equivalence between i = 1 |xA/cm2 and the corrosion rate. Mi Corrosion rate = C • — np = 0.424 mpy = 0.43 mpy (mils/year) 3.2 CORROSION KINETICS where C is the constant for conversion depending Thermodynamics gives an indication of the tenon unit. dency of electrode reactions to occur, whereas electrode kinetics addresses the rates of such reactions. The reactions of concern are mainly Mi ,„ r = 0.129 ( C = 0.129 for units corrosion reactions, hence, it is more appropriate np of mpy) to call the kinetics of such reactions as corrosion kinetics. In order to understand the theory Since, we are given the composition of the steel of aqueous corrosion, it is important to develop 316, we will find the corrosion rate of steel as a complete understanding of the kinetics of reaction proceeding on an electrode surface in contact below: with an aqueous electrolyte. Methods which are Corrosion rate: used to study the rate of a reaction involve the determination of the amount of reactants remain52.01 \ / 58.68 \ ing in products after a given time. In aqueous 0.18 + 0.08 1x7.1/ V2x8.9/ corrosion, it is very important to appreciate the nature of irreversible reactions which take place / 95.95 \ / 55.85 \ 1 on the electrode surface during corrosion. + 0.03 + 0.701 \ 1 x 10.2/ \2x7.86/ J K = 0.129 [1.318 + 0.263 + 0.282 + 2.48] = 0.129 x 4.343 = 0.55 mpy (mils/year) Example 8 Prove that for Al alloy 1100, the penetration rate of the alloy equivalent to 1 |xA/cm2 is 0.43 mpy. 3 .2.1 ANODIC AND CATHODIC REACTIONS Consider a piece of iron corroding in an acid solution. The following are the basic reactions: = Fe < ± F e 2+ + 2e H ? ^ 2 H + + 2e (anodic) (cathodic) (3.11) (3.12) 62 Principles of Corrosion Engineering and Corrosion Control Such a driving force is provided by the free energy of the reaction (AG*). 3 .2.2 ENERGY-DISTANCE PROFILES These are plots of free energy (AG*) vs the distance representing the progress of reaction through the double layer. The metal atoms are located in energy wells associated with lattice structure. A metal atom cannot be detached from the lattice and pass into the solution until it crosses an energy barrier, called activation energy. Figure 3.2 shows an energy-distance (from metal surface to the outer Helmholtz plane of double layer) profile. The solid circle in the figure represents the metal atom in the lattice and the circle with (+) charge, the cation in aqueous solution. The left-hand side of Fig. 3.2 is the lattice side and the right-hand side is the solution side. The hump in Fig. 3.2 represents the activation energy barrier, AG*. The location of the hump is described by a symmetry factor /3, and for all practical purposes it is taken as (P = ^). The thermal energy of the ions in a metal crystal makes them vibrate with a frequency, generally of the order of 1012 per second. Any ion with sufficient thermal energy to reach to the top of the hump would vibrate with a characteristic frequency given by (/). An explanation of the Figure 3.1 Anodic and cathodic processes in a corrosion cell The overall reaction which takes place in the iron surface is Fe + 2H + -> Fe 2+ + H2 (3.13) The above process is illustrated in Fig. 3.1. The anodic reaction shown above involves the transfer of a metal atom from a metal lattice to the aqueous solution at the electrode/electrolyte interface. The reactions (3.11) and (3.12) shown above are called charge transfer reactions as they proceed by transfer of charge (electrons). The metal reaction like M T± M z + + ze is reversible as the rate of forward reaction is equal to the rate of reversible reaction. An anodic reaction like M ^ M + z + ze, is, however, not a reversible reaction as the rate of forward reaction (if) is greater than the rate of reduction (ir) and a net current if flows, M T± M z + + ze, if = i{ - \ ir |. The anodic reaction shown above, involves transfer of a metal atom from a metal lattice M to a metal cation having a positive charge M z + . The water dipoles are attracted to the positive ions in solution and form a hydration or solvation sheath around each cation. The cations in solution are, therefore, hydrated or solvated. The charge transfer reactions described above cannot proceed until a driving force is available. Figure 3.2 Energy-distance profile Corrosion kinetics reversible and irreversible processes illustrated by energy profiles is described below. 63 3 .2.3 REVERSIBLE REACTIONS Consider an electrode at equilibrium: M +± M + z + ze The free energy of activation for the anodic (forward) and cathodic (reverse) reactions are located at the same level (Fig. 3.2). The dissolution and discharge reactions need the same energy of activation (AG*). At equilibrium there is no net current, as i — i = 0. The electrode potential assumes its equilibrium value Erev. Here, ?f = rr = io, io is called 'Exchange Current Density.' It is the current density associated with an electrode at equilibrium. Every reversible electrode has a characteristic exchange current density. exchange current density. There is no net transfer of charge as shown above. Each reversible electrode reaction has its own exchange current density. Consider a reaction, such as Cu = C u + + + 2e. Although at equilibrium no net current will flow through the circuit but the interchange of Cu atoms and copper ions would take place at the electrode surface. Hence, a current would be associated with the anodic and cathodic partial reactions. An electrode will not achieve equilib= rium electrode potential for M z + + ze < ± M, unless its io is much greater than io value of any other reversible reaction in the system. A high value of io represents a high rate of reaction. The value of io ranges between 10 - 1 and 10~5 A/cm2 for different materials. The relationship between exchange reaction rate and current density can be derived from Faraday's law discussed earlier: 3.2.4 REVERSIBILITY CURRENT AND EXCHANGE where ff is the rate of oxidation and rr is the rate of reduction expressed in terms of current density (io). It is a convenient way of expressing ?f or rr at equilibrium. In order to understand reversibility, consider the following (Fig. 3.2): The rate of reaction in forward direction, ?f is given by rf = Kie-{AGVRT) (3.14) 3 .2.S IRREVERSIBLE PROCESS where K\ is the constant depending on temperature, time and activity. The rate of reverse process is rr = K2e^G*clRT) (3.15) Figure 3.3 represents a situation when the electrode is irreversible, displaced from its reversible potential (Erev)- This irreversibility can be brought about by connecting the electrode to an external source of current. By connecting the electrode to the positive pole of an external source of current, the electrode is made the anode, and similarly it can be made a cathode by connecting to the negative terminal. M^>Mz++ze, z Ct = But if the system is at equilibrium rf = rr. Equating the two processes forward (anodic) and reverse (cathodic), we get -(AG*A-AG*C)/RT = ezEFfRT = if>ir a k ~ IhI> ? e t = net anodic current density M £ = ± M z + +ze \k\ ^2 Ki ,- _ , ianet =\ - ir, Thus, each reversible process has a characteristic potential, called electrode potential. When the reaction irreversible, if < * ir = 0, no net = current flows, if = ir = io(exchange)> h is called the The electrode is no more at equilibrium. The atom is, therefore, shifted from its equilibrium position to a new energy well in the direction of the positive potential. The magnitude of shift 64 Principles of Corrosion Engineering and Corrosion Control hence T/C must be negative. Thus the electrode potential moves to a more negative value. Summarizing the discussions, on connecting the electrode at equilibrium to the positive terminal of an external current source, the electrode becomes the anode and the potential shifts in the noble direction; AG^ for the anodic process is made more favorable. On connecting the electrode to the negative terminal, the electrode becomes the cathode and the potential becomes more negative, AG£ for the cathodic process becomes more favorable. H ±=+ w*+ze p)ZFn* 2 Pn A f , J': *. -:. j jl EXOBSS — w~^ — 1 „-„ "- 1 : ^ : i " —. ' * \A/r»rlr J energy >t~-1~ .? - "'! |(l-p)ZFm|- { JK done 3.3 HELMHOLTZ DOUBLE When a metal dissolves in an aqueous solvent by release of cations (positively charged ion), it becomes negatively charged. As more Figure 3.3 Energy-distance profile when the elec- and more ions are released, the metal surface becomes increasingly negatively charged. This trode is irreversible process continues until an equilibrium is reached of potential in the noble direction is called 'over- such that for any cation formed there must be potentiaV(riA)- It is given by 7 A = E — £rev> where a metal atom formed by the reverse process 7 subscript A is the anode, n is the over-potential, simultaneously. E is applied potential and Erev is the reversible M ^ M z + + ze potential. The potential shifts in the noble direcan tion, because E > £rev d * is positive. Consider 7 (Metal atom (Hydrated ion now the electrode being connected to the negative in solution) in lattice) pole of the external current source. The following reaction now proceeds in the reverse direction: At this point, the excess negative charge at the surface of the metal balances out the excess posz+ Z+ + ze M< - M +ze, M ^ : M itive charge due to cations in solutions adjacent \ic\ to the metal. Two layers having opposite charges, therefore, exist, one being negatively charged The reverse process now prevails over the forward and the other positively charged (Fig. 3.4a). The process: separation of charges exists like in a capacitor. The two oppositely charged layers constitute ; net net ir > if, V = | ir|-*f or the Helmholtz double layer. The double layer was compared to parallel plates by Helmholtz. Cathodic reduction now prevails over anodic A potential difference is thus created between the oxidation. The activation energy for the cathodic metal and solution. Under standard conditions, process AG£ is made more favorable than the this potential difference is the standard electrode rate of anodic process and the rate of transfer potential at the metal solution interface. Due to of charge by cathodic process is faster, i"et = the separation of charges, a strong electric field M ^ - M z + + ze, rjc = E — Eey where E<Eey in the space between the two charged layers is » v -— - - i-~\ ' *•.* . -."". ."". ,, LAYER Corrosion kinetics 65 Adsorbed water Diffuse layer (Guoy-Chapman) Sofvated anion f ! <> Plane (position o f closest approach for cation) f2 Compact layer Diffuse layer £ Figure 3.4a Electrical double layer showing inner Helmholtz, outer Helmhotz and diffuse layer (0i = inner, 02 = outer) u Distance Distance Figure 3.4b Electrical double layer set up. In the absence of external current, the electrode has a charged double layer, the capacitor Qi is charged (Fig. 3.4b). The total impressed current, /total? i s divided into two parts, one crossing the double layer, through Faradic resistance (Faradic current Ip) on the other, Ich (current of the charge), with a very small value. Faradic current is used in the electrode reaction, and /ch is stored in the capacitor, C&\. In a few seconds or less, the electrode charge reaches a steady value and it is proportional to the charge Q = (7 x t) of the double layer (Fig. 3.4b), which is high enough to pull the ions across the double layer. It is known that water is a dipolar molecule. The oxygen-end of a water molecule forms the negatively charged end and the hydrogen-end, the positively charged end. Due to the dipole nature, the water molecule is attracted towards the electrode and contributes to the potential difference across the double layer. The orientation of the water molecule depends on the charge on the metal surface. If the metal has large negative charge, the H2O molecules would be oriented with the positive ends (hydrogen) towards the metal, and negative ends (oxygen) towards the large positive charge. The water molecules are attracted towards the metal electrode and contribute to the potential difference. The water molecules form the first adsorbed layer on the metal surface. The cations which are hydrated and attracted towards the metal surface are limited in their approach to the metal surface because of the presence of water molecules on the metal surface. When electrostatic interaction operates, ions from the solution phase approach the electrode only as far as their solvation sheath allow. The surface arrays of these ions are thus cushioned from the electrode surface by a layer of solvent molecules. The line drawn through the center of such solvated cations at a distance of closest approach is called the 'outer Helmholtz plane.' The anions are specifically adsorbed sometimes in the water molecules (solvent) (Fig. 3.4b). 66 Principles of Corrosion Engineering and Corrosion Control 1 |xm from the outer Helmholtz plane in the bulk solution, the net charge being equal and opposite to that of the Helmholtz double layer. The size of the ions forming the OHP are such that the sufficient ions cannot fit these to neutralize the electrode, and the remaining charges which cannot fit that are held in increasing disorder. This less orderly arrangement constitutes the diffuse layer. As the concentration of charged ions in the Guoy-Chapman layer (Fig. 3.4b) increases, the thickness of the layer decreases. The ions are thus pushed down into the outer Helmholtz layer. It is analogous to a space charge layer in a semiconductor; where the thickness of diffusion layer decreases with an increase in the number of charge carriers. In case of concentrated solution, this layer is completely eliminated. To summarize, the double layer consists of three constituents: (a) An inner layer (inner Helmholtz layer) in which the potential changes linearly with the distance comprises the adsorbed water molecules and sometimes the specifically adsorbed anions. (b) An outer Helmholtz layer. It comprises hydrated (solvated) cations and the potential varies linearly with the distance. (c) An outer diffuse layer (Guoy-Chapman layer). It contains excess cations or anions distributed in a diffuse layer and the potential varies exponentially with the distance, F. The water molecules sometimes contain the specifically adsorbed anions. The water molecules form the inner Helmholtz layer. The line drawn through the center of these molecules is called the c inner Helmholtz plane* The outer Helmholtz plane cf2 (OHP) represents the locus of the electrical centers of the positive charges. This plane resides at a fixed distance from the metal due to the water molecules that are between the surface of the metal and ions. The outer Helmholtz plane (OHP) is significantly affected by hydrated cations M z + (hydrated). In the simple model of the Helmholtz double layer developed earlier, the adsorption of dipoles was not considered. When a metal surface has a slight excess charge, the dipoles are adsorbed. This process contributes significantly to the potential difference across the double layer. Two factors seriously affect the potential difference across the double layer: (a) the magnitude of charges at the interface and (b) the presence of a layer of adsorbed dipole at the interface. The simple model presented above, however, takes no account of the presence of dipole layers, on the metal surface. It suggests that when the surface has excess negative charge, it would adsorb cations and when it is positively charged it would adsorb anions. However, it fails to take account of specific adsorption of ions (chemisorption) in contact with metal surface. The anions are held on the metal surface of the same charge either by chemical bonding or by electrostatic forces exerted by cations. Hence, they may be present in a solvent water molecule. The model discussed above shows clearly two Helmholtz planes, the inner Helmholtz plane and the outer Helmholtz plane and it also shows how the anions are specifically absorbed. The Helmholtz double layer model is only applicable to a concentrated solution. Guoy and Chapman observed that the net charge in the compact double layer does not balance the charge on the metal surface. There is an additional region on the solution side of the layer, where either the cations are in excess of anions or anions are in excess of cations. These ions are distributed in a diffuse layer which extends from the outer Helmholtz plane to the bulk solution. Guoy and Chapman, therefore, proposed another charged layer, Guoy-Chapman layer, extending up to 3 .3.1 FACTORS AFFECTING E XCHANGE C U R R E N T D E N S I T Y A. Forward reaction: As described above, only those atoms which are energetically at unfavorable positions, such as at grain boundaries, dislocations, half planes, are able to detach themselves and participate in the reaction. Atoms are more easily pulled from the kink sites than terrace sites. The number of surface atoms available (Ns) in a given area can be calculated. B. Electrode composition: It depends upon the composition of electrode (see Table 3.1). The exchange current density for Pt is 10~3 A/cm2, Corrosion kinetics Table 3.1 Reaction 2H+ + 2 e = H 2 2H++2e=H2 2 H + + 2 e = H2 2H+ + 2 e = H 2 2H+ + 2 e = H 2 2H+ + 2 e = H 2 2H+ + 2 e = H 2 2 H + + 2 e = H2 2H+ + 2 e = H 2 2H+ + 2 e = H 2 2H++2e=H2 02 + 4H++4e=2H20 02+4H++4e=2H20 F e +3 + e = F e +2 Ni = Ni + 2 + 2e Exchange current densities Electrode Al Au Cu Fe Hg Hg Ni Pb Pt Pd Sn Au Pt Pt Ni 0.5 N NiSQ4 Solution 2 N H 2 S0 4 1 NHC1 0.1NHC1 2 N H 2 S0 4 1 NHC1 5NHC1 1NHC1 1NHC1 1NHC1 0.6 N HC1 1NHC1 O.lNNaOH O.lNNaOH to (A/cm2) 10-io 67 10-6 2 x 10~7 i o- 6 2 x 1 0" 12 4 x 1 0" 11 4 x 10-6 2 x 10~12 IO"3 2 x 1 0" 4 1 0" 8 5 x 10~13 4 x 10~13 2 x 1 0" 13 10"6 Source: Bockris, J. O'M. (1953). Parameters of electrode kinetics. Electrochemical Constants, NBS Circular 524, U. S. Government Printing Office, Washington, DC, 243-262. whereas for mercury Hg it is 10 3 A/cm2 in pure platinum. It is, therefore, widely used as a catalyst because of the rapidity of the reactions dilute acid. C. Surface Roughness: Large surface areas pro- which occurs on its surface. Consider a forward anodic oxidation process: vide a high exchange current. D. Impurities: The exchange current density is M - > M z + + ze reduced by the presence of trace impurities, such as As, S and Sb which are catalyst Note: The term anodic oxidation and anodic poisons. dissolution is used interchangeably in this text. The total number of moles of atoms dissolving Example 9 in a unit time, according to classical rate theory is > The exchange current density for 2H + -\-2e — H2 given by on mercury is 10~12A/cm2, but is nine orders AG* of magnitude higher on platinum at 10~3 A/cm2, Forward rate. exp which indicates Pt provides a favorable catalytic RT surface for the above reaction to occur rapidly on r= m ol-cm~ 2 -s _1 (3.18) platinum. A large current density means a very stable potential which cannot be easily disturbed, where hence, platinized platinum is made a standard a = fraction of active atoms, which are hydrogen electrode, so that the potential remains chemically active stable. / = frequency factor (1012 s _1 ) It is well-known that most reactions occur r = rate of forward reaction (mol/cm2 • s) very rapidly on the platinum surface. Platinized platinum has a larger surface area and conseThe rate r can be converted to a curquently a higher exchange current density than rent density term by multiplying f by zF ' rf=/ ( a t) 68 Principles of Corrosion Engineering and Corrosion Control (zF = coulombs/mole) (z = No. of charges atom by the reverse process. Energy-distance protransferred) file for a reduction reaction is shown in Fig. 3.5. The rate of the reverse process is now given by iv. r (in mol/cm2 -s)xzF (in coulomb/mol) Suppose Cs (mol/cm2) is the concentration of metal cations, and VL i s t n e volume of this part of As 1 coulomb (C) is equivalent to 1 ampere- the Helmholtz layer, and if it is further assumed second, we obtain A/cm2 after multiplication, that all the CSVL ions in the double layer are available for the reverse reaction io> the magnitude which is the unit of current density of the cathodic partial current would be i = zFr = A/cm2 AG* io(rev) = zFfQVLexip Equation (3.18) can now be rewritten as: ~RT~ * 3.4 ,= z F/a CXP (t) [" AG* ~RT~ (3.19) where io is the exchange current density. where Cs = 10~3 mol/cm2, and VL = 10" c nr (typical values). At equilibrium iQ = if = — *(rev)The reverse current has a negative sign, however, io is the magnitude of both forward and reverse process. As the cations are involved in the reverse process, NS/NQ (equation (3.18)) is replaced by (Cs VL). '(rev) = ~zFfQ VL exp I •--zFfa(QVL)exip AG* RT (3.20) REVERSE REACTION (CATHODIC REACTION) Let us now consider the conditions under which a cation would be transformed back to a metallic AG* " RT Distance Figure 3.5 Energy-distance profile for a reduction reaction (r = reduction, f = forward reaction) Corrosion kinetics At equilibrium: io = i-- 69 --zFfQV^exp /-AG*\ \ RT ) (3.21) Equation (3.21) shows that at io, the rate of dissolution is equal to the rate of deposition as discussed in Section 3.2.2. 3.5 DEPARTURE FROM (I/)] EQUILIBRIUM [ACTIVATION OVER-POTENTIAL When a metal electrode is in equilibrium, the partial current if for forward reaction and partial current ir for reverse reaction are precisely equal and opposite. There is no net current flow. If the potential drop across the double layer (Section 3.3) is altered by superimposing an external electromotive force (emf), the electrode is polarized and hence there is a deviation from the equilibrium condition. The extent of polarization is measured by the change in the potential drop (A£) across the double layer. The shift of potentials from their equilibrium value on application of an external current is called polarization. The magnitude of the deviation is termed 'overvoltage (77) which is directly proportional to the current density. Due to polarization, an imbalance is introduced in the system. If d(AE) is = positive, a net anodic current flows, i£et = if < ± ir, where if ^>i r , and if d{AE) is negative, the electrons are pushed in the metal by connecting the electrode to the negative terminal of an external source of emf, a net cathodic current would flow, icet(^l);herezr»!f. To summarize, electrons are supplied to the surface and, hence, the surface potential (E) becomes negative to the equilibrium potential (Ee)- On the other hand, in anodic polarization, electrons are removed from the metal surface which cause the potential (E) to become positive to the equilibrium potential, Ee. At the corrosion potential (Ecorr), the rate of forward reaction (oxidation) if equals the rate of reverse reaction, ir and hence the over-voltage (r] = 0) (Fig. 3.6). I(+) E o Q m C 1 Log(i) Figure 3.6 Polarization diagram (reversible electrode) 70 3.5.1 Principles of Corrosion Engineering and Corrosion Control ACTIVATION POLARIZATION A ctivation polarization can be a slow step in the electrical reaction for which an activation energy in the form of potential is required for the reaction to proceed. When a certain step in a half cell reaction controls the rate of electron flow, the reaction is said to be under activation charge transfercontrol and activation polarization occurs. For example, consider the reduction of hydrogen ions: 2 H + + 2<T H2 T he rate of hydrogen reduction is determined by the slowest of the steps. The rate, controlling s tep, varies with the metals, current density and environment. There is a critical activation energy needed to surmount the energy barrier associated with the slowest step. The rate of transformation is controlled by the magnitude of the energy barrier that an atom or ion must surmount to transform from metal to ion or from ion to metal. The energy that must be acquired is the activation energy AG*. The relationship between activation polarization and the rate of reaction is given by the Tafel equation: Hydrogen evolution occurs in four major steps: (1) 1 st Step - H is absorbed in the surface at the electrode. (2) 2nd Step - H + + e - - > H a c l , the species is reduced on the surface. (3) 3rd Step - The two reduced species combine to form a hydrogen molecule Ha(j + Had-+H2. (4) 4th Step - Hydrogen bubbles are formed by combination of hydrogen molecules. + riA'- = / U o g - and T]C •• =/Uog^ to w here rj is over-potential. A plot of overpolarization (?7act) vs log(i) is linear for both anodic and cathodic polarization. /3 a a nd /3C a re called the Tafel slopes and io is the exchange current density (Fig. 3.7). (+) •f" 0.059 i O 5 m s. „ -i 0.059 • 4 ^ ' Z(l-B) (-) Log (i) Figure 3.7 Polarization diagram of zinc in contact with its own ions Corrosion kinetics 3 .5.2 FACTORS AFFECTING 71 required to shift the position of the atom from the reversible electrode is given by rj. ACTIVATION POLARIZATION In the forward reaction (anodic polarization), M -> M z + -\-ze, the electrode is connected to the A. Current Density: Activation polarization increases with current density i, in accordance positive terminal external emf and the potential superimposed T/A is positive. It is greater than with the equation: zero and the electrode polarizes in the noble direction. For an activated ion on the top of an 7 = ±£log/ energy hump, the magnitude of activation energy AG^ required is the total activation energy of B. Materials: Activation polarization varies with the barrier [ AG* + (1 — P)ZFT]A] minus the overone metal to another because of the specific potential (ZFIJA) applied. The magnitude of the total activation energy A G^ (the energy state of effect of current density. C. Surface Roughness: Activation polarization metal is increased) is given by: is high on a smooth surface compared to a shiny surface. A G J = A G * + [ (1-J8)ZFI;A]-ZFI7A (3.22) D. Temperature: Increased temperatures decrease polarization as less activation energy Hence would be needed and the exchange current density would be increased. AGl = AG-PzFriA (3.23) E. Pressure: Hydrogen over-voltage increases rapidly with decreasing pressure. But F. p H: Over-voltage increase initially and decreases with increased pH value. (3.24) G. Agitation: It has no effect on activation polarRT ization, because it is a charge transfer process involving electrons and not a mass transfer. After substitution H. Adsorption of Ions: The hydrogen overvoltage is decreased by adsorption of anions A G * - PZFTIAI (3.25) and increased by adsorption of cations. exp RT J Wi Kt) r Ns 3 .5.3 ANALYTICAL FOR FORWARD REACTION The above expression can be written in terms of equation (3.23), hence if = zFfa—exp No PZFTJA' AG* ' RT x exp RT (3.26) EXPRESSIONS IRREVERSIBLE AND REVERSE It has been shown that when the rate of forward reaction if is equal to the rate of reverse reaction ir, there is no net accumulation of charges and no current flows in the external circuit, M «=* M z + + ze, The potential across the metal solution interface in the double layer is the equilibrium potential. The energy-distance profile for the reversible electrode is shown by broken lines (Fig. 3.5). The original position of the atoms is shown in the energy wells. By supplying an external emf, the atom has been pushed in a new position in a higher energy well. The over-voltage AG* to = zFfa— exp — by equation No |_ ~RT~ (3.19) (where io is the exchange current density). Therefore, But zf = *oexp fizFru RT (3.27) Equation (3.27) suggests that as the anodic activation over-voltage increases, the rate of forward reaction also increases. 72 3 .S.4 Principles of Corrosion Engineering and Corrosion Control REVERSE REACTION (CATHODIC DEPOSITION) If the reaction is now reversed, by connecting the electrode to the negative terminal of external emf: Here, the rate of reverse reaction ir is much higher than the rate of forward reaction if . The over-potential r)c is less than zero. The activation energy for the forward reaction is higher than the reverse reaction. The total available energy for cathodic reaction is: AG£ = AG* + (l-l3)zFric As shown earlier, io = zFf(QVL)exp (equation (3.24)) (3.28) Let us now consider the departure of the anodic and cathodic reactions from the equilibrium potential at large enough voltages, approximately > 0.12 volt. Under the above conditions one of the two terms (partial current) becomes negligible, only one reaction would prevail, and the other would become negligible. Consider anodic polarization (n > 0) only at voltages higher than 0.1 volt. The current (^ et ) would be equal to the anodic partial current density and ic would be negligible (the reverse cathodic reaction). Under this condition the right-hand expression of equation (3.31) would be eliminated, however, for the anodic reaction £*=«>«? PzFrjA RT (3.32) RT which relates the partial anodic current density to the over-potential. By similar arguments, at a large enough cathodic polarization [rj (negative)], the anodic partial current becomes negligible and the current density is tr =• Replacing AG£ in equation (3.24) by the expression in equation (3.28), we obtain: | zc | = zo exp (1 -^C-] RT (3.33) T _.. r /AG*+a-^cy -jfi 1 (3.29) (3.30) zr = z F/QV L expl-l zr = -zoexp — which shows that the anodic or cathodic current densities vary approximately as the exponential of the over-voltage. When \ir\ > 7 f ; (for nc > 0.03 V). I - net I „• _ __. Equation (3.30) shows that as the cathodic over-voltage is increased, the rate of cathodic reaction, ir, is increased and the rate of forward reaction is decreased. To sum up, the net current density under the two conditions when if ^>/r and |ir|^>*f> is given by \ic = zo exp {\-P)zBric\ RT (3.33a) The expression derived above for z"et and z"et equations (3.32) and (3.33) can be written in terms of (rj), the over-voltage. Taking logarithms of equations (3.32) and (3.33) In ia — In io = ^^ RT r]k'- > I when i ^> | ir | j -lnip+lnza fizFIRT ^net — ^a -exp (l-P)zFricm (3.31) which is the general form of current vs potential relationship. The above equation is called the Butler-Volmer equation. *-(-£)-(*) RT\J VA = pzFJ ; l nto+1 — . {RT\1 \PzFj llni , C orrosion kinetics For over-potential: 73 3.6 TAFEL EQUATION ^ (>0) ) , A =(--g)lnio+(^)lni a ~(£)-G) -P)zFr]c In | zc | — In io = — — RT RT -In r]c (3.34) T he relation between the over-voltage (rj) a nd the reaction rate is extremely important. Consider anodic polarization only. At high anodic overvoltage, ir^>if,ic is negligible and the current density (i£ et ) becomes equal to the anodic partial current density, C e t = *a = ioexp Similarly for cathodic polarization, on taking logarithms, RT J (equation (3.31)) T aking logarithms, we get: fizFrjA l nz a = lnzo + - — = a-P)zF ln ip — In | zc 1 VC-(l-P)zFIRT RT ,(l-£)zF for (r] < 0) RT rjc = (l-)8)zF. -RT r]c = (l-/5)zF l nio- >r]A (3.36) RT , . RT 1 . In zn In I* PzF PzF RT (l-j8)zF l lnlicl hence, (3.37) * A= 7 -m(-) ^ )zP. In lie O n the other hand, at large enough cathodic polarization (rj n egative), the anodic partial current becomes negligible, hence ^et = ic = ioexp ( l-j8)zFy/ c ' RT ( from equation (3.33)) (l-P)zFr)C RT lnio (3.38) (3.35) w here r) = (E-Erev) rjA > 0, anodic polarization r]c < 0, cathodic polarization. The experimental equations obtained by Tafel are very similar to equations (3.34) and (3.35), where the Tafel constant, fra a nd bc c omprise of equation (3.32) by equation (3.34), and equation (3.33) by equation (3.35), are 23RT U-j8)zF bc = 23RT (l-P)zF Taking logarithms, In ic = In i$ + RT r]c = (l-P)zF RT I n ic - (l-j8)zF Equations (3.37) and (3.38) are Tafel equations. The above expressions can be written in a more generalized form of equation (Tafel equation) r] = a+b\o%I (3.39) T he matter is further explained in Section 3.6. where a a nd b a re constants and b is the Tafel slope. The left-hand term of equation (3.38) represents the constant a a nd the right-hand term constant b. T he following are the empirical 74 Principles of Corrosion Engineering and Corrosion Control 23RT relationships between the current I and overvoltage r] for the anodic and cathodic over-voltage as shown in equations (3.34) and (3.35). The slope b- ~JzT (equation (3.39d)) (3.39a) inserting appropriate value again as in the previ23RT 0.0592 , ous case, n = — - — , hence, Pz PzF 0.0592 23RT\ >c 7 (3.39b) 1^~ ~JzTJ 0.059 ac- (3.41) Similarly for cathodic polarization RT\ From equation (3.36), constant a is I — JzF) RT . log io and constant b, the slope is ——. It is more PzF convenient to express these in the form of log to base of 10: (1-/0* logio (3.42) bc = - 0.059 (3.43) The Tafel equation is generally written as ?7A = ±/31og — (r/A = activation polarization) *o where P = constant, being positive for anode and negative for cathodic reaction i = the net rate of reaction io = exchange density. The situation is different if the over-voltage is low ( ^ ^ f p - ) ; the electrode potential lower than 0.03 V. Under such conditions, the partial current densities are negligible. Equation: a= ~vw)logzo 3RT\ / 2.3KTV (3.39c) b=(^^\ J V / zF) (3.39d) In the expression (3.39c), appropriate values are inserted. T = 25+273 = 298°K R = 8.314 J/deg.mol F = Faraday's constant in coulombs/mole 2.303 is a conversion factor to convert natural log to the base 10 2.303#r/F= 0.5916 V. and On inserting these values, the term in equation 0.059 (3.39c) at 25°C becomes —— log ioPz Hence, /23RT\ 0.059 log k = aa = - ( -T-TT J log h fiz =ioexp h^_rexpL—^^J x2 x3 x2n +-1+ — 21 3 n1 can be expressed into a series: ex = l+x+- (3.40) ex = l + n The symmetry factor (P) represents the location of the energy hump. It ranges from 0 to 1 as shown in the energy profile diagram, however, to locate it precisely, it is extremely difficult. Generally, it is assumed that it is located in the center; here 0 = 0.5. z = *ofexp PzFr] RT -[-T]) (l-p)zFr]l RT J i = to + PzFn RT Corrosion kinetics tozF a RJ 75 / . _ ipzF -+kRTr] The mixed potential theory partly mentioned earlier, is used with advantage to predict the rate of corrosion of metals and alloys in given environment. It was postulated by Wagner and Traud in 1938. It has two basic assumptions: (a) Electrochemical reactions are composed of two or more partial anodic and cathodic reactions. (b) There cannot be any accumulation of charges. Thus, the over-voltage is proportional to current density. Consequently, polarization resistance can be defined as £ p: (dn\ __J T R iozF \*'/i-*o""« This theory was not applied until 1950, when A low field approximate, r\ < 0.002, the hyperbolic Stern applied it in analysis of corrosion. sine function, approximates to a linear function Consider a piece of zinc immersed in ZnSC>4. so that The reactions would be Zn->Zn 2 + +2e~ (oxidation) and Z n 2 + + 2 e ~ - > Z n (reduction). For zFr] iozFr] (3.44) the two reactions at rest an equilibrium poten*net = 2*o s inh = —— = — — £T tial would be observed which would be given by RT Nernst equation. Consider now an electrode, such which is the lower field approximation for the as zinc, immersed in HC1. The following could be Butler-Volmer equation. the possible reactions: (a) Z n ^ Z n 2 + + 2 e " (anodic) 2H+2e~ -> H2 (cathodic) Zn 2 + +2e~ - » Zn (cathodic) H 2 -> 2H + +2e~ (anodic) 3,7 MIXED POTENTIAL (b) (c) (d) THEORY A N D ITS APPLICATION 3 .7.1 MULTIPLE ELECTRODES In a corrosion cell (electrolyte, anode, cathode and a metallic path), multiple reactions occur. For instance, when zinc corrodes in a dilute acid, the following reactions occur: Zn < * Z n 2 + +2e = (oxidation) (Forward reaction, if) H 2 ^ 2 H + +2e (reduction) (Reverse reaction, iY) The total reaction is Zn+2H+ -* Zn 2+ + H 2 f. Accordingly, the metal constitutes a multielectrode, because at least two different reactions occur on its surface simultaneously, one oxidation and one reduction. Bubbles of hydrogen are observed from the surface of zinc electrode, and formation of bubbles of hydrogen is a cathodic reaction. Hydrogen is reduced and not oxidized. Similarly, zinc is oxidized and not reduced. Hence, only the two reactions (a) and (b) proceed. Under the condition of rest (no outside current), the potential of the electrode cannot be computed by the Nernst equation as it is not reversible. Also, the above electrode would not corrode in the absence of an external current. The potential assumed by the electrode under the above condition is the mixed potential and its value lies between the value of equilibrium potential of hydrogen and zinc. The value of the potential would depend on the metal and the environment. It is to be observed that the corrosion potential (£Corr) is not the equilibrium potential of either of the reactions, but some intermediate potential determined by the two partial anodic and cathodic reactions. Both the reactions 76 Principles of Corrosion Engineering and Corrosion Control -£ EH 0 w > Rest potential ofZh B -0.76 2 Log(i) Figure 3.8 Diagram illustrating mixed potential theory direction. The upper curve, anodic polarization, corresponds to the reaction M «=t M z + + ze and hence i"et = if — iT and rjA > 0. The variation of r\ with \l\ is linear. The departure from the equilibrium is shown by the over-potential (rj). 3 .7.2 CONSTRUCTION OF Similarly, for cathodic polarization, starting A POLARIZATION DIAGRAM from Erev, the over-voltage (rj), is increased in the A Tafel plot can be obtained by plotting the more active cathodic direction (negative) rate of corrosion in terms of current (I) vs the applied potential (rj = over-voltage), either above M ^ M z + + ze or below the reversible potential (£rev) (Fig. 3.6). The x-axis is the potential axis and /-axis is the current axis. Firstly, a zero over-potential point hence, (rj = 0) is located on the axis. This corresponds /-net —{_{ to the reversible potential of the system to be kathode — l l investigated, because at Erev> rj = 0. There is no r)C<0 driving force when n = 0, and so it provides a good starting point. At £rev> (a) and (b) are polarized to a common potential, the mixed potential (£mix) (Fig. 3.8). As r] is increased, the rate of cathodic reduction (cathodic polarization) increases. The slope of the anodic portion of the curve is given by io = i-un = 0 /3a = — 0.059/z/? and that of the cathodic portion b y£ c = - 0.059/z(l-£). znet = 0 A linear relationship is observed in the Tafel plot. When the activity of the ion is unity, the For anodic polarization, a predetermined over- reversible potential is the same as the standard potential (rj) is applied step-by-step in the noble potential measured against hydrogen. If zinc Corrosion kinetics in contact with its ions at unity is polarized, it would be oxidized at more noble potentials: M « ± Mz+ + ze and reduced at more active = potentials M z + + ze <=^ M. £reV(Zn) =-0.763 VSHE (aZn2+ = l). It is an example of a single electrode reaction when the metal is in contact with its ion in the solution. Unit activity of ions is 1 g ion per dm3 or 1 g ion per liter. A theoretical polarization diagram for the anodic and cathodic polarization of zinc is shown in Fig. 3.7. The linear Tafel relationship is illustrated in both diagrams. The mixed potential diagram or E vs I diagram allows accurate prediction of corrosion rate to be made in specific environments. It also allows the prediction of current based on current density higher than the corrosion current (iCorr) and utilization of advanced technique for corrosion measurement, such as galvanostatic and potentiostatic techniques. These techniques will be described at the end of the chapter. 77 A corroding electrode constitutes a polyelectrode because a minimum of two electrode reactions, oxidation and reduction, occur simultaneously. In case of a half cell, such as Fe/Fe2+, H2, Pt, parameters, such as current and individual electrode potentials, may be measured as a function of change in the external resistance. It is, however, impossible to measure the two parameters on the tiny anodic and cathodic areas of corroding metal. It is, however, possible to measure the mixed potential and the net current. In case of zinc corroding in an acid, k = Zn < ^ Zn ~l~+2e, the following are the values of £° k (potential, V) and i$ (exchange current density): £ ° , Z n | Z n 2 + = -0.76V io,(Zn|Zn 2+ ) = 10~7A/cm2 and for H2 +± 2H+ +2e, on Zn = 10~10 A/cm2 = 10"6A/m2, £ £ 2 /H+ = O v o l t s The difficulty in the measurement of individual current and potential is resolved by Evans diagram. The ratio of anodic to cathodic areas is taken as unity and current (J) is replaced by current density (z). On polarizing anodically, the potential is displaced in the positive direction, and on cathodic polarization, it is displaced in the negative direction. The intersection of anodic and cathodic curve gives £Corr> the corrosion potential, which can be measured experimentally. The potential corresponding to icorT (corrosion current) is corrosion potential (EcorT). -fccorr represents the contribution of both anode and cathode to polarization, hence, it represents a mixed potential. The rate of corrosion is given by icorr> which represents the total current associated with the dissolution of zinc and evolution of hydrogen. The following are the reactions that take place as shown in Fig. 3.8: (A) Z n ^ p * Z n 2 + + 2 e ( l ) o r Z n ^ Z n + + + 2e k 3.8 EVANS DIAGRAMS The Evans diagram shows the electrode potential in volts in the ordinate and the reaction rate (ampere) in the abscissa (Fig. 3.8). Consider a base metal, such as iron or zinc, placed in acidic solution; the metal will dissolve at the same rate as hydrogen is evolved and the two reactions mutually polarize each other. This is shown in a very simple form by Evans diagram. In Evans diagram, either current or current density can be plotted against potential. If the ratio of anodic to cathodic areas is taken as unity, current density (i) maybe used rather than current. A negligible resistance is assumed between the anode and cathode. As the change in the anodic and cathodic polarization has the same effect on corrosion current, the system is considered to be under mixed control. Consider zinc corroding in dilute HC1. The following two reactions occur: Zn<=^Zn 2+ +2e (oxidation) 2 H + + 2 e ^ H 2 (reduction) E°Zn = - 0.763 V(SHE) and Zn < -> Zn + + + 2e( l a) or k Z n 2+ + 2e -> Zn(i f >i r ) 78 Principles of Corrosion Engineering and Corrosion Control or H 2 ^ 2 H + + 2e £corr> which is the mixed potential of the system. The corresponding current at the current axis is called 'corrosion current? represented by iCorr- At the corrosion potential, the main anodic and cathodic reactions are two different reactions, whereas, at an equilibrium potential, a single partial reaction occurs in both anodic and cathodic direction. The corrosion rate given by icorr is equal to the total anodic and cathodic rates of corrosion. (B) H 2 ^ > 2 H + + 2 e ( 2 ) k and H 2 «|>_2H + +2e(2a) or 2H++2e-> h H 2 (i r >if) (C) Zn+2H+ -> Z n 2 + + H 2 (3) (total reaction) Consider polarization (forward reaction) of zinc: Zn^Zn2++2e(4) In the forward reaction, the oxidation rate of Zn 2+ exceeds the reduction rate of zinc (equations (1) and (la)). Similarly, the rate of reduction of hydrogen ions exceeds the rate of oxidation of hydrogen (equations (2) and (2a)). H 2 ^ 2 H + + 2 e (5) (cathodic reduction of hydrogen) 3.9 PREDICTION OF CORROSION TENDENCY ON THE B A S I S OF MIXED POTENTIAL THEORY According to the emf series, zinc has a more active potential than iron, £°(Zn) = -0.76 V, £°(Fe) = Equation (4) represents the anodic oxidation of —0.44 V, hence zinc should corrode more easily zinc and equation (5), the cathodic reduction of than iron, but, surprisingly, iron corrodes faster hydrogen. than zinc, as explained below. As observed above, four separate corrosion Figure 3.9 shows a comparison of the elecreactions are shown for zinc: trochemical parameters for iron and zinc in acid solution. (1) Zn ^ Z n 2+ + 2e (zinc oxidation) (if > ir) For iron: = (2) Zn 2 + +2e~ * ± Zn (zinc reduction) (ir > if) Fe->Fe 2 + +2e (anodic) For hydrogen: 2 H + + 2e -> H2 (cathodic) + = (3) H 2 < * 2 H + 2e (hydrogen oxidation) For zinc: (if > ir) + = (4) 2H +2e~ < * H 2 (hydrogen reduction) Zn 2+ + 2e -+ Zn (anodic) (ir>if) Of the four curves corresponding to the above reactions, only reactions (1) and (4) are of practical interest. Reaction (3) hydrogen oxidation, and reaction (2) zinc reduction, can be ignored because zinc can only be reduced at potentials more negative than —760 V and it continues to generate electrons with time. Hydrogen evolution can be clearly observed and it does not oxidize. Hence, only reactions (1) and (4) are of practical interest. The curves corresponding to the above reactions converge at a point of intersection called ''corrosion potential? represented by 2 H + + 2e -> H2 (cathodic) £ ° n = - 0.763 V, £H+/H = 0.00V £ ° e = -0.443 V, £H+/H = 0.00V The polarization diagram of zinc in acid is given by dashed lines and that of iron by solid line. It is observed that icorr of zinc is less than that of iron, because of the lower io value of hydrogen reduction on zinc (10~7 A/cm2) compared to on iron (10~6A/cm2), so iron corrodes faster than zinc as shown in Fig 3.9. Corrosion kinetics 79 10 i 0 10 10~8 10"6 10"^ 2 w* Current density (A/cm ) Figure 3.9 Comparison of electrochemical parameters for iron and zinc in solution solutions, such as the effect of Fe 3+ ions on the corrosion of zinc in hydrochloric acid. Log i vs E diagrams can be constructed to predict the effect of environment on the corrosion rate of metals. Examine diagram (Fig. 3.10). When multiple reactions take place, £corr and icorr can be obtained by summing the currents at different potentials at which reactions occur (anodic and cathodic) to obtain total anodic and cathodic polarization curves. For instance, two partial anodic and two partial cathodic reactions are shown in Fig. 3.10. The total rate of oxidation and reduction is shown by broken line. Charge neutrality must be maintained. 3.10 APPLICATION OF Consider a metal, M, in HC1 to which F e 3+ ions have been added. The corrosion rate MIXED POTENTIAL THEORY of metal M in the absence of ferric salts is given by *corr(M). The corrosion increases from 3.10.1 E F F E C T O F A N *corr(M) to z corr (M-> Fe 3+ /Fe 2+ ), on adding the oxidizer as shown in the diagram. The OXIDIZER rate of hydrogen evolution is decreased from It is of interest to observe the effect of oxidizing iH2 = i corr (M) to ift2(M-+ F e 3+ /Fe 2+ ) as shown metal ions on the corrosion rate of a metal in acid in the figure. This is due to depolarization. The reason for the success of Evans diagrams in corrosion is that they combine thermodynamic factors (E values) with kinetics factors (i values). The usefulness of corrosion kinetics in the study of corrosion rates is, therefore, obvious. The exchange current densities have been included in the polarization diagram by Stern, and such diagrams are called Stern diagrams. Evans diagrams do not include exchange current densities. 80 Principles of Corrosion Engineering and Corrosion Control Platinum has a very noble potential and it does not oxidize in HC1. In fact its driving force is negative. Zinc dissolves in acid with the liberation of hydrogen. Platinum does not react. On connecting zinc with platinum electrically, it is observed that: (a) The rate of hydrogen evolution on zinc (H on Zn) decreases. (b) The rate of oxidation of zinc, Z n 2 + + 2 e ^ Zn, in the acid solution increases. (c) The rate of hydrogen evolution on platinum surface (H on Pt) increases vigorously. Consider first, the oxidation reaction of zinc (Zn 2+ + 2e —• Zn). The oxidation reaction polarizes zinc in the noble direction and the hydrogen > reduction reaction (2H + + 2e —• H2) in the active direction. The anodic and cathodic curves converge at a point d. The point of convergence yields *corrforzinc. The potential corresponding to icorr for zinc is £corr« Now consider the oxidation of platinum. There is no oxidation of Pt in HC1, hence, no oxidation curve for platinum is shown in Fig. 3.11. However, the reduction of hydrogen on platinum is clearly shown by the cathodic polarization curve for platinum. It is observed that there is only one oxidation process (Zn—>• Zn 2+ + 2e) and two reduction processes H 2 ^ 2 H++2e on platinum and H 2 ^ 2 H + + 2 e on zinc. The situation is analogous to Case 3.10.1 above. The total cathodic process is the sum of H2 on Zn and H2 on Pt, resulting in the dotted line. ** i (+) X X & S^tf^^jy^^ ^e vX V ^ ^S -''' J °HV^ " ^^ ^Ss,s> V_ ^^ sc„'''l<M\ V rate of *v^ S__ c ?v tcprr oxidation ^ (M-Fe^VFe 2 *) I (-) x^xC • s. •z J V/MJ^ \N^ ^ N > > Total rate of v \ " reduction — *^^ ^N * ^/^ ^\ Log(i) I *»(M«IWfi*) Figure 3.10 Effect of oxidizer on a metal in an air solution The corrosion potential of the metal £Corr(M) is shifted to a more noble value on addition of the oxidizing ions as shown by the £corr arrowed on the figure obtained by the intersection of total anodic and cathodic curves. Thus, there are three consequences of adding an oxidizer: (a) The corrosion rate of the metal is increased. (b) The corrosion potential is shifted to a more noble direction. (c) The rate of hydrogen evolution is decreased. It is to be noted that the effect of F e 3+ ion is pronounced on the metal M, because of its high exchange current density. If the exchange current tendency is small, there would be little effect on the corrosion rate of metal M. The exchange current density of the oxidizing ions must be higher than the exchange current density of hydrogen on the metal surface, to have a significant effect. E«z„ = -0.76 3.10.2 C O U P L I N G OF AN ACTIVE METAL TO A N O B L E M E T A L ( F I G . 3 . 1 1) Consider the coupling of zinc (E° = — 0.76 V) with platinum (E° = 1.2 V) in a hydrochloric acid Figure 3.11 Galvanic coupling of active to noble solution. metal Corrosion kinetics As observed in Fig. 3.11, the icorr of zinc (uncoupled) increases from 10~4 A/cm2 to a very high value at a when coupled with platinum; consequently, on connecting zinc with platinum, the rate of corrosion of zinc is vigorously increased. As shown by the points of intersection, a and c, the rate of hydrogen evolution on zinc is drastically reduced (from d to c), whereas the rate of evolution of hydrogen on the platinum is vigorously increased. The current density at b is significantly higher than at c as shown in Fig. 3.11. Thus the rate of hydrogen evolution is decreased is shown by the reduction in the current from point d to point c. What does it all amount to? It is shown clearly that zinc is a very poor catalyst for reduction of hydrogen as shown by a very small exchange current density io(H.2 on Zn) = 10 - 1 0 A/cm2, whereas platinum is an excellent catalyst for reduction of hydrogen as shown by a very large exchange current density for reduction of hydrogen io(H2 on Pt) = 10~3 A/cm2. To summarize the effect of coupling of zinc to platinum, the following are the main points of interest: (a) The rate of hydrogen evolution is decreased on zinc and increased on platinum. (b) The rate of oxidation of zinc is increased significantly on coupling and zinc dissolves vigorously. Nothing happens to platinum. 81 (c) Platinum is an excellent catalyst for reduction of hydrogen and zinc is a poor catalyst. 3.10.3 E F F E C T OF GALVANIC COUPLING (FIG. 3.12) It is a common practice to use the potential of a metal in an emf series to predict its corrosion tendency. In case of galvanic couples, the difference of potential between the couples is taken as a measure of corrosion tendency of the couple. In general, the larger the difference between the thermodynamic potential of the metals forming a galvanic couple, the more severe would be the magnitude of galvanic corrosion. This is on the basis of thermodynamics. Consider coupling of zinc to gold and zinc to platinum. According to the thermodynamic approach, the difference between the potential of zinc (E^n = —0.76V) and gold (E£u = 1.50V) is higher than the difference between the potential of the zinc and platinum (£° n = - 0.76V,£ Pt = + 1.2V). According to the thermodynamic approach, the Zn-Au couple should corrode faster than Zn-Pt couple. This is certainly not true and contrary to the experimental observation. The purely thermodynamic approach could not, therefore, be a good basis for prediction. Let us observe now the accuracy of prediction made on the basis h, H2 on Zn k Hi o n Au l «H 2 o n Ft t» Zn ^**<s Ecorr™ fccorr ~ Un/Au Icorr ioorr IziVPt Icorr E»zn - - 0.76 Lava** icorr i ! 1, I 1 1 -10 -8 -6 -4 -2 0 Log ( i) — * Current density, A/cm2 Figure 3.12 Effect of galvanic coupling of zinc with gold and platinum 82 Principles of Corrosion Engineering and Corrosion Control couple is higher than that of Zn-Au couple. This result in contrary to that obtained on the basis of only thermodynamic potentials but is true, due to the effect of kinetics. Gold is a poor catalyst of hydrogen evolution io (H on Au) = IO -6 A/cm2, compared to io (H on Pt) = IO - 3 A/cm2. Zinc when coupled to platinum, therefore, corrodes at a much higher rate than when it is coupled to gold. The above predictions are based on kinetics of reactions and are, therefore, more accurate and complete than the predictions based on thermodynamic potentials. The latter can be often misleading and if an accurate prediction is made, such as in case of active metal coupling, it may be more of a coincidence than the rule. of kinetics. Examine Fig. 3.12 showing the galvanic coupling of zinc with gold and platinum. The oxidation of zinc polarizes the electrode in the noble (positive) direction and the cathodic reduction of hydrogen in the active (negative) direction. The intersection of the two curves gives *corr for Zn. The exchange current density for hydrogen on zinc, Au and Pt is shown in the diagram. The exchange current density for zinc is shown in the first curve in Fig. 3.12. The oxidation curves for gold and platinum are not shown in the diagram as the above metals do not oxidize in HC1. The intersection of hydrogen reduction curve for gold with zinc oxidation curve yields the zCorr and £corr of zinc coupled to gold (Au) and the intersection of the oxidation curve of zinc with platinum yields the icon and £corr of zinc when coupled to platinum. It is clearly observed that the highest value of icon is shown by Zn—Pt coupled. The icorr value of Zn-Au couple is clearly very much lower than Zn-Pt couple. The hydrogen reduction reaction rate is the highest on a platinum surface io(H on Pt) = 10 _3 A/cm 2 followed by gold io(H on Au) = IO - 6 A/cm2. The reduction rate of hydrogen is very low on zinc io ( H + /H2 on Zn) = 10 _10 A/cm 2 . As observed in Fig. 3.12, the icorr of Zn-Pt couple is higher than that of Zn-Au couple, hence, the corrosion rate of Zn-Pt 3 .10.4 (FIG. EFFECT OF A R E A RATIO 3.13) The effect of anode-cathode area ratios has an important bearing on the rate of corrosion. This can be explained by a plot of log I vs E. In the plot (Fig. 3.13) current lis plotted vs Eand not i (current density) to establish the effect of area ratio. To define the conditions, the reversible potential £° of Zn (-0.760 V) and hydrogen (0.00 V) are located in the diagram. The values for hydrogen reduction on zinc, io (H) on zinc (—1 cm 2 ), i0, H on Zn kH on Pt io H on Pt Jm re(juctf0n ^ Log (1) — » Figure 3.13 Effect of cathode anode area ratio on corrosion of zinc-platinum galvanic couples Corrosion kinetics 8 3 hydrogen reduction on platinum, io(H) on Pt (10 cm2) are shown in Fig. 3.13. The intersection of hydrogen reduction curve (cathodic) on Zn and anodic oxidation curve of zinc gives icorr and £corr of zinc. The points of convergence of the cathodic reduction curves for Pt (1 cm2) and (10 cm2) with the zinc oxidation curve gives the icorr value for Zn coupled to Pt (1 cm2) and Zn coupled to Pt (10 cm 2 ), respectively. The icorr of uncoupled zinc is lower than the icorr of zinc coupled either to Pt (lcm 2 ) or Pt (10 cm 2 ). The icorr of Zn coupled to Pt (10 cm2) is highest. The corrosion potential of coupled platinum shifts to more noble values as the area of platinum surface is increased from lcm 2 to 10cm2. The above observations prove something very important. As the area of the cathode increases, icorr increases, hence, the rate of corrosion also increases. In short, the smaller the anode to cathode ratio as in the case of Zn coupled to Pt (10 cm 2 ), the larger is the magnitude of corrosion. A valuable rule: Avoid a small anode to cathode area ratio to minimize the risk of serious galvanic corrosion. and later oxygen is introduced into the solution. £ rev (02) is the reversible potential of oxygen and £rev(H2), the reversible potential of hydrogen, as shown in the above figures. The reversible potential of any metal is represented by £rev(M/Mz+). In the deaerated condition, the following reactions take place: Fe^Fe 2 + 4-2e (anodic) (if > ir) H2 <=*2H+ +2e (cathodic) (ir > if) The oxidation reaction polarizes iron in the anodic direction and the reduction reaction in the cathodic direction. The point of intersection gives iCorr- The magnitude of over-voltage from the reversible potential is given by r]c for cathodic polarization and by T)A for anodic polarization. Hydrogen is liberated and iron is corroded as Fe 2+ . Consider now an aerated condition. Oxygen is passed into the acid until it becomes saturated with oxygen. The reduction of oxygen is shown by the reaction, l/20 2 +2H++2e -> H 2 0 . In Fig. 3.14b, two partial reduction reactions are shown, i.e. 2H+ + 2 e ^ H 2 -02+2H+ + 2e->H20 3 .fo.s E F F E C T O F O X Y G E N (FIG. 3 . 1 4 ) Figures 3.14a and b show the reactions which occur when iron is placed in a deaerated acid L0g ( j) * Log (i) (b) Aerated solution (a) Deaerated solution Figure 3.14 Effect of aeration and deaeration 84 Principles of Corrosion Engineering and Corrosion Control or 02 + 2H20+4e^40H~ (oxygen reduction) In order to achieve either an anodic or cathodic process, it is necessary to apply an overvoltage (77). If AG* (activation energy) for the reaction is small, the over-voltage required is small and if it is large, the over-voltage required is also small. In the Evans diagram, the relationship between log I and n is observed to be linear for activation polarization: The summation of two partial reactions is shown in the figure. There are two cathodic processes and one anodic process. The exchange current density of oxygen on iron is very low (—10~10 A/cm2). Even on a metal like platinum the exchange current density of oxygen is very low. The reason is that charge transfer is slowed down considerably due to film formation on the metal surface by oxygen. The metal electrode under oxidizing conditions becomes a less efficient catalyst for reduction. Oxygen reduction on platinum is approximately ten times slower than hydrogen reduction. The intersection of cathodic reduction curve of oxygen with the anodic polarization curve is given by fcorr. The intersection of cathodic reduction curve with the oxidation curve for iron is given by the appropriate icorr (Fig- 3.14b). It is observed that upon aeration of the acid solution, rate of reaction (iCorr) increases and the corrosion potential (£Corr) shifts to more noble value. To summarize the results: (a) (b) (c) (d) (e) Fe is oxidized to Fe 2+ . Hydrogen is reduced. Oxygen is reduced. The rate of corrosion increases on aeration. The rate of corrosion decreases on deaeration. and r£ = bc\o%(^\ W where A is activation polarization, subscript a and c represent anode and cathode, respectively. As described in detail, in Section 3.1, the Tafel equation is n = a+blogI (iii) 3 .1 1 CONCENTRATION POLARIZATION In an earlier discussion pertaining to activation polarization it was assumed that dissolving metal ions move directly into solution and there is a plentiful supply of ions to be deposited during cathodic reduction. The above assumption is often not correct especially for oxygen reduction because O2 takes time to diffuse in solution to the corroding interface, and metal ions take a definite time to cross the double layer, hence, in oxidation and reduction reaction there exists a metal ion concentration across the double layer. In a polyelectrode system, two separate electrode processes occur: = M < * M 2 + + 2e H2^2H++2e (metal oxidation) (hydrogen reduction) which indicates a linear relationship (Figs 3.6 and 3.7). As long as there is no concentration built up, an E vs log J plot shows a linearity and it is activation controlled (called activation polarization). But if the current density is further increased there is a deviation from linearity. Here, the rate of diffusion is not fast enough to carry all the cations from outer Helmholtz plane (OHP) of the double layer by diffusion because the rate of generation of cations becomes very high at higher current densities. The concentration of ionic species is no longer at equilibrium and a concentration imbalance is created. The cations generated, therefore, accumulate and exert a repulsive force on the ions to be transported from the metal surface to OHP. A back emf is, therefore, created. This process, therefore, blocks the further progress of anodic dissolution. No further current can be carried by the cations, whatever the over-voltage is applied. Any amount of applied voltage at this stage has no bearing on the current, which has reached a limiting value. The current at this point is called 'limiting current, or Z'L,' at this stage the Corrosion kinetics current become independent of potential. The repulsive forces do not allow further oxidation and release of cations and the process becomes self-limited. It has been shown earlier that the rate of corrosion is controlled either by an anodic or a cathodic charge transfer process. However, if the cathodic reagent falls in short supply, mass transport becomes rate controlling. Mass transport control is much more noticeable for the cathodic oxygen reduction process (limiting O2 diffusion rate in the solution). Let us now consider a cathodic reaction. A limiting current density may be reached by a cathodic process, where the ions from bulk solution are unable to diffuse across the double layer to replace the ions that have been discharged. At this point, infinite concentration polarization is reached and the current becomes independent of potential (Fig. 3.15). The over-potential is, therefore, the sum of two components: (a) activation polarization, and (b) concentration polarization: *7Total = * 7 a + ^ 85 and in the bulk electrolyte x = B, C = CB. CB = bulk electrolyte concentration Q=o < CB, and Ex==o < £ B (for reduction) (3.46) where Q is the concentration at the electrode surface, and Q=o > CB and Ex=o > £B (for oxidation) (3.47) Applying Nernst equation ^conc = £ x=0 - £ B = - 7 7 In (3.48) nF m Fick's first law of diffusion dc dx T_ (3.49) (3.50) ( CB=X=O) X , k RT, ( k\ (3.45) 7T = A : l 0 g - + 2 . 3 — log 1 - ? * o nF \ i LJ Concentration polarization over-potential (rjc) is Where obtained as shown below: Consider a reduction process taking place on a metal surface. At the metal surface x = 0, Cx=o> i dQ , (3.51) / = — = — (in terms — nF dt of electrodes, Faraday's law) i = -DnF (CB-CX=O) (3.52) E volts p activation polarization concentration polarization ntotai = r}a + % iF = bclogii + H £ [ i o g ( l - J i io nF Vk Log (i) Figure 3.15 Combined activation and concentration polarization 86 Principles of Corrosion Engineering and Corrosion Control the corrosion potential shifts in the noble direction because of a decrease in cathodic over-potential caused by concentration polarization. Temperature. As the temperature rises, the thickness of diffusion layer is decreased and the corrosion current is increased. V elocity. T he higher the velocity, the less is concentration polarization. At a sufficiently high velocity, concentration polarization becomes zero because the ionic flux is now sufficient to maintain the surface concentration of ions at the electrode/electrolyte surface equal to the bulk concentration. At high velocity, the driving force for the concentration polarization becomes zero. The process at high velocity becomes totally activation controlled and there is no concentration polarization at high velocity (Fig. 3.16). C oncentration of ionic species. C oncentration polarization generally results from the depletion of ions in the vicinity of the cathode. When the concentration of species at the cathode reaches zero, Q = 0, the reaction is completely mass transport controlled. The higher is the concentration of species, the greater would be the concentration polarization. Geometry. T he geometry of fluid flow and the design of the cell (horizontal or vertical) affects concentration polarization. H ence, when Cx = o> t he concentration gradient is maximum, and i = i^. H ence, DnFCB *L = " (3.53) w here D = t hickness of diffused layer (mm) n = n umber of electrodes F = F araday, 96485 C(gequiv.) _ 1 C = c oncentration of electrolyte (molar concentration). Dividing C B — Cx=o by C B from equations (3.52) and (3.53) CB - Q;=o _ _ . x / DnF CB Cx=o _ z CB k -, D nP (3.54) (3.55) (3.56) i CB *L Replacing in the Nernst equation term W = ^ln^l--j £T / i\ (3.57) w here ZL is the limiting current density (A/cm 2 ), ' and J7Conc is concentration over-potential. " oxygen 3 .1 2 E F F E C T OF Joxy J \Q/ Increasing velocity V A R I O U S F A C T O R S ON CONCENTRATION POLARIZATION A gitation. By agitation, the thickness of the diffusion layer is decreased, and the rate of diffusion of ions increases. There is no build up of any concentration gradient between the corroding surface and bulk electrolyte. The end result is a decrease in concentration polarization and an increase in the rate of corrosion. As the rate of agitation is increased, c, 8. A, ew i I 1 !5 i LogO) Figure 3.16 Effect of velocity on limiting current density, for 0 2 + 2H 2 0 + 4 e ^ 40H~ Corrosion kinetics 8 7 3.13 RESISTANCE (OHMIC POLARIZATION POLARIZATION) (FIG. 3.17) It is known that current is transported from the anode to the cathode by ions in the electrolyte and in the metallic path from the anode to cathode. Because of the normal high conductivity of metals, almost no resistance is offered to the current flow in the metallic path. However, resistance can be encountered if the distance between the anode and the cathode is appreciable. The effect of ohmic polarization may be significant where this current flows from the anode to the cathode in an electrolyte, depending on the resistance (q) of the electrolyte. Consider an electrolyte in seawater with a low resistivity or high conductivity represented by R\. In the seawater, as electrolyte, the anodic and cathodic polarization curve would intersect at the point R\, where the potentials of the anode and the cathode are polarized at the same value. If the resistivity of the solution is high, a potential drop (IR) would result from the flow of the current, in the resistive solution and the potential of the anode and cathode would differ slightly in the case of seawater. The reason is that the potential drop (IR) diminishes the driving force of activation polarization and the anodic and cathodic reactions are not polarized to the same potentials. This situation is represented by R2 in Fig. 3.17, and further increasing the electrolyte resistance the magnitude of the ohmic drop would further increase as shown by R3 in the figure. With an increasing resistance offered by the electrolytes, the magnitude of the corrosion current would decrease as shown by R\, R2 and R$ in the figure. Incidentally, increasing the solution resistance offers a good method of controlling corrosion since decreasing corrosion current means decreasing corrosion rate. Also, painting the metal surface inserts a high resistance into the corrosion circuit and is also illustrated by Fig. 3.17 (ohmic polarization). 3.14 M E A S U R E M E N T OF CORROSION 3.14.1 CORROSION POTENTIAL AND CORROSION CURRENT When a metal, M, corrodes in a solution, there must be at least one oxidation and one reduction process. What is measured is the sum total of all partial cathodic processes and partial anodic processes occurring during corrosion of a metal. An anodic curve represents the sum total of all partial oxidation processes and a cathodic curve, the sum total of all partial reduction processes. The point of intersection of anodic and cathodic polarization curves in an Evans diagram gives the mixed potential £Corr (corrosion potential), also called the compromise potential, or mixed potential, or free corrosion potential, and the corrosion current (iCorr). An Evans diagram for a metal M is shown in Fig. 3.8. The lower limit of the potential in the diagram is given by the reversible potential of metal (£M/M2+) and the upper limit by the reversible potential of hydrogen ( £H+/H)- 1*9(0 Figure 3.17 Effect of ohmic resistance and current The anodic process is M^M^++2e 88 Principles of Corrosion Engineering and Corrosion Control working electrode and the positive to the reference electrode. The open circuit potential in the freely corroding state is shown by the voltmeter. The corrosion potential is also referred to as the open circuit potential as the metal surface corrodes freely. It is called mixed corrosion potential as it represents the compromise potential of the anodes and the cathodes. At the rest potential no drive force is applied. It is important to realize that a metal in an electrolyte will have a characteristic potential even if the salt bridge and voltmeter are removed. and the cathodic process is given by H2^2H++2e The corrosion rate is given by iCOrr (corrosion current). The current cannot be measured between local anodes and local cathodes on freely corroding metal surface, however, this can be measured as shown below. 3.14.2 M E A S U R E M E N T OF Ecorr (CORROSION POTENTIAL) The experimental arrangement for the measurement of corrosion potential is shown in Fig. 3.18a. The potential of the metal electrode (working electrode) is measured with respect to a standard Calomel electrode, which is non-polarizable. The reference electrode is kept in a separate container and it is connected electrically with the working electrode placed in a container in contact with the electrolyte via a salt bridge. A high impedance voltmeter is connected between the working electrode and the reference electrode. The negative terminal of the voltmeter is connected to the 3.14.3 M E A S U R E M E N T OF CORROSION C U R R E N T (iCOrr) Arrangements for polarization measurements are shown in Fig. 3.18b. In anodic polarization, an over-potential (E—Ee) is impressed in the noble direction starting from the open circuit potential (natural corroding potential without any impressed current). The over-potential is a measure of how far the reaction is from the equilibrium where the over-voltage, n> is zero. If the potential is made more positive than the equilibrium potential (irreversible potential) then the rate of forward anodic reaction if is greater than the rate of reverse cathodic ir reaction and metal dissolution continues. In the above case ia > | i c |. In cathodic polarization ia < \ic\. At small overHigh impotence voltmeter voltage both the anodic and cathodic reaction oppose each other, however, at a sufficiently large over-potential, only one reaction, either anodic or cathodic, takes place depending on which direction the potential is impressed. Consider a freely corroding metal in an acid. The anodic partial process represented by M -> M 2 + +2e intersects the cathodic partial process at -Ecorr? the corrosion potential. The current corresponding to £corr is i con The current between the local (microscopic) anodes and cathodes cannot be obviously measured by conventional means, such as by placing an ammeter. At £Corr (freely corroding potential), the current is icorr and the rate of forward process (if) is equal to the rate of reverse process if = ir. What can, therefore, be done to measure the curFigure 3.18a An experimental arrangement for makrent? An experiment can be designed to measure ing corrosion potential measurement Corrosion kinetics 8 9 Potentiostat Power supply Ammeter Electrometer Test electrode Counter electrode Reference electrode Figure 3.18b Circuitry associated with controlled potential measurements. A feedback circuit (not shown) automatically adjusts the voltage held between test electrode and reference electrode to any pre-set value the current as well as the potential, which are the necessary electrochemical parameters. Impress a potential E\ in the noble direction. Actual (theoretical) and measured (experimental) curves are shown in Fig. 3.18c. The corrosion potential (£Corr) at the intersection of the anodic and cathodic curves is shown in the actual diagram (Fig. 3.18c). This value is measured experimentally and plotted in the measured diagram. The potential lines E\, £2 and £3 intersect the anodic curves at z'i, ii and i$. As the overvoltage is shifted to £3, the current becomes completely anodic, as shown by z'3. These three values are plotted in the measured diagram. On connecting the three points, a measured anodic polarization curve is obtained. In a similar manner but in an opposite direction, a cathodic measured polarization curve is obtained on increasing the potential from £ corr to £1, £2 and £3 in the negative direction (active potential direction), points iCl, iC2, corresponding to cathodic currents similarly obtained in the measured polarization curve. The points are connected and a measured cathodic polarization curve is obtained. The portion of the curve near £corr shows an asymptotic behavior. The linear behavior is shown only above 50 mV If the linear portions of the measured anodic and cathodic polarization curves are extrapolated to £COrr> the point of intersection yields, icorr> the corrosion current obtained from the measured diagram corresponds accurately with the icorr in the actual diagram. In the previous discussion, it was shown that a net anodic current is generated when the potential is raised from £1 to £3 in the noble direction. As the potential is raised from E\ to £3, an excess of electrons is being generated because of oxidation. A platinum electrode or a graphite electrode is connected with the working metal and a voltage is impressed on it by a power source. An ammeter is placed in the circuit to measure the current. In anodic polarization, the platinum electrode is the cathode and the working electrode is the anode, whereas in cathodic polarization it is the reverse. With this arrangement, the metal can be anodically and cathodically polarized and the net current can be measured. In advanced corrosion measurement systems presently used, icon is automatically calculated 90 Principles of Corrosion Engineering and Corrosion Control Log (i) (I) (Actual) Log (i) (Measured) (II) Figure 3.18c Actual and measured polarization curves for active metals after the curve is recorded by the instrument upon the input of required parameters, such as the over-voltage (n), slopesfoa>K and scan rate. 3 . 1 5 . 2 ADVANTAGES AND D I S A D V A N T A G E S OF _ _ T AFEL T E C H N I Q U E S (1) The specimen geometry requires a strict control to obtain a uniform current. (2) The specimen is liable to be damaged by high current. (3) The Tafel region is often obscured by concentration polarization and by the existence of more than one activation polarization process, e.g. as seen in Fig. 3.10. 3.15 DETERMINATION OF C O R R O S I O N R A T E S BY ELECTROCHEMICAL MEASUREMENTS 3 .15.1 T A F E L METHOD EXTRAPOLATION For a complete description of Tafel plots refer to 'ASTM, Designation G3-74 (Re-approved In this technique, the polarization curves for the 1981), Annual Book ofASTM Standards, Volume anodic and cathodic reactions are obtained by 03-02, 2001.' applying potentials about 300 mVscE well away from the corrosion potential and recording the current. Plotting the logarithms of current (log I) vs potential and extrapolating the currents in the 3 . 1 6 POLARIZATION two Tafel regions gives the corrosion potential and RESISTANCE (LINEAR the corrosion current icon. A hypothetical Tafel POLARIZATION) plot is shown in Fig. 3.19. Knowing icorr> the rate of corrosion can be calculated in desired units by using Faraday's law. The theoretical relationship between a metal The modern techniques for measurement of cor- freely corroding potential (£), current density rosion rates are based on the classical work of (0 and the rate of corrosion was developed by Stern. Stern and Geary. Corrosion kinetics 91 Tafel slope = 171 mV/decade (active) 100 nA 1 pA 10 MA 2 100 MA Log current density (pA/cm ) F igure 3.19 A h ypothetical Tafel plot Polarization resistance (£ p ) of a corroding potential-current relationship is obtained. This metal is defined using Ohm's Law as the slope of a technique is quick and reliable. potential (E) vs current density (log i) plot at the corrosion potential (EC0TT). Here Rp = (AE)/(AI) at A£ = 0. By measuring this slope, the rate of corrosion can be measured. The correlation between 3 , 1 7 THEORETICAL icon and slope (dI)/(dE) is given by BACKGROUND AE AI hbc 2.3icorr(2?a + frc) (3.58) (ELECTROCHEMICAL MEASUREMENTS) In a corroding system: M ^ Mz++Ze~ X z+ +Ze~ -X2 (corroding metal) (1) where fra and bc are Tafel slopes. The potential-current density plot is approximately linear in a region of within ±10 mV of the corrosion potential. As the slope of the plot has the units of resistance, this technique is called 'Polarization Resistance' technique. The current measured in the external circuit equals the change in corrosion current, ±A7, caused by a small perturbation. As both anodic and cathodic reactions proceed in the vicinity of corrosion potential (Ecorr), they are exponentially dependent upon the applied potential. Over a small potential range (20 mV), however, the exponentials are linearized (because mathematically Logx—>>x as %—>*0) and an approximate linear (species in solution, such as hydrogen, oxygen) (2) The reversible potential (Erev) of the M in equation (1) is £rev> M and that of X in equation (2) is jErev, X. By increasing the over-voltage, a condition of irreversibility is created in the system. If £COrr is sufficiently removed from £rev(M)> m e reduction of M becomes insignificant compared 92 Principles of Corrosion Engineering and Corrosion Control to the rate of oxidation of M. M - ^ M Z + + Ze~ and the rate of oxidation of X becomes insignificant compared to the rate of reduction of X (species in solution). The corrosion rate zcorr becomes equal to either ZM or ix at £Corr' If the potential of a metal M is held at the next negative value, and is then made more positive, the linear (Tafel) part of this right-hand graph will be ascended. But on reaching Ejjj, an increasing current is added to the cathodic 2H + + 2e~ -> H2 line, but in the opposite direction, due t o M ^ M z + + Z e ~ starting. This increasing anodic current becomes equal (and opposite) to the cathodic current when potential £corr is reached and then the net (measured) current becomes zero. As the graph is of Log / (measured), the experimental line becomes asymptotic to the horizontal potential level of £corr> because log 0 is 00. Thus the experimental line becomes curved and cannot follow the cathodic line 2H + +2e—• H2 up to its origin at E^ o n m . Similarly, if the metal M were to be polarized to a very positive potential, starting from the top right point on the right-hand graph, it begins down a linear (Tafel) part of the curve, if made less positive, but deviation from linearity occurs as £corr is approached, due to an increasing current in the opposite direction starting at potentials below H2 on m ' 3.18 MODERN DEVELOPMENTS The development of microprocessor based measurement systems have expanded the capability of corrosion scientists, relieved operator's tedium, and improved the accuracy and speed of the data required for experimenting. In these corrosion measurement systems, corrosion rates are calculated by using constants for areas, equivalent weight and density and Tafel slopes which are measured or keyed-in. The microprocessor examines data on both the anodic and cathodic sides of the first corrosion potential to find the semi-log straight line segment which will yield a Tafel constant. The straight line segment is extrapolated to intersect the corrosion potential and the value of current density is utilized in computing the corrosion rates. A typical output of a polarization resistance plot obtained by a microprocessor based on corrosion measurement system is shown in Fig. 3.21. Data from potentiodynamic polarization plots can be replotted as polarization resistance. Once icorr is determined, the corrosion rates are calculated using Faraday's law as shown in Example 1. Example 1 From a given plot of polarization resistance (Fig. 3.22), calculate the rate of corrosion. Solution: Data obtained: Rp = 1.111 xl0 2 ohm-cm 2 , ft = lOOmV/decade, ft = lOOmV/decade Calculations: ^ __ _ 1 2.303£D Hence, ^ measured = *M — *X z= 0 a t -CCorr where Imeasured is the net current. To apply the linear polarization method for obtaining icorr> an important condition is that the over-voltage must be small. Only perturbations of up to ± 30 mV are allowed. The b value may be determined in separate experiments or obtained from literature. If fra = K = 0.1, the error will not be more than 20%. The polarization resistance plots are recorded from £corr to, say ±20 mV f r o m ECOTY. P A+flc J l.lllxlO2 [ ftfl For example, use a resistance of 0.1 ohm and let the applied potential on the instrument be E — 30mV(SCE). The maximum current available is now 300 mA. Record E and current density i. Figure 3.20 shows a hypothetical linear polarization plot. J 2.303 x r.100x100 200 = 1 9.54xlO" 5 A-cm" 2 = 1 .954xlO" 4" 2 C orrosion kinetics 93 f |(mV) Noble (+) , 'Slope = RP (~) Cathodic current density Figure 3.20 Hypothetical linear polarization plot Polarization resistance of Fe Volts -0.475 -0.485 .2 c -0.495 SAMPLE DATE AREA & EF MV/SEC 4 30 02,15 9.000 - 0.520 -0.480 0.166 -0.505 RESULTS RP ICORR -0.515 -266 MPY ECORR 1.75261 1.30366 6.02062 -0.496 -50 -40 -30 -20 -10 0 10 20 30 40 50 +266 Current density, jiA/cm 2 Current density, | iA/cm 2 Figure 3.21 Typical polarization resistance plot Figure 3.22 tance plot Output of a typical polarization resis- 94 Principles of Corrosion Engineering and Corrosion Control Although the above reaction is possible, it is more likely that Fe2C>3 (5) formed as a result of electrochemical oxidation at a sufficiently high Mi potential followed by interaction of F e 3+ by OH~ r = 0.00327— rjp ions, the latter being formed by cathodic reaction 0 2 + 2 H 2 0 + 4 e ^ 4 0 H ~ . Iron oxidizes as where Fe(s) —• F e 3+ (ag) + 3e. As suggested by equation i = current density (|xA/cm2) (1), the oxide formation is increased by increasp — density (g/cm3) ing the activity of O H - ions. When a surface film of an oxide or hydroxide develops, corrosion is M, Eq. wt. = 27.92 g/equiv. 3 either eliminated or retarded. The passive films p = 7.87 g/cm maybe as thin as 2-10 nm, and they offer a limited electronic conductivity, and behave like semi0.00327x27.92x195.4 conductors with metallic properties rather than Corrosion rate, r = 2x7.87 the properties shown by bulk oxides. The films also allow a limited amount of conductivity of r= 1.13 mm/year cations because of lattice defects and a slow anodic dissolution. Because of film formation, there is a reduction in the current density for cathodic 3.19 KINETICS OF reduction and an increase in the current density for an anodic polarization. The above factors lead PASSIVITY to retardation of corrosion. The important point is that once a film is 3 .19.1 I N T R O D U C T I O N formed, the corrosion rate sharply declines. The passivity on the metal surface which develops Faraday published a theory in which he suggested due to film formation on metal surface causes that the metal surface of certain passive metals inhibition of the anodic dissolution process. was oxidized and the oxide layer was very thin (Phil Mag., 9 (1836), p. 5). If a metal is converted to an oxide and the oxide which is formed is stable, the metal is considered passive as this oxide forms 3 .19.2 A C T I V E - P A S S I V E a barrier between the metal and the environment. M E T A L S A N D C O N D I T I O N S For example, iron is not attacked in concentrated F O R P A S S I V A T I O N HNO3, as a very thin film of an oxide is formed on its surface and causes loss of reactivity. A film Transition metals, such as Fe, Cr, Ni and Ti, of solid hydroxide or oxide may be precipitated demonstrate an active-passive behavior in aquefrom an aqueous solution if there are metal ions ous solutions. Such metals are called activein the solution passive metals. The above metals exhibit 5-shaped polarization curves which are characteristic of such metals. Consider, for instance, the case of Later, the hydroxide converts to oxide by giving 18-8 stainless steel placed in an aqueous solution of H2SC>4. If the electrode potential is increased away the water molecule: then the current density rises to a maximum, with 2Fe(OH)3 = Fe203(s) + 3H20(5) (2) the accompanying dissolution of the metal taking place in the active state. The current density assoIf there are no metal ions in solution the film may ciated with the dissolution process indicates the also be formed by chemical combination with magnitude of corrosion. At a certain potential, adsorbed oxygen in solution the current density is drastically reduced as the metal becomes passivated because of the forma2Fe(s) + |o 2 (ads) = Fe 2 0 3 (s) (3) tion of a thick protective film. Iron shows passivity F e 3+ (aq) + 3 0 H " (aq) = Fe(OH)3 (s) (1) Corrosion rate in mm/year can be found from *corr- Here 0.00327 is a constant for mm/year. Corrosion kinetics in acids containing S O ^ N O ^ C r O ^ - , etc. The above metals show a transition from an active state to a passive state and passivity occurs when the corrosion potential becomes more positive than the equilibrium potential of the metal and its metal oxide. The anodic dissolution behavior of a metal demonstrate an active-passive behavior as shown in Fig. 3.23. More details are shown in Fig. 3.24. The reversible potential of a hypothetical metal M is designated by £(M/Mz+)> oxygen by £ ^ ( 0 ) . The anodic and cathodic polarization curves in the active region intersect at £Corr> the corrosion potential. The anodic curve shows a Tafel behavior with a slope ba. Similarly, the cathodic curve can be obtained by deviation from £ corr in the active direction and the cathodic Tafel slope is given by bc. The point of intersection gives EC0TT. On polarizing the metal in the noble direction from £COrr> it is observed that as the potential increases, the rate of dissolution of the metal also increases. The highest rate of corrosion is achieved at a maximum current density, called critical current density (/critical)- 95 The lower portion of the anodic curve (nose of the curve) exhibits a Tafel relationship up to fcritiCai which can be considered as the current required to generate sufficiently high concentration of metal cations such that the nucleation and growth of the surface firm can proceed. The potential corresponding to /'critical is called the primary passive potential (£pP) as it represents the transition of a metal from an active state to a passive state. Because of the onset of passivity, the current density (log i) starts to decrease beyond £pP due to the oxidefilmformation on the metal surface. Beyondluminum, show a passive behavior in atmospheric corrosion. Corrosion kinetics Metals, like Fe, Cr, Ni and Ti, show a strong active-passive behavior. Whether passivity is achieved or not by the system depends on the cathodic reaction. Thus, cathodic reaction is a deciding factor in the establishment of passivity. A metal not showing any passivity will exhibit a linear E vs log i relationship. On the other hand, a metal exhibiting passivity would exhibit a nonlinear anodic polarization. Anodic polarization curves for active-passive metals, such as Fe, Cr or Ni, show a highly polarized passive region (Fig. 3.24). The rate of corrosion depends upon the degree of polarization of the anode, contrary to the observation in active metals where the rate of corrosion depends on the degree of polarization of both the anode and the cathode. 99 3.20.2 AND ACTUAL (THEORETICAL) MEASURED FOR POLARIZATION DIAGRAMS ACTIVE-PASSIVE METALS reduction in the current density occurs at a particular current density called the critical current density, iCritical- It represents the onset of passivation due to the formation of a barrier film of Fe2C>3 explained earlier. The current finally reaches a minimum value at a particular potential, the flade potential (Ep). Once it is reached the current does not show any response to any further increase in over-voltage until the reversible potential of oxygen is reached. The metal is in a passive state because of the formation of a barrier film of Fe2C>3 which restricts the electrical conductivity. The current required to maintain the passive layer is called the passive current density (i p ). Both ip and £p are shown in the actual diagram. The measured diagram follows completely the kinetics shown in the actual diagram. The measured anodic polarization curve approximates the actual anodic dissolution behavior over a wide range of potential. However, in the measured diagram, the starting potential for the anodic and cathodic processes (£M/MZ+ a n d ^ H ) are not shown. All the points in the actual diagram are identified in the measured diagrams. The actual polarization behavior is not generally obtained by anodic or cathodic polarization curves. At potentials more noble than 50 mV above £corr and below 0.8-0.9 volts, the measured curves correspond closely with the actual polarization curves. Locate £M/MZ+ (reversible potential of the metal) and £R in the actual diagram (Fig. 3.27). The diagram shows the exchange current density for metal oxidation io> M /M z+ and hydrogen reduction io (H on M). The anodic oxidation and hydrogen reduction curves intersect at £corr (corrosion potential) which is the first point on the measured diagram (Fig. 3.27). The starting point of the partial anodic oxidation (M < ^ M z + + ze) = and cathodic reduction (H2 < * 2 H + +2e) are not = shown in the measured diagram. As the overvoltage (77) is increased in the noble direction, the rate of dissolution of the metal is simultaneously increased. The dissolution kinetics observed initially is asymptotic, which becomes linear after 30-40 mV and shows a Tafel behavior. The slope isfoais the same for both the actual and measured polarization diagrams. As the over-voltage is further increased, the curve bends back to the left to very low values of current. The onset of the 3.21 M E A S U R E D VS ACTUAL POLARIZATION BEHAVIOR OF ACTIVE-PASSIVE METALS Consider three conditions which can arise when a cathodic process is superimposed on the curve. The three cases are discussed below to illustrate the polarization behavior of activation metals. The three curves representing different activation controlled process with different exchange current densities are shown in Fig. 3.27. The three reactions represent different rates of hydrogen reduction on the metal surface. The three exchange current densities associated with these reduction reactions are shown in Fig. 3.27 by *o(H)P *o(H)2 a n d *o(H)3Casel The cathodic curve intersects the anodic polarization in the active region (Fig. 3.27). The point of 1 00 Principles of Corrosion Engineering and Corrosion Control ULt m I 1 . j i * . - Active, *M/M»Cathodic Current density (nA/cm 2 ) Actual Applied Current density (j*A/cm2) Measured Figure 3.27 Three curves representing different activation controlled process with different exchange current densities intersection of the cathodic partial process with the linear Tafel kinetics gives £corr> the corrosion potential. It is the first point which is identified in the measured diagram. As the over-voltage is further increased in the noble direction, the measured curve follows the active-passive kinetics shown in the actual diagram. Initially, the curve in the measured diagram approaches asymptotically and later it follows the linear Tafel kinetics. The slope of the Tafel curve is given by ba. As the potential is further increased in the more noble direction, the measured curve follows the characteristics of the actual curve. Both the actual and measured diagrams have the same slopes ba> the only difference being that the measured curve approaches £corr asymptotically. On deviation from £corr in the more active (downwards) direction, the Tafel slope bc can also be measured as shown in the measured diagram. The above case is exemplified by titanium in deaerated H2SO4 or HC1 or Fe in H2SO4. Titanium has a more active primary passivation potential (£pP) than the reversible potential of hydrogen (£rev(H))« This example shows that a metal is stable in the active, active-passive or passive state. This condition is the least desirable because of the accompanying high rate of corrosion. Case 2 The partial cathodic process represented by io(H)2 shows a higher current density than the partial process (Fig. 3.28). The curve representing the partial cathodic process intersects the anodic polarization curve (actual) at three points, a, b and c (Fig. 3.28). The first point of intersection in the active region c corresponds to the corrosion potential (£Corr) which is the first point identified in the measured diagram. The other points of intersection are a and b at more noble potentials. It is, however, observed that although more and more over-voltage is applied in the noble direction, the cathodic process proceeds at a higher rate than the anodic process. The points of intersection a and b corresponds to a high magnitude of cathodic current rather than the anodic current Corrosion kinetics 1 01 Jo(H)2 S^ #_ _ ^ ^ r l \| Actual J Measured \ J ;. \ ICathodic loop a I B \N Metallic dissolution ^ s M/W* ^P> z * Current density (^A/cm2) Applied Current density (pA/cm2) x = some arbitrary value for current density Figure 3.28 Actual and measured polarization curve for a cathodic curve intersecting the passive region of an anodic polarization curve as shown in the actual diagram. The cathodic currents between the points a and c are identified as negative loops in the measured diagram. Hence between potentials a and c, cathodic loops are observed in the measured diagram. Below point c, the metal is in an active condition and above a it is in a passive condition. From point a onwards, the measured curve follows the characteristics of the actual curve in the passive region. The above case is typical for chromium in deaerated H2SO4 or stainless steel in H2SO4 containing oxides. Case 1 stated before is undesirable and Case 2 stated above is less desirable. Once the surface film becomes active it may not passivate again. Case 3 In this case the exchange current io(3) is higher than either of the two partial reduction processes represented by io(i) and io(2) (Fig- 3.29). The partial cathodic reduction curve represented by io(3) intersects the anodic polarization curve in the middle of the passive range. The point of intersection identifies the corrosion potential (£Corr) and it is the first point in the measured polarization diagrams. As the over-voltage is increased in the noble direction, the curve approaches £Corr asymptotically, followed by a quick shift in the passive range. The measured diagram shows the same characteristics as the actual diagram. There is no linear kinetics suggesting anodic dissolution observed in the measured diagram. The metal, therefore, simultaneously passivates. The above case is exemplified by Ti-Pt, Ti-Ni and Ti-Cu alloys. The above alloys do not undergo active dissolution and passivate instantaneously. Case 3 offers the best condition for passivity. 3.22 CONTROL OF PASSIVITY Two general rules can be applied to control passivity. If corrosion is controlled by an activation control reduction process (l^) an alloy which exhibits a very active primary potential must be selected. Conversely, if the reduction process is under diffusion control an alloy with 1 02 Principles of Corrosion Engineering and Corrosion Control / Io(H)3 \/ 1 Passive region s>^ I Transpassive Passivation Spontaneous passivation I 33 C > ^ Active region M»rr "Mm2" „ „ _ _ „ ^ 1 * Current density (jiA/cm2) Applied Current density (MA/cm2) # x « some arbitrary value for current density Figure 3.29 Actual and measured polarization curves for a cathodic curve intersecting the anodic polarization curve in the passive region a smaller critical current density must be selected Critical (critical current density) of iron (Fig. 3.31). (Fig. 3.30). Thus, if activation polarization is the Addition of up to 18% chromium reduces /'critical controlling factor, alloy 2 must be selected and iron. Similarly, addition of more than 70% nickel if concentration polarization is rate controlling, to copper reduces /'critical a n d /p. alloy 1 must be selected. The following factors have a major bearing on the passivity. 3 .22.2 C A T H O D I C R E D U C T I O N 3.22.i ALLOYING ADDITION Those alloying additions which decrease ^critical are effective in increasing the passivating tendency. Consider alloying additions of Mo, Ni, Ta and Cb to Ti and Cr. The critical current density of Ti and Cr is reduced on addition of Mo, Ni, Ta or Cb. The potential of the above elements is active and their rate of corrosion is low. Generally, those alloying elements are useful which show low corrosion rates at the active potentials. Alloying with metals which passivate more readily than the base metal reduces /'critical a n d induces passivity. Elements, like chromium and nickel, which have a lower /'critical a n< i ^passive than iron, reduce the Passivity can also be increased by increasing the rate of cathodic reduction rather than changing the behavior of the anodic polarization curve. When the reduction process is made faster, the metal spontaneously passivates. Metals having a high hydrogen exchange current density tend to passivate spontaneously without any oxidizer being added. If alloying additions of noble metals are to be made to the matrix metal, it is preferable to select metals with active £pP, such as titanium and chromium. In highly oxidizing conditions, the addition of a noble metal causes an increase in the rate of corrosion of matrix metal as exemplified by the addition of platinum to chromium. This happens C orrosion kinetics 103 -rev(H) § EM/M*+(2) ~M/M2+(i) Current density Figure 3.30 process Behavior of active-passive alloys in a system containing an activation controlled cathodic reduction 2.0 8 1.0 *crt$cti I mini 10s lo-i juuii i H Innl i n nml i i 1Q« mi Current density (pAfcm2) Figure 3.31 Effect of chromium content of anodic polarization curve of Fe-Cr alloys in 10% H2SO4 at 21°C 1 04 Principles of Corrosion Engineering and Corrosion Control if the alloying element has an exchange current density greater than the matrix metal and the mixed potential of the metal is in the transpassive region. Chromium-platinum alloys have a low resistance to corrosion in oxidizing acids and a high resistance in non-oxidizing acid. Noble metal alloying additions are beneficial in only weak or moderately oxidizing acids. In highly oxidizing conditions the corrosion potential of chromium is very near the transpassive region. If platinum is added to titanium, the rate of corrosion of titanium is not increased because, unlike chromium, titanium does not exhibit a transpassive region. If the mixed potential of the metal is close to the transpassive region, the rate of corrosion is increased. The alloying metal must be selected carefully. The above effect is illustrated in Fig. 3.32. Suppose platinum is coupled to titanium in an acid solution. The point of intersection of hydrogen reduction reaction and titanium oxidation, give icorr for titanium (uncoupled). There is no oxidation of Pt. The reduction curve for Pt is shown in Fig. 3.32. The total rate of reduction is shown by the broken line. It is observed that iCOrr for the Ti-Pt couple is much lower than zCorr of titanium alone. Titanium spontaneously passivates and the rate of corrosion is reduced. 3.23 E F F E C T OF ENVIRONMENT A. E F F E C T OF CHLORIDE ION CONCENTRATION Chloride ions damage the protective films and cause the metal surface to be pitted. Stainless steel is subjected to serious pitting by stagnant water containing a high concentration of chloride ions. Steel pipes are subjected to pitting in brackish water and seawater. The higher the concentration of chloride the greater is the tendency of pitting. The effect of chloride concentration on the passivation of steel is shown in Fig. 3.33. Chloride ions break down the passivity and increase the rate of anodic dissolution. B. E F F E C T O F T E M P E R A T U R E An increase in temperature generally decreases the passive range and increases the critical current density (Critical) • An increase of temperature decreases polarization and enhances the dissolution kinetics. Log (I) Figure 3.32 Spontaneous passivation of titanium by galvanically coupling to platinum Corrosion kinetics 1 05 ( Ecritfcal(D) ^—D Increasing S O ^ ^—C c O 13 M Ecitbcal(B) Ecritical(C) h- t 0.1M NaCI y Increasing CI* ^-~~B \ Ecrttical(A) U Log current density 1 Figure 3.33 Schematic representation of the effect of aggressive ions (CI )and inhibitive ions (SO4 ) on passivation of stainless steel C. V E L O C I T Y When a metal is under cathodic diffusion control, agitation of the electrolyte increases the current density and the rate of corrosion increases up to a certain point. At a particular velocity (critical velocity), the rate becomes activation controlled rather than diffusion controlled, and, hence, the rate of corrosion becomes independent of velocity (Fig. 3.34). Beyond a certain point agitation has no effect. As /'critical is exceeded, the metal attains passivity and corrodes only very slowly. Mxcygen Log current density D. O X I D I Z E R C O N C E N T R A T I O N Figure 3.34 Effect of velocity on the electrochemical behavior of active-passive metals corroding under diffusion control Addition of oxidizing agents, like Fe +2 or C rO^ ions significantly affect the corrosion rate of metals which exhibit passivity. In the case of an active metal, the rate of corrosion is gen- value of iCOrr is exhibited and the metal tends to erally increased by an increase in the oxidizer passivate spontaneously. On increasing the oxiconcentration. Consider six curves (1-6) repre- dizer concentration represented by curve 6, the senting the oxidizer concentrations (Fig. 3.35). rate of corrosion is increased as icorr is shifted It would be observed that the oxidizer concen- in the transpassive region. Oxidizer concentratration having a concentration represented by tion C3 and C4 produce a stable passive state curve 6, would allow passivity to be achieved that does not depend on oxidizer concentration. quicker, because at this concentration, the lowest The corrosion rate, therefore, remain low and 1 06 Principles of Corrosion Engineering and Corrosion Control Transpassive -M/M2+<2) Passive Active Log current density Figure 3.35 Effect of oxidizer concentration on electrochemical behavior of an active-passive metal constant at concentration C3. As the oxidizer concentration increases, the transpassive range is intersected resulting in a sudden increase in the rate of corrosion. (4) mg dm 2 day l to mils per year x (1.144)/p. (5) mils per year to mg d m - 2 d ay - 1 x (1.144)/p. (6) mg d m - 2 d ay - 1 to mmy" 1 x (36.52)/p x mdd. (7) mdd to ipy x (0.365)/p. (8) gmd to mm/year x (0365)/p. Also remember: 1 mpy = 0.0259 mm/year = 25.9 |xm/year = 2.90 nm/year 1 cm year -1 = 0.0394 mpy Table 3.3 summarize the above conversion factors. 3,24 CONVERSION FACTORS (a) For corrosion rates (1) To convert g/m2/day to mm/year, multiply by 3 65/d. (2) To convert mm/year to mdd, multiply by 27.4 d. (3) Inches per year to mils per year x 1000. Table 3.3 Unit (1) (2) (3) (4) Summary of conversion factors mdd 1 27.4 x d 0.696d 696d mm/year 0.0365/d 1 0.254 25.4 mils/year 1.144/d 39.4 1 1000 in/year 0.00144/d 0.0394 0.001 1 Milligrams per square decimeter per day Millimeters per year (mm/year) Mils per year Inches per year Note, mdd = mg/dm2/day, d = day Corrosion kinetics (b) For current density (1) To convert mA/cm2 to A/m2, multiply by 10. (2) To convert |xA/cm2 to A/m2, multiply by .010. (3) To convert A/cm2 to A/m2, multiply by 10 000. (c) Relationship between magnitude of current and rate of penetration of corrosion The relationship between the magnitude of current (current density, |xA/cm2) and the rate of penetration of corrosion (mpy) is important. The rate of corrosion of different metals and alloys can be equated with 1 |xA/cm2 of current generated by a corroding metal as shown in Table 3.4. For instance, a conversion rate of 0.540 mpy for 316 stainless steel and 0.52 mpy for type 304 stainless steel corresponds to 1 |xA/cm2 of current. Such factors are called electrochemical conversion factors. They can be calculated for any desired alloy or metal as shown below. Consider, for example, conversion factor for AISI type 316 steel (non-magnetite). The following is the nominal composition of the steel: Cr-18% Ni-10% Mo - 3 % Mn - 2% Fe - balance luA/cm 2 = 0.128 9 1 07 + { (2)(7.80) J ^ ( m p y ) = 0.169+0.0442+0.00945 +0.018 + 0.318 = 0.559 mpy 1 |xA/cm2 = 0.56 mpy / 0.128(55.85) \ 0.70, 3.25 ILLUSTRATIVE PROBLEMS Problem 1 From an anodic polarization diagram, the value of korr recorded is 4.21 x 10 7 A/cm2. Find the rate of corrosion in (a) mdd, and (b) in inches per year. Solution: Applying Faraday's law: M = — x 1 nF 4.21 x l O - 7 A/cm2 55.8g/mol lOOOmg X 96490A.s./mol 2 g [In S.I. units, atomic mass = 55.8 x 10~3 kg/mol] = 1 .21xl0- 7 mg/cm 2 /s (b) Convert to mg/dm2/day 1.21 x 10~7mg/cm2/s x 3600 s/h x 24h/day x 100 cm 2 /dm 2 = 1.052 mdd / at.wt. of the element \ \ valence x density / x percent of alloy x mpy 0.128(52.01)\ 0.18 1(7.1) 7"Cr" (c) To get the answer in inches/year , , 0.00144 mdd x = 1.044 x 0.000183 7.87 = 1.924 x 10~4inch/year The above conversions can be done in S.I. units as shown below. It is suggested that S.I. units be used in all further problems. Unit (Dimensions) Density, kg m - 3 Current Density, Am" 2 Conversion factor 1 g/cm3 = 103 kg/m3 1 A/cm2 = 104 A/m2 •( N /'0.128(58.69) 0.10 0.128(58.69)\ + V (2X8.9) ) "NT (2)(8.9) + 128(54.94) \ 0.C 1.02 Mn (23)(7.43) ) M 128(95.95) \ 0.03 + ( - (2X10.2) / Mo" The atomic mass (amu) is given in kg/mol and molality is given in mol/kg. 1 08 Principles of Corrosion Engineering and Corrosion Control Table 3.4 M etal/alloy Electrochemical and current density equivalence with corrosion rate E lement/oxidation state D ensity ( g/cm 3 ) E quivalent weight Penetration rate equivalence t o 1 | xA/cm 2 ( mpy) Pure metals I ron Nickel Copper Aluminum Lead Zinc T in T itanium Zirconium Fe 2 + Ni2+ Cu2+ Al 3 + Pb 2 + Zn2+ Sn2+ T i 2+ Zr 4 + 7.87 8.90 8.96 2.70 11.34 7.13 7.3 3.51 6.5 2.71 2.68 2.71 2.70 2.80 8.96 8.39 8.39 8.52 8.33 7.9 7.9 7.9 8.0 7.7 8.89 8.84 8.51 8.14 27.92 29.36 31.77 8.99 103.59 2.68 59.34 23.95 22.80 8.99 9.05 8.98 9.01 9.55 31.77 32.04 32.11 32.00 30.29 25.12 25.13 24.62 25.50 25.30 29.36 30.12 26.41 25.52 0.46 0.43 0.46 0.43 1.12 0.59 1.05 0.69 0.75 0.43 0.44 0.43 0.43 0.44 0.46 0.49 0.49 0.48 0.47 0.41 0.41 0.41 0.41 0.42 0.43 0.44 0.40 0.50 A luminum alloys AA1100 ' Al3+ A l 3 +, M g 2 + AA5052 Al3+,Mg2+ AA6070 Al3+,Mg2+ AA6061 A l 3 +, M g 2 + , Z n 2 + , C u 2 + AA7075 C opper alloys C DA110 CDA260 CDA280 CDA444 CDA687 S tainless steels 034 321 309 316 430 N ickel alloys 200 400 600 825 Cu2+ Cu2+,Zn2+ Cu2+,Zn2+ Cu2+,Sn4+ Cu2+,Zn2+,Al3+ Fe2+,Cr3+,Ni2+ Fe2+,Cr3+,Ni2+ Fe2+,Cr3+,Ni2+ Fe2+, Cr3+, Ni2+, M o 3 + Fe2+,Cr3+ Ni 2 + Ni2+, Cu2+ Ni2+,Fe2+,Cr3+ Ni2+, Fe2+, Cr3+, M o 3 + Source: Adapted from Proposed Standard, ASTM G 1.11. O Corrosion kinetics Problem 2 An electrode has a potential of —0.80 V relative to a 1 N Ag Silver-Silver chloride electrode. What is the electrode potential on the hydrogen scale? Solution: AgCl + e^Ag + C i~£ 0 = - 0.22 V y±0.00lNKC\ = 0.901(y = activity coefficient) [flcl_] = (0.01)(0.901) = 9 x l 0 " 3 Applying Nernst equation: £Ag = E°- 0.0592 log [acr ] n=1 ic = ioexp 1 09 Problem 4 Consider the reaction M — M z + + ze on an elec> > trode surface. An over-potential of—0.155 V (rjc) is applied to the electrode. The exchange current density (*O,M/MZ+) a t 25°C is 5 x 10~7A/cm2. Determine the current density, i, if the change in the oxidation state is one unity. Solution: •(l-flzFi/cl RT J P = 0.5 (symmetry factor) C Z = 2, F = 9 6500—-, ?7c = 0.155V mol £ = 8.314 (gas constant) T = 2 5+273 = 298K, = 0.222 - 0.0592 log [9 x 10"3] = 0.343 V The potential on hydrogen scale -0.80 + 0.343 = -0.457 V r ( i -MzFricY] RT -0.5) x2x96500x ( -0.155) 8.314x298 ic = ioexp[6.037] ic = 5 xl0~ 7 xexp[6.037] ic = 2.093 x l O - 4 A/cm2 xio Problem 3 The potential of a platinum cathode at which hydrogen is evolved is —0.85 V relative to a saturated calomel electrode. If the pH of the aqueous electrolyte is 2.00, determine the hydrogen over-potential (T/C). Problem 5 A nickel electrode is corroding in a deaerated elecIt is necessary to convert the potential given, trolyte which has a pH of 3.0 and a concentration —0.85 volt (SCE) to the hydrogen scale. The of nickel ions of 0.003 at 25°C. conversion factor is (a) Determine the zcorr of nickel. SHE = SCE + 0.242 (b) Determine also £CorrSHE = - 0.85 + 0.242 (0.242 is the potential of standard calomel electrode) The following data is provided = - 0 . 6 0 8 V (SHE) The over-potential r]C = E-Erey = - 0 . 6 0 8 - (-0.11840) =-0.4896 V io(HonNi) = 10"6A/cm2 i o(OonNi) = 10-14A/cm2 fra = 0.04V/decade 2?c = -0.13V/decade Solution: £rev(H) = - 0.059 x pH = - 0.592 x 2 = -0.11840 V 1 10 Principles of Corrosion Engineering and Corrosion Control Problem 6 The rate of corrosion of a steel pipe in Arabian Gulf water is 3.5 gmd. The corrosion proceeds mainly by the reaction 40H~ <=t 0 2 + 2 H 2 0 + 4e~. The pH of water is 8.0. The pipe needs to be protected cathodically. Calculate: (a) The voltage required for complete protection. (b) The minimum initial current required for protection. Solution: (a) Find out the mass in kg/m2 • s m = 3.5 gmd x lday lh x -7—r- x 1000g 24h 3600s 1kg Solution: Reaction: Anodic: Ni -> Ni 2 + +2e Cathodic: H 2 -> 2H++2e Total: N i + 2 H + -> Ni 2 + +H 2 F = - 0.25 V (a) £ N i = £° + _ _ l o g ( 4 + ) = -0.25+0.0296log(0.003) =-0.325 V (1) 7 A = £corr-£rev(Ni) = ^ a l o g ( - ^ j ? (2) nc = Ecovx-Eu2 = bc\o%[^- J EH2 = - 0.0592 xpH = -0.0592x2 = -0.1184V rjA = £ corr (-0.325) = -0.04log x 10" 7 + 0.04logi corr £corr = - 0 . 0 4 5 + 0.04 log icorr ^C = £corr-(-0.1184) = 0 . 1 3 l o g i x l 0 ~ 6 - 0 . 1 3 l o g iconi c =£corr+0.1184 = 0.13 log XlO" 6 log W £corr = -0.828-0.13logi C O rr (2) = 4.051 x lO" 8 kg/m2-s (b) Find *corr from Faraday's law: (1) mxnxF Icorr == 'At. mass 4.051 x 10 - 8 x 96490 (A.s./mol) ~ 0.055847kg/mol where m = mass (kg) n = number of moles of electrons F = Faraday (coulombs/mole) W = 0.140 A/m2 Taking & = 0.08 V/decade * o ; ^ = 10-4A/m2 iCFe(OH)2 = 1.64 x 10"14mol/liter [H + ] = 1 x 1 0 - 8 - » [OH - ] = 1 x 10~6 1.64 x l O - 4 [Fe 1 = — = 0.0164 L J (lxl0~6)2 aFe = 1.0(y)x 0.0164 = 0.0164 r 9,, 2+ Equating equations (1) and (2), we get - 0.045+0.4 log icorr = -0.898 - 0.13 log fCorr *corr = (-5.0176) 1 0 i corr = 9.6 x l O - 6 A /cm 2 (b) By substituting the value of icorr either in equation (1) or (2), £corr can be obtained £ corr = - 0.045+0.04log9.6 x 10 - 6 = - 0.246 V(SHE) r]A = Econ - £ Fe 2+ = j8a l og -^ Corrosion kinetics , / 0.140 \ From Faraday's law: *corr x -A-twt 1 11 £corr = "0.493 + 0 .08log ( - ^ Z I J £ corr = —0.241 V, which is the required voltage. AE (in) (C) S i n c e — = l o g ^ Pa *corr m- nxF 0.5478 A/m2 x 0.055847kg/mol m2x96490(A.s./mol) m = 1.585 xl0~ 7 kg/m 2 -s The rate of corrosion in mm/year can be determine as below: CR (mm/year) = 1.585 x 10~7kg/m2 -s 1 7 .87xl0~ 3 (kg/m 3 ) 103mm x x lday lm 365 days x lyear 86400s -0.2412+0.493 +log0.14 = logi p O08 ip = 195.94 A/m2, which is the minimum current density required for complete cathodic protection. Problem 7 An iron pipe is used for transporting of 1N H2SO4 (pH = 0.3). The rate of flow of acid in the pipe is 0.3 m/s at 25°C. The relationship between the limiting current density Z'L and velocity is given by Z'L = V0,5. The following additional information is provided: ft = 0.100 ft = -0.060 . / Fe \ = 1.585 xl0~ 7 kg/m 2 -s The rate of corrosion in mdd: 1.585 x l 0 _ / k g / m z - s x lm 2 100 dm2 : 136.95 mdd 3600s lh 106mg 1kg 24h lday "W* = 10"5A/m2 io(H) = 10~2A/m2 If the surface acts as the cathode, determine: (a) The corrosion potential of iron. (b) The corrosion rate of iron in mm/year. (c) The corrosion rate of iron in mdd. Solution: icon = ( 0.3)^ = 0.5478 A/mz 77c = £ c o r r - £ H 2 - A : l o g -^ Problem 8 Calculate the concentration over-potential for silver depositing at a rate of 3 x 10~3g d m - 2 from a cyanide solution at 25°C. The limiting current density is 3A/dm2. The concentration over-potential is given by W = -^-log^l--j Solution: First to obtain i 3 x 10~3g.A.s dm2 x 60s x 0.00118 g = 4.23 Am" 2 (£corr)Fe = "0.0592 X 3 - 0.0592 X 3 - 0.06 log = -0.122 volt 0.5478 1 x 10~2 ] 1 12 Principles of Corrosion Engineering and Corrosion Control Problem 10 The maximum corrosion current density in a steel sheet coated with zinc and exposed to seawater is found to be 5mA/m 2 . What thickness of the layer is necessary so that the coating may last for one year? Solution: Consider lm 2 of steel coated with zinc. Suppose the steel sheet is (X) meters (thick). The mass lost is = 7.13 mg/m2 x (1 m 2 )(t x ) = (7.13X)mg. The corrosion current can be determined as , (7.13X)mg 106g * 9 —- x 5 x 10~3 A/m2 x l m 2 = lyear mg 0.6023 x lO 24 atoms x (one mole of Fe produces 2 moles of electrons) 65.38 g 0 .16xlO" 18 C X X (0.00118 g of silver is liberated by 1 coulomb) iL = 3 A /dm - 2 = 300 Am" 2 * = W = 6i0 b( g ^ l - -— \j = 60 log[l-0.0141] = 60log(0.986) = -0.37mV Problem 9 What volume of oxygen gas at STP must be consumed to produce the corrosion of 200 g of iron? Solution: Fe^Fe2++2e 4 23 2e x • tom a \A C/s lyear (365)(24)(3600) 40H"->02+2H20+4e (V2 mole O2 = 1 mole Fe) mole O2 gas = 200 g 55.85 gFe/mol-Fe l/2mol-0 2 1 mol • Fe 1 .89mol0 2 Using ideal gas equation nRT V= P X = 7.50xl0"°m = 7.50|xm Problem 11 An over-potential of 250 mV is applied to platinum in an aqueous and electrolyte. What would be the rate of hydrogen evolution if the exchange current of hydrogen on platinum is 1 x 10~4 A/cm2 and 0 = 0.5? Solution: RT RT . RT , . RT , . in ic = In tn In ir fizF fizF fizF fizF 0.059 , 1 x0.5 "°' . w , lX = (8.314J/mol) 8.314J/molx(273fc) (1 atm) (1 Pa/9.869) x 10" 6 atm r1\ = 0.059 , 1 x0.5 . = 0.042 J/Pa = 0.042 J/Pa = 0.42 N • m/Nm2 = 0.0201m3 = 0.12logio-0.12logi c logic = 0.25 = 0.12logx(-4)-0.121og*c logic = -0.23 • =.1.2 xlO" 2 A/cm2 0.12 Corrosion kinetics 113 QUESTIONS A. M U L T I P L E C H O I C E QUESTIONS Select one best answer: (More than one answer may be correct in some question.) 1. Which one of the following reactions is a reversible reaction? [ [ [ [ ] ] ] ] M-»Mz++ze H2^2H++2e M^M3++3e 2H+ + 2 e ^ H 2 The above equation is called [ ] Guoy-Chapman equation [ ] Nernst equation [ ] Tafel equation [ ] Butler-Volmer equation 6. On the basis of corrosion kinetics iron corrodes faster than zinc. Which of the parameters given below is used for prediction? [ ] The exchange current density (io H onFe) [ ] The exchange current density (io(H) Zn/Zn 2+ ) [ ] E° Fe/Fe2+ and £° Zn/Zn 2+ L J * corr 2. When a current flows as a result of impressing a voltage in the noble direction, the impressed current is [ ] net anodic [ ] net cathodic [ ] mixed current 3. The term over-voltage is given by [ ] ri = E-E° [ ] r] = E-ECOTT L J ' H = A :orr ^corr (Fe) and iCorr(Zn) 7. One corroding metal (M) is connected to another corroding metal (N) in an acid electrolyte. Metal M has a relatively more noble potential EM/M+ and iCorr(M) while metal N has a less noble potential EN/N+ and corrodes at arateof icorr(N). [ ] The corrosion rate of metal M is decreased whereas the corrosion rate of metal N is increased [ ] The corrosion rate of metal M is increased whereas the corrosion rate of metal N is decreased [ ] The corrosion rate of the couple M-N is less than the corrosion rate of metal M and N individually [ ] The corrosion rate of metal N is higher than the corrosion rate of the couple 8. Consider the corrosion behavior of zinc (lcm 2 ) coupled to platinum (1cm 2 ) and platinum (10 cm2) in an acid electrolyte. [ ] Zinc coupled to platinum (10 cm2) will show the highest rate of corrosion [ ] Zinc coupled to platinum (lcm 2 ) will show the highest rate of corrosion [ ] n = E°-E 4. When r\ = 0, and E = Eeq [ ] h > -k > h [ ] h <k <* o ic = io **=' -'o-*o 5. The polarization curve for the forward anodic reaction follows the general equation i=iv\expl\p—r]A - ioexp - ( l - j 8 ) —?fe J J 1 14 Principles of Corrosion Engineering and Corrosion Control [ ] Zinc uncoupled will show a higher rate of 13. Activation polarization is present corrosion than zinc coupled with 1 cm2 of platinum [ ] at low reaction rates [ ] at high reaction rates [ ] Zinc uncoupled will show a higher rate of corrosion than zinc coupled with 10 cm2 [ ] when the limiting current density is of platinum surface reached 9. Which of the following statements is not true about the linear polarization technique? [ ] Measurements begin at — 20 mV from OCP and end around +100 mV from OCP [ ] The slope of the plot is AE/AC is given in volts/amperes or mV/mA [ ] This technique can be used to determine very low rate of corrosion [ ] It can be used to monitor corrosion rate in process plant 10. Passivation occurs when [ ] the corrosion potential (£Corr) becomes more positive than the potential corresponding to the equilibrium between the metal and one of its oxides/hydroxides [ ] £corr<(£eq)M/M0,M/M0H [ ] when an oxidizing is introduced in the corroding system 14. In activation polarization [ ] the current varies linearly with the applied potential [ ] the current varies exponentially with the potential [ ] up to 50 mV, a linear relationship between E and log lis observed followed by an exponential relationship [ ] none of the above is observed in a plot indicating activation polarization 15. Concentration polarization occurs when [ ] the concentration of electroactive species at the metal/electrode surface and the bulk solution is the same [ ] a concentration gradient is built up between the electrode/electrolyte interface and the bulk solution [ ] the rate of formation of an ion is balanced by the rate of its arrival by diffusion from the outer Helmholtz plane [ ] the solution is continuously stirred or agitated L J A:orr > -^passive [ ] £corr = Eeq, M / M 0 11. The most important criteria to compare passivity between two metals are []k L J ^critical ^passive *corr LJ I J B . How AND WHY Q U E S T I O N S 1. Explain why: 12. Cathodic polarization takes place when: (rj = over-voltage) [] [} rj>0 n<0 a) Only the atoms at the kink sites are chemically active. b) Anodic polarization proceeds when the over-voltage applied is greater than zero. c) No net currentflowswhen rate of forward reaction (if) is the same as for the reverse reaction (i r ). I ] fl = 0 [ ] r) > £ rev Corrosion kinetics d) The rate of charge transfer by the cathodic process i is faster than by the anodic process when a negative voltage is impressed on the metal. e) Cathodic polarization proceeds when the over-voltage is less than zero. 2. Answer the following questions with regard to the energy vs distance profiles. a) What renders the dissolutions of a metal M more rapid when the reaction is irreversible and why the reverse process does not take place? b) What causes a decrease in the free energy of activation required for the forward process of dissolution? c) The general form of an anodic dissolution process is given by * = *oexp RT -*oexp -(l-P)zr]F RT 5. Answer the following questions: 115 a) Why anodic polarization occurs when 7A > 0 and cathodic polarization takes ? place when 77c < 0? b) Why the relationship between E vs log I is exponential if concentration polarization occurs? c) Why activation polarization takes place at low applied over-voltages? d) Why activation polarization is not affected by agitation? 6. a) If i£et = if = I ir I write the full expression for anodic activation. b) When rjA > 0.03 V, write an expression for ianet. c) If i j e t = | i r | —if, write an expression for cathodic activation. d) If ia = io exp Rjk write an expression for t]A. e) As asked in Problem (d), write an expression for in terms of rjc (consult energy distance profiles in the text). If the irreversibility of the electrode pro- 7. Why iron corrodes faster than zinc in an acid electrolyte, according to kinetics? What is the cess is high (rj = 0.05 V), why the second basis of this prediction and why it is different term from the above equation is dropped? from the prediction based on thermodynamic potentials? 3. Answer the following question with regard to polarization. 8. When zinc is galvanically coupled to platinum: a and b ? a) What is the numerical value of 2 3RT\ ) a) Why the rate of corrosion of zinc coupled with platinum increases? (Consult 2 3RT the diagram in the text.) c) What is the significance of — ———-—? b) Why the rate of corrosion of zinc couWhy the polarization term r] is multiplied pled to Pt (1 cm2) is lower than the rate by zF\ What are the units for zF ? of corrosion of zinc coupled to platinum d) Why the current density is a linear func(10 c m 2 ). tion of over-potential (rj) in activation c) Why zinc coupled to gold corrodes faster polarization? than zinc coupled to platinum? ( 4. a) What is the significance of i£et = if — | i r | ? k b) Consider the reaction M+^Mz + ze. If AGX = AG* + [(l-p)FrjrA]-zFrjAy show that if = io exp l " ^ ] . 4 9. What is the effect of the following on concentration polarization? a) Agitation b) Temperature 1 16 Principles of Corrosion Engineering and Corrosion Control 13. Answer the following questions related to passivity of metal: a) Why zinc does not show an activepassive behavior whereas iron shows such a behavior? b) Why the current starts to recede as soon as the critical current density (^critical) is reached as observed in a 5-shaped anodic polarization diagram for active-passive metals? c) What is the function of passive current density (ip)? d) What is the significance of transpassive region? What reaction generally takes place in this region? e) What reaction is likely to take place above the reversible electrode potential of oxygen? f) What would be the effect of addition of 10-15% chromium on the Flade potential (£p). Distinguish between the following: a) Aeration and deaeration. b) io and /criticalc) £p and £pP. d) Outer Helmholtz plane and the diffused layer. e) Active region and passive region in a polarization plot of an active-passive metal. c) Velocity d) Concentration of species 10. State the reasons for the following questions on concentration polarization: a) For small shifts in potential, charge transfer is completely controlling and the over-potential is purely an activation over-potential. b) For larger shifts in potential, the current is less than expected. c) If sufficiently large shift, the current becomes independent of potential. 11. A metal surface exposed to an aqueous electrolyte corrodes. The metal has several tiny anodic and cathodic areas. a) Why it is not possible to measure the current between a local anode and a local cathode? b) How the corrosion potential is measured experimentally? c) When a net anodic reaction is generated, 14. electrons are released. How are these electrons consumed? What is the method which is used to collect these electrons. d) Why an auxiliary electrode is used in the setup for measurement of corrosion current? What is its function? e) What are the functions of the reference, working and the auxiliary electrodes? 12. Explain why: a) It is essential to polarize the electrode sufficiently away (at least 100 mV) either in C. C ONCEPTUAL Q U E S T I O N S the anodic direction or cathodic direction 1. Explain clearly the difference between in the linear polarization method. activation polarization and concentration b) Tafel relationship is dependent on activapolarization. Under what conditions would tion control method and not on diffusion activation polarization change to concentracontrol. tion polarization? c) A luggin probe is used in the polariza- 2. Explain clearly the difference between the tion cell and the reference electrode is inner Helmholtz plane, outer Helmholtz plane not generally introduced directly to the and diffuse layer in electrical double layer. electrolyte. 3. Explain on the basis of kinetics why zinc corrodes more rapidly in an aerated solution of d) The Tafel method is applied to systems HC1 than in a deaerated solution. containing one reduction reaction only. Corrosion kinetics 4. Explain the effect of the following on the anodic polarization curve for active-passive metal: a) b) c) d) Effect Effect Effect Effect of adding an oxidizer. of chloride in concentration. of temperature. of addition of inhibitors. 117 tank cathodically by using an external current. Estimate the minimum current density (A/m2) and the voltage required for complete cathodic protection of the storage tank. Assume that corrosion occurs by oxygen depolarization. The following constants are provided: £ a = 0.06V/decade 5. Explain how the passivating tendency can be increased by the following. Give examples. a) Increasing the rate of cathodic reduction. b) Reducing the critical current density. c) Noble metal addition. D. PROBLEMS Fe *> ^ 2 T = 1 0- 2 A/cm 2 o iC,Fe(OH)2 = 1 .64xlO" 14 mol/l 6. A bimetallic platinum copper couple is immersed in an acid solution at 25°C and oxygen is passed into the solution rapidly. Calculate the anodic current density on the copper assuming that the corrosion rate of uncoupled copper in the same solution can be neglected and that the area of platinum and copper are 10 and 1 cm2, respectively. The solubility of oxygen in the solution is 1.4|xmol/cm3. The diffusion coefficient of oxygen is 1.75 x 10 _ 5 cm 2 -s _ 1 . The thickness of the diffusion layer is 0.05 cm. The Faraday's constant is 96 500 C. 7. Convert 0.122 mg/mm2/second to a) mg/dm2/day b) mg/m2/day c) mm/year 8. A tin immersed in seawater shows a current density of 2.45 x 10~6 A/cm2. What is the rate of corrosion in mdd? 9. In a polarization resistance experiment, an applied over-voltage of 10 mV results in a current density of 5 mA. a) What is the current density for corrosion iffoa= 0.06 V and bc = - 0.12 V? b) Using Faraday's law, calculate the rate of corrosion in mdd. 10. In a linear polarization experiment, a current increase of 10|xA for a voltage increment of 4 mV is recorded. The area of the specimen 1. The potential of a cathode at which hydrogen is generated is —0.80 volts with respect to a standard Calomel electrode. If the pH = 4, determine the hydrogen over-potential. 2. An electrode has a potential of —0.70 V with respect to a silver-silver chloride in 0.01 N KC1 at 25°C. Determine the potential on the hydrogen scale. 3. If the anodic over-voltage of an electrode is 0.12 V, and the exchange current density io> Fe/Fe2+ = 10~6A/cm2, calculate the anodic current density. 4. From the following data on the corrosion of a nickel electrode in an acid electrolyte (pH = 2.0), calculate a) icon- of Ni. b) EcorrOfNi. Data: 1) io(HonNi) = 10- 6 A/cm 2 2) fo(H) = 10- 14 A/cm 2 3) 2?a = 0.04 4) bc = - 0.13 5) aNi++ at 25°C 5. A steel tank used for the storage of water of pH = 7 corrodes at the rate of 10mg/dm2/day. It is desired to protect the 118 Principles of Corrosion Engineering and Corrosion Control is 10 cm 2 . Calculate (a) the polarization resistance, and (b) the rate of corrosion. Assume fca = 0.01 V, bc = 0.02 V/decade. 1 1. A m etal M has been anodically polarized to 0.50 mV during anodic polarization. The following values for electrochemical parameters have been obtained: 2?a = 0.20V, bc = - 0 . 2 V , and I = 20mA/m 2 . Determine the current density. 12. W hat would be the rate of corrosion of copper in aerated seawater (pH = 7)? The solubility product for Cu(OH)2 = 5.6 x 5 x l(T20mol/l. 13. T he value of zcorr o btained for iron corroding in an aqueous solution (electrolyte) by the Tafel extrapolation technique is 3.74 x 1 0~ 4 A /m 2 . Find the rate of corrosion in (a) mm/year and (b) mdd. 14. F rom the following data, calculate the thickness of the diffusion layer: S U G G E S T E D READING Davis, J.R. ed. (2000). Understanding the Basics, Materials Park, Ohio: ASM International, USA. [2] Stansbury, E.E. and Buchanan, R.A. (2000). Fundamentals of Electrochemical Corrosion, Materials Park, Ohio: ASM, USA. [3] Bockris, J.O.M and Reddy, A.K.N. (1973). Modern Electrochemistry, New York: Plenum Press. [4] Pourbaix, M. (1974). Atlas of Electrochemical Equilibria, NACE, 1974. [5] Vetter, K.J. (1967). Electrochemical Kinetics, New York: Academic Press, USA. [6] Stern, M. and Geory, A.L. (1957). Electrochemical polarization I: A theoretical analysis of the shape of polarization curves, /. Electrochem Soc, 104, 56-63. [7] Stern, M. and Geory, A.L. (1957). Electrochemical polarization II, /. Electrochem Soc, 104, 559. [8] Stern, M. (1957). Electrochemical polarization III, /. Electrochem Soc, 104, 645. [1] KEYWORDS 15. 16. 17. 18. 19. 2 0. Activation polarization Polarization of an electrode controlled by a slow step in reaction sequence of steps iL = 1 2A/dm 2 , c = l m o l / l , at the metal/electrolyte interface. There is a critical activation energy needed to surmount the energy barrier associated with the slowest step. D = 1 0 ~ 5 / c m 2 / s , n = 2. Anodic polarization The shift of the potential of an I n a corrosion cell involving chromium, electrode in a positive direction by an external current. Concentration polarization (Diffusion or transport which forms C r 3 + , an electrical current of over-potential) It is the change of potential of 20 mA is measured. How many atoms per an electrode caused by concentration change near second are oxidized at the anode? the electrode/electrolyte interface. The concentration A p iece of steel corrodes in seawater. The changes are caused by diffusion of ionic species in the corrosion current density is 0.2mA/cm 2 . electrolyte. Calculate the rate of weight loss in mdd units. Corrosion potential It is the potential of a corroding surface in an electrolyte with reference to a reference If zinc surface is corroding at a current denelectrode. It represents the mutual polarization of the 5 2 sity of 2 x 10~ a nd current of 0.2 mA/cm , potentials of the anodic and cathodic reactions which what thickness of metal would corrode in constitute the overall corrosion reaction. Corrosion rate It is the rate showing slow or fast cor8 months. D etermine the potential of an electrode on rosion proceeds on a metallic surface. Corrosion rate is commonly expressed in terms of millimeters per year hydrogen scale which has a potential of 0.87 (mm/year), mils per year (1 mil = 1/1000 of an inch), relative to an Ag/AgCl electrode. or milligrams per decimeter square per day (mdd). T he corrosion rate of iron in deaerated HCl Critical current density The maximum current density is 40 mdd. Calculate the £COrr of iron with exhibited in the active region of a metal/alloy system respect to 0.1 N Calomel electrode (b a = 0.1) which exhibits an active/passive behavior. A m ild steel cylindrical tank 1 m h igh and Exchange current density It is the rate of exchange of electrons (expressed as electrical current) when 50 cm dia contains aerated water to the 60 cm an electrode reaches equilibrium at the equilibrium level and shows a loss in weight due to cor- potential. At the equilibrium potential, the rate of forrosion of 304 g after 6 weeks. Calculate the ward reaction —^(anodic) balances the rate of reverse reaction <—. corrosion current density involved. C orrosion kinetics Faraday It is the quantity of electrical charge required to bring a change of one electrochemical equivalent ( F = 9 6 500C). Flade Potential It is the potential signifying the onset on passivity in an active/passive metal/alloy system. Helmholtz double layer Ions at the metal/electrolyte interface create separation of changes due to repulsive forces. The separation of negative and positive charges like in a capacitor constitute a double layer. It was introduced by Helmholtz, hence it is known as Helmholtz double layer. 119 Mixed polarization The combined potential of a specimen where two or more electrochemical reactions proceed on the surface of the specimen. Passive current density The current density at which the metal or alloy is pushed in the passive region. Polarization The shift of the potential of an electrode from its equilibrium potential (rj = c). Transpassive region The region of anodic polarization above the passive potential range, which shows a sudden increase in the current density due to breakdown of passivity. s OF CORROSION: MATERIALS AND ENVIRONMENTS 4 .1 INTRODUCTION the rate of corrosion of steel will not be the same in Dhahran and Riyadh,1 because the former has a sea-coastal (marine) environment and the later, a typical desert environment. Generally, the variation in severity of corrosion at different geographical locations can be attributed to the variation in moisture, temperature and air-borne substances found in the atmosphere. The marine environment is considered the most corrosive of natural environments. The environmental areas are classified on the basis of the degree of atmospheric contamination. Air pollutants are found in liquid, solid and gaseous forms. The familiarity of environment and the type of corrosion is very important for design engineers. The engineer should recognize the potential hazards of corrosion and should be familiar with the methods used to mitigate various types of corrosion attacks. This chapter, therefore, will be devoted mostly to the study of localized corrosion. The scope of localized corrosion is very extensive and the literature is abundant on the theoretical aspect of localized corrosion. The main object of this chapter is to familiarize engineers with the type of localized corrosion so that they may identify the various forms of corrosion and suggest remedial measures. Contradictory mechanisms will, therefore, be left to a minimum. Each form A wide spectrum of corrosion problems are encountered in industry as a result of combination of materials, environments and service conditions. Corrosion may not have a deleterious effect on a material immediately but it affects the strength, mechanical operations, physical appearance and it may lead to serious operational problems. Corrosion may manifest itself as a cosmetic problem only, but it can be very serious if deterioration of critical components is involved. Serious corrosion problems, such as the pitting of condenser tubes in heat exchangers, degradation of electronic components in aircrafts and corrosion fatigue of propellers can lead to catastrophic failures. When catastrophic failures occur, the cost in terms of lives, equipment, and time is very high. While evaluating the long range performance of materials, it is essential for an engineer to consider the effects of corrosion along with other characteristics, such as strength and formability. In order to be able to select suitable materials, it is important for an engineer to understand the nature of corrosion, types of corrosion and methods of prevention of different types of corrosion. The mechanism of corrosion has already been discussed at length in Chapter 3, and so is not repeated here. Environment plays a very important part in corrosion. The severity of corrosion varies considerably from one place to another. For instance, ^Both are located in Saudi Arabia. Dhahran has a marine environment, and Riyadh has a desert environment. Types of corrosion: materials and environments of corrosion will be discussed in the following manner: (1) (2) (3) (4) (5) Definition. Environment. Mechanism. Examples from industry. Methods of prevention. 1 21 SO3 in the upper atmosphere. The reaction is: S 0 3 + H 2 0 -> H 2 S0 4 (Sulfuric acid) (4.1) S 0 2 + H 2 0 -> H2SO3 (Sulfurous acid) (4.2) S 0 2 + 2 Fe 2 0 3 -> FeS0 4 + F e 3 0 4 (4.3) 4.2 4 .2.1 UNIFORM CORROSION DEFINITION It is the uniform thinning of a metal without any localized attack. Corrosion does not penetrate very deep inside. The most familiar example is the rusting of steel in air. The FeS04 formed accelerates corrosion. Once rusting has started, corrosion cannot be stopped even after SO2 is removed from the air. Iron corrodes faster than any other engineering material in an industrial and marine atmosphere (Fig. 4.1). Other contaminants are nitrogen compounds, H2S and also dust particles. CO2 does not play a significant role in uniform corrosion. Sulfur compounds are abundant in an atmosphere where petroleum industry is located. 4 .2.4 E F F E C T OF HUMIDITY 4 .2.2 ENVIRONMENT (1) (2) (3) (4) (5) (6) (7) (8) (9) Dry atmosphere. Damp atmosphere. Wet atmosphere. Acids (HC1, HCIO4, H3PO4). Atmospheric contaminants. Process water containing hydrogen sulfide. Brines. Industrial atmosphere. Hydrocarbon containing wet hydrogen sulfide. Corrosion can be caused in the atmosphere when about 70% of the humidity is present, as this is the value in equilibrium with saturated NaCl solution and NaCl is commonly present on surfaces. In the presence of such humidity, an invisible thin film of moisture is formed on the surface of a metal. The thin film of moisture acts as an electrolyte for the passage of current. Structures which are exposed to open air, are affected by damp environments. Beyond 80% relative humidity, a sharp increase in the rate of corrosion is observed. Each metal has a critical value of relative humidity beyond which the rate of corrosion increases significantly. 4 .2.5 WATER LAYERS 4 .2.3 E F F E C T OF POLLUTANTS Corrosion can proceed in a dry environment If visible water layers are formed on the metal without any moisture if traces of sulfur com- surface, corrosion initiates. Splashing of seawater, pounds or H2S or other pollutants are present in rain and drops of dew provide the wet envithe air. Tarnishing of silver in dry air in the pres- ronment. The water layer on the metal surface ence of H2S traces is an example of dry corrosion. acts as an electrolyte and provides a passage for Industrial atmospheres contain SO2 as the major the flow of current, similar to the situation in a corrosion cell. contaminant. The rate of corrosion in the presence of SO2 D E W FORMATION increases in the presence of moisture. The sul- 4 .2.6 fur dioxide released in the atmosphere reacts with the rust formed on the metal surface as shown in If the dew becomes acidic, due to the presence reaction(4.3). Much of the SO2 is converted to of SO2, it increases the rate of corrosion. 1 22 Principles of Corrosion Engineering and Corrosion Control AIR High HumidHy Temperature and Wind Speed Automobiles HC#CO,NO / OH OH Refineries S02,SGi / Sea Dust storms Dust particles \ SOs*l/2 Os - * SO* - * H£Ot fo + SQt + Q t ^ f t g S a i a \ SQr*Qr-» SQi+HaO —* HMk ^OOH O H OH FeOOH FeOOH OH OH SU% I 1* 11 iifc C t t & + <A» *& 1 i....|pi. | jM 4^ft M* v2?S Aadinptts HO 4FeaSO# + Oi * 6H3O—* 4FeOOH * 4H*SO* O i * 2 Hrf>*4e — * FeOOH* 4H r% + O t * 2HK> — * FfcOOH* 2H* 2H**2e—*Ni F ed** 2HiO - * 4 f^OH) * 2H* 4 20* Figure 4.1 Corrosion of iron in marine atmosphere Automobiles left open in the air may be subject to corrosion through acidic dew formation. 4.2.7 CORROSION PRODUCT If the corrosion product on the metal surface is microporous, it can condense the moisture, below the critical value. Corrosion proceeds rapidly in such a case, even if the moisture content is below the critical limit. 4.2.8 M E C H A N I S M OF UNIFORM CORROSION Corrosion mechanism in aqueous solution has been amply demonstrated. In atmospheric corrosion which also exemplifies uniform corrosion, a very thin layer of electrolyte is present. It is probably best demonstrated by putting a small drop of seawater on a piece of steel. On comparing the atmospheric corrosion with aqueous corrosion, the following differences are observed: On a metal surface exposed to atmosphere, only a limited quantity of water and dissolved ions are present, whereas the access to oxygen present in the air is unlimited [1], Corrosion products are formed close to the metal surface, unlike the case in aqueous corrosion, and they may prevent further corrosion by acting as a physical barrier between the metal surface and environment, particularly if they are insoluble as in the case of copper or lead. The following is a simplified mechanism of aqueous corrosion of iron (Fig. 4.2): At the anodic areas, anodic reaction takes place: Fe -> Fe + + + 2e (4.4) Types of corrosion: materials and environments 1 23 Aqueous Corrosion rust Fe(OH) 3 —*• Fe304 *— Fte(OH)3 © 4e+Q2+ 2HiO—MOrf —&~~~~~~~J 4 e f O 2* 2H 2 0—*>40*t etecfctm airrent electron current Figure 4.2 Aqueous corrosion of iron At the cathodic areas, reduction of oxygen takes to insoluble hydrated oxides of ferric called rust. place: The rust is formed away from the corroding site. The corrosion rate is very high if the ferrous ion is 0 2 + 2 H 2 0 + 4e -> 40H" (4.5) oxidized to ferric oxide rapidly. Fe(OH)3 is insoluble and if it forms away from a metal surface, The OH ions react with the F e + + ions produced the corrosion reaction speeds up as equilibrium is at the anode: to be maintained by supplying more ferrous ions (Fe ++ ) from the surface. If, however, Fe(OH)3 is 2+ F e + 2 0H" (4.6) formed on the surface of a metal very rapidly, the Fe(OH)2 corrosion is prevented (a passive film). With more access to oxygen in the air, Fe(OH)2 If S0 2 is present as a pollutant, in air, FeSC>4 oxidizes to Fe(OH)3 and later it loses its water: is produced (equations (4.1-4.3)). The corrosion of iron is significantly affected by the presence 4Fe(OH)2 + 0 2 + 2 H 2 0 -> 4Fe(OH)3 (4.7) of soluble sulfate ion in solution. The sulfate ion continues to attack iron and the surface becomes Ferrous hydroxide is converted to hydrated ferric uneven and even pitted. In this case, layers of oxide or rust by oxygen: porous rust are formed. As no protection is provided by the porous rust to the metal, corrosion 4Fe(OH)2 + 0 2 - » 2 Fe 2 0 3 • H 2 0 + 2 H 2 0 continues to take place. The effect of S 0 2 , SO3 (4.8) and Cl~ ions is illustrated in Fig. 4.3. Rust (Fe 2 03-H 2 0) is formed halfway between the drop center and the periphery which is alkaEXAMPLES OF UNIFORM line. The electrons flow from the anode (drop 4 .2.9 center) to cathode (periphery) in the metallic cir- C ORROSION cuit. The current flow is shown in Fig. 4.2. The ferrous ions on the surface of iron are soluble (1) Tarnishing of silver ware. whereas those in solution are oxidized by oxygen (2) Tarnishing of electrical contacts. 1 24 Principles of Corrosion Engineering and Corrosion Control Atmospheric Corrosion AIR Oh SOi © Cathode t Fe(OH}3 t FeCOH)* 4HaS04 + 4Fe + 20* 4FeS04 + 0 2+6H20 - 4FeSO**4HbO 4FeOOH * 4HaS€M 0 *+2H*0 + 4e Figure 4.3 Effect of SO2 and humidity on metallic corrosion. Reaction occur in a very thin (invisible) aqueous layer Rusting of steels in open air. Corrosion of offshore drilling platforms. Corrosion of galvanized steel stairways. Failure of distillation columns. Corrosion of electronic components. Corrosion of underground pipes (composite asphalt coated). (9) Corrosion of automobile bodies. (10) Corrosion of heat exchanger tubes. (11) Corrosion of structural steels. (3) (4) (5) (6) (7) (8) The pressure was 0.1 MPa. The insulation was wetted by exposure to air. (b) Micro and macro examinations A macro examination with a magnifying lens indicated severe corrosion of the external surface. The surface was thinned to the point of perforation. Micro examination was not considered necessary. (c) Remarks The operating temperature of the base of the distillation tower is about 60° C, whereas the top operating temperature is 0°C. Because of the leakage of water in the insulation material and ice formation, moisture penetrated below the insulation surface. As a result of wetting of insulation, the corrosion rate substantially accelerated. The situation could have been worse, had the temperature range been 60-80°C. 4 .2.10 FAILURE CASE HISTORIES Case 1 - Failure of a Distillation Column Wall A distillation column wall, originally 20.0 mm thick failed in a humid atmosphere. The operating temperature varied between 0 and 15°C (32-59°F). After a service of three years, the column failed. (d) Remedy Coat the entire column of steel with epoxy-phenolic coating. Cathodically protective pigments should not be used in the coating system, particularly in the hotter section as the polarity of zinc can be changed at temperatures above 60° C, the zinc becomes cathodic and (a) Environment Ninety percent relative humid- iron becomes anodic. Iron corrodes and zinc is ity. Conditions: condensed moisture and oxygen. protected. Types of corrosion: materials and environments 1 25 80- Case 2 - Severe Corrosion of Carbon Steel Pipe (c) Conditions Operation 130°F (26-54°C). temperature: (d) Water quality Brackish water with high A carbon steel pipe carrying brackish water was chloride contents (1300 ppm). Conductivity of found to be severely corroded after six months of brackish water is 550 micromhos. service. (e) Insulation material Glasswool. Also, long (a) Equipment Six inch diameter carbon steel shutdown periods were involved. pipe. (f) Remarks There is a large probability of entry (b) Environment Brackish water containing of humidity and moisture in the annular space 900 g/1 chlorides and 1500 T.D.S. (Total Dissolved between the metal surface and the inner side of Solids). the insulation. A physical examination of chemical analysis showed water absorption to take place. (c) Conditions The heater was corroded to the point of perfoMajor species - water ration and the external surface of the tank was Minor species - oxygen, T.D.S. heavily corroded. A microscopic examination of Temperature 25-35°C (ground temperature in the steel surface showed general corrosion to be summer) the main cause of failure. Pressure - 10 psig Time to failure - 6 months (g) Remedy (d) Observations and remarks Both the internal and external corrosion was observed. The internal corrosion was caused by high chloride contents and dissolved oxygen. High summer temperature accelerated the damage. The external surface was damaged by high corrosivity and low conductivity soil. (e) Remedy Use PVC pipes. This history is reported from a town in the Arabian Gulf. The use of PVC pipes solved the problem. Most of the carbon steel pipes have now been replaced by PVC pipes. This is suitable for non-potable water. (1) Corrosion allowance. A popular remedy is to make a corrosion allowance. After calculating the rate of corrosion penetration and knowing how much thickness of the metal would be reduced after a specified period, an equivalent amount of thickness maybe added to prevent the loss of designed thickness. A corrosion allowance is, therefore, made at the design stage to prevent the loss of thickness by general corrosion. For instance, if the predicted rate of corrosion for a new carbon-steel product cooler is 4 mpy, as the unit is designed for 15 years, the required corrosion allowance is 4 mpy x 15 years = 60 mils (0.060 in). The general practice is to allow 1/8 in, or 3.2 mm minimum corrosion allowance. Estimation of remaining life (years) Remaining corrosion allowance (mils) Present corrosion rate (mpy) For example, consider a crude oil tower overhead pipeline. Due to some accident, the crude tower was shutdown. The general wall thickness measured was 0.14 in on a 35 in diam carbon steel overhead line. The minimum thickness required is 0.15 in. The current corrosion rate is 40 mils/year. Another Case 3-Failure of Water Heaters (Glass-lined) Approximately 200 water heaters failed in one year as the outer jacket of the heater containing the insulation severely corroded to the point of thinning, and the external surface of the glass-lined tank corroded severely. (a) Material Heaters were made of carbon steel. The tank was glass-lined. (b) Environment Heaters were located in rooms with poor ventilation with access to dust particles and moisture. 1 26 Principles of Corrosion Engineering and Corrosion Control shutdown for inspection is scheduled to take in which the chemical change is the source of place after two years. energy, is called a galvanic cell The corrosion Use the above equation to estimate the which is caused due to the formation of the galvanic cell is, therefore, called galvanic corroremaining life sion. The driving force for corrosion is a potenCorrosion allowance remaining tial difference between different materials. This force was described by Luigi Galvani, late in the eighteenth century. Between the two different So corrosion allowance remaining materials connected through an electrolyte, the = 80 mils (0.080 in) less noble will become the anode and tend to corrode. The required remaining thickness The tendency of a metal to corrode in a gal= 0.15 in + 0.080 in = 0.23 in. vanic cell is determined by its position in the galThe best solution, therefore, would be vanic series of metals and alloys. A galvanic series to install reinforcing plates of above 0.20 inch is a list of metals and alloys arranged according to in thickness. Use a magnesium anode rod their relative potentials in a given environment inside the tank containing brackish water or (Table 4.1). Galvanic series have already been discussed in Chapter 3. In the galvanic series, the high conductivity water. a metal tends to corrode when connected to a (2) Reduce operating temperatures. metal which is more cathodic to it. The further (3) Coat the steel surface with epoxy-phenolic apart the metals or alloys are in the series, the coating. more rapid will be the corrosion of the more (4) Do not allow the insulation to become wet. anodic (baser) metal. For instance, chromium (5) Inspect the insulation periodically. steel (12-14% chromium) in the active condi(6) Avoid exposure to dust storm. During the tion cannot be joined to chromium steel in the Gulf war, helicopter rotor blades were subpassive condition, because they are far apart in jected to severe corrosion by dust storm. The the galvanic series. Similarly, aluminum cannot general corrosion was of such a high magnibe joined to silver or copper, without the risk of tude that the thickness of some rotor blades galvanic corrosion. was reduced to 5 mm. In the galvanic series, there are many stainless steels at the anodic end of the series (active end), and many stainless steels at the cathodic end (noble end) of the series. The dual behav4 . 3 G A L V A N I C C O R R O S I O N ior of stainless steel is related to its ability to form protective films on the surface in the presence of oxygen or other oxidizing agents, such 4 .3.1 DEFINITION as nitric acid or sulfuric acid. These films are Galvanic corrosion occurs when two metals with destroyed and the steels corrode fast in acids, different electrochemical potentials or with dif- such as HC1 or HF or other non-oxidizing ferent tendencies to corrode are in metal-to-metal acids. Before selecting stainless steels for application in a particular environment, it must be contact in a corrosive electrolyte. determined whether the environments will cause them to be in the passive state or in the active state. Galvanic corrosion may not occur if two 4 .3.2 DESCRIPTION metals close to each other in the galvanic series When two metals with different potentials are joined, such as copper and brass. Metals are joined, such as copper (+0.334 V) and close to each other offer a minimum risk of iron (—0.440 V), a galvanic cell is formed. A cell corrosion. Table 4.1 Galvanic corrosion chart Anodic (least noble) corroded Magnesium Magnesium alloys Zinc Beryllium Aluminum 110, 3003, 3004, 5052, 6053 Cadmium Aluminum 2017, 2024, 2117 Mild steel 1018, wrought iron HSLA steel, cast iron Chrome iron (active) 430 Stainless (active) 302, 303, 321, 347, 410, 416 Stainless steel (active) Ni-resist 316, 317 Stainless (active) Carpenter 20Cb-3 stainless (active) Aluminum bronze (CA687) Hastelloy C (active) Inconel 625 (active titanium (active)) Lead/tin solder Lead Tin Inconel 600 (active) Nickel (active) 60% N i-15%Cr (active) 80% Ni-20% Cr (active) Hastelloy B (active) Naval brass (CA464), yellow brass (CA268) Red brass (CA230), admiralty brass (CA443) Copper (CA102) Manganese bronze (CA675), Tin bronze (CA903, 905) 410, 416 Stainless (passive) phosphor bronze (CA521, 524) Silicon bronze (CA651, 655) Nickel silver (CA732, 735, 745, 752, 754, 757, 765, 770, 794) Cupro nickel 90-10 Cupro nickel 80-20 430 Stainless (passive) Cupro nickel 70-30 Nickel aluminum bronze (CA630, 632) Monel 400, K500 Silver solder Nickel (passive) 60% N i-15% Cr (passive) 302, 303, 304, 321, 347 Stainless (passive) 316, 317 Stainless (passive) Carpenter 20Cb-3 Stainless (passive), Incoloy 825 (passive) Silver Titanium (passive), Hastelloy C and C276 (passive) Graphite Ziconium Gold Platinum Cathodic (most noble) protected 1 28 4 .3.3 Principles of Corrosion Engineering and Corrosion Control MECHANISM OF GALVANIC CORROSION To understand the mechanism of galvanic corrosion, caused by joining of two metals differing in potential, such as iron and copper, consider a galvanic cell shown in Fig. 4.3. For the formation of a galvanic cell, the following components are required: (1) (2) (3) (4) A cathode. An anode. An electrolyte. A metallic path for the electron current. In the case of copper and steel, copper has a more positive potential according to the emf series, hence, it acts as a cathode. On the other hand, iron has a negative potential in the emf series (—0.440 V), hence, it is the anode. As a matter of principle, in a galvanic cell, the more noble metal always becomes the cathode and the less noble always the anode. Moisture acts as an electrolyte and the metal surface provides a metallic path for the electron current to travel. Thus, when a piece of copper is joined to iron, all qualifications required for the formation of a galvanic cell are fulfilled and galvanic corrosion proceeds (Fig. 4.4). The positive ions (Fe ++ ) flow from the anode (iron) to cathode (copper) through the electrolyte, which is water. Iron, therefore, corrodes. The hydrogen ions (H + ) are discharged at the copper cathode, and ultimately hydrogen is released. The Fe + + ions travel towards the cathode and OH~ towards the anode. They combine to form insoluble iron hydroxide, Fe(OH)2. Positive electricity (conventional current) flows from the cathode (+) to the anode (—) through the external metallic path. In the electrolyte the electric current flows from anode to cathode by positive ions (cations). From the cathode to the anode, it is carried by negative ions (anions). In the external circuit, the current (conventional current) is actually carried by electrons from anode to cathode. The electrons, after being released by the anodic dissolution of iron, participate in the reduction process, such as 2H+ + 2e -> H2 or Cu 2+ + 2e -> Cu (4.9) The analogy of the galvanic cell described above, with a copper and iron galvanic couple, illustrates why iron corrodes and copper does not corrode. Naturally, copper does not corrode because it acts as a cathode whereas iron corrodes and generates a more negative potential and it acts as an anode. Consider a steel pipe of A" OD joined to a copper pipe of the same diameter and exposed to soil containing some moisture. The steel pipe would become the anode and, therefore, corrode. Steel $pe (anode) Copper pipe (cathode) Current flow No corrosion occur on copper pipe Corroding area Pipe flange Figure 4.4 Formation of galvanic cell by joining of two dissimilar metals Types of corrosion: materials and environments 129 Copper pipe (inlet) 7 %* Pits dye to plating of Cu tons. ) Galvanized steel tank Stagnant area J Figure 4.5 Galvanic corrosion in a hot water tank Figure 4.5 shows a galvanized steel tank with a copper inlet pipe and an aluminum tank with an aluminum water inlet pipe. The steel tank would corrode as shown in the figure. Copper ions in the tank which are leached from the copper pipe would deposit on the wall of the tank and form a galvanic cell, hence the galvanized steel tank would corrode as shown in Fig. 4.5. To summarize, a more active metal in potential series tends to corrode, whereas a less active or more noble metal does not corrode. series to minimize galvanic corrosion. Active metals should not be joined with passive metals. Thus, aluminum should not be joined to steel, as aluminum being more active would tend to corrode. B. The Nature of Environment Due consideration must be given to the environment that surrounds the metal. For instance, water containing copper ions, like seawater, is likely to form galvanic cells on a steel surface of the tank. If the water in contact with steel is either acidic or contains salt, the galvanic reaction is accelerated because of the increased ionization of the electrolyte. In marine environments, galvanic corrosion may be accelerated due to increased conductivity of the electrolyte. In cold climates, galvanic corrosion of buried material is reduced because of the increased resistivity of soil. In warm climates, on the other hand, it is the reverse because of the decreased resistivity of the soil. 4 .3.4 FACTORS AFFECTING GALVANIC CORROSION The following factors significantly affect the magnitude of galvanic corrosion: A. Position of metals in the galvanic series. B. The nature of the environment. C. Area, distance and geometric effects. A. Position of Metals in the Galvanic Series As mentioned earlier, the further apart the metals are in the galvanic series, the greater is the chance for galvanic corrosion. The magnitude of galvanic corrosion primarily depends on how much potential difference exists between two metals. For a particular environment, the metals selected should be close to each other in the galvanic C. Area, Distance and Geometric Effects Effect of Area The anode to cathode area ratio is extremely important as the magnitude of galvanic corrosion is seriously affected by it. The area ratio can be unfavorable as well as favorable. 1 30 Principles of Corrosion Engineering and Corrosion Control the conduction path. Most corrosion damage is caused by current which cover short paths. Hence, the greatest galvanic damage is likely to be encountered near the junction of the two metals and the severity would be decreased with increased length. If two different metals are far away from each other, there would be no risk of galvanic corrosion, because of very little current flow. This is possible in designs, like oil rigs and other complex structures requiring a very large variety of material. The effect of area and distance may be best understood by the examples of utility lines in a large building. Consider, for instance, the copper tubes transporting water and natural gas. Coated carbon steel pipelines are laid in the same trench in a soil of low resistivity. In the trench, a corrosion cell would be formed if the pipes touch (metal-to-metal contact), or if they are bonded together somewhere (for electrical earthing requirements). Copper acts as cathode and steel acts as anode in an electrolyte of soil and galvanic corrosion would initiate. Now consider the area effect. Being coated, steel pipe would have a small anodic area at the sites of coating defects, whereas copper pipes would have a large cathodic area. Due to the small anode area and large cathode areas, galvanic corrosion would initiate and leaks in the carbon pipe would soon start. The problem also illustrates the effect of distance (Figs 4.7a and b). If the above utility pipe were in two different trenches, a sufficient distance between them would not have allowed the galvanic current to flow. Unfavorable Area Ratio The area ratio of the anode to cathode plays a dominant role in galvanic corrosion. As a given amount of current flows in a galvanic couple, the current density at the anode or cathode controls the rate of corrosion. For a given amount of current, the metal with the smallest area has the largest current density and, hence, is more damaged if corrosion occurs at it. For similar reasons, the current density at a large metal is very small. The rate of corrosion increases with the ratio of cathodic to anodic areas (Fig. 4.6). Take the example of steel plates joined by aluminum bolts (Fig. 4.6). Aluminum has a smaller anodic area and steel, a larger cathodic area. Aluminum is more active in the galvanic series than steel. The current density on aluminum is, therefore, extremely large and serious galvanic corrosion of aluminum takes place. This shows the end result of an unfavorable anode/cathode ratio. The other ratio, large anode/small cathode, would only slightly accelerate the rate of galvanic corrosion. Effect of Distance It is a known principle that the solution conductivity varies inversely with the length of Insulating sleeve Insulating insert Insulating washer Figure 4.6 Avoidance of galvanic corrosion Figure 4.7a Representation of galvanic corrosion in less conductive and more conductive solution Types of corrosion: materials and environments 1 31 j ^ >/ JHUL. J&£_ V/ ^L. J JKL more severe less severe more severe less severe distance * ~ - > distance Figure 4.7b Dissimilar metal couple mechanism Galvanic attack would be restricted for two dissimilar metals in contact with soil of high pH and low in carbon dioxide. can cause galvanic corrosion of the steel shaft if it is wet. B. Metallic Coatings Effect of Geometry Two types of metallic coatings are generally used, Geometry of components and their design also noble and sacrificial type. Zinc coating is an examinfluence galvanic corrosion. As current does not ple of the sacrificial type. Zinc corrodes eventually flow around the corners, the geometry of the and it protects the steel substrate both by its barcircuit affects the degree of galvanic corrosion. rier effect and also by providing electrons (Zn -> Polarization may be affected by a break in the Zn 2+ + 2e) into the steel which prevent Fe + + ions continuity of the current. from escaping from the steel (cathodic protection Component design is also a factor in gal- - see below). Noble coatings act as a barrier only vanic corrosion as the current circuit geometry between the metal substrate and the environment. affect the magnitude of galvanic corrosion and the Nickel, silver, copper, lead and chromium are polarization process. Any obstacle to polarization called noble metal coatings. Formation of pores would accelerate galvanic corrosion. and damage to the noble coating can cause galvanic corrosion of the substrate (Fig. 4.8), as there is no sacrificial cathodic protection of the substrate. 4 .3.S APPLICATION OF P R I N C I P L E S OF GALVANIC CORROSION C. Cathodic Protection A. Non-metallic Conductors As mentioned above, a positive use of the principle of galvanic corrosion is cathodic protection. In Many non-metallic materials are cathodic to met- a sacrificial system of cathodic protection, anodes als and alloys. For example, impervious graphite of active metals, like Zn, Mg and Al, are used for used in heat exchanger applications is noble to protection of steel structures. The sacrificial galmore active metals. The nature of non-metallic vanic anodes provide protection to the less active conductors must be known before their applica- metals, like steel because they corrode and release tion. Graphite packing around a steel pump shaft electrons. The electrons which are released by the 1 32 Principles of Corrosion Engineering and Corrosion Control Exposure t o air Breakdown of noble coating Z&Z Noble coating (catftode) Point of contact between active metal and noble metal Anode Figure 4.8 Initiation of galvanic corrosion by breakdown of coating corroding metals enter the steel structures, which become cathodic and, therefore, do not corrode. This system of cathodic protection is based on galvanic corrosion, however, in this case a beneficial use is the result of galvanic corrosion. (9) Galvanic corrosion inside horizontal stabilizers in aircrafts. 4 .3.7 FAILURE CASE HISTORIES 4 .3.6 EXAMPLE OF FAILURES OF GALVANIC CORROSION A. Case 1 - Failure of Aluminum Alloy Spacers by Galvanic Attack Numerous examples of galvanic corrosion are Description of the problem Several rebuilt encountered as it is one of the most frequently hydraulic actuators had been in storage for three encountered forms of corrosion. The following years. At each joint, there was an aluminum are common examples: alloy spacer and a vellum gasket. The mounting flanges of the steel actuators were nickel (1) Galvanic corrosion of aluminum shielding in plated. While assembling the actuators, a lubriburied telephone cables. cant containing molybdenum sulfide had been (2) Galvanic corrosion of steel pipe with brass applied to the gasket to serve as a sealant. The fittings. galvanic attack occurred on the aluminum alloy (3) Galvanic corrosion of the body of the ship in spacer [2]. contact with brass or bronze propellers. (4) Galvanic corrosion on a gannet wheel where Identification The three major components: aluthe steel securing the bolts is in contact with minum spacers, vellum gasket and actuators housing were examined in a laboratory. The folthe magnesium wheel. (5) Galvanic corrosion between the tubes and the lowing were the conclusions of the examination: tube sheet in heat exchangers. (6) Aluminum conduit buried in steel reinforced (a) Corrosion had penetrated to a depth of 64 |xm in some areas of the aluminum spacers concrete. which had been stained badly. (7) Galvanic corrosion of steel coated with copper due to the defects in copper coating. (b) The vellum gasket was electronically conductive. (8) Galvanic corrosion of the Statue of Liberty. Types of corrosion: materials and environments (c) A potential of 0.1 V was measured between the aluminum alloy spacers and the actuator housing. (d) No potential was detected after breaking contact between aluminum spacer and the actuator housing. (e) Deposits of molybdenum sulfide were found on the aluminum surface of aluminum spacers. Conclusion Galvanic corrosion was the reason for failure. Molybdenum disulfide acted as an electrolyte between the aluminum spacers and the nickel plated steel actuator housing. In order to remedy the situation, dry vellum gaskets were used and the use of molybdenum disulfide was discontinued. Remedy 1 33 (1) The operating temperature of the boiler was lowered. The water temperature in the pipe was not allowed to exceed 60° C. (2) The water was treated with sodium, hexametaphosphate, before it entered the piping system of the house. (3) Normally this problem is avoided by using copper for hot water pipes and tanks. 4.3.8 METHODS OF GALVANIC PREVENTION OF CORROSION (1) Select metals, close together, as far as possible, in the galvanic series. (2) Do not have the area of the more active metal B. Case 2 - Failure of Hot Water smaller than the area of the less active metal. (3) If dissimilar metals are to be used, insulate Galvanized Pipes in a Housing Area them. Description of the problem In a housing society (4) Use inhibitors in aqueous systems whenbuilt for an aluminum company, the household ever applicable and eliminate cathodic plumbing was severely affected. All hot water depolarizers. pipes were galvanized steel. The temperature of (5) Apply coatings with judgment. Do not coat the water inside the pipe was about 80°C. Every the anodic member of the couple as it would morning, the color of the water becomes red reduce the anodic area and severe attack and it gave an offensive odor. The residents of would occur at the inevitable defect points the houses could not use this water for washing in the coating. Therefore, if coating is to be purpose as it stained their hands [3]. done, coat the more noble of the two metals in the couple which prevents electrons Identification Upon metallographic examinabeing consumed in a cathodic reaction such tion of the internal surface of the pipe it was found as 2H + + 2e —> H2, which is likely to be that the pipe was seriously corroded. The corcorrosion rate controlling. rosion was of uniform type. A chemical analysis (6) Avoid joining materials by threaded joints. of the water showed the leaching of iron from (7) Use a third metal active to both the metals in the pipe. the couple. Conclusion It was concluded that iron was (8) Sacrificial material, such as zinc or magneseverely corroding. The amount of iron increased sium, may be introduced into this assembly. overnight when the water was relatively unused. For instance, zinc anodes are used in cast iron In the morning, when the taps were opened, a water boxes of copper alloy water-cooled heat substantial amount of iron leached in the water exchangers. rendering it unsuitable for use. The reason was (9) In designing the components, use replaceable the reversal of polarity which occurs in hot water, parts so that only the corroded parts could be zinc became the cathode and iron the anode. The replaced instead of the whole assembly. zinc coating did not, therefore, protect the steel pipe which continued to corrode by becoming Above all, understand materials compatibility anode. which is the key to control galvanic corrosion. 1 34 Principles of Corrosion Engineering and Corrosion Control The two types of attack can be observed in Figs 4.9 and 4.10. In the uniform type of dezincification, the active area is leached out over a 4 .4.1 INTRODUCTION broad area of the surface and it is not localized It is a form of corrosion in which zinc is selectively to a certain point of the surface. On the other attacked in zinc-containing alloys, like brasses. hand, the plug type of attack is localized, at a It mainly occurs in alloys containing less than certain point on the surface and the surround85% copper. De-alloying and selective leaching ing area remains unaffected. Dezincification can are broader terms which refer to the corrosion of occur on grain boundaries, such as a-p brasses one or more constituent of a solid solution alloy. (Fig. 4.9). Dezincification is a form of de-alloying. As the phenomenon was first observed in brass in ENVIRONMENT which zinc separated by dissolution from copper, 4 .4.3 the term dezincification is still used. Ordinary brass consists of about 30% zinc and Dezincification generally takes place in water 70% copper. Dezincification can be observed by under stagnant conditions. Copper-zinc alloys naked eyes, because the alloy changes in color containing more than 15% zinc are susceptible to dezincification. from yellow to red. 4 .4.2 TYPES OF ATTACKS 4.4.4 MECHANISM 4.4 DEZINCIFICATION Two types of dezincification are commonly There is a disagreement between the workers on the mechanism of dezincification. One group observed: contends that firstly the entire alloy is dissolved and later one of its constituent is re-plated from (1) Uniform type (layer type) (Fig. 4.9) the solution which leached the alloy. This is (2) Plug type (Fig. 4.10) Figure 4.9 Dezincification of a-p brass (500 x 300) Types of corrosion: materials and environments 1 35 Figure 4.10 Dezincification plugs in 70-30 Brass exposed to NaCI. Evidence of two mechanisms operating simultaneously can be observed in the picture. (From Vernik, R.D. Jr. and Heiderbach, R.H. Jr. (1972). In Localised Corrosion, Cause of Metal Failure, STP 516, ASM. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) It has been shown that zinc has a high tendency to dissolve, whereas copper has a high tendency to plate. The E° for zinc is —0.763 V, Zn -> Z n + + + 2e whereas the E° for copper is 0.337 V, which shows the above tendencies. Zinc replaces copper in the 2H+ + 2e -> H 2 solution because copper is far nobler than zinc 2+ in the EMF series. As copper is redeposited as a Zn -» Zn + 2e porous mass, the strength of brass is significantly (I) dissolution CuCl + e -> Cu + CI lowered. The alloy is mechanically weakened to (4.10) the extent that a slight increase in the load causes the alloy to rupture. Dezincification causes only M -> M n + + ne a slight change in the appearance and does not H 2 -> 2H° affect the dimensions. Contrary to the opinion of one group of workH+ + e - * H ers, the other group believes that one species is Cu -> C u + + + 2e (II) dissolution (4.11) dissolved selectively from the alloy, and a porous residue of the more noble metal is left [4]. A third +2 +2 group believe that both the mechanisms operate. (III) plating Cu + Zn -> Cu + Zn The evidence of that both mechanisms may oper(4.12) ate is shown in Fig. 4.10. In thefigure,a dezincified Copper is deposited as a fine copper dust region is shown below the metal surface. The perwhich readily dissolves once more in any elec- turbations shown above are deposits of copper. trolyte. Zinc is leached out of the brass leaving The evidence that both mechanisms may operate behind a highly porous mass. The three steps at the same time, is shown in the above figure. involved above are: Use has been made of electrochemical hysteresis methods to evaluate the mechanism Step I Dissolution of Cu and Zn of dezincification. On the basis of the infor(equations I and II) mation generated by electrochemical methods, Step II Zinc stays in solution pH-potential diagrams have been constructed Step III Copper plates back (equation III) the basis of the dissolution and redeposition mechanism. 1 36 Principles of Corrosion Engineering and Corrosion Control r ~ ~~--^ Co : ZnO *** <*•»» ** "^ hH ' *w * { ~ §hr -is 4 «J> ^ ^ - i i ] (~- -s- "~ ,i .1.,,« ~JU. ^ - a x .^ • l Zn i. * 1 1 L„ 1 iJ id pi n -a n Small dots militate the domain m nhivft selective removal of zim h expected institutions free uf copper turn. Larger dots indicate the domain in which both copper ami zinc dissolve CruAshatefimg hulk ales the region m H hich copper is expet ted lo deposit. Figure 4.11 Potential-pH diagram of 70-30 Cu-Zn in NaCl Solution. Figure shows the domains of copper leaching. (From Vernik, R.D. Jr. and Heiderbach, R.H. Jr. (1972). In Localised Corrosion, Cause of Metal Failure, STP 516, ASM. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) (Fig. 4.11). From these diagrams, the domains of selective leaching of zinc, and domains of dissolution of both zinc and copper can be observed. The regions where copper deposition takes place are also shown in the figure. From the diagram it is observed that at potentials between 0.000 and +0.200 VSHE> both zinc and copper dissolve, but in stagnant conditions copper may deposit on the specimen without any evidence of selective leaching. Above +0.200 VSHE> both alloys dissolve but there is no redeposition of copper. If the potential is held above +0.200 VSHE for a long period of time and then lowered below +0.200 VSHE> deposition of copper would take place (CuCl + e —• Cu + C P ) . In spite of the development and evolution of new techniques, such as microprobe analysis and advances in the electrochemical methods, it is not clearly known which mechanism is operative. It, however, appears on the basis of the literature published that both mechanisms may operate, i.e. selective leaching or dissolution of both components and re-plating of one of the Types of corrosion: materials and environments 137 component, or selective dissolution of one species dezincification, and it should, therefore, be added only leaving a porous residue of the more noble to the tubes. A small amount of Mg is also required with the As. High copper alloys (copper above species. 85%) are immune to dezincification and they can be used safely. If dezincification is very severe, 4 .4.5 EXAMPLES OF the use of more expensive copper-nickel 70-30 is recommended. DEZINCIFICATION (a) Layer Type Uniform layer type dezincification occurs in tooth of gear wheel. It also occurs on the inner surface of admiralty brass heat exchanger tubes when exposed to water at pH = 8.0 and temperature range 31-49°C. (b) Plug Type It is found particularly in a-brass heat exchanger pipes. If the heat exchanger is not cleaned and dried, differential aeration cells are formed in which the brass dissolves. The corroded region is filled with the re-precipitated copper. 4 .4.7 DE-ALLOYING Dezincification is a special case of de-alloying. The phenomenon of de-alloying by selective leaching can occur in various materials (as shown in Table 4.2). Selective leaching (de-alloying) of some important materials is summarized below. (a) Graphitic Corrosion Gray cast iron sometimes show the effect of selective leaching out of iron in mild corrosive environments. The surface layer of the iron becomes like graphite and it can be easily cut with a knife. Because of the attack, the iron or steel matrix is dissolved and an interlocking nobler graphite network is left. The graphite becomes cathodic to iron and a galvanic corrosion cell is formed. Iron is dissolved and a porous mass of voids and complex iron oxides is left behind. This graphitized cast iron loses its strength and other metallic properties (Fig. 4.12), but to a casual view it looks dirty but unchanged in shape, which can lead to dangerous situations. Graphite corrosion does not occur in nodular and malleable cast iron. One common mistake in the books is to use the term'graphitization rather than graphitic corrosion. Graphitization occurs when a low alloy steel is subjected to high temperature for an extended time period. Graphitization results from the decomposition of pearlite into ferrite and carbon, whereas in graphitic corrosion the gray cast iron is selectively attacked. The presence of graphite is necessary for leaching to take place. 4 .4.6 CASE HISTORY Failure of copper alloy (C27000) inner cooler tubes for air compressors. Description of the problem Yellow brass tubes (65% copper) in air compressor showed leaking in cooling water after 17 years of service. Conditions The cooling water was chlorinated well-water. The water was recirculated. Physical examination A visual examination showed a thick layer of porous brittle copper on the inside surface. A plug type deposit had penetrated deep in the well of the tube. At many points the wall of the tube was completely damaged [5]. Analysis of the tubes showed that they were fabricated from copper alloy C27000. It was observed that the alloy before dezincification contained 35% zinc. The analysis showed that only a trace of the original 35% zinc remained. It is obvious that most of the zinc was lost as shown by the analysis of the base metal and the brittle layer. This was due to the leaching of zinc by dezincification. (b) De-aluminification Copper alloys containing more than 8% Al may Recommendations Arsenic addition in the be subjected to the preferential dissolution of alurange 0.02-0.06%, provides high resistance to minum component of the alloy. The a-phase 1 38 Principles of Corrosion Engineering and Corrosion Control Alloys subjected to leaching Environment Many waters, especially under stagnation conditions Soils, many waters Hydrofluoric acid, acids containing chloride ions Not reported High heat flux and low water velocity (in refinery condenser tubes) Hydrofluoric and other acids Nitric, chromic and sulfuric acids Molten salts Not reported Oxidizing atmospheres, hydrogen at high temperatures High-temperature oxidizing atmospheres Oxygen at high temperature Element removed Zinc (dezincification) Iron (graphitic corrosion) Aluminum Silicon Nickel (de-nickelification) Table 4.2 Alloy Brasses Gray iron Aluminum bronzes Silicon bronzes Copper nickels Monels Alloys of gold or platinum with nickel, copper or silver High-nickel alloys Cobalt-tungsten-chromium alloys Medium-carbon and high-carbon steels Iron-chromium alloys Nickel-molybdenum alloys Copper in some acids, and nickel in others Nickel, copper or silver (parting) Chromium, iron, molybdenum and tungsten Cobalt Carbon (decarburization) Chromium, which forms a protective film Molybdenum reliable technique acceptable to researchers has been developed. The technique of superimposing an electrochemical hysteresis curve on the pH-potential Pourbaix diagram has shown a fair degree of (c) De-nickelification promise. As shown in the text related to dezincification, this technique provides a basis for the Although, not common, the de-alloying of nickel prediction of the tendency for de-alloying as a in 70-30 Cu-Ni alloy has been observed under low function of potential. From a knowledge of potenflow conditions. tial, it can be predicted whether zinc is being leached or copper is being deposited or copper is being leached. 4 .4.8 T E C H N I Q U E S FOR However, more work is needed to improve the existing technique and to develop a new technique EVALUATION OF to make such predictions. DEZINCIFICATION AND DE-ALLOYING 4 .4.9 PREVENTION of aluminum-bronze is attacked and a porous residue of copper is left behind. Since the phenomenon of de-alloying is extended over a long period of time, accelerated techniques have been in demand, but still no single (1) Use copper alloys with copper content above 85%. Types of corrosion: materials and environments 1 39 differential aeration cell for crevice corrosion to occur. This phenomenon limits the use, particularly of steels, in marine environment, chemical and petrochemical industries. 4 .5.1 CAUSES A pump Impeller section showing graphite corrosion attack (a) Presence of narrow spaces between metal-tometal or non-metal to metal components. (b) Presence of cracks, cavities and other defects on metals. (c) Deposition of barnacles, biofouling organisms and similar deposits. (d) Deposition of dirt, mud or other deposits on a metal surface. 4 .S.2 MATERIALS AND E NVIRONMENT The conventional steels, like SS 304 and SS 316, can be subject to crevice corrosion in chloride containing environments, such as brackish water and seawater. Water chemistry plays a very important role. Factors affecting crevice corrosion (Fig. 4.13): (a) (b) (c) (d) (e) (2) Use brass alloys with tin, arsenic or antimony (f) addition. (g) (3) Avoid environments where the solution (h) becomes stagnant and deposits accumulate on the metal surface. Crevice type. Alloy composition. Passive film characteristics. Geometry of crevice. Bulk composition of media. Bulk environment. Mass transfer in and out of crevice. Oxygen. Figure 4.12 Graphite corrosion in gray iron pipe. (From Ports, R.D. (1987). Nalco Chemical Company) (a) Crevice Type 4.S CREVICE CORROSION Crevice type means whether the crevice is between metal-to-metal, metal to non-metal or a marine growth, like barnacles or other marine biofouling organisms, on the metal surface. It is important to know whether factors affecting crevice are man-made or natural in order to select appropriate methods for prevention. This is a localized form of corrosion, caused by the deposition of dirt, dust, mud and deposits on a metallic surface or by the existence of voids, gaps and cavities between adjoining surfaces. An important condition is the formation of a 1 40 Principles of Corrosion Engineering and Corrosion Control L Sulk solution composition 2 . Bulk solution environment 3» Nass transport in m6 out of crevice - convection 4 . Crevice solution 5. Electrochemical reactions - metal dissolution - Oa reduction 7. Passive film characteristics - passive current - film stability 8. Crevice type - metal/metal - metal/non-metal - metal/marine growth 9. Crevice geometry -gap - depth 10. Total geometry - exterior to interior crevice area ratio * number of crevices - Hi evolution 6. Alloy composition - major constituents • minor constituents - impurities Figure 4.13 Factors affecting crevice corrosion. (Oldfield, J.W. and Sutton, W.H. (1978). Br. Corros. Jr., 13, (1). Reproduced by kind permission of British Corrosion Journal) (b) Alloy Composition It is important to know whether or not the alloy is resistant to crevice corrosion. For instance, work on the various grades of steels, such as SS 304, carpenter alloy, Incoloy (Alloy 825), Hastelloy (Alloy g), and Inconel (625) showed that the later two alloys were highly resistant to crevice corrosion in ambient and elevated temperature seawater [6]. The alloying elements in various grade of steel affect both the electrochemical and chemical processes, such as hydrolysis, passive film formation, passive current density and metal dissolution. Hence, the effect of major alloying elements, such as Fe, Cr, Ni and Mo and Fe-Cr-Ni-Mo class of steels on crevice corrosion initiation is important. It has been shown that iron, chromium, nickel and Mo improve the resistance of steels to crevice corrosion [7]. Crevice corrosion is not usually observed with 6% molybdenum. The effect of chromium can be harmful if it increases the acidity in the crevice by hydrolysis. At a higher chromium concentration, it generally increases the stability of the passive film, lowers the pH below that required for crevice to initiate, and reduces the passive current density. The rate of entry of chromium ions in the crevice is, therefore, minimized, hence hydrolysis which causes acidic conditions in the crevice is minimized by chromium. Types of corrosion: materials and environments 141 (c) Passive Film Characteristics The type of passive film formed is important, as the breakdown of a passivefilmresults in the onset of crevice corrosion. The improved performance of certain steels can be attributed to enrichment of the surface film by chromium. The quality of surface films affects the susceptibility of steels to crevice corrosion. The formation of passive films is dependent on oxygen and hence, its concentration affects the magnitude of crevice corrosion. It has been reported that as the seawater temperature is raised from ambient to 70° C, the resistance of steel types 304 and 317 is increased. This has been attributed to the enrichment of the passive film by chromium which increases its stability. (d) Geometry of Crevice The magnitude of crevice corrosion also depends on the depth of the crevice, width of the gap, number of crevices and ratio of exterior to interior crevice. It has been shown for types 316 and 304 stainless steel that smaller the gap, the less is the predicted time for initiation of crevice corrosion (Fig. 4.14). The reason is that when the ratio of crevice solution volume to creviced area is small, the acidity is increased and the critical value for initiation of crevice is achieved rapidly. The ratio of the bold area to the creviced area also affects crevice corrosion (Fig. 4.15). Generally, the larger is the bold area (cathodic) and smaller the creviced area (anodic), the larger is the probability of crevice corrosion. This has been shown by work on types 304 and 316 stainless steel and 1,0 Type 304 L0 10 Predicted time (hours) *00 1000 Figure 4.14 Effect of crevice gap on predicted times to onset of crevice corrosion for Type 304 and Type 316 stainless steel. (Kain, R.M. (1979). NACE Paper No. 230, Corrosion, March) [10] 1 42 Principles of Corrosion Engineering and Corrosion Control m I I T-304 stainless steel T-316 stainless sfceet | IiicoloyaMciy825 so r WM 40 h | m 30 h 20 10 L 04/1 10/i Bold cmviced area ratio mmn Figure 4.15 Probability of crevice corrosion initiation in one month. (From Andreason D.B. (1974). R&D INCO, Paper presented on ASTM-ASM Symposium in Pitting Corrosion, Detroit, MI, October 23. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) Incoloy alloy 825 [7]. If the depth is increased, acidity is also increased in the crevice. (e) Effect of Temperature the electrolyte. All chloride containing solutions are highly aggressive and contribute to onset of crevice corrosion. Seawater and brackish water are high aggressive and promote crevice corrosion of steels. The rate of crevice corrosion propagation of stainless steel alloys, such as types 304 and 316, is decreased in natural seawater, when the tempera- (g) Mass Transfer in and out of Crevice ture is increased from ambient to 70°C [8,9]. The tendency for initiation of crevice corrosion is not There are three forms of mass transport: migraaffected by increase in temperature. The above tion, diffusion and convection. Most of the trend may be attributed to the decreased solu- current is carried by migration and diffubility of oxygen with increased temperature and sion. The effect of bulk chloride concentration changes in the nature of the passive film which is on predicted time for onset of crevice corrosion is shown in Fig. 4.16. The prediction formed. for onset is shortened with increased chloride concentration. (f) Bulk Solution Composition The process of bringing chloride ions increases the concentration of chloride ions in The breakdown of a passive film on a metal the small crevice, hence, the aggressiveness of surface largely depends on the aggressiveness of the electrolyte inside the crevice is increased. Types of corrosion: materials and environments 1 43 0.2 0.3 0.4 0.5 Chloride concentration (mol/L) Figure 4.16 Effect of bulk chloride concentration on predicted initiation of crevice corrosion for Type 304 and Type 316 for a given set of severe crevice conditions. (From Kain, R.M. Tech. Paper No. FP-37, LCCT, Corrosion/80, NACE, Chicago. Reproduced by kind permission of NACE, Int., Texas, USA) The solution at a stage becomes highly acidic and 4 . 5 . 3 M E C H A N I S M O F C R E V I C E the pH is lowered. CORROSION Most textbooks give only an over-simplified picture of the mechanism of crevice corrosion, The onset of crevice corrosion is strongly linked hence, one may not appreciate its complexity. with the nature of the passive film on the metal In the last fifty years, a significant progress on surface. If the passive film is very stable, crevice the understanding of the mechanism of crevice corrosion is blocked. Oxygen is essential for the corrosion has been made. The phenomenon is formation of passive film on the metal surface. extremely complex, and a unified mechanism Hence, it has an important influence on the does not exist presently. Most of the mechanism onset of crevice corrosion. As oxygen is consumed is based on certain type of concentration cells within the crevice, a potential difference is set up [10,11]. The following is a summary of some of between the creviced areas and the boldly exposed the concentration cell mechanisms, which have areas open to oxygen. A differential concentra- contributed to the understanding of the crevice tion is, therefore, set up which accelerates oxygen. corrosion. However, as the temperature is increased, oxygen solubility decreases which retards the crevice (a) Metal ion concentration cells [11]. Accordprocess. The exact relationship between oxygen ing to this old theory, a difference in metal and temperature is not understood. Thermal ions exists between the crevice and outinsulation in process plants provide water and side, hence, a corrosion cell is formed. The oxygen for crevice corrosion to develop. area with low metal concentration becomes (h) Oxygen 1 44 Principles of Corrosion Engineering and Corrosion Control the anode and other, the cathode. Anodic dissolution at the anode, therefore, initiates. (b) A high concentration of oxygen on the surface outside the crevice and a low oxygen concentration inside a crevice creates a differential aeration cell, which initiates crevice corrosion. A unified mechanism of crevice corrosion was given by Fontana and Greene [12]. A unified mechanism which is rather over-simplified is given below (Fig. 4.17): (1) The site at which a crevice is formed, becomes the anode and the site outside the crevice, the cathode. The reason for the above can be attributed to the formation of differential metal ion, oxygen concentration or active-passive cells. The following reactions take place: Anode (in the crevice) M -> M + + + 2e (M represents a metal) Cathode - 0 2 + 2 H 2 0 + 2e -> 40H~ (oxygen reduction outside the crevice) (4.14) (4.13) (2) After sometime, the oxygen in the crevice is consumed, but the concentration of oxygen at the cathode remains unchanged, hence, the reaction continues unabated. (3) Within the crevice, the following processes continue to occur: Cr —> C r + + + + 3e (chromium contained in the stainless steel) (4.15) Fe -» Fe + + + 2e (4.16) To preserve electroneutrality, the chloride ions are attracted by C r + + + or Fe + + ions and metallic chlorides are formed: C r + + + + 3C1" -> CrCl3 F e + + + 2 C r -> FeCl2 (4.17) (4.18) With the formation of metallic chlorides, the process of anodic dissolution continues, and the cavity becomes deeper and deeper. (4) Hydrolysis of these chlorides takes place immediately which results in the production of acid conditions in the pit. Hydrolysis increases the level of acidity in the crevice. The geometry of a crevice High O2 concentration 0 2 — > 4QH- Low O2 and high M*, CI*, H* concentration <* M* or M* M* *r M*aQ*H»q- wqH* Figure 4.17 Mechanism of crevice corrosion. (From France, M.W.D. (1987). Localized Corrosion, cause of metal failure, ASM STP 516. Reproduced by kind permission of ASTM, Philadelphia, USA) Types of corrosion: materials and environments 1 45 limits the exchange of solution between 4 . 5 . 4 D E V E L O P M E N T O F A the structure and the crevice in the bulk, M A T H E M A T I C A L M O D E L thus creating acid conditions in the pit. The above is generalized by: An outstanding contribution has been made by Oldfield and Sutton [13]. They have developed MOH + HC1 a mathematical model of crevice corrosion in M + + CT + HOH (4.19) which observations, such as the fall in pH, localized breakdown of passive films and the onset of In case of 18-8 steel: crevice corrosion in stainless steels, is presented. A summary of the model developed by Oldfield CrCl3 + 3HOH -> Cr(OH) 3 + 3HC1 and Sutton is given below. For calculation and (4.20) detailed discussion, reference must be made to the original paper. In the model developed, crevice FeCl2 + 2HOH -> Fe(OH)2 + 2HC1 corrosion is considered a four stage mechanism. (4.21) The four stages of crevice corrosion suggested are shown in Fig. 4.18. It can be observed that acid is produced and hence acid conditions are (1) Depletion of oxygen in crevice due to conproduced inside the crevice. The pH may sumption in the cathodic reaction. attain a value of as low as 1.0 inside the (2) Increase of acidity in the pit by the process of crevice. Once the acid conditions are hydrolysis. generated, the process continues until (3) Breakdown of the passive film on the surface the reaction is terminated. The mechat a critical value of pH. anism described above is self-generative (4) Propagation of crevice corrosion with further and once it starts, it continues. hydrolysis and production of acidity. The above mechanism is, however, purely qualitative and does not provide an explanation of the following: (a) Why crevice corrosion takes place even in non-aggressive environments? (b) What is the critical concentration of chloride ions necessary to induce crevice corrosion? (c) The major emphasis is on the formation of a differential aeration cell, whereas other differential cells also may affect crevice corrosion. (d) The relationship of time, chloride concentration and passivity is not explained clearly. (a) First Stage The following is the initial reaction when stainless steel is placed in a oxygenated neutral chloride solution: Anode: M -> M 2 + + ze Cathode: H 2 0 + 2e + - 0 2 -> 2 0 H " Overall reaction: 1 2M 2+ + - 0 2 + 2 H 2 0 + 4e (4.22) (4.23) 2M(OH)2 The mechanism is, however, far more com(4.24) plex than given by a unified mechanism. This phenomenon is highly unpredictable because of Due to passivation, the potential of the specmany of design, metallurgical and environmental imen shifts into the passive region where the factors associated with it. It is, therefore, essen- small /passive is equal to the cathodic current. The tial to identify the parameters before any reliable film becomes thicker by the driving force cremechanism could be given. The development of ated by the cathodic reaction. If steel containing mathematical models have been a useful step in a crevice is now placed in seawater, the reacthis direction [13]. tion would occur all over the surface. However, 1 46 Principles of Corrosion Engineering and Corrosion Control S TAGE-I STAGE - II STAGE-III STAGE - IV Depletion of oxygen In the crevice solution Increase in acidity and chloride content of the crevice solution Permanent breakdown on the passive film and then onset of rapid corrosion Propagation of crevice corrosion Figure 4.18 The four stages of crevice corrosion if the crevice is very small, oxygen diffusing into ( b) Second Stage it is slower than its removal by the cathodic reaction [14]. The solution in the crevice becomes In this stage, the cathodic reduction of oxygen slowly depleted in oxygen (Fig. 4.19). In the first proceeds outside the crevice and slow dissolustage: tion of the metal takes place inside the crevice. The concentration of metal cations produced by (a) Oxygen is depleted and the solution inside the reaction M —• M z + + ze inside the crevice becomes deoxygenated. increased until the solubility of one of the hydrox(b) The metal ion concentration is increased. ides is exceeded. The entry of metal ion through Initial condition Passive current consumes dissolved Oz Figure 4.19 Stage I of crevice corrosion. (From Kain, R.M. and Lee, T.S. (1979). Technical Paper No. FP 40, Corrosion. Reproduced by kind permission of NACE, Int., Texas, USA) Types of corrosion: materials and environments the passive film has two effects: (1) chloride ions migrate from the bulk solution to the crevice to maintain charge neutrality and (2) hydrolysis of metal chloride immediately takes place which causes an increase in acidity inside the crevice (Fig. 4.20). 1 47 The aggressiveness of the solution inside the crevice is increased. (c) Third Stage In the third stage (Fig. 4.21), accelerated corrosion takes place due to the breakdown of the passive M"+ + HOH -> M(OH) ( n _ 1 ) + H+(hydrolysis) film, because the solution inside the crevice in the (4.25a) second stage is highly aggressive. The concentration of solution at which the passive film breaks M"+(cr) + HOH -> MOH + H + cr down is called 'critical crevice solution.' (4.25b) or (d) Fourth Stage M" + (C1~) 2 + 2 H 2 0 -> M(OH) 2 + 2 H + + C I" (4.25c) In the final state (fourth stage), the crevice corrosion continues to propagate. The propagation is terminated when the metal is perforated. Rapid dissolution of the alloy inside the crevice continues. The process is autocatalytic; once it starts, it continues until termination. The acidity within the crevice in increased and the pH inside the crevice is reduced. The cations (M n + ) are moved out of the crevice and the anions (Cl~) are moved inside the crevice. Critical crevice solution break down passivity Breakdown occurs Rapid corrosion begins Figure 4.20 Stage II of crevice corrosion. (From Ka L, R.M. and Lee, T.S. (1979). Technical Paper No. FP 40, Corrosion. Reproduced by kind permission of NACE,it., Texas, USA) 1 48 Principles of Corrosion Engineering and Corrosion Control They are: critical crevice solution (CCS), passive current (Ip) and composition of the alloy. (5) What value of passive current for stainless steel is assumed? From 10 - 1 to 10 A-cm"2. 4 .5.6 CASE HISTORIES [IS] (a) Case 1 Development of leaks at the rolled joint of the pipe and pipe bottom in a heat exchanger. Description of the problem A heat exchanger Figure 4.21 Stage III of crevice corrosion. (From was made from a pipe of 35-29 stainless steel, Kain, R.M. and Lee, T.S. (1979). Technical Paper No. 25/30 mm diam. The oil at the pipe was heated FP 40, Corrosion. Reproduced by kind permission of externally from 90 to 170° C by superheated steam NACE, Int., USA) of8-10atm. 4 .S.5 SOME IMPORTANT ON Critical crevice solution break down passivity Breakdown occurs Rapid corrosion begins QUESTIONS AND ANSWERS THE MATHEMATICAL TO P R O V I D E A BASIC MODEL Investigation A physical examination of the pipe showed that the surface was pitted all around the rolled joint very close to the steam chamber. Also grooves were noticeable in the vicinity of this spot. UNDERSTANDING (1) (2) (3) (4) Identification of localized corrosion The attack was due to crevice corrosion. The rolled joint was not very tight, and, therefore, the steam conWhy pH fall is so important in the model? densate penetrated into the gap. A differential The passive current generates metals ions aeration cell was formed due to the depletion of within the crevice, which hydrolyze and gen- oxygen in the crevice formed by the rolled joint erate hydrogen ions. This is indicated by a because of its not being tight. fall in pH. Hence, fall of pH is very important to understand what is happening within Prevention (1) Inconel 625 could be used as an alternate the crevice. alloy in such a situation. Alloy 20 Cr-25 What can be predicted by a plot of crevice Ni-4.5 Mo-1.5 Cu could also provide a useful solution composition and the change of pH? The change in the concentration of metal ion service. can be predicted in the crevice. For instance, (2) Minimize the gap by a change in design of the as pH falls, the concentration of Cl~ within rolled joint. the crevice increases. These changes show the exit of positive ions from the crevice and entry of negative ions into the crevice. (b) Case 2 At what pH value generally is the passive film assumed to be broken? Crevice corrosion of a tubing in a hydraulic oil For most stainless steels, a pH between 1 and cooler. 2 indicates the breakdown of a passive film. What are the three important parameters Problem Leakage occurred in horizontal heat which form a basis for determination of exchanger tubes in an electric power plant after 18 months of service. crevice corrosion susceptibility of the alloy? Types of corrosion: materials and environments Service conditions (a) Coolant - river water. (b) Tube dimensions: 9.5 mm, 0.65 mm wall thickness. (c) Material - Cu-10% Ni. Investigation Visual examination revealed nodules on the inner surface and holes through the nodule. On metallographic examination of a pit, a high rate of attack was observed. On examining a nodule, it was found that corrosion had penetrated 65%. A greenish residue of copper carbonate hydroxide [CUCO3, Cu(OH)2] was observed. The formation of nodules is very characteristic of crevice corrosion. They are generally isolated from each other. The nodules are generally black from inside and rusted on the outside. Microbial corrosion is very likely in such situations, enhancing crevice corrosion. Conclusion The tubes failed by crevice corrosion. Deposits on the surface of the tubes were formed from the dirt in the river water. 1 49 (8) Use alloys resistant to crevice corrosion, such as titanium or Inconel. Increased Mo contents (up to 4.5%) in austenitic stainless steels reduce the susceptibility to crevice corrosion. Use appropriate alloy after prescribed service tests for a specific application. (9) Apply cathodic protection to stainless steels by connecting to adjacent mild steelstructure. (10) For seawater service, maintain a high velocity to keep the solids in suspension. (11) For better performance of steels in seawater, allow intermittent exposure to air to allow the removal of protective films. (12) Use inhibiting paste, wherever possible. (13) Paint the cathodic surface. (14) Remove deposits from time to time. (15) Take precautions against microbial corrosion, which creates crevices and is very damaging to low Mo stainless steels. Prevention and remedial measures The following 4 . 6 measurements were adopted: 4 .6.1 PITTING CORROSION DEFINITION (1) The cooling water supply was changed by a It is a form of localized corrosion of a metal cleaner water not containing dirt. (2) The tubings were replaced. Even those nod- surface where small areas corrode preferentially ules which were not leaking, were also leading to the formation of cavities or pits, and the bulk of the surface remains unattacked. Metals replaced. which form passive films, such as aluminum and steels, are more susceptible to this form of corro4.5.7 P R E V E N T I O N O F sion. It is the most insidious form of corrosion. It causes failure by penetration with only a small CREVICE CORROSION percent weight-loss of the entire structure. It (1) Use welded joints in preference to bolted or is a major type of failure in chemical processing industry. The destructive nature of pitting riveted joints. (2) Seal crevices by using non-corrosive mate- is illustrated by the fact that usually the entire system must be replaced. rials. (3) Eliminate or minimize crevice corrosion at the design stage. ENVIRONMENT (4) Minimize contact between metals and plas- 4 .6.2 tic, fabrics and debris. (5) Avoid contact with hygroscopic materials. Generally, the most conducive environment for (6) Avoid sharp corners, edges and pockets pitting is the marine environment. Ions, such as Cl~, Br "and I - , in appreciable concentrations where dirt or debris could be collected. (7) In critical areas, use weld overlays with tend to cause pitting of steel. Thiosulfate ions also induce pitting of steels. highly corrosion resistant alloy. 1 50 Principles of Corrosion Engineering and Corrosion Control are more susceptible to pitting than steels with large grain sizes. Aluminum also pits in an environment that cause the pitting of steel. If traces of Cu 2+ are present in water, or Fe +3 ions are in water, copper or iron would be deposited on aluminum metal surface and pitting would be initiated. Oxidizing metal ions with chloride, such as cupric, ferric and mercuric, cause severe pitting. Presence of dust or dirt particles in water may also lead to pitting corrosion in copper pipes transporting seawater. With soft water, pitting in copper occurs in the hottest part of the system, whereas with hard waters, pitting occurs in the coldest part of the system. 4 .6.3 CONDITIONS 4 .6.4 SHAPES The most important condition is that the metal must be in a passive state for pitting to occur. Passive state means the presence of a film on a metal surface. Steel and aluminum have a tendency to become passive, however, metals which become passive by film formation have a high resistance to uniform corrosion. The process of pitting destroys this protective film at certain sites resulting in the loss of passivity and initiation of pits on the metal surface. It may be recalled that passivity is a phenomenon which leads to a loss of chemical reactivity. Metals, such as iron, chromium, nickel, titanium, aluminum and also copper, tend to become passive in certain environments. The following are the conditions for pitting to occur: Studies on stainless steels have shown that sulfide inclusions are the most probable sites for pit nucleation. Pits may grow in several forms, such as circular, square, pyramidical and hexagonal. However, it is the depth of the pit which matters more than the shape of the pit. Crystallographic pits may be observed in certain alloys, such as Al 5052 associated with very low dissolution rates. Pits generally grow downwards from horizontal surface. Several months or sometimes even years may be needed before the pits become visible. The period intervening between their initiation and becoming visible is called induction period, which depends upon a particular metal and environment. 4 .6.5 MECHANISM OF PITTING In order for pitting to take place, the formation of an anode is a prerequisite. With the formation of an anode, a local corrosion cell is developed. The anode may be formed as a result of (a) lack of homogeneity at the metal corrosive interface. Lack of homogeneity on the metal surface is caused by impurities, grain boundaries, niches, rough surface, etc. The difference in the environments can cause, for (1) Breaks in the films or other defects, such as a instance, the formation of concentration cells lack of homogeneity in the film on the metal on the metal surface. surface. (b) destruction of a passive film. The destruc(2) The presence of halogen ions, such as Cl~, tion of a passive film results in the formation Br~ and I~, and even S20^~. of a small anode. Therefore, a breakthrough (3) Stagnant conditions in service. Steel pumps of the protective film on the metal surface in seawater serve for a good number of years results in several anodic sites and the suras long as they keep on working. When taken rounding surface acts as a cathode. Thus, an out of service even for a short period stainless unfavorable area ratio results. steel pumps tend to pit. For this reason, stainless steel pumps must run for a few minutes (c) deposit of debris or solids on the metal surface. This generally leads to the formation of several times per week even if not required anodic and cathodic sites. for pumping duty. (d) formation of an active-passive cell with a Sites which are most susceptible to pitting are large potential difference. The formation of grain boundaries. Steels having small grain size a small anode on the passive steel surface, for Types of corrosion: materials and environments instance, leads to the formation of the above cell. 1 51 Conditions (1) The passive metal surrounding the anode is not subject to pitting as it forms the cathode and it is the site for reduction of oxygen. (2) The corrosion products which are formed at the anode cannot spread on to the cathode areas. Therefore, corrosion penetrates the metal rather than spread, and pitting is initiated. (3) There is a certain potential characteristic of a passive metal, below which pitting cannot initiate. This is called pitting potential, £p. With the above considerations, it should be possible to suggest a simple mechanism for pitting corrosion. 4 .6.6 PITTING PROCESSES surface. The metal dissolution takes place at the anode. Pitting is, therefore, initiated. The anodic metal dissolution reaction is represented by M -> M n + + ne (4.26) This is balanced by the cathodic reaction of oxygen on the adjacent surface. 0 2 + 2 H 2 0 + 4e -> 40H" ( 4.27) (1) The formation of anodic sites by disruption of the protective passive film on the metal Initially, the whole surface is in contact with the electrolyte containing oxygen so that oxygen reduction as shown above takes place (Fig. 4.22). (2) Due to the continuing metal dissolution, an excess of positive ions M + is accumulated in the anodic area. The process is self-stimulating and self-propagating. Conditions are produced within a pit which are necessary for continuing the activity of the pit. To maintain charge neutrality negative ions (anions), like chloride, migrate Electrolyte surface 02 02 Ha Na+ a Na* Na+ Cathode *• W a- a / o2 ii OH" OH" N 0 H a* MOH OH' QW Cathode NCi+H^O Figure 4.22 v *MOH + HCI Schematic of an active corrosion pit on a metal in a chloride solution 1 52 Principles of Corrosion Engineering and Corrosion Control from the electrolyte (consider, for example, 4 .6.7 MAIN S T E P S INVOLVED seawater or a 5% NaCl solution) IN THE P ITTING O F C ARBON M + cr + H2O -> MOH + H + + or (4.28) S T E E L A ND S T A I N L E S S S T E E L (3) (4) (5) (6) The following are the main steps involved in the pitting of carbon steel and stainless steels. According to Wranglen [16], sulfide incluOH~ ions also migrate (but more slowly) sions are responsible for the initiation of attack in to neutralize the positive charges. The reaction of M + C1~ with H 2 0 resulting in the for- both carbon steel and stainless steels. The followmation of M + OH~ + H + C1~. This process ing reactions occur in the pit interior (Fig. 4.23): is called hydrolysis (equation (4.28)). Fe -> Fe 2+ + 2e (4.29) The presence of H + ions and chloride content, prevents repassivation. The above proF e 2+ + H 2 0 -> FeOH + + H + (4.30) cess generates free acid and the pH value at the bottom of the pit is substantially lowered. MnS + 2H+ -» H2S + Mn + 2 (4.31) It has been measured between 1.5 and 1.0. The above reactions (equations (4.29-4.31)) The pH value depends on the type of steelvalues for stainless steel pits which are lower are anodic. Reaction (4.30) results in the decrease + than for mild steel and have a pit pH of 4. of pH due to the production of H ion which This is due to the solubility product effects reacts with the inclusion (MnS) and results in the +2 and depends on the presence of chromium. formation of H2S and Mn , as shown in equa+ In stainless steels, the pH is reduced by tion (4.31). Hydrogen is evolved as H accepts 2_ 3+ 2+ hydrolysis of Cr and Fe as well as by electrons released by the anodic reaction. S and HS~ stimulate the attack. Because of an excess accumulation of chloride. The increase in the rate of dissolution at of positive ions as shown by equation (4.29), the the anode increases the rate of migration concentration of the Cl~ ions increases with time. of the chloride ions and the reaction becomes As the process continues with time, the pit depth time dependent and continues, resulting in becomes more and more. Rust is formed at the pit mouth. It may be the formation of more and more M + C1~ and, therefore, generation of more and more either Fe3C>4 or (FeOOH). Further oxidation of F e 2+ and FeOH may take place at the pit mouth: H + C1~ by hydrolysis (equation (4.28)). The process continues until the metal is perforated. The process is autocatalytic and it 2FeOH+ + - 0 2 + 2H+ -> 2Fe(OH) 2+ + H 2 0 increases with time resulting in more and (4.32) more metal dissolution. Finally, the metal is perforated and the 2Fe2+ + - 0 2 + 2 H + -> 2Fe3+ + H 2 0 reaction is terminated. (4.33) As shown above, basically three processes are involved: (a) pitting initiation (b) pitting propagation (c) pitting termination. Step (1) of the mechanism shown above, describes the initiation of pitting, steps (2-5) describe the propagation of pitting, and step (6) termination of pitting. The rust formed does not allow the mixing of the products of the anodes and the cathode (anolyte and catholyte). The pH of the solution falls at the pit mouth as H + is formed by hydrolysis of FeOH + or Fe 3+ formed (equations (4.34) and (4.35)): FeOH 2+ + H 2 0 -> Fe(OH)+ + H + F e 3+ + H 2 0 -> FeOH 2+ + H + (4.34) (4.35) Types of corrosion: materials and environments 1 53 , 02 402 IO2 Jo* io* Neutral aerated NaCI solution o J^ 02+2H20+4e - JO2 o o Porous crust 1=teOOH !° CI Fe3 4 a 40H- J02 °* Ha0 Jo2 Fa —* * 3Fe0 + ^ Steei MnS+2H Concentrated acid pit solution containing HjS 2H* + 2 e—*Ha Fe — * Fe2* + 2e* Occasional precipitation of solid salt Feth.4H20 Figure 4.23 Electrochemical reactions that occur when a pit is initiated at sufide inclusion in a carbon steel. (From Wranglen, A. (1974). Corrosion Science, 14, 331) The process continues, the pH falls, and the reaction becomes autocatalytic. Outside the pit, the surface is cathodically protected, the main reactions being the reduction of oxygen and the formation of rust. The surface adjacent to the pit acts as the cathode as shown earlier, and, therefore, it does not pit. The main cathodic reactions are: 4.6.8 MECHANISM OF PITTING CORROSION OF ALUMINUM One important mechanism suggests the following steps which lead to pitting [16,17]: (1) Absorption of the halide on the oxide film (at the oxide solution interface): Cl~ (in bulk) = Cl~ (adsorbed on AI2O3, n H 2 0 sites) (4.38) (2) Formation of basic hydroxychloride aluminum salt which separates from the lattice and goes into solution: Al + + + (in AI2O3,nH2 lattice) + 4C1" (adsorbed) = A1C1~ in solution of lower pH (4.39a) 0 2 + 2 H 2 0 + 4e -> 40H~ (4.36) 3FeOOH + e -> Fe 3 0 4 + H 2 0 + OH" (4.37) Production of O H - ions increases the pH of the cathodic area. Because the electrons are consumed at the surface adjacent to the pit, the surface around the pit is also cathodically protected to a certain extent. 1 54 Principles of Corrosion Engineering and Corrosion Control A l + + + (in A1203, nH 2 0 lattice) + 2C1~ + 2 0 H " = Al(OH)2Cl^ (readily soluble) (4.39b) Al(OH)2Cl~ is formed by: Al(OH)3 ^ Al(OH)+ + O H Al(OH)+ + Cl~ -> Al(OH)2Cl polarization curve, as shown in Fig. 4.24. The diagram is conveniently divided in three regions as below: (1) Region I: This region represents immunity to pitting and at a potential more negative than £pP (protection potential) pitting is not expected to propagate. (2) Region II: In this region, new pits do not initiate, but the pits already initiated in Region III continue to propagate. This region is, therefore, the propagation region for pits. (3) Region III: In this region, pits initiate and propagation take place on and above a certain potential, called breakdown or critical pitting potential (Ep). The critical pitting potential signifies the onset of pitting. Corrosion potential is the potential at which the rate of oxidation equals the rate of reduction. It is a potential that a corroding metal exhibits under specific conditions of time, temperature, aeration, velocity, etc. The pitting potential, the protection potential and corrosion potential, can be read from the polarization hysteresis diagram. The pitting resistance of aluminum alloys may be compared and predicted by the magnitude of the above values and the relative position of the pitting potential (Ep) compared to the corrosion potential. For instance, if the value of pitting potential (Ep) is positive to the corrosion potential (Ec), the resistance of the alloy to pitting is high. Generally, the more negative the value of Ec compared to Ep, the high is the pitting resistance. There is a disagreement between the workers on the application of the above parameters for prediction of pitting resistance of aluminum alloys, however, it is useful to understand the significance of these parameters. In contrast to the situation for aluminum, electrochemical studies on the pitting of stainless steels are fraught with contradictions because of the effect of experimental factors. Several investigations have used breakthrough potential or breakdown potential, as a measure of pitting susceptibility. As pointed out earlier, the current increases rapidly at the breakdown potential (Fig. 4.24), this is considered as an indication of a breakdown of the passive, corrosionresistant steel film and initiation of pitting. (3) The thinning of oxide films: The oxide film is thinned to the extent that aluminum ions can pass from the metal to the solution interface. The chloride ion gets entry by one of the following possible methods: (a) By competitive absorption of halide ions. (b) Penetration of halide in the oxide film. (c) Diffusion of halide through the oxide film and attack on the metal. (d) Formation of soluble complexes with the halide ions [17]. For a complete understanding of the mechanism of corrosion of aluminum, refer to literature cited at the end of the chapter. 4 .6.9 ELECTROCHEMICAL PARAMETERS OF PITTING The classical techniques involving exposure to a corrosive medium and metallographic examination by optical microscopy have been supplemented by time-saving electrochemical techniques in recent years. One such electrochemical technique is the polarization hysteresis technique used for observing the anodic polarization behavior of steels, aluminum and their alloys. The anodic polarization scan is commenced from the open circuit potential of aluminum and continued in the noble direction at a slow scan rate. The scan is reversed on reaching a predetermined current value and continued in a reverse direction until the loop is closed at a more active potential and the current drops to a minimum value. The plot obtained is called a cyclic Types of corrosion: materials and environments 1 55 r < 4 Critical breakdown potential. Chloride ions breakdown passlvatfng film. Region III Initiation of Surface starts to passivate pitting corrosion Region I I Propagation of pit and crevice corrosion Region I Critical protection potential Reverse polarization back to more active potentials. Immunity - No corrosion Anodic polarization from active to noble potentials. Passivating surface film inhibits corrosion. JO 2 o -L 1Q? 10 « 10 * 1(r* 10 3 1 0* 2 10* Current density mA/cm* Figure 4.24 Shape of the polarization curve for a typical stainless steel in a solution showing how corrosion behavior changes as the passivating surface film forms Figure 4.25 shows the effect of chloride ion concentration on the polarization behavior of steel 304. The effect of various factors, such as concentration, heat treatment and composition of the solution, can be determined by breakdown potentials. For instance, in Fig. 4.24, an increase in the chloride concentration from 1000 ppm to 40 000ppm decreases the breakdown potential from +400mVscE to about — lOmVscE- It is to be pointed out that below £p, the metal is in the passive state and pitting does not occur. Above £p the passivity breaks down and pitting occurs. The most accurate method of measuring Ep is by the potentiostatic or potentiodynamic technique. In this technique, the potential is held at an assigned value which can be changed gradually at a constant rate, to plot potential against current. The potentiostat keeps the potential difference between specimen and reference electrode at a set value, by applying any necessary voltage to an auxiliary electrode and the voltage and current are recorded. 4 .6.10 FACTORS AFFECTING T HE B REAKDOWN P OTENTIAL ( PITTING POTENTIAL) (a) Chloride ions concentration: A certain concentration of chloride ions is necessary to initiate pitting of steel and aluminum alloys. 1 56 Principles of Corrosion Engineering and Corrosion Control LO 0,8 h n 10 ppm 0 0.6 100 ppm CJ1(KK) ppm C! K o > 0.4 m S3 C \a.0000 ppm €1 1 40000 ppm 0 t\ 2 ' t -i i " i i i i i i i mm i i i ,mill t i i mm i > i n un t t < mm i i i i,mii 0.001 0,01 0,1 1 10 100 Current density, mA/cm2 Figure 4.25 Potentiodynamic polarization curves of type 304 stainless steel in various chloride ion concentration using NaCl as the salt (b) Alloying elements: The effect of certain alloying elements on the breakdown potential is important. Generally, the alloying elements which have a beneficial effect shift £p to a more positive value, whereas the elements having detrimental effects shift £p to more negative value. Generally, Mo, Cr, Ni, V, Si, N, Ag and Re improve pitting resistance of stainless steels. (c) Effect of electrolyte composition: Ions, such as SO^",OH-,C10~,CO?" and MoO2 , w n , ^ w >^^3 <"iu x v i u w 4 4 shift Ep to more positive value. Nitrates and chromates when added to a solution containing Cl~, inhibit pitting. In the case of 18-8 stainless steels, the inhibitor efficiency of ions decreases in the order: OH~ > NO^~ Acetate > SO 2 " > CIO". (d) Effect of pH: In the acid range of pH (pH value less than 6), the pitting potential remains unaffected by pH changes as shown by 18-8 stainless steel in 0.1 NaCl. In the alkaline range, however, the pitting potential shifts in the noble direction. The pitting potentials obtained in this region are generally non-reproducible and general dissolution take place in a limited range. Thus, in the alkaline region, the dissolution pattern is changed from pitting to uniform corrosion. It is to be observed from the effect of pH on critical potential, that pitting is stopped by addition of hydroxyl (OH") ions. As shown above, the measurement of pitting potential (Ep) can be utilized to predict the tendency of steels to pitting in chloride containing environment. (e) Effect of temperature: It has been shown that an increase of temperature by 10°C shifts the critical pitting potential (£p) by 30 mV for certain grades of steels. This relationship is, however, differs for other grades of steels. Generally a rise in temperature shifts the critical pitting potential (£p) in the more active direction, which leads to acceleration of pitting. Critical pitting potential (£p) is synonymous with the breakdown potential (£b)- There is a critical pitting temperature below which pitting will not occur. Types of corrosion: materials and environments (f) Effect of heat treatment and cold working: It is not possible to make a general statement on the effect of heat treatment on the pitting potential of steels as different types of steels respond differently to cold working or hot working. The nature of alloying elements, their concentration and microstructural changes that they introduce, affects the response of various grades of steels to cold working and hot working. For instance, the effect of cold working on the value of £p for 18 Cr-10 Ni, 25 Cr-25 Ni, steel is negligible. In case of 18 Cr-2 Ni-Ti steel, alloyed with 3% Si, the pitting resistance is appreciably decreased. (g) Induction time (T): This is the time required for the first pit to initiate. The induction period of pitting varies from one steel to another and depends on the environment. Mainly, the induction time depends upon the concentration of Cl~ ion in the solution. The higher the concentration of chloride, the less the induction time [18,19]. It has been observed that the induction time decreases with increasing potential or the induction time increases with increasing potential. Studies on the induction time of 18 Cr-8 Ni steel have shown that the rate of breakdown of oxide film by the chloride ions resulting in initiation of pit is proportional to the chloride ion concentration [18]. It has also been observed that Ep[t (current density inside pit) decreases with decreasing potential, decreasing chloride concentration and increasing temperature for steel [20,21]. 1 57 (3) Surface finish. A smooth surface does not allow accumulation of impurities and stagnancy to be built up on the metallic surface, thus minimizing the risk for the formation of differential aeration cells and minimizing corrosion and pitting. However, the pits which are formed on a smooth and shiny surface are deeper and larger compared to a rough surface. (4) Sensitizing temperature. The temperature zone which promotes intergranular corrosion of steels also promotes pitting, because of the weakening of grain boundaries. (5) Velocity. It is a matter of common observation that a stainless steel pump handling seawater gives a good service if it is used to run continuously, but it would tend to pit if stopped because of the formation of differential aeration cells. Pits are often associated with stagnant conditions. High velocity may cause passivation of the steel surface by greater control of the oxygen with the steel surface, thus enabling the integrity of the passive film to be maintained. (6) Environmental contamination. An environment contaminated by dust particles would be more conducive to pitting. Dust particles on metals absorb moisture and lead to the formation of differential aeration cells. Salt particles in a marine environment when deposited on a metal surface lead to intensive pitting of equipment and appliances. 4 .6.12 I N F L U E N C E OF ALLOYING E L E M E N T S ON THE PITTING OF ALUMINUM 4 .6.11 FACTORS AFFECTING PITTING (1) Surface defects, such as inclusions, second phases and surface heterogeneity, have been identified with initiation of pits. It has been established that sulfide inclusions are mainly responsible for pitting of carbon and stainless steels. (2) Degree of cold work. Severe cold working increases pitting susceptibility as shown by austenitic steels in ferric chloride solution. (1) Chromium. It is usually added to Al-Mg alloy and Al-Mg-Zn alloys in amounts of 0.1-0.3%. It has a beneficial influence on pitting resistance in seawater. (2) Copper. It reduces the corrosion resistance of aluminum alloys and increases the susceptibility to pitting. The corrosion potential of Al-Cu alloys become cathodic in direct proportion to the amount of Cu in solid solution. (3) Iron. It exists in the form of FeAl3 and promotes pitting of aluminum alloy. At 1 58 Principles of Corrosion Engineering and Corrosion Control low copper levels (<0.05%), addition of 1.25% Mn (e.g. AA-3003), minimizes the detrimental effect of Fe due to the formation ofMn(Fe)Al6. Magnesium. It provides substantial strengthening with good ductility in addition to excellent corrosion resistance. The aluminum-magnesium family (5xxx) have the lowest pitting probability and penetration rates. Alloy 5052 containing 2.5% Mg is found to possess an excellent resistance to pitting in seawater. In the presence of Si, Mg2Si is formed which leads to slight enhancement of pitting and may lead to intergranular corrosion. Pitting is generally combined with intergranular corrosion in Al-Mg alloys. Manganese. Up to 1.25% Mn can be added to AA 1099 without increasing pitting probability. It is normally used in heat treatable alloy AA 5182 in concentration ranging from 0.20 to 0.50%. It reduces the pitting probability of 99.5 or 99.8% aluminum silicon. It does not increase the pitting probability of AA 1099 in water. It has a less pronounced effect on pitting than Fe. At a higher level (0.3%) it can have an adverse effect. Zinc. Up to 1.0% zinc can be added to AA 1099 without increasing the pitting probability. At higher levels, its influence would depend upon other constituents, such as Mg and Cu, and the microstructure produced. Scandium. Aluminum alloys with 0.15 to 0.6% scandium have been reported in recent years. Scandium addition up to 0.6% has a dramatic influence on strength and it does not adversely affect the resistance of Al-2.5 Mg alloys to pitting. 4 .6.14 P I T T I N G O F C O P P E R Copper tubes (alloy C-106) are widely used in water distribution system. One of the serious problems encountered in these tubes containing a small carbon residue, is pitting. The following mechanism of pitting has been suggested. (4) A. Anodic Dissolution Cu -* Cu + + e (4.40) Copper ions (positive charged) combined with chloride ions present in water to form CuCl: C u + + C l " -> CuCl (4.41) (5) Cuprous chloride is not stable in near-neutral pH ranges and hydrolyzes to form cuprous oxide which is precipitated on the metal surface. 2CuCl + H 2 0 = Cu 2 0 + 2HC1 (4.42) B. Cathodic Reduction The main cathodic reaction is the reduction of oxygen on the surface adjacent to the pit: 0 2 + 2 H 2 0 + 4e -> 40H~ (4.43) (6) (7) If the water is acidic, the O H - produced in equation (4.43) is removed. If bicarbonate is present in the water, OH~ is also removed. Pitting occurs in copper tubes at a certain potential, the critical pitting potential The pH inside the pit is substantially lowered by the production of HC1 as shown in reaction (4.42) (Fig. 4.25). 4.6.13 A L L O Y A D D I T I O N IN I RON A ND S T E E L THEORIES O F PITTING Alloying additions of chromium, nickel, molyb- The pitting corrosion process is a highly complex denum, increase the pitting resistance, whereas process with a sequence of steps. The following silicon, sulfur, carbon and nitrogen, decrease are the steps: the resistance to pitting. Addition of titanium increase resistance of steels only in solutions of (1) Breakdown of passivity FeCl3 and not in other mediums. (2) Early pit growth Types of corrosion: materials and environments (3) Late pit growth (4) Repassivationofpits. The primary step in pitting is the breaking of the passive layer on the metal surface which is explained by (A) (B) (C) (D) (E) ion penetration [22] adsorption [23,24] film breakdown [25-27] repassivation mechanism [28,29] point defect model [30]. 159 A. Ion Penetration Theory It is based on the assumption that the tendency of the chloride ions to penetrate and break the passive layer is because of its small diameter. The penetration of the aggressive anions, however, is not an isolated process as it is combined with the transport of O 2 - to the electrolyte to maintain charge neutrality. The transport of O 2 - is, however, a slow process and, therefore, if the above assumption is taken to be true, the charge neutrality would not be maintained. It has also been shown that NOJ has an identical rate of penetration to Cl~, However, pitting is not caused by the former. Also, it has been shown by some studies that the thickness of the passive layer does not affect pitting and hence the rate of penetration of Cl~ ion cannot be a decisive factor. It is known that SO^ - , ClO^and SCN~ cause pitting, whereas according to the penetration theory they should not cause pitting because of their relatively larger diameter compared to Cl~ ion. passivity breaks down resulting in the initiation of pits. It has been observed that the breakdown of a passive surface is accompanied by high current densities. If values of currents are plotted against potential after a pre-determined time, it would be observed that at the instant a pit initiates, the current at a certain value of potential shoots to a very high value. This high value of current indicates the initiation of a pit on a passive surface. This theory, like other theories on pitting, has its own points of weakness. For instance, the displacement of a doubly charged negative ion by a singly charged anion should be theoretically favored by a shift of the potential to more negative values and not to move positive values of potential as suggested by the adsorption theory. If the above suggestions were true, PO^" should be more aggressive than Cl~, which is contradictory to the observations. C. Film Breakdown Theory This theory assumes a direct access of the aggressive anions to the metal surface by breaking the passive film. In order to break the passive layer, a high concentration of aggressive anions is necessary which could be caused by high concentration of the aggressive anions in the bulk solution. The high concentration stabilizes the adsorption of salt layer which prevents the repassivation of pits. Pitting, therefore, is likely to occur at a certain value of concentration (critical concentration). It has been experimentally shown that below a certain concentration, pitting does not initiate. It is very difficult to decide which of the two mechanisms, adsorption or film breakdown, explains the mechanism more accurately. As discussed above, the concentration of aggressive anions and the effect of electrolytic composition on pitting seem to suggest that the adsorption theory is more appropriate to explain the phenomenon observed in pitting. If inhibiting anions are absorbed in place of aggressive anions, pitting would not be initiated because the oxide layer would not be broken as it would be by aggressive anions. This observation can be best explained by the film breakdown theory; the pitting does not initiate until the passive layer is broken. It is, therefore, not possible to B. The Adsorption Theory This theory is based on the assumption that aggressive anions which are absorbed at the energetically favored sites on the passive layer cause the nucleation of pits. In order to overcome the repulsive forces between anions, there should be an exchange of O H - or O 2 - of the passive oxide layer by the aggressive anions. Experiments have shown that the passive layer (oxide layer) is removed and replaced by an adsorbed salt layer containing the aggressive anions. Where O 2 - is absorbed the metal is passivated and where Cl~ is adsorbed, the 1 60 Principles of Corrosion Engineering and Corrosion Control of pitting beyond the basic material provided in most of the books on corrosion. For a better and more detailed understanding of the mechanism of pitting, references are provided at the end of D. Repassivation Theory the chapter. Although the theories outlined above are In repassivation, the aggressive ions at the pit are conflicting, there are certain points which are replaced by a passive layer - at more negative generally agreed upon by most of the investigapotentials. Pitting would, therefore, not initiate. tors. These are: In the process of repassivation, the layer of aggressive anions is replaced by a passive layer. The (a) Pits initiate above a certain critical potenaggressive ions are removed by diffusion from the tial called breakdown, rupture or pitting pit to the electrolyte. Repassivation is expected potential. only if the pitting potential becomes more than (b) The breakdown potential is a function of the flade potential (the potential in a polarizahalide ion concentration. tion curve of an active-passive metal at which (c) The breakdown of a passive film takes place the current density is minimum). Repassivaat a highly localized site. tion is prevented if the concentration of chloride (d) There exists an induction time for a pit to exceeds 1M. appear on the surface. (e) A passive surface is essential for pitting to take In spite of a substantial progress made in the place. field of pitting corrosion, no single theory exists which can explain all the known observations on (f) The induction time is potential dependent. Pitting does not appear at values negative to pitting. the pitting potential. (g) The two processes, breakdown of a passive film, and adsorption of chloride ions, play a E. Point Defect Model major role in the initiation of pits. Whether pits are initiated by film breakdown or by This model assumes the transport of both anions adsorption of chloride or by formation of (oxygen ions) and cations (metal) ions or their complex of chloride is not clear, however, all respective vacancies. The anions diffuse to the the three factors appear to affect the process metal surface with a resultant thickening of the of pitting. film. Cation diffusion cause film dissolution and metal vacancies are created at the metal/film interface. When the rate of production of metal Some other salient features of the pitting mechavacancies in the metal exceeds their rate of migra- nism suggested by various workers are: tion, the vacancies pile up at the metal/film interface and result in void craters. This is the pro- (a) Three or four halide ions are jointly absorbed cess of pit incubation. At a certain critical size, the on the oxide film surface around a lattice passive film suffers local collapse and the incucation and a transitional complex is formed bation period ends. The collapsed site dissolves which possesses high energy, and separates much faster than at any other place and leads to from the oxide. Once the complex is formed, pit growth. the process is repeated until the passive layer is completely destroyed. There are other mechanism, such as that of Pickering [31], however, discussion of all of them (b) Pits develop on the site where oxygen is beyond the scope of this chapter. absorbed on the metal surface is displaced by Cl~ ions. The breakdown potential (pitThe theories described above are summaries ting potential) is the potential at which of the work done by various investigators. As pitting is an extremely important form of localized the aggressive anions produces a reversible corrosion, it is necessary for engineering studisplacement of the oxygen from the metal dents to have some ideas about the mechanism surface. distinguish between the merits of the two theories clearly. Types of corrosion: materials and environments (c) The bond between the metal and oxygen is weakened at many points and exchange of chloride ions with oxygen takes place at these points. (d) Voids are created by accumulation of vacancies at the metal/film surface and at a certain critical void size, the film suffers damage. 1 61 B. Case 2 An impeller was made of steel, ASTM designation A296, Grade CF-8, having the following composition: C = 0.06, Si = 1.47, Mn = 0.98, P = 0.02, S = 0.019, Ni = 8.88, CI = 18.81, Balance Fe. The impellers were inspected after three years of operation. Pitting was observed to be the major problem. Conditions The same as for the casing in Case 1 above. Material 18-8 Cr-Ni steel. 4 .6.16 C A S E H I S T O R I E S A. Case 1 A seawater pumping plant had been designed to Remedial measures The following measures provide seawater for cooling purpose to a thermal were recommended: power plant (TPP), a turbo-blower station (TBS) and other plants, such as refrigeration, oxygen, (1) Steel 18-8 with 2% Mo be used instead of rolling mills and refractory plants, in a steel mill. 18-8 Cr-Ni steel. Pitting was observed to be the major problem in (2) The pH of the seawater be maintained at welds of the outer casing of the main pump and 11.00. in impellers. Conditions The seawater contained 36 020 ppm TDS, 24 200 ppm chloride and 3 076 ppm SO4 ions. The average pH was 8.0. C. Case 3 Pitting of Type 321 stainless steel aircraft fresh water tanks caused by retained metal cleaning Inspection After three years of operation, the solution. inside of the casing was examined. The surface was found to be covered by a thin dark brown layer Conditions One tank had been in service for 32 h of uniform thickness over the surface. Removal and the other for 10 h. The tanks contained fresh of the layer showed a large number of pits on water. the weld metal. On analyzing the problem, it Service history The sodium hypochlorite soluwas found that the following factors caused tion used for cleaning of the tank was three times pitting: the prescribed strength. After sterilizing, a small amount of cleaning solution remained in the (a) High contents of dissolved salt bottom of the tank. (b) Suspended solids like sand, silt, etc. Observations Metallographic examination (c) Seepage of seawater through concrete. showed a large number of pits, most of the pits Remedial measures The following remedial were about 6-13 mm (1/4-1/2 in) from the weld bead at the outlet. measures were adopted: (1) The pH of the water was maintained at 12.5. (2) Seepage of the seawater through the concrete was stopped. (3) The casing was cathodically protected by galvanic anodes. Remedial measures It was recommended that (1) the cleaning solution of hypochlorite be made according to the recommended strength. Material The casing material was stainless steel (304 L). 1 62 Principles of Corrosion Engineering and Corrosion Control is related to the sample size, it is often used as a criteria of pitting resistance in laboratory tests. The prediction should always be made on a combination of results obtained from number of pits per mm 2 and average pit depth. The development of pit depths follows a time function: D = Ktc (4.44) (2) the drainage after cleaning be thoroughly done and no amount of cleaning solution be retained in the vessel. (3) welding be properly controlled and with an inert gas to avoid the formation of a black scale. (4) the design of the tank be changed to achieve 100% drainage of the liquid. 4 .6.17 P R E V E N T I O N O F PITTING CORROSION where (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) D is depth of the deepest pit Kis an alloy surface area and environment-dependent constant Use materials with appropriate alloying elet is time and ments designed to minimize pitting susc is an environment-dependent parameter ceptibility, e.g. molybdenum in stainless often close to a. steel. For alloys showing a good resistance to pitting Provide a uniform surface through proper cleaning, heat treating and surface finishing. with t in years and D in mm, K can be typically Reduce the concentration of aggressive close to 0.75. Estimation of the life time of matespecies in the test medium, such as chlo- rial on the basis of controlled laboratory tests is of little use. However, by extreme value statistical rides, sulfates, etc. treatment of the data quantitatively, the life and Use inhibitors to minimize the effect of time of materials can be satisfactorily predicted pitting, wherever possible. Make the surface of the specimen smooth [32,33]. and shiny and do not allow any impurities to deposit on the surface. Minimize the effect of external factors on INTERGRANULAR those design features that lead to the local- 4 . 7 ized attack, such as the presence of crevices, C O R R O S I O N sharp corners, etc. Apply cathodic protection, wherever pos- Intergranular corrosion refers to preferential sible. corrosion along the grain boundaries. Grains are Coat the metals to avoid the risk of pitting. 'crystals5 usually on a microscopic scale, that conDo not allow the potential to reach the stitute the microstructure of the metal and alloys. critical value. By analogy, they are like the grains of sand which Add anions, such as OH~orNO^ to chlo- constitute a sandstone. ride environment. Operate at a lower temperature, if service conditions permit. 4 .7.1 DEFINITION It has been defined commonly as a form of localized attack on the grain boundaries of a metal or alloy in corrosive media, which results in the loss of strength and ductility. The localized attack may lead to dislodgment of the grain. It works inwards between the grains and causes more loss of strength than the same total destruction of metal uniformly distributed over the whole surface. The attack is distributed over all the grain 4 .6.18 E V A L U A T I O N O F PITTING DAMAGE Conventional weight-loss techniques cannot be used to evaluate pitting because of its localized nature. What matters in pitting is the depth of the deepest pit, which is an indicator of the resistance of a material to pitting. Although the depth Types of corrosion: materials and environments boundaries cutting the surface. Intergranular corrosion is less dangerous than stress corrosion, which occurs when stress acts continuously or cyclically, in a corrosive environment, producing cracks following mostly intergranular paths. The difference between the two, i.e. intergranular corrosion and stress corrosion cracking is important. There are many materials which are susceptible to intergranular corrosion and not to stress corrosion cracking. The attack is very common on stainless steel, nickel and aluminum alloys. Metals are crystalline materials. They consist of grains. As corrosion proceeds along the grain boundaries, the grains become weaker particularly at the grain boundaries and they eventually disintegrate (Fig. 4.26). All metals and alloys are joined together by grain boundaries. The intergranular corrosion of steels, brasses, bronzes and aluminum alloys containing copper is of particular interest to engineers. Because of the importance of steels, the largest amount of work reported in the literature is on steels. It would be, therefore, appropriate also here to review the phenomenon of intergranular corrosion with a particular emphasis on steels. It would be appropriate, hence, to review 1 63 the types of steels and some important technical terms related to intergranular corrosion. 4 .7.2 TYPES OF STEELS On the basis of their structure and hardening response, stainless steel may be divided into a number of classes which are summarized in Table 4.3. A. Austenitic Stainless Steels Austenitic stainless steels have a face centered cubic structure that is attained by the addition of further alloying addition to chromium and iron, such as nickel, manganese, carbon and nitrogen. Nickel is added in concentration ranges of 3.5-3.7%, Mn in the range of 1.15%, nitrogen in the range of 0.1-0.4%, and carbon from 0.02 to 1.0%. The following are the main characteristics: (1) Strength. Increase in strength obtained by cold working. For maximum work hardening low chromium and nickel concentrations are recommended. Figure 4.26 SEM micrograph of grain boundary corrosion in HAZ of 316L SS Weldment (x 900) 1 64 Principles of Corrosion Engineering and Corrosion Control Table 4.3 Steel type Austenitic Classification of stainless steels Strength range (MN/m2) 200-500 Characteristic properties Non-hardenable except by cold working. Very high toughness. Very good formability. Creep resistant. Non-magnetic. Very good corrosion resistance. High cost. Good formability. Inferior corrosion resistance to austenitic steels. High temperature oxidation resistance. Lower cost than austenitic steels. High strength and wear resistance. Inferior toughness. High strength. High toughness. Optimum creep properties. High cost. Ferritic 250-500 Martensitic Precipitation hardened 400-2000 500-1640 (2) Toughness. Very high impact toughness. They are used for cryogenic applications. (3) Corrosion resistance. Excellent resistance to atmospheric corrosion and scaling at elevated temperatures. Corrosion resistance in acids and chloride solution is increased by addition of molybdenum. Addition of titanium and niobium increases resistance to intergranular attack. Scaling resistance is increased by addition of chromium and silicon. (4) Creep resistance. For high temperature application, steels stabilized with titanium or niobium are used. (5) Weldability. To avoid intergranular attack steels containing niobium or titanium are used. Austenitic steels are easily welded by inert gas welding. On the whole, austenitic chromium nickel steels are used with the advantage for pipings, pressure vessels and other numerous applications, because they posses a unique combination of properties summarized above. The composition and mechanical properties are shown in Tables 4.4 (a-d). with lower chromium contents. AISI Type 409 and AISI Type 430 ferritic steels are used for vehicle mufflers and automobile trims, respectively. Ferritic steels have BCC structure. The following are the main characteristics. (1) Strength. They possess relatively lower strength than austenite steels. They are used in either the annealed or cold work condition. (2) Toughness. Impact toughness decreases with increasing strength and carbon content. Optimum toughness may be obtained by tempering above 650° C. Upon heating in the range (399-482°C), for prolonged periods, the notch toughness is reduced. (3) Corrosion resistance. They do not offer a good resistance to atmospheric corrosion as chromium nickel austenitic steels. Their resistance to corrosion is increased by addition of molybdenum and tantalum. (4) Formability. The formability is not as good as that of austenitic steels. (5) Weldability. Is not as easy as for austenitic steels. Post-welding treatment is necessary to avoid cracking. B. Ferritic Stainless Steels They usually contain 16-30% chromium, although it is possible to have a ferritic structure C. Martensitic Stainless Steels They contain 11-16% chromium as alloying element, although it is possible to develop Types of corrosion: materials and environments Table 4.4a The composition and mechanical properties of steels - Austenitic stainless steel Nominal composition (%) Cr 18 17 18 19 19 18 23 25 24.5 17 18 18 18 Ni 8 7 8 10 10 11.5 13.5 20.5 20.5 12 10.5 11 3.5-6.0 Mo C max 0.15 0.15 0.15 0.08 0.03 0.12 0.20 0.25 0.25 0.08 0.08 0.08 0.15 Si 2.75 S or Se 0.15 min Other elements AISI - lype Effect on properties 1 65 302 301 303 304 304L 305 309 310 314 316 Basic steel Lower Cr and Ni for greater work hardening S added for enhanced machinability Lower C for improved ductility and weldability Lower C for welding of thicker sections Higher Ni for increased formability Higher Cr and Ni for corrosion and scaling resistance Highest Cr and Ni to increase scaling resistance Si higher to increase scaling resistance Mo added for more corrosion resistance Ti added to avoid intergranular corrosion Nb, Ta added to avoid intergranular corrosion Substitution of nickel by manganese gives similar corrosion resistance at lower cost. Strength is approximately 50% higher but weldability is impaired Ti = 5 x C min 321 Nb + Ta = 10 xCmin Mn 5.5x10.0 N 0.25 max 347 martensitic structure with even higher chromium levels. The martensitic structure of the steels is obtained by rapid quenching from the austenite range. (1) Strength. Steels with medium strength and good toughness contain 0.15-0.35% C. Steels with very high strength and wear resistance may contain up to 0.12% C. (2) Toughness. Optimum toughness is obtained D. Duplex and Precipitation at temperature of 650° C or above. Hardening Steels (3) Corrosion resistance. They are the least resistant members of the steel family. Resis- There are some stainless steels in addition to the tance to corrosion is increased by tempering above three classes, which exhibit a wide range of in the range 360-650°C. The corrosion high strengths because of the formation of carbide resistance is increased with increasing chromium content. (4) Weldability. Post-treatment and pretreatment is necessary to avoid cracking. Welding of martensitic steels is not generally recommended. 166 Principles of Corrosion Engineering and Corrosion Control C/5 d G G a> G 03 +e o3 0> M a> co o3 "" ^ G o3 O 1# s 44 •S ta * • ?«v &* a •S.S « 8 « •^i q> « -H *-* • S . S 3 ^ S ^ *- « 2 o3 o3 o3 <U a3 6 M 03 T3 O O 5 -1 j5 3 "O _o ^ O a CO CO <D rt 43 4S O 3P3 H 6 §£ O o3 G id 03 O o m o OH o <L> O t3 S a) J3 3 O O O O C 13 H G o3 43 O O / CO ^-^ H-» ^ o m m m *"~s * tb 8 i 2 in S O co - —' ^h (N CO (N in o o O O o m m o m m m t^ m G * H—» col O u G 43 < H 03 < col OH G o T 3 4 4 -a 4 4 JH o <L> O 73 £ 13 £ 0> <U -a G T3 G G G XJ JH T3 <u 13 (U G G 13 <D G G < < u X 1 £3 O i in 1 T3 G o3 G s o 1 I o Oq ^H ^ 0 0H T' " ^ T" mm r^ <N d 1-5 1 o o OH qq qq u O OH <N i—i vd od ^H ^H rn K <N (N H 6 U G O q s i—i s q ^ 00 q i—i s ^ s Bq B X! 03 o. 6 ^ S X qS T—1 03 s u o 03 oS ^ <N 03 in ^ X X! d s d B (N o -i B ^ o <N d Types of corrosion: materials and environments 167 G DM O 3* . 3 «3 G 03 C/5 *V5 X& Su 3m <u V-i .3 ° a •> &• B CO G <U V5 <-k a a 15b o Co G 45 G ^ G 45 bO G O +-> 45 <> L CO T3 £ (75 G 03 P 2i G c« •? o f) bJD 4 5 4G 5 <U tb 175 GG oO VH +-* -a w ^ .§ Of) 3 53 i£ 45 45 O G £ cc ery > > ^o J-i O «« 6 o i5 45 45 b O O 5 <L> <> L CM 6oA O m o 13 *«3 O ^ co 13 45 '3 T—I tbjj I o m I o m oo V£> r—1 ^ o m O o •6 00 r—1 si o o CN 00 1400 m 1 m r^ ON <—1 o m o 00 ON T-H 3 13 45 1—1 ^ o CN o T-H CN O ^ 1 o o o m o <—i 1 o o o m cn r -H o CN t^ o m 1 ^o o m T-H o o ON T-H CN ON T-H o OH o m [>; o 1 1 1o mm * 1 CN 1 1 1o mm <~1 CN © 3o 3o 3 ^ ^ a m 2m 2 X! X 'S co G co G o 1 T-H 1 1m 1m ^ m* ^ cn T-H 1 1 G O O u ^ ^ CO T-H T-H q ^ ° . t>: o CN m H <—< 1o 1o 1 © 00 1o1o ° . 00 ° . 00 vd T-H \Q T-H V.O »-H o OH 3 o <u CO CO CO 03 03 CO 3 O ogqg qgqgqgqgqg H u G CO CO CO CO CO cd cd ogqg qgqgqgqgqg 3 u tt « m^^S o o Io mcN o o * Im |m | o osot^mONmcN o o o 1 68 Principles of Corrosion Engineering and Corrosion Control atui atui atui nee rx T3 <> L PN (Li VH rH CO CO E C NJ resi stance ood high t< stre ngth igh jstrengt G »•*-! E ^2 cor dati 8§ ^ r—i "lo <) L rG rG •s rG "6b G CO r G 45 bG bo n3 O O T3 O O rG TJ £ xi o o «§ G <> - G rG bo G O rt T3 O O E 0> rt a CO CO O X o o a ao pact G ^> CO <r> 4-> Q-> E ^S LO CO §1 o CN ^ CN o 1 00 r—1 "<3 5 D 1 'fi <D CX J-i OH CO CO <L> o * C0 tat ion ha dened tainl 13 Mechani CO G <> L CO >—' H r-H si O o CO 1H — r-H © CO r—1 o vo CO o r-H o vo vo ON VO o © © © CN © CN © CN © S-H o o PH CN © CO >—" si LO © o LO O [^ ^H © t^ CN © 00 © CN LO LO © CN o CO ^ r-H rec •< S o U 1 © LO <-H 1 LO CO 1 1 1 1 I CO G U CO* 1 1 1 © 1 © <L> 4-» CO <> L * ^} • ^ CO o r£> 0s- 3 CM o UH ex v —' G 5 LO CN £ H © ^ © © © 1 1 1 m LO T}H 1 1 1 1 r—i i—< 1 t\ 1 1 1 o © 1 CN 1 1 chani a u "lo O PM o > 1 LO 1 LO 1 1 -* 1 © 1 1 LO CD T3 G G O +-> rH s 2 U o U E o CN 1 LO 1 CN ^ t^ CN VO LO vo ^ LO 1 t^ 00 *co OH o vo r_H ^ N 2N 2 t^ 00 © © LO ^ ON § LO 6 <> L rG 2 ^U © o o q Eq S o o r-H © q E© E q © C uMo H -8 Mo abl 4.4d ation W £ o > G ffi * aS • b /) CO ^ M I C/} '• a •4—» r-H 8 LO 0> l >s LO £> P-t £ LO -H PH t^ C^ VO VO ffi PH U £H t^ < r -H r^ PH ^ r -H X lion resi; o« LO CO <U CH J-i U u *H <> L <L> PH ctf +-> G «3 en o G fr d VO 1 CO CN 1 CO © CN r-H 1 1 ON CN CO © © LO LO E o co G U Types of corrosion: materials and environments and intermetallic compounds. Elements, like aluminum, molybdenum, copper and titanium, are added to obtain the intermetallic strengthening effect (increase of strength by the formation of an intermetallic compound). Elements, like titanium, molybdenum and vanadium, are added to obtain the carbide strengthening effect. In these steels the strengths is derived from the formation of intermediate precipitates, such as Ni3Al, N13T1, N13M0 and C11AI3. Their high temperature strength is superior to that of austenitic steels. (1) Strength. Very high strength. (2) Toughness. The toughness is superior to that of martensitic steels. The toughness decreases with increased strength. (3) Corrosion resistance. Good resistance to stress corrosion cracking. Superior to the martensitic stainless steels, inferior to ferritic and austenitic steels. (4) Weldability. They are welded by inert gas welding technique. These steels may be used in aged conditions or as-rolled condition. In the as-rolled condition, they are strengthened by niobium carbide precipitation and further hardening is obtained on aging by the precipitation of copper. 1 69 (E) Weld decay zone. It is a band in the parent plate removed from the weld. 4 .7.4 WHAT HAPPENS W H E N THE S T E E L IS S E N S I T I Z E D When stainless steel, for example 18-8 steel, is cooled slowly through 550-850°C, it becomes susceptible to intergranular corrosion when exposed to a corrosion medium. The following observations characterize the intergranular attack: (1) Sensitization occur only in the temperature range 550-850°C because of the diffusion of the carbon to the grain boundaries. The sensitization band is usually 1/8-1/4 in thick adjacent but slightly removed from the weld. (2) Fast cooling rate through the sensitizing temperature range does not provide adequate time for the process of sensitizing to occur. (3) The degree of sensitization increases with increasing carbon contents and decreasing chromium contents. (4) Chromium carbide precipitate at the grain boundary. The precipitation of carbide is a time-temperature dependent phenomenon. The carbide precipitated is mainly Cr23C6, so that a single atom ties up almost four chromium atoms. 4 .7.3 TECHNICAL TERMS (A) Sensitization. When steels, such as austenitic type, are slowly cooled through the range 550-850°C, they frequently become susceptible to intergranular attack in a corrosive medium. (B) Stabilized steel. Steels containing stabilizers, such as titanium and niobium, which form carbides that are more stable than chromium carbides, are called stabilized stainless steels. (C) Chromium depletion. When the concentration of chromium adjacent to the grain boundaries becomes less compared to its concentration away from the grain boundaries, the phenomenon is termed chromium depletion. (D) Quenching. It consists of heating the alloy in the temperature range of 1950-2000°F (1056-1110°C) and cooling rapidly by quenching in water. 4 .7.5 E F F E C T OF VARIOUS FACTORS ON THE SENSITIZATION OF S T E E L S A. Effect of Carbon Carbon plays a very crucial role in the precipitation of carbides and a decrease of carbon content of steels restricts the formation of harmful carbides, such as chromium carbide, which is primarily responsible for the depletion of chromium adjacent to the grain boundaries. The effect of carbon is illustrated in Fig. 4.27. The ordinate is the temperature, and the abscissa the time of annealing (slow cooling in the furnace). Two types of steels are shown, one steel containing 1 70 Principles of Corrosion Engineering and Corrosion Control 1380 750 h 1290 700 1200 650 h 1110 600 1020 550 0.06 % C 0,5 10 Time, hours 100 1000 Figure 4.27 Comparison of time-temperature sensitization curves for 0.06 and 0.03 carbon austenitic stainless steel. (From Sandvik, Steel works) 0.06% C and another steel containing 0.03% C. Steels containing 18% chromium, 9-12% It can be observed that above 725° C and below Ni and 0.02-0.03% C which are sensitized 400° C, these steels do not become sensitized. at 500° C for 100 h showed a maximum Steel containing 0.06% C becomes sensitized at resistance sensitization at 0.04% nitrogen. 600° C in less than an hour, whereas the steel (5) Titanium and Niobium. These are added containing 0.03% C takes about 25 h to become to give stabilization against the harmful presensitized. The above observations show the effect cipitates, such as chromium carbides. These of time and temperature on the precipitation of elements combine with carbon and nitrogen carbides. and do not allow carbon and nitrogen to precipitate as carbides and nitrides which lead to intergranular corrosion. The optimum B. Effect of Alloying Elements temperature for formation of TiC (titanium carbide) is 900-950° C. (1) Nickel. The susceptibility to sensitization (6) Silicon. It is added to increase corrosion decreases with increased nickel content resistance in sulfuric acid. because the carbon solubility decreases with (7) Sulfur and Selenium. These are added to increased nickel content. increase machinability, however, the corro(2) Molybdenum. The susceptibility of steels to sion resistance is decreased. intergranular attack decreases with increased Mo content. The Mo also increases the 4 .7.6 CARBIDES resistance to pitting and crevice corrosion. (3) Chromium. It generally increases the resisCarbides play a very important role in intergrantance to sensitization. (4) Nitrogen. The effect of nitrogen depends ular attack. Intergranular attack does not occur upon the amount contained in a steel. without the formation of chromium carbides. The Nitrogen is an austenite stabilizer. carbides are present in dendritic shapes. All the Types of corrosion: materials and environments 1 71 carbon in the alloy does not precipitate carbide as range (sensitization temperature range) because exemplified by &23C6 in which only half of the of the grain boundary precipitation of chromium amount of carbon present in the steel is tied up as carbides (0*23 C6). carbide. 4 .7.7 SIMPLIFIED MECHANISM The Attack Carbon has a strong tendency to form carbides with chromium present in the steel (Cr23C6) in the sensitizing temperature zone during slow cooling from annealing treatment, stress relieving in the sensitizing temperature zone or during fabrication procedures involving prolonged welding and flame cutting. Hence, chromium carbide is precipitated at the grain boundary and the zone next to the grain boundary becomes depleted of chromium, thus making this zone anodic to the remainder of the surface. In austenitic steels, a minimum of 12% chromium is required to resist corrosion, however, depletion of chromium below this level causes weakening of the grain boundary and may lead ultimately to dislodging of grains. As localized attack progresses through the grains, the attack is more often called 'intergranular attack.' It must be noted that it is not the chromium carbide which is attacked but a zone adjacent to the grain boundary which is depleted of chromium and is attacked. The term 'weld decay is of historic origin when the mechanism of intergranular corrosion was not understood. The welders called the localized corrosion attack adjacent to the welded zone [heat affected zone (HAZ)], as weld decay. The heat affected zone (a narrow zone) becomes susceptible to intergranular attack because it reached the sensitization temperature range during welding (400-900°C), whereas the temperature of the weld region is much higher. Hence, the region adjacent to the weld fails by intergranular corrosion, commonly called 'weld decay' The rest of the base metal remains unaffected. O F I NTERGRANULAR C ORROSION Various theories and models have been proposed to account for the intergranular corrosion of steels. These are summarized below. For in-depth understanding of the theories, references are cited at the end of the chapter. The theories are reviewed here to provide a basic understanding of the mechanism of intergranular corrosion to engineers. The following three steps lead to the intergranular corrosion of austenitic steels: (1) Sensitization in the temperature range 550-900° C. (2) Diffusion of carbon to a grain boundary and formation of carbide. This leads to depletion of chromium content along the grain boundary. (3) Weakening of the grain boundary and disintegration of the grains. Preferred Site Preferential attack at the grain boundaries occurs because of the higher free energy of the grain boundary region. This is enhanced by the segregation of grain boundary impurities and by the precipitation of second phase particles which may be noble and lower the resistance of the matrix. The grain boundaries are the preferred location of localized attack compared to the matrix. Conditions Leading to Attack A majority of intergranular attacks on stainless steels can be attributed to the metallurgical changes which occur in the temperature range of 427-816° C due to slow cooling, or to prolonged welding of steel. The steels become sensitive to intergranular corrosion in this temperature Effect of Time and Temperature Time and temperature effects are important in welding. For instance, electric arc welding produces intense heating in a shorter time compared to gas welding. The metal will be in a sensitizing 172 Principles of Corrosion Engineering and Corrosion Control z one for a longer time if gas welding is used, hence, a greater amount of carbide would precipitate. T hus, steel which has been welded by gas flame would be more sensitive to attack by intergranular corrosion. It was observed from a t i m e temperature sensitization diagram (Fig. 4.27) that a stainless steel (SS 304) with 0.05% carbon content remained in the sensitizing zone for 10 s. The minimum time required to cause susceptibility to weld decay was 40 s. Therefore, theoretically, this alloy should not have undergone weld decay (intergranular attack). This alloy was, however, subjected to a rapid weld decay attack in nitric acid in the heat affected zone, contrary to the theoretical assumptions. To explain the above contradiction, it was suggested by Tedmon [35] that the nucleation of chromium carbide may initiate at a higher temperature, and the growth of carbide may take place at a lower temperature. T his explains why weld decay may occur even after an exposure period of 10 s. The time required for sensitization includes the period of initiation as well as the period of growth and propagation. A substantial loss of mechanical strength accompanied by a loss in corrosion resistance occur. Figure 4.28 shows the chromium carbide, grain boundary and low chromium areas. The grains of austenitic steels contain 18% chromium. Figure 4.29a shows the chromium depleted zone Precipitate (about 60% Cr) Cr level ( 18% f or matrix) Cr Depleted Grain Boundary Regions Figure 4.29a Cr depleted zone around a grain boundary precipitate of secondary carbide Carbides at grain boundaries Figure 4.29b Sketch showing the occurrence of carbides at grain boundaries around a grain boundary precipitate of a secondary carbide. Figure 4.29b shows carbides at grain boundaries due to Cr depletion. Grain boundary. j/~^-^^ /C ^^^><L ^* Chromium carbide Precipitate 6G% Method of Prevention T he following are the methods of prevention of austenitic nickel chromium stainless steels from intergranular corrosion: Low chromium area Grains (18% Cr) level for matrix (a) Purchase and use stainless steel in the annealed condition in which there is no harmful precipitate. This only applies when the steel is not to be exposed to the sensitizing temperature. (b) Select low carbon grade steel with a maximum of 0 .03% C, such as 304 L. T his would prevent the formation of harmful chromium carbide Figure 4.28 Grain boundary in sensitized type 304 during fabrication. stainless steel Types of corrosion: materials and environments (c) Select stabilized grades of stainless steels, such as type 347 (niobium stabilized) or steel type 321 (titanium stabilized). Stabilized grades of steels do not allow the depletion of chromium to occur adjacent to the grain boundary, as carbon has a greater affinity to combine with either titanium or niobium than with chromium, hence chromium carbide, the harmful precipitate is not formed. It is possible to reclaim steel which suffers intergranular corrosion by heating above 1000°C followed by water quenching. (d) Solution treatment at 1121 °C re-dissolves the carbides into the matrix, followed by rapid quenching to prevent re-precipitation of the carbides. (e) Cold working of the steel prior to sensitization. The application of cold work causes an increase in the precipitation of chromium carbide on dislocations, and reduces the amount of grain boundary carbide. (f) Modification of analysis of steel to produce delta-ferrite. The precipitation of carbides at the delta-ferrite/austenite interface reduces the severity of grain boundary precipi tation. The success of the remedial measures would depend upon the selection of materials, design and cost. The second remedy can be rejected because it may not be practical for large welded structures. There are, therefore, two options: use of low carbon steels or ferritic steels. The main disadvantage of low carbon steels is their lower strength. Remedial measure (C) is also not practical because the stabilized grades suffer from another localized form of corrosion called 'knife-line attack,' discussed in Section 4.7.9 below. Ferritic steels may be used with advantage. The ferritic steels may become sensitive to weld decay while cooling from 900° C, hence, they can be conveniently annealed by a torch in the temperature range 600-800°C to avoid sensitization. There are, however, other limitations, such as hydrogen embrittlement (diffusion of hydrogen in the metal and loss of mechanical strength), which may restrict the use of ferritic steels. The fifth remedy could be applied only in the context of the mechanicalproperties. 4 .7.8 THE CHROMIUM 1 73 D E P L E T I O N T HEORY The above theory was developed by Bain, et at [34]. According to the assumptions made in this theory a passive film is not formed on the surface of the material, if the chromium content is less than 12%. It means that if the amount of chromium in the austenitic steel is less than 12%, it would be subjected to localized corrosion. As a result of sensitization between the temperature range of 550-900°C, carbon diffuses towards the grain boundaries and forms chromium carbide, which is insoluble and precipitates out of solid solution if the carbon content is higher than 0.02%. The bulk alloy contains only 18% chromium. However, on analyzing carbides it has been observed that the chromium content in carbide can be as high as 94% which suggests that chromium is depleted from the areas of the alloy matrix immediately adjacent to the carbides (Fig. 4.30). The corrosion resistance of the steel which is primarily due to chromium (optimum concentration for corrosion resistance is 12%) is, therefore, lowered in the region adjacent to the grain boundary. When steel is in the temperature range (500-900°C), the diffusion rate of chromium from the matrix to the depleted area in the grain boundary is too slow to replenish the amount of chromium depleted. The rate of recovery of austenitic steels from intergranular corrosion, however, is greater at a higher temperature than at a lower temperature. As chromium is depleted from an area adjacent to grain boundary, it becomes mechanically weaker and is a preferred site for corrosion attack. Grain boundary attack is used synonymously with intergranular corrosion. A. Positive Points of the Theory According to the above theory, the susceptibility of the steels to intergranular corrosion should decrease with: (a) decreased carbon content, (b) increased chromium content. 1 74 Principles of Corrosion Engineering and Corrosion Control Figure 4.30 Grain boundaries in 316 L stainless steel etched in acidified C11G2 (x 500) Both these facts have been found to be true and verified by several investigators. (c) The existence of chromium depleted regions has been experimentally verified electrochemically and by microanalysis using electron microprobes. observed to take place at +350 mV; hence it was concluded that chromium depletion precedes intergranular corrosion. B. Quantitative Model A quantitative model was developed by Strawstrom, Hillert and Tedmon et al. [35,36]. This further strengthened the chromium depletion theory. Figure 4.31 shows a comparison of Cr-C-Cr23C6 chromium carbide equilibrium data for 304 stainless steel with experimentally obtained corrosion data for several temperatures. The bulk carbon concentration is shown on the abscissa. The chromium content in equilibrium with Cr23C6 in the steel is shown as the ordinate. If 12% chromium is assumed to be the minimum quantity to form a passivefilm,it is observed Proof Sensitized steels were maintained at potentials between 0 and +350 mV in 2N H 2 S0 4 and the products of dissolution were analyzed. On analysis of the products, it was found that the chromium content decreased from 18.6% at 0 to 13.7% at +350 mV. It has been shown by different tests on steels that a decrease in chromium content is accompanied by an increase in passive current density (the density required to achieve passivation). Intergranular corrosion was Types of corrosion: materials and environments 1 75 30 ill! M M| IMS illl| I I I ! IIM| 1—! M M ! Fe~18%Cr~10%Nf 2S 500°C 20 6G0®C 700°C 800<C u # % IS 10 0I (Ml Mill J aoooi 0.001 I 1 I Mill 0.1 1 I 1 >Mill I I I » Mil 10 wt%c Figure 4.31 Comparison of Cr-C-Cr23C6 equilibrium data for type 304 stainless steel with data for several temperatures that if the alloy contains more than 0.15% carbon it would become sensitized at 800° C. For the alloy annealed at 700° C, the carbon content from the plot is determined to be 0.08 and 0.02% at 600°C. If the carbon contents exceed the levels indicated above, sensitization is likely to occur. It is not possible to sensitize 304 stainless steel at 500°C, even after a prolonged exposure period, because carbide does not precipitate. It is, however, possible to sensitize the samples below 600°C, if the alloy has been equilibrated at a high temperature where carbide first precipitated. A sample of type 304 stainless steel containing 0.075% carbon was annealed at 700° C for one hour and quenched in water. When this sample was put in a test solution of boiling acid copper sulfate, no intergranular attack was observed although the sample contained precipitates of chromium carbides. The answer is simple. The chromium content was below the required limit of 12-15%, to cause intergranular corrosion. This proves the point made earlier in the chromium depletion theory that an optimum concentration of 12-15% of chromium is necessary to cause intergranular attack which is now proved. The same sample was slowly annealed for several hours at 500° C and water quenched. This treatment caused an increase in chromium carbide content. This sample was exposed again in the acid copper sulfate solution and rapid grain boundary attack occurred. The carbide was already precipitated on exposing the sample to 700° C as shown above, hence, the Cr-C-Cr23C6 equilibrium is rapidly attained even at 550°C. The amount of maximum allowed carbon at 550° C was determined to be 0.01%. As this amount of carbon contained in the alloy exceeds 0.075%, intergranular attack was observed. Thus, evidence is provided in favor of chromium depletion theory by the above work. Why would sensitized stainless steel held at sensitizing temperature for a very long period of time not be susceptible to intergranular corrosion? Because sufficient time is provided for the diffusion of chromium from the grain matrix to the depleted region. Of course, thousands of hours may be required to achieve this. Strawstrom and Hillert [36] obtained good agreement between the theoretical and experimental results. The chromium depletion model has been widely accepted. 1 76 Principles of Corrosion Engineering and Corrosion Control at the interface. The strains arise from a distorted lattice adjacent to the carbide precipitate. It has been suggested that the chromium carbide (Cr23C6) formed at the grain boundary is only in registry with one grain and not with other grains. The magnitude of the knife-line attack is found proportional to the magnitude of mis-orientation at the carbide austenite interface. It was found by Streicher [40] that the rate of grain boundary attack on a particular grain was dependent on the orientation of that grain. It was confirmed also by other workers that the density of carbide precipitated was dependent on the mis-orientation between the grains. An explanation of the healing effect was offered by this theory. It was postulated that the knife-line attack does not occur above 800° C because the total strain reduces due to the precipitation carbides at this temperature. The above explanation, however, was not very convincing. It was through the work of Stickler and Vinckier [38] that sufficient evidence was provided in support of the strain theory. It was shown that in intergranular corrosion, healing only occurred in the absence of a continuous grain boundary film of carbide and the growth of carbide particles was essential for the process of healing to occur. The knifeline attack does not propagate in the absence of a continuous grain boundary film even if the grains are not in registry, because of the healing effect. It was pointed out earlier that the magnitude of the intergranular corrosion depends on the de-registry of the grain accordingly to the strain theory. Thus, the latter finding gave a reasonable explanation of the healing effect. The later observations also support the chromium depletion theory. C. Objections Against the Chromium Depletion Theory (1) Lack of direct proof of the impoverished areas. With the rapid progress in electron microprobe analysis, the objection has been largely resolved. The difficulty in direct detection is due to the narrowness of the depleted area in the atomic dimensions, which can be up to 2 |xm. The strongest evidence in favor of chromium depletion has been provided by anodic polarization studies mentioned in the text to follow. (2) After sensitizing a 302 stainless steel containing 0.10% C for 150 h at 650°C, only 45% of the carbon precipitates as carbide and recovery from the intergranular attack is possible. The above observation, if correct, means that depletion does not occur as suggested, as only 45% of the carbon is precipitated. Also, the process of self-healing should not occur while the carbon particles are still growing. The above objections have been largely resolved. According to Baumel [37], the precipitation of carbide at the grain boundary stops at every stages and the process of healing can start while the precipitation of carbon occurs in the interior of the austenitic grains. Recovery from intergranular attack is, therefore, possible after sensitizing the alloy for 150 h at 650°C. Thus, the objection raised by Stickler and Vinckier [38] is not valid. Also, the theoretically calculated annealing time for developing a chromium depleted zone is in good agreement with the experimental annealing time. 4 .7.9 STRAIN THEORY [ 3 9 ] T H E ELECTROCHEMICAL THEORY On the basis of chromium depletion theory, a uniform attack occurs on both sides of the grain It was shown that a piece of M23C6 (metallic boundary carbide. Therefore, the observation that carbide), when attached to 18-8 steel (annealed) the knife-line attack (knife-line attack occurs in a and immersed in nitric acid, developed intervery narrow band in the parent metal immedi- granular attack. From the above experiment, it ately adjacent to the weld), occurs only on one was concluded that a potential difference existed side of the carbide, austenite interface cannot between the metallic carbide and stainless steel. be explained by the chromium depletion theory. It was shown that M23C6 was more noble The knife-line attack arises as a result of strains than the steel matrix. It was further suggested Types of corrosion: materials and environments that the intergranular attack was accelerated in the presence of residual stresses. The evidence described above shows that: (1) A corrosion cell is formed between an anode and cathode. The metallic carbides being more noble act as cathode and the grain matrix surface as anode. It was suggested that whether the carbide is noble or the steel (boiling 65% HNO3) would depend on the electrolyte used. The Huey solution attacked M23C6, whereas the Streicher solution attacked the grain matrix. (2) An emf is set up between the metallic carbide and the grain surface in the presence of an electrolyte and intergranular attack takes place. Objections (a) The electrochemical theory was questioned by Baumel, et a\. [37]. They suggested that localized corrosion could not be confined to a very narrow zone (2 |xm) and the corrosion should extend to the matrix. (b) The potential measurement of 18-9 steel, platinum, M23C6, and copper in Strauss solution (boiling 16% H 2 S0 4 + 5.7 % CUSO4 + Cu) showed that the potentials of the four materials is nearly the same within an accuracy of ± 1 % . This work was in contradiction of the results obtained on the effect of electrolyte on potential of M23C6 and austenite. From the above description, it is clear that the electrochemical theory [41] is controversial and it does not present a better picture of intergranular corrosion than predicted by the chromium depletion theory. 1 77 (2) Soluble impurity segregates resulting from solute vacancy interaction. The studies were conducted on annealed material. This model is concerned primarily with intergranular attack on non-sensitized steel and only secondarily with carbide forming sensitized steel. According to the theory, the resistance to intergranular attack is improved if discontinuous carbides are precipitated. This can be achieved by heating between 800 and 900° C followed by water quenching. By this treatment, chromium carbides would be precipitated in a discontinuous form and phosphorus and silicon would be precipitated. Although the dissolution rate of carbides is increased, the portion of the material in between the carbides would be low in phosphorous and silicon, hence the rate of corrosion would be decreased. This improvement is attributed to the incorporation of a major part of the segregating solute into the carbide phase. It is a general opinion that grain boundary segregation of elements, such as silicon and phosphorus, is the cause of intergranular corrosion but the exact mechanism is not known. It has been further suggested that the driving force for the attack is provided by the potential difference which exists between the grain matrix and the grain boundary. Experiments on hardness measurements showed the boundaries to be harder due to the presence of solute segregates. Intergranular attack was observed on alloys in a highly oxidizing solution, likeHN0 3 + K 2 Cr0 4 . The evidence that intergranular attack occurs on non-sensitized steels in nitric acid containing oxidizing ions, like Cr +3 , is strong evidence in favor of grain boundary segregation of impurity solute. The steel does not remain passive in the oxidizing solution of nitric acid and potassium dichromate as the protective film possibly breaks down. It would be difficult to otherwise explain why intergranular attack is observed on an alloy which is not subjected to sensitization (formation of metallic carbide in the range of 550-850°C). Grain boundary segregation may increase the electronic conductivity of the oxide, at the boundary. It increases the attack because Cr +3 is oxidized to Cr +6 which is soluble, whereas Cr +3 is not soluble. A film of oxide which is not soluble will prevent intergranular attack. 4 .7.11 S O L U T E S E G R E G A T I O N THEORY [ 42-45] This theory was developed by Aust [43] and suggested the presence of the following as the primary cause for the intergranular corrosion of non-sensitized austenitic steels: (1) Continuous grain boundary path of a second phase. 1 78 Principles of Corrosion Engineering and Corrosion Control 825 becomes susceptible to intergranular corrosion between 650 and 750° C in the Huey test. The reason of sensitization is stated to be the depletion of chromium at the grain boundary because of the formation of M23C6. It was further shown that nickel base alloys, such as alloys containing 38-46% nickel, 19-25% chromium, 2-3% molybdenum, 1-3% copper, 0-5% C, 0.6-1.2% titanium and balance iron, could be sensitized in 5 h at 680°C. By the above treatment, titanium carbide was precipitated instead of chromium carbide and the chromium distribution was not affected. The alloys, therefore, showed a good resistance to intergranular attack because of the addition of titanium. Experimental work related to chromiumcarbide-carbon equilibrium on a nickel base alloy (15% chromium, 10% iron, balance Ni), showed that the carbon solubility in nickel base alloys is lesser than in austenitic steels. Nickel-based alloys containing chromium can be sensitized in the temperature range 500-700°C. In high nickel alloys containing molybdenum, such as 15% Cr, 15% Mo, 4% W, 5% Fe, 0.06% C and balance Ni, intergranular attack is generally attributed to the depletion of molybdenum and chromium, because of the formation of M6C (molybdenum carbides and chromium carbides ( M 23 C 6 ). Sufficient evidence is not available to present a more acceptable mechanism of intergranular corrosion of nickel base alloys. The attack takes place at the grain boundaries because they are high energy regions. The de-adsorption of impurities from the grain boundaries at elevated temperatures above 1100°C causes a reduction of the rate of intergranular attack which is in conformity with the solute segregation theory. Objections (1) It does not seriously consider the importance of oxidizing power of the corroding environment. In nitric acid solution, the insoluble Cr +3 ions can be transformed to soluble Cr +6 state. The stainless steel does not remain passive in this solution and general corrosion along with localized intergranular attack is likely to take place. Under such conditions, the presence of solute segregate is not, therefore, necessary. The above observation, therefore, contradicts the solute segregation theory. (2) The postulate states that the presence of a continuous second phase leads to intergranular attack, does not seem to be valid because of the observation that steel 316 in which sigma phase precipitated at the grain boundary showed no acceleration of intergranular corrosion until the solution was made highly oxidizing. Generally speaking, the solute segregation theory is only valid for non-sensitized steel and attempts to extend it to the sensitized steels have not been successful because the tests were carried out in highly oxidizing solution in which general corrosion as well as intergranular takes place as discussed above. 4 .7.13 S E N S I T I Z A T I O N OF F ERRITIC S T A I N L E S S S T E E L S Sensitization in ferritic stainless steels is introduced by high temperature heat treatment (above 925°C) and relieved by heating for a short time between 650 and 815°C, which is opposite to the observations on austenitic steels. Annealing sensitized ferritic steel at 788°C for several minutes will eliminate intergranular attack. The susceptibility is reduced by the addition of titanium or niobium. The presence of carbon or nitrogen is necessary to cause sensitization. It was postulated by Houdermont and Tofante [46,47] that austenite is formed at the grain 4.7.12 HIGH SENSITIZATION NICKEL ALLOYS OF The nickel base alloys are typically singlephase, multi-component and contain iron and chromium. Carbon is present in the range of 0.03-0.08 wt%. From several investigations, it is clear that chromium increases the corrosion resistance of nickel base alloys by virtue of formation of a passive film containing chromium. Inconel Types of corrosion: materials and environments boundaries during heating in the sensitization range because of the higher solubility of carbon in austenite than in ferrite. At 927° C, there is a transfer of carbon from ferrite to austenite, and ferrite is enriched in chromium. The grain boundary austenite becomes rich in carbon. During cooling, austenite is decomposed to give chromium carbide, and these carbides are responsible for the intergranular attack. In another theory, Bond [48] postulated that: (1) The ferritic steels become immune to intergranular corrosion if the carbon or nitrogen content is made very low. (2) Intergranular corrosion in alloys containing a very high amount of carbon and nitrogen becomes less severe than observed in alloys containing a smaller amount. He demonstrated that on increasing the carbon content of an alloy (containing 17% Cr and 0.0031% C) to 0.012% C, and raising the nitrogen content from 0.0095 to 0.022%, intergranular corrosion was observed. If the carbon or nitrogen content is increased to sufficiently high values as suggested [49], the austenite phase is extended; hence, when the alloy is cooled martensite precipitates. As martensite contains most of the carbon and nitrogen in solution, the precipitation of carbides (chromiumrich) and nitrides is reduced. The magnitude of sensitization would be consequently reduced in alloys containing sufficiently high amounts of these elements compared to alloys containing lower amounts of these elements. The above model is based on the chromium depletion theory described earlier. The mechanism proposed by Hodges [50] is based on the following observations on high purity ferritic steels. (1) Alloys containing 17-26% chromium, C = 0.001-0.002%, N = 0.004-0.009%, when water quenched from 1000°C, were not found to be sensitive to intergranular attack. On the contrary, the same sample when air cooled, showed sensitivity to intergranular attack. (2) He further observed that alloys containing molybdenum were more sensitive to 179 intergranular attack than molybdenum-free ferritic steels when furnace cooled. These results were explained in terms of chromium depletion theory. Because the alloys contained small amounts of carbon and nitrogen, diffusion occurred over long distances to the grain boundary to form chromium carbides. In conventional low purity alloys, the distances involved for diffusion are smaller. Thus, when the alloy is water-quenched sufficient time is not allowed for diffusion of either carbon or nitrogen to form carbides or nitrides; hence the intergranular attack does not occur under this condition. The immunity from intergranular attack, therefore, depends on how much time is available for carbon to diffuse to the grain boundaries which depends on the type of heat treatment, i.e. annealing, normalizing or quenching. Thermal treatment which produces susceptibility in ferritic alloys are those which are used to minimize it in the austenitic range. The following table (Table 4.5) illustrates the point. It can be observed that heat treatments beneficial for austenitic steel 304 is harmful for ferritic 430, and the heat treatment beneficial for ferritic (430) is harmful for austenitic 304. 4.7.14 SENSITIZATION STEELS OF DUPLEX Steels which contain both the body-centered cubic alpha-phase and the face-centered cubic gamma-phase, are called 'Duplex Steels.' Austenitic ferritic steels containing about 5% or more delta-ferrite show a higher resistance than Table 4.5 steels Heat treatment of ferritic stainless Heat treatment Corrosion rate Ferritic 430 Austenitic 304 400 10 1300°F, 40 water-quenched (1 h) 2000°F, 400 water-quenched (1 h) 1 80 Principles of Corrosion Engineering and Corrosion Control During the process of welding, the metal at the weld pool and the base metal interface is held at a temperature above that required for the precipitation of either niobium carbide or titanium carbide (1220°C). During the process of cooling, the metal at the weld pool-metal interface passes through the temperature range at which the precipitation of carbide of niobium or titanium occurs. At room temperature, the heat affected area becomes rich in carbon and on reheating this area (550-780°C), the chromium carbide precipitates faster than the niobium or titanium carbide, which accounts for the knife-line attack. The precipitation of carbide leaves a narrowband susceptible to intergranular corrosion adjacent to the fusion line. It is also been suggested that eutectic carbides of niobium and titanium are formed at the grain boundary during cooling (600-900°C). The carbides dissolve and cause intergranular corrosion. One good method of prevention against knife-line attack is to heat the stabilized steel after welding to 1061.5°C followed by cooling to 891°C. the austenitic alloys only with the same carbon content, because the islands of ferrite minimize the formation of continuous grain boundary precipitate in austenitic grain boundaries. They can be either low carbon low chromium (0.03% C, <20% Cr) or higher carbon high chromium alloys ( C>0.03%, C r>20%). The low carbon, low chromium alloys are typified by alloys, such as 308 L; the steel has a matrix of austenite and the amount of ferrite is between 5 and 10%. The other class is typified by alloys, such as AISI 326, which has a ferrite matrix and contains 40-50% austenite. Chromium-rich duplex alloys, such as 326 and Uranus, are highly resistant to intergranular attack because: (a) the alloys offer more resistance to intergranular corrosion by virtue of their higher chromium contents which cause the formation of a well adherent and stable chromiumrich oxide film, (b) a better resistance to sensitization is offered because one particular heat treatment affects only one phase; a heat treatment which sensitizes the ferrite phase has no adverse effect 4 .7.16 I N T E R G R A N U L A R on the austenite phase. A TTACK W ITHOUT Alloys, such as type 329 (Carpenter 7 Mo), is used in the form of annealed heat exchanger tubing. The other important alloys developed is Sandvik 3 RE60 which is in the form of annealed tubing. Stabilizing elements are added to duplex steels to improve intergranular resistance. These alloys are quite resistant, if a network of continuous carbides is not formed. E NVIRONMENT S ENSITIZATION: E F F E C T S O F It has been observed that even those steels which are not susceptible to intergranular attack undergo intergranular attack in the presence of strong oxidizing agents, like Cr0 4 ~, MnO^,Fe 3 + ,Ce 4 + ,V 5 + , and high temperature water containing FeC^, CuCl2 or KOH. It has been shown already that intergranular attack can be avoided by adopting the following means: (a) Reducing carbon content to 0.2%. (b) Solution treatment. (c) Addition of stabilizers, like niobium and titanium. In spite of these measures, commercial ironchromium-nickel alloys are found susceptible to intergranular attack under highly oxidizing condition. What could be the reason? Ions, like Cr +6 , depolarize the cathodic reaction and increase the 4 .7.15 K N I F E - L I N E A T T A C K This phenomenon is restricted to the stabilized grades of steel, such as 321 and 347. The knife-line attack occurs immediately adjacent to the weld and shows as a thin line of intergranular corrosion. It results from intergranular corrosion like the weld decay. It may be noted that weld decay develops at some distance away from the weld. The following is the mechanism suggested for the knife-line attack of stabilized steels. Types of corrosion: materials and environments 1 81 rate of anodic dissolution. The grain boundaries Material Welded 18-8 stainless steel pipe. being regions of high energy are preferentially Environment Seawater. attacked. It is also likely that the elements segregated at grain boundaries induce a driving force Analysis of failure In order to analyze the failfor corrosion because of a difference of poten- ure, test pieces of the pipes were prepared tial between the grain boundary and the matrix. and subjected to metallographic examination. Armaijo [51,52] who showed that the elements The metallographic examination showed macroresponsible for the above behavior were phospho- scopic welding defects in the transveric weldrus and silicon. ing seam. The welding did not penetrate the The attack on non-sensitized alloys can be transversal joints of stainless steel pipes. Corrosion was observed adjacent to the longitudinal attributed to the following factors: weld. Deep cavities were observed on the trans(1) Surface heterogeneity of grain boundaries. verse section of the weld about 3 mm from the (2) Impurity segregation at the grain boundaries. weld metal corresponding to the external points This point was proved by hardness measure- of the attack. A detailed microscopic examiment in which alloys containing phospho- nation of the hole showed intergranular attack rous showed increased hardness due to grain and grain boundary carbides (Fig. 4.32). Furboundary segregation. No grain boundary ther optical studies showed weld edge misalignhardening effect is observed in alloys of high ment. It was also observed that the weld metal purity. did not penetrate sufficiently at some points of (3) Increase in the dissolution rate in the the seam. transpassive region due to high chromium contents and decrease due to low chromium Prevention content [53]. (1) Selection of stabilized steels for the piping (4) Acceleration of dissolution by lower silicon systems, such as AISI 321 and 347. concentration and retarding of dissolution (2) Eliminating welding defects as far as possible by higher silicon concentration [54]. by adopting proper technique of welding. (5) The change of character of Cr +3 from being (3) Using ferritic stainless steels, wherever protective to being soluble Cr +6 at more possible. noble potentials. The impurities segregates affect the flow of electrons and the rate of corrosion. Dissolution would depend on how conductive are the impurities at the grain B. Case 2 boundaries. Failure of Type 304 stainless steel fused salt It is not very clear which of the above factors pot [55]. contributes more to the acceleration of intergran- Material Stainless steel type 304. ular attack on non-sensitized steels. It is clear that the rate of intergranular attack is affected by all Environment Molten eutectic mixture of sodium, factors stated above, however, the effect of each potassium and lithium chloride at melt temperatures ranging from 500 to 600° C. The pot used factor cannot be quantitatively ascertained. was a weld cylinder with 3 mm thick steel walls and 305 mm in height. 4 .7.17 FAILURE CASE HISTORIES A. Case 1 Failure of pipeline system for the feed of a swimming pool with seawater [55]. Analysis of problem Six specimens were prepared for evaluation and examination. All specimens exhibited carbide precipitation in the grain boundaries. Specimens taken about 100 mm from the melt level showed no attack as these regions were not exposed to the sensitizing temperature. Similarly, specimens taken about 45 mm 1 82 Principles of Corrosion Engineering and Corrosion Control H 7>* I * " *f > 45. •>• Figure 4.32 Inter-crystalline attack from the melt also showed no evidence of intergranular cracking as in these regions also the sensitization temperature was not reached. Specimens taken from the regions which were in direct contact with the fused salt and were exposed to temperature of 500-650 °C and showed intense grain boundary attack. Welding followed by exposure to a temperature in the sensitizing range caused intense intergranular attack. Ferritic steels (1) The most important method is to control the amount of interstitial carbon and nitrogen to less than 50ppm and 150 ppm, respectively. Lowering the carbon to 0.03% does not appreciably help, because of the lower solubility of carbon in ferrite. It has been shown for 18 Cr-2 Mo alloys that the maximum level of C + N is 60-80 ppm, and for 26 Cr-1 Mo steel, the level required is 150 ppm. Use ferritic steels in annealed Prevention conditions (788°C). Stabilized grade steel (2) Addition of stabilizing elements. Both titanium and niobium can be used as stabilizing Stainless steel 347, stabilized grade, could be elements. Both elements minimize intergranused with advantage. After welding, the steel ular attack of ferrite steels, hence, titanium should be stress relieved to about 900° C for 2h stabilized alloys are now recommended for and rapidly cooled in order to minimize the applications in nitric acid environments. The stresses. Also, Hastelloy (70 Ni-17 Mo-7 Cr-5 Fe) amount of stabilizing elements, titanium or could be used with advantage. It should, howniobium, required for immunity to interever, be solution treated to 1120°C and cooled granular corrosion by CuS04-16% H2SO4, rapidly. Types of corrosion: materials and environments a test for 18 Cr-2 Mo alloys, is given by Ti + Nb = 0.2 + 4 ( C + N ). These limits are valid for carbon contents ranging from 0.02 to 0.05%. Duplex stainless steels (1) Avoid the formation of harmful phases by rapid cooling through 900-700°C. These phases, however, affect the mechanical properties mainly and do not significantly affect intergranular corrosion. (2) Keep the carbon lower than 0.03%. As the duplex stainless steels are not as widely used as other grades of steels, sufficient information on preventive methods is not available. 1 83 in wrought, nickel-rich, chromium-bearing alloys. ASTM G34 - test method for exfoliation corrosion susceptibility in 2xxx and 7xxx series aluminum alloys (EXCO Test). ASTM G66 - test method for visual assessment of exfoliation corrosion susceptibility of 5xxx series aluminum alloys (Asset Test). ASTM G67 - test method for determining the susceptibility to intergranular corrosion of 5xxx series aluminum alloys by mass loss after exposure to nitric acid (Namlt Test). 4.8 S T R E S S CORROSION CRACKING AND HYDROGEN 4 .7.18 DAMAGE TEST OF INTERGRANULAR CORROSION The following tests are used to check the susceptibility of steels to stress corrosion cracking: (1) (2) (3) 4 .8.1 INTRODUCTION (4) Stress corrosion is the failure of a metal resulting from the conjoint action of stress and chemical attack. It is a phenomenon associated with a comHuey test. It is the nitric acid test described bination of static tensile stress, environment and in ASTM A262-68 practice. in some systems, a metallurgical condition which Acid ferric sulfate test. It is described in leads to component failure due to the initiation ASTM designation A262-68. and propagation of a high aspect ratio crack. It Strauss test. It consists of exposing a sam- is characterized by fine cracks which lead to failple to a boiling solution of 6 wt% CUSO4 + ure of components are potentially the structure 16 wt% H 2 S0 4 for a period of 72 h. concerned. Stress corrosion cracking is abbreviStreicher test. It consists of electrolytically ated as SCC. The failures are more often sudden etching in a polished sample ( 3% grid paper) and unpredictable which may occur after as little in a room temperature solution consisting of as few months or years of previously satisfactory 100 g of H2C2O4, 2 H 2 0 dissolved in 900 ml service. of distilled water. The current required is 0.1 A/cm2 for 15 min. It is described as ASTM designation A262-68 Practice A. 4 .8.2 REQUIREMENTS FOR S CC 4 .7.19 S TANDARD TEST M ETHODS • ASTM A262 - practices for detecting susceptibility to intergranular attack in austenitic stainless steel. ASTM G28 - test methods for detecting susceptibility to intergranular attack The following conditions are necessary for SCC to occur: (1) A susceptible metal. (2) A specific environment. (3) A tensile or residual stress. Figure 4.33 shows a stress corrosion triangle. • 1 84 Principles of Corrosion Engineering and Corrosion Control Specific Environment Tensile Stress Susceptible Alloy Figure 4.33 The stress corrosion triangle Failure occurs when the stress intensity factor reaches a critical value, K\c. This critical The development of SCC can be historically value is called 'fracture toughness.' The relation divided into three periods; the first period between fracture stress to fracture toughness is between 1930s and 1950s, being the identifica- given by: tion period, the second period between 1960s and 1970s, being the mechanistic period in ^lc (4.46) Of = which important basic mechanisms of SCC were y^frca explained and various alloys were evaluated for application in selected environments and the where third period being the application and developK\c = fracture toughness (psiVin or MPa^/rn) ment period of the eighties. G{ = nominal stress at fracture (psi or MPa) a = crack length (one-half of the crack length depending on geometry) 4 .8.4 APPLICATION OF y = correction factor. FRACTURE MECHANICS For an edge crack, the crack length is a, for a One important characteristic of fracture in met- center crack the crack length is 2a. Most of the past als is the occurrence of plastic deformation at research of stress corrosion cracking utilizes the the tip of the crack. The fracture toughness is fracture mechanics approach in which the relation proportional to the energy consumed in plas- between the initially applied stress, stress intensity tic deformation, which is difficult to measure factor K{ and the time to failure of pre-cracked specimen is observed. The time to failure increases accurately. as K{ decreases from the level required for the The fracture toughness of most materials is determined by a parameter called 'stress intensity fracture to occur (K\c, fracture toughness) to a factor,' K[. It is a measure of the concentration of threshold level (Kiscc) below which SCC does not occur. stresses at the tip of the crack. It is given by If the stress intensity factor K[ is plotted K{ = yGyy/na (units : MPaVm) (4.45) against the time to failure as shown in Fig. 4.34, it is observed that below a certain value of K[ (stress where intensity factor), failure does not occur, or there is a threshold limit for a failure to occur. The <7y = nominal applied stress value of K{ below which failure does not occur is, a = crack length therefore, designated -K'isccy — correction factor. 4 .8.3 HISTORICAL Types of corrosion: materials and environments 185 Figure 4.34 A typical curve showing failure time as a function of stress for normal fatigue (no corrosion) 4 .8.5 TYPES OF SCC (a) Intergranular Cracking Various types of SCC are distinguished as below: (a) Chloride stress corrosion cracking. It occurs in austenitic steels under tensile stress in the presence of oxygen, chloride ion and high temperature. (b) Caustic stress corrosion cracking. Cracking of steels in caustic environments where the hydrogen concentration is high, for instance, cracking of Inconel tubes in alkaline solutions. (c) Sulfide stress corrosion cracking. Cracking of steels in hydrogen sulfide environment as encountered in oil drilling industry. (d) Seasonal cracking. The term is now obsolete. It had a historical significance only. (b) Transgranular Cracking It refers only to SCC of brass in ammoniacal environment, but still occasionally occurs in refrigeration plant using ammonia Figure 4.35 Micrographs of intergranular and transgranular cracking refrigerant. 4 8 6 INTERGRANULAR A N D .. TRANSGRANULAR CRACKING Intergranular cracking. If the cracking proceeds along the grain boundaries, it is called intergranular cracking (Fig. 4.35a). Transgranular cracking. The crack proceeds across the grain. It is the most common type in concentrated chloride environment (Fig. 4.35b). 1 86 4 .8.7 Principles of Corrosion Engineering and Corrosion Control IMPORTANCE OF STRESSES The tensile stresses, a key requirement for SCC to occur, are 'static and they may be residual, thermal or applied. It is generally agreed that residual stresses are dangerous, because of their greater magnitude, which may sometimes approach yield strength, such as in fusion welds. For SCC to be caused alone by applied stress, it must be of very high magnitude. Stress corrosion can be caused by either residual, thermal or applied or a combination of all. Cracking caused by cyclic stresses is called'corrosion fatigue? however, the two have different fracture mechanisms. Compressive residual stresses are used either to (a) Surface discontinuities. Cracks may initiate prevent SCC or corrosion fatigue or to delay at surface irregularities, such as grooves, laps their onset. Table 4.6 shows the sources for stress or defects arising from fabrication process for SCC. (Fig. 4.37). The question, 'what is the threshold stress for (b) Corrosion pits. SCC can also initiate at the pits which are formed on the surface due to a certain alloy for SCC to occur,' has no definite breakdown of passivity by chloride ions. Pits answer because the magnitude of stress required are formed at a certain value of potential, would vary with environment. In literature, the the critical pitting potential. In many cases, values are quoted under specific conditions of stress corrosion cracks have been observed to stress and temperature and environment. initiate at the base of the pits by intergranular corrosion. The electrochemistry at the base of the pit is the controlling factor in the 4 .8.8 SCC SITES initiation of cracking at pit sites. Stress corrosion cracking is a deleterious phe- (c) Grain boundaries. Intergranular corrosion resulting from the sensitization by impurinomenon which occurs under a tensile stress, either residual or applied in a corrosive environties, such as phosphorus or sulfur, at the grain ment. The cracks are initiated and propagated boundary which makes the grain boundaries by the combined effect of stresses and the envivery reactive to SCC. Dissolution of the slip ronment. The mechanism of stress corrosion planes is caused by the breakdown of the cracking is highly complex and despite extensive protective film and initiates SCC. research, it is not conclusively understood. However, various important factors which lead to SCC are given below. A simplified diagram of SCC is shown in Fig. 4.36. The SCC cracks can be both intergranular or transgranular, depending on the alloy, stress conditions and the environment. Stress corrosion cracking initiate and propagate without any outside evidence of corrosion. The failures can take place without any previous warning. They often initiate at pre-existing flaws or flaws formed during the service period of the component. The following are generally the sites for crack initiation. Table 4.6 Sources Sources of stress for SCC Applications Rapid heating and quenching Thermal expansion Vibrations Rotation Bolting (flanged joints) Pressure (internal or external) Structural loading Residual Welding Shearing, punching, cutting Bending, crimping, riveting Machining (lathe/mill/drill) Heat treatment Straightening, breaking, deep drawing Types of corrosion: materials and environments 187 Hetai surface CrsC2 opt CrsCa Crack Areas low In protective Gr or grain tXHjndarles Figure 4.36 A schematic sketch showing crack propogation along grain boundaries (intergranular stress corrosion cracking) 4 .8.9 Surface irregularities E X A M P L E S OF TYPICAL ENGINEERING MATERIALS WHICH U N D E R G O Defects ~\~r~ (1) High strength steel in water. (2) High strength aluminum alloys in chloride solutions. (3) Copper alloys in ammonical solutions. (4) Mild steels in hydroxide and nitrate solutions. (5) Austenitic steels in hot chloride solution and hydroxide solution. (6) Titanium alloys in chloride solutions and hot solid chloride. (7) High nickel alloys in high purity steam. Table 4.7 shows stress corrosion cracking environments. / S CC Non-uniform surface Grain boundary precipitate 4 C H A R A C T E R I S T I C S O F S CC (1) Stress, either residual or applied, is required. There is a simultaneous action of stress and corrosion. Figure 4.37 Initiation of crack at the pit Table 4.7 Metals Stress corrosion cracking environments Environment NaOH solution NaOH + Na2SiC>3 solutions Nitric acid (concentrate) Ammonia ~| Calcium Nitrate solutions Sodium HCN solutions HCN + SnCl2 + AsCl2 + CHC12 solutions CH3COOH solutions NH4CI solutions CaCl2 solutions FeCl2 solutions H2S solutions NH 4 CNS solutions Hydrofluoric acid H 2 S0 4 - HNO3 solutions Mixture of antimony chloride, hydrochloric acid and aluminum chloride in hydrocarbon MEA (mono-ethanolamine), DEA, etc. (amines) Carbon steel Low alloy steels Ferritic and martensitic stainless steels (Fe-Cr) H2S solutions May also be cracked by media listed for carbon steel Seawater H2S solutions MgCl2 solutions NaCl solutions NaCl + H 2 0 2 solutions NaOH solutions NH3 solutions HNO3 H 2 S0 4 H 2 S0 4 + HNO3 solutions High temperature and high pressure water (high purity) May also be cracked by media listed for carbon steel NH 3 Ba Co Ca Fe Mg Hg K Li Na Vinyl Ethyl Methyl Chloride solutions (Contd) Table 4.7 Metal (Contd) Environment Seawater NaCl + H2O2 solutions Dichlorethane CCI4 solutions Chloroform Aniline chlorohydrate Ortho-dichlorobenzene Oxychlorobenzene Aluminum sulfate N a 2 S0 4 Sodium aluminate Ammonium carbonate KOH solutions NaOH solutions Molten NaOH Air NaCl + H2O2 solutions NaCl solutions Seawater Austenitic stainless steel (Fe-Cr-Ni) Aluminum alloys Al-Zn Al-Mg Al-Mg Al-Cu-Mg Al-Mg-Zn Al-Zn-Cu Al-Zn-Mg-Mn Al-Zn-Mg-Cu-Mn Titanium alloys NaCl solutions NaCl + H 2 0 2 solutions Seawater HNO3 (fuming red) HC1 Molten NaCl Seawater Salt water Molten N 2 0 4 Trichloroethylene Acetic acid HCN HF Fluosilicic acid H2SO3 Naphthenic acid Polythionic acid NH 3 Calcium Mercurous Nitrate solutions Sodium (Contd) 1 90 Principles of Corrosion Engineering and Corrosion Control (Contd) Environment Nitric acid Formaldehyde Steam (260°C) H2SO4 + C11SO4 solutions NaOH + sulfide solutions H2SO4 + chloride solutions N a 2 C0 3 + 0.1% NaCl solutions Cone, boiler water NaHC0 3 + N H 3 + NaCl solutions Nickel alloys (Monel, etc.) Table 4.7 Metal Nickel alloys (Monel, etc. HF Molten NaOH Hg salts Hydrofluosilicic acid Chromic acid NH 3 Ammonia compounds H2S solutions Moist S0 2 HNO3 fumes Some amines Copper alloys (mainly brass and bronze) (2) Generally all alloys are susceptible to SCC, however, there are a few pure metals which have been observed to undergo SCC, such as 99.999% Cu and high purity iron. (3) SCC of a specific alloy is caused by only a few chemical species in the environment. (4) There is a period called 'induction period' which is necessary to produce crack initiation, similar to the induction period required for producing pitting. (5) Conditions of cracking are specific to alloy and environment. An alloy maybe corroded in one corrosive medium while it may not under SCC. All the specific environments for a particular alloy are not known. (6) The mode of cracking may be intergranular or transgranular. The transition from intergranular to transgranular depends upon factors, such as heat treatment, corrosive medium, stress level and temperature. (7) The rate of attack is very rapid at the crack tip and very low at sides of the crack. (8) There is a particular corrosion reaction critical for SCC to occur. (9) SCC cracks are microscopically brittle in appearance. (10) The fracture mode of an alloy in SCC is always different from its fracture mode in a plain strain fracture. (11) For some systems, there appears to be a threshold value of stress below which SCC does not occur. (12) There may exist a critical potential below which SCC does not occur. 4 .8.11 ENVIRONMENTAL AND M A T E R I A L S F A C T O R S IN S T R E S S CORROSION CRACKING The cracking environments are specific because not all environments promote stress Types of corrosion: materials and environments Table 4.8 Common environmental stainless steel alloy systems susceptible to SCC • • Concentration of chloride (evaporation) Elevated temperature 1 91 classified in two categories: (1) grain boundary precipitation and (2) grain boundary segregation. A. Grain Boundary Precipitation This is best illustrated by the formation of chromium carbide (&23C6) on the grain boundary and depletion of chromium adjacent to the grain boundary in stainless steels, such as AISI 304. The grain boundary region is subjected corrosion cracking. The specific environments to corrosion attack called intergranular attack. which promote SCC of alloys are shown in Carbide precipitation also takes place in nickel Table 4.8. Alloys which are inherently corro- alloys, such as alloy 600. In the case of aluminum sion resistant, such as austenitic steels which alloys intermetallic compounds, such as Mg2Si, develop protective films, require aggressive ions are precipitated. For instance, the intermetallic to promote SCC, whereas alloys of no inherent compound which is precipitated in aluminum corrosion resistance, such as carbon steels, require alloy (5086-H34) containing more than 3% Mg an environment which is partially passive. Thus, is beta phase (Mg2Al3) which is highly anodic to for cracking of Mg-7% Al, a mixture of CrO^ 2 the aluminum magnesium matrix. and Cl~ ions is required, and no SCC would be caused by either of the species alone. In order for SCC to occur, it is important to B. Grain Boundary Segregation maintain a balance between passivity and reactivity. If either of the above is not met, there would Impurities, such as phosphorus, sulfur, carbon be no SCC but localized corrosion or general cor- and silicon, segregate at the grain boundary and rosion on the one hand, and complete passivity cracking contribute to the SCC steels and nickel base alloys. on the other hand. Cracking is only achieved in certain specific Examples of grain boundary precipitation environments and not in all environments. It is, therefore, not very practical to suggest specific (1) Precipitation of chromium carbide in stainsolution requirements for cracking to occur. less steel in temperature range 500-800°C. • • Oxygen Time • pH>2 4 .8.12 AND MATERIALS CHEMISTRY (2) Chromium carbide precipitation in nickel base alloys, such as alloy 600. MLCROSTRUCTURE The effect of materials chemistry and micro- 4 . 8 . 1 3 T H E R M O D Y N A M I C structure of materials in SCC and the interrela- R E Q U I R E M E N T S F O R tionship between the two is highly complex. The composition of the alloy has a significant bearing A N O D I C A L L Y A S S I S T E D on the properties of the passive films and phase D I S S O L U T I O N distribution. For example, a high amount of carbon in steels tends to form chromium carbide (a) The dissolution or oxidation of the metal which causes sensitization of steel and leads to must be thermodynamically possible. intergranular corrosion. Similarly, impurity ele- (b) The protective film formed on the metal ments in steels segregates and affects the corrosion surface should be thermodynamically stable. dissolution process. (c) On the basis of (a) and (b), it has been The effect of materials chemistry and suggested that a critical potential exists at microstructure on intergranular corrosion can be which SCC occurs. The values of critical 192 Principles of Corrosion Engineering and Corrosion Control p otential for SCC of alloys are shown in Fig. 4.38. The critical values are experimentally obtained from the anodic polarization curves as discussed in Chapter 3. Zones for intergranular and transgranular cracking of alloys 600 and 800 are shown in Figs 4.39 and 4.40. The zone of intergranular cracking of AISI Type 304 stainless steel in 10% NaOH at 288°C is shown in Fig. 4.41. It is observed that intergranular corrosion occurs over a much wider range compared to transgranular cracking. In zone 1, the metal is in a transition state from active to passive formation (Fig. 4.38). The conditions of film formation on crack wall and film dissolution at the crack tip exist in this zone. An active surface is necessary for the process of dissolution to occur, such as active surface provided by the anodic zones formed in the material. Similar conditions also exist in zone 2. The above conditions are ideal for transgranular cracking which takes place in the two zones, 1 and 2. Zone 2 Cracking zones I Zonel log current density Figure 4.38 Potentiokinetic polarization curve and electrode potential values at which stress corrosion cracking appears. (From lones, R.H. Battelle Pacific Northwest Laboratory, Ricker, R.E. N.B.S., Metals Handbook, 13, ASM. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) 1.20 1.00 ^ 0.80 0,60 A3 f 0.40 jTransgranular/intergrariular SCC | £ p. 020 j ~—-_ -0.20 1E+05 1E*06 1E+07 Currenty density, rnA/cm 1E+08 2 1E+09 Figure 4.39 Alloy 600 in 10% NaOH solution at 288° C showing transgranular/intergranular SCC Types of corrosion: materials and environments 193 1.20 _-_— 1.00 _i _ _ < |lntergranular$CC | 0.80 1 J 0.60 > 15 w § 0 u40 | j Transgranular SCC [ u~ , S. 0.20 0 —-^ -0.20 1E+0S 1E+06 1E+07 1E+08 1E+09 Currenty density, mA/cm* Figure 4.40 Alloy 800 in 10% NaOH solution at 288° C showing regions of inter granular and transgranular SCC. (From Jones, R.H. and Ricker, R.E. (1987). Metals Handbook, 13, Corrosion. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) 1.20 1.00 0.80 j f 0.60 > 18 W U *W c 0 .40 |lntergranular SCC | 0.20 i 0 ——^ -Q2Q 1E+01 1E+02 1E*03 1E*04 a 1E*0S Currenty density, mA/cm AISI Type 304 stainless steal In 10% NaOH at 288* C Figure 4.41 Electrode potential values at which intergranular and transgranular stress corrosion cracking appears. (From Jones, R.H. and Ricker, R.E (1987). Metals Handbook, 13. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) [64] 1 94 Principles of Corrosion Engineering and Corrosion Control Intergranular cracking takes place over a (d) Grain orientation is affected by plastic deformation. much wider range of potential because of a larger difference of potential between the segregation of impurities at the grain boundary 4.8.17 MECHANISM OF S T R E S S and the metal matrix. CORROSION CRACKING The first recognized case of environmentally induced corrosion was that of seasonal cracking of brass in the rainy season. It was a classical examFollowing are the sources of formation of anodic ple of SCC of brass in ammoniacal environment. zones which assist dissolution: Soon upon being realized as a major mechanism, it became the subject of intense research (a) Composition and microstructural differ- and a brief review of literature would show that ences. thousands of papers have been written on the (b) Grain boundaries and sub-boundaries. subject. The most intriguing is the mechanism (c) Stress disorder effects. of stress corrosion cracking and stress corrosion (d) Local rupture of surface films due to stress. crack propagation. A full discussion of the mech(e) Heterogeneous effects of stress, or prior cold anism is beyond the scope of this chapter. An work. online search would reveal thousands of articles and update references. Four basic categories of mechanisms have been proposed: ZONES 4.8.14 S OURCES OF ANODIC 4 .8.15 T H E I N F L U E N C E O F S T R E S S O N R ATES O F L OCALIZED C ORROSION Localized corrosion as a result of stress or deformation may arise from one of the following factors: (a) Rupture or damage of surface film. (b) Changes in the metallurgical characteristics of the alloy. (c) Change in the polarization characteristics. (d) Change in the electrode potential. (a) (b) (c) (d) Mechano-electrochemical model [56]. Film rupture model [57-59]. Embrittlement model [60]. Adsorption model [61,62]. (a) Mechano-Electrochemical Model This suggests that there are pre-existing paths in an alloy which become intrinsically susceptible to anodic dissolution. For instance, a grain boundary precipitate anodic to the grain boundary would provide an active path for localized corrosion to proceed. Similarly, if a more noble constituent is precipitated along a grain boundary, the impoverished zone adjacent to the precipitate would provide an active path for localized corrosion. Also, the removal of the protective film at the crack tip by plastic deformation would facilitate the onset of localized corrosion. For instance, the precipitation of C11AI2 for an Al-4 Cu alloy depletes of copper along the grain boundaries and provides active paths for localized corrosion. Application of cathodic protection stops crack propagation and its removal restarts the process, thus establishing conjoint action of mechanical and electrochemical processes. 4.8.16 C H A R A C T E R I S T I C S O F S T R E S S A ND P L A S T I C D EFORMATION (a) The internal energy of the materials is increased by plastic deformation. (b) The amount of plastic deformation varies from grain to grain and even within the grain. (c) Fabricating processes also generate residual stresses. The residual stress influences the susceptibility to SCC very significantly. Types of corrosion: materials and environments 1 95 (b) Film Rupture Model This is based on a repetitive cycle comprising (a) localized film disruption, (b) localized attack at the point of disruption and (c) film repair. Plastic strain plays a major role. This mechanism has several variations. mechanism to another. The above mechanisms are summarized below. 4 .8.19 PRE-EXISTING ACTIVE P ATH M ECHANISM [67-69] (a) Mechanism (c) Embrittlement Model This is based on the postulate that an electrochemical process embrittles the materials in the vicinity of a corroding surface. This mechanism was based on a study of high strength martensitic steels in chloride media [63]. If solute segregation or precipitation occurs at the grain boundary, such as the segregation of impurities like sulfur and phosphorus, and the precipitation of chromium carbide as discussed under intergranular corrosion, the electrochemical properties of the matrix and segregate are changed. The area adjacent to the grain boundary is depleted of by one of the alloying elements and it is preferentially attacked. Under such conditions, localized galvanic cells are created with the segregate being generally anodic to the matrix of the grain (Fig. 4.42). The polarity can also reverse in some situations. Such structural features lead to intergranular corrosion. The changes brought about by change in the grain boundary structure, produce compositional difference and provide active paths for SCC to occur. This mechanism is predominant in cases where SCC is governed by electrochemical or metallurgical factors rather than stress. (d) Adsorption Model Adsorption of damaging species causes weakening of cohesive bonds between surface metal atoms by specific damaging species. The surface energy associated with crack formation is lowered by adsorption of species and the probability of a metal forming a crack under tensile stress is increased. A universal theory does not exist and several theories have to be examined for results and analyzed to reach a plausible explanation. As observed in the above discussions, it appears that there is not a single but two or three different mechanisms which operate. The mechanism of crack propagation falls into two basic categories, the dissolution model and the mechanical fracture model. (b) Evidence in Support (1) The cracking susceptibility can be altered by altering the ratio of anode to cathode area at the grain boundary. For instance, the susceptibility to SCC is reduced if the grain boundary regions become less anodic with respect to the grain matrix, i.e. the difference of potential between the active area of the grain boundary and the grain matrix is reduced by alloying. (2) Steel samples are etched when immersed in a cracking environment at predetermined potentials. Corrosion attack takes place only in certain selective regions, i.e. upon preexisting paths determined by the changes in the grain boundary structure. In nitrate solutions, the grain boundary attack only penetrates to a depth of few millimeter in the 4 .8.18 E X I S T I N G C RACKING MECHANISMS O F S T R E S S C ORROSION On the basis of the progress made in the understanding of the SCC phenomenon in recent years, the existing mechanisms of SCC can be divided into three categories: (1) pre-existing active path mechanisms, (2) strain assisted active path mechanism and (3) adsorption related mechanisms [64-66]. These subdivisions are only from a point of view of understanding of one continuous process with only shift of emphasis from one 1 96 Principles of Corrosion Engineering and Corrosion Control Hi ~Lf—I Anodic phase Cathodtc phase Figure 4.42 Galvanic cell mechanism. (From Perkins, R.N. (1972). Br. Corros. Jr., 7, 15. Reproduced by kind permission of British Corrosion Journal, London, UK) absence of a stress. However, on applying anodic polarization, the steel is virtually destroyed by intergranular corrosion. The stress and applied potential both appear to contribute to the dissolution of the active path [70]. The low susceptibility of low carbon steels can be explained by the change in the distribution of carbon which occurs after a deformation processes. Carbon is not generally present at the ferrite grain boundaries [71]. The carbon distribution is intimately related to the active path, and it appears to fix the site of the active path. Pure iron containing about 0.005% impurity, is not susceptible to SCC. It becomes susceptible when it is carburized [72]. Aluminum alloys, such as Al-Mg, Al-Cu and Al-Zn-Mg, are susceptible to intergranular corrosion in the absence of stress after appropriate heat treatment [73]. If the heat treatment produces precipitates, such as MgAl3, MgZn2 and C11AI2, intergranular corrosion and SCC become very severe. Galvanic cells are set up as a result of differences in the electrochemical properties between the precipitates and the grain matrix. The cracking susceptibility can be altered by altering the grain boundary precipitate volume (anode/cathode area ratio). A precipitate-free zone exists next to the grain boundaries in Al-Zn-Mg alloys [74]. It has been observed that the susceptibility of alloy having a narrow precipitate-free zone is more than the alloy having a wide precipitate-free zone. It has been observed that deformation occurs more readily in the precipitate-free zone and causes the oxide film to rupture. Although the later is disputed, it appears that the precipitate-free zone influences SCC by its effect on deformation at the crack tip. This shows that mechanical effect is also important but it only plays a secondary role in SCC of aluminum alloys. (3) 4 .8.20 STRAIN GENERATED (4) A C T I V E P ATH M E C H A N I S M S [74-76] (a) Mechanism In contrast to the active paths mechanism, there are many systems where strain is the controlling influence, e.g. high strength steels in chloride solutions or tin alloys in methanol. Such instances are best explained by strain-generated active path mechanisms. It is based on the idea of straininduced rupture of the protective film. Various theories have been proposed which relate crack propagation to dissolution of the crack tip and the existence of strain/stress conditions existing in that region. These theories depend on the existence of strain/stress regions for crack propagation and hence, they are classified as (5) (6) Types of corrosion: materials and environments 1 97 a strain-generated active path mechanism. The (5) The rate of cracking is controlled by the rate of film growth. A process of competitermlactive path'has been explained earlier. tive adsorption between species promoting The most outstanding mechanism which has passivity, and species promoting Cl~, takes found a wide acceptance is the film rupture mechplace. If repassivation occurs too quickly, anism. This mechanism has been extensively only a small amount of corrosion would take studied in stress corrosion cracking of alphaplace and the crack would grow. brass in ammoniacal environment, although it was originally proposed for caustic cracking of (6) The crack propagates by successive dissoluboiler steel. Here are some salient features of the tions of the crack tip when it successively mechanism. The items are illustrated in Fig. 4.43. becomes bare due to the rupture of the oxide film due to plastic strain in the underlying metal. (1) The theory assumes the existence of a passivating film on a metal surface. The exisIn some other theories, it has been suggested tence of such films has been experimentally that the crack tip always remains covered with an verified. (2) The passivating film is ruptured by plastic oxide and the film is only periodically ruptured by emergence of slip steps. It has also been suggested strain in the underlying metal. (3) After the film is ruptured, the bare metal is that the crack tip remains bare because the rate of rupture of the oxide film is higher than the rate exposed to the environment. (4) The processes of disruptive strain (disrup- of repassivation of the film. In general, the rate of tion of protective film) and film formation attack is determined by stress (applied or resid(due to repassivation), alternate with each ual), electrochemical potential, total strain rate and specific ions and effect of solute segregates. other. Oxide metal Rupture deformation Oxide Oxidation Dissolved region Slip line Rupture (B) (A) Figure 4.43 Strain generated active path mechanisms (A) Film rupture model (B) slip step dissolution model. (From Sedriks, A.J., Slattery, P.W. and Pugh, E.M. (1969). Trans. Am. Soc. Met., 62, 1238. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) 1 98 Principles of Corrosion Engineering and Corrosion Control Figure 4.44 Schematic representation of crack propogation by thefilmrupture model. (From Staehle, R.W. (1971). in Stress Corrosion Cracking in Alloys, NATO, p. 223) A generalized mechanism based on a slip rupture This is because arsenic addition directly model is shown in Fig. 4.44. influences the film forming characteristics of the brass. The formation of passive film is most important, because SCC takes place commonly in alloys (3) It has been shown that the preferential that are covered by a highly protective film, such growth of film occurs along grain boundas aluminum or steel alloys. Under a tensile strain, aries, as shown by the preferential oxidation the slip plane breaks the protective film as shown of copper at the grain boundary by FeCi3 in in Fig. 4.45a, a small part of the film undergoes copper-gold (Cu3Ag) alloys. The removal of dissolution as shown in (b) and later repassivacopper leaves gold in a spongy state and the tion takes place as in (c). As pointed out earlier, if grain boundary is attacked due to the ruprepassivation occurs too rapidly, corrosion attack ture of the film. The film growth, therefore, would be too small and the crack would propais a structural dependence process, and its gate slowly. On the other hand, if repassivation rupture leads to crack propagation [78]. occurs very slowly, excessive metal dissolution (4) A strain rate which produces a bare metal occurs on the crack tip and sides. This widens at a faster rate than the film formation rate and blunts the crack tip, and the crack growth is sufficient to cause deformation. The curis arrested. The greatest damage is caused by rents associated with straining electrodes are moderate repassivation rates. higher than at a static surface as shown by Hoar [79]. (5) Generally, the segregation of solute on dislocation results in localized corrosion. (b) Evidence in Support Under such conditions the corrosion progresses along the active paths. If such seg(1) Ellipsometric studies on the effect of inregates are not present, corrosion takes place crease of zinc content on the film growth because of moving dislocations which generin ammoniacal solution confirm the film ate chemical activity similar to that produced rupture mechanism [77]. by segregation leading to stress corrosion (2) It has been shown that addition of arsenic cracking. to brass increases its resistance to SCC. Types of corrosion: materials and environments 1 99 , Environment Crack arrests markings s Slip plane K \ Slip plane <€) Figure 4.45 Schematic of film rupture model (a) illustrates the main feature of the model - the protectivefilmis ruptured at the crack tip by localized slip, permitting propagation by anodic dissolution. The crack tip is depicted in greater details in (b) which illustrates the view that crack advance results from a number of independent film rupture and transient dissolution events. (From Staehle, R.W. (1979). Stress Corrosion Cracking in Alloys, NATO, p. 223) (6) The observation that the rate of crack Active paths for SCC may also be generated by the propagation observed is greater than that rupture of an oxide film or by the dissolution of accounted for by electrochemical observa- slip plane. tion, has been supported by the evidence that To conclude, it is observed that in mechacrack propagation may be discontinuous. nisms based on a pre-existing active path, the It has also been suggested that the discrep- susceptibility to SCC at one end is predominantly ancy in the rate of crack propagation actually controlled by localized corrosion where stress is observed and accounted by electrochemical not necessary, such as carbon, steel in NO^~, and process, is due to the crack formation by Al-Zn-Mg in CI - , brass in NH3 solution, to the mechanical factors and crack growth by elec- other limit where the stress influences the proptrochemical factors. The mechanical crack agation of a crack from an intergranular notch, propagation is followed by electrochemical such as in Mg-Al alloys in CrOJ,Cl~ and high crack growth. strength steel in water. It has been stated earlier that active paths are generated by formation of segregates, precipitates at the grain boundary. There exists a difference of potential between the segregates, precipitates and the grain matrix and hence, galvanic cells which are formed by the precipitates/segregates and metal matrix provide favorable sites for continuous paths and the active corrosion to proceed. 4 .8.21 A D S O R P T I O N PHENOMENON RELATED [80-85] (a) Mechanism This model is based on the assumption that the adsorption of environmental species lowers the 2 00 Principles of Corrosion Engineering and Corrosion Control species are present and they are adsorbed at the crack tip, the bond strength is reduced, the surface energy is effectively lowered, and fracture takes place. The species may also diffuse into the metal and become absorbed in some region in advance of crack tip where the stress/strain conditions are favorable for the nucleation of a crack. In the latter case, hydrogen is the only known species which diffuses and causes SCC referred to as hydrogen embrittlement. The latter is, however, considered as a particular case of SCC. Other mechanisms suggest that hydrogen atoms are formed by the reduction of hydrogen ions within the crack. These hydrogen atoms cause the weakening of the bonds beneath the surface of the crack tip. This can also be achieved by formation of metal hydrides which are known to be brittle. The formation of hydrogen gas in small quantities formed by H atoms diffusion through the metal lattice leads to the development of enormous pressures in micro-cavities where H + H combine to H2 gas. This leads to the rupture of the metals. Pressures as high as 3000-20 000 atmospheres (0.3 to 2 GPa) are developed. The adsorption mechanism, however, fails to explain the phenomenon of SCC of metals where considerable plastic deformation occurs. SCC progresses without any significant plastic deformation, but localized plastic deformation occurs at the crack tip, and under such conditions the surface energy term is significantly lower than the plastic work term, hence, the reduction of surface energy would be insignificant and it would exert only a negligible effect on fracture stress (3Fr). Equation (4.47) can be modified to equation (4.48): [E(s-p)]± b interatomic bond strength and the stress required for cleavage fracture (a brittle fracture in which the crystallographic planes are separated). There are several features which identify a cleavage fracture, such as river patterns, tongues, herring bone (Fig. 4.46). The theoretical fracture stress required to pull apart two layers of atoms of spacing b is given by [80]: 3Fr = (!j\ where E = modulus of elasticity y = surface energy b = spacing between atoms. (4.47) The above theory implies that if the surface energy term y is reduced, then 3pr will also be reduced. This led Orwan [80] to suggest that delayed fracture of glass occurs by adsorption of environmental species which lowers the surface energy, and hence the stress required to cause fracture. The same principles were later adopted to account for the SCC of metals. If environmental aFr = where (4.48) s = specific surface energy p = plastic work term. In the above condition, p > s, hence, fracture would only be slightly easier with the lowering of the surface energy. The effect of lowering of surface energy is, therefore, not significant. Figure 4.46 Typical cleavage fracture in iron Types of corrosion: materials and environments It can be concluded that specific adsorption does not account properly for SCC of metals where plastic deformation is associated with fracture. 201 (c) Evidence of Support (1) Studies on Al-Mg alloys have shown that the process of adsorption assists the mechanical part of crack propagation. (2) The failure of engineering materials by adsorption of hydrogen causing embrittle(b) Grain-size Relationship ment of metal in advance of the crack tip. It has been shown that coarse grain size materials (3) The stress corrosion cracking of a-titanium alloys occurs by nucleation of hydride (by the are more susceptible to SCC than fine grain size interaction of absorbed hydrogen with the material. A patch-type of relationship connecting metal). grain diameters and the stress required to initiate Vl (4) The failure of high strength steels by hydroa SCC crack is well-known: d?r = 3o + Kd~ , gen adsorption lends strong support to where 3o and K are constants. K is related to the hydrogen adsorption mechanism; only spesurface energy associated with formation of new cific environments promote stress corrosion surface: cracking, and this goes in favor of an adsorption mechanism. (5) Lack of systematic explanation of the envi(dnGy) K=— (4.49) ronmental aspect of cracking. Specific envi1—v ronments are more consistent with the adsorption theory. (6) Failure of materials by absorbed hydrowhere gen provides strong evidence in favor of y = surface energy adsorption theory. The cracking of high v = Poissons ratio strength steels is not dependent on speG = modulus of rigidity. cific environment and failure would occur as long as a source of hydrogen is availMeasurement of the relation between the able. The hydrogen lowers the surface grain size and stress corrosion fracture, allows energy by adsorption and lowers the cohethe surface energy to be calculated. A lowersive forces of the lattice. The lowering of ing of surface energy was observed for mild cohesive forces causes the lowering of the steel in nitrate solution [62]. The surface enerstress required to cause SCC of the metal. gies so calculated [80] were found to be lower This particular mode of cracking by hydrothan in other conditions. Hence, it was deduced gen is called 'hydrogen embrittlement" A that surface energy is lowered by adsorption of sufficient reduction in the hydrogen gas active species during stress corrosion cracking. pressure surrounding the specimen conThis supported the adsorption mechanism. Howtaining a crack at a given intensity would ever, later it was found that the yield stress and cause arrest of the crack but a subsequent work hardened flow stress after 'plastic deforincrease in the pressure would restart the mation are also dependent on grain size similar crack. That cracking results from the forto the surface energy. It is, therefore, observed mation of a hydride phase [85] due to that grain size is related to plastic flow as it the adsorption of hydrogen by titanium is is to the lowering of surface energy. If plastic supported by a strong evidence. The SCC deformation accompanies stress corrosion crackof a-titanium alloys constitutes an examing, the plastic strain term (yp) becomes more ple of slow strain rate hydrogen embrittlesignificant than (ys), the surface energy lowment. Other researchers believe that SCC ering, hence adsorption would be insignificant. in titanium alloys results from dissolution. The adsorption theory appears to be more valid Thus, adsorption theory accounts well for where the yv term is not much higher than hydrogen induced cracking (HIC). ys term. 2 02 Principles of Corrosion Engineering and Corrosion Control known. Electrical conductivity is a very important characteristic and if a particular ion of certain electrical conductivity promotes SCC, other ions of similar electrical conductivity should also promote SCC. The specificity of ions can be explained better by an adsorption mechanism than other mechanisms. The propagation of SCC requires the reaction which occurs at the crack tip to proceed at a much faster rate than any dissolution process. A balance between passivity and reactivity has, therefore, to be maintained. For those alloys which develop a protective film, an aggressive ion is required to promote SCC. Other metals with a low corrosion resistance, such as carbon steel or Al-Mg alloy, require a partially passivating balance between chemical reactivity and passivity and if this balance is not maintained, the result would be general corrosion or pitting rather than SCC. The evidence for the above is provided by anodic polarization of mild steel in ( N H ^ CO3 at 75°C by potentiodynamic method [86] (Fig. 4.47). It can be observed from Fig. 4.47 that between —600 and — 800 mV, the specimens suffered pitting and between —500 and —600 mV cracking (d) Limitations (a) The adsorption mechanism accounts mostly for metals where yp ^> ys and it does not account for the SCC of metals that undergo significant plastic deformation. (b) It does not explain how the crack maintains a sharp tip in a normally ductile material because it does not include a provision for limiting deformation in the plastic zone. (c) The discontinuous nature of cracks observed in some instances is not explained. 4 .8.22 E F F E C T ON OF ENVIRONMENTAL FACTORS S CC (a) Specific Ions Not all environments promote SCC and hence, the environments causing SCC are specific. However, all environments that cause SCC are not fully Anodic protection Cathodic protection 0.1 r 1 0.01 h 1000 mV/min 20 mV/min 0.001 0.0001 -400 -500 Log(current density) -800 Figure 4.47 Fast and slow sweep rate anodic polarization curves for steel in (NH^CC^ at 75° C. (Perkins, R.N. (1972). Br. Corros. Journal, 7, January. By kind permission of British Corrosion Journal, London, UK) Types of corrosion: materials and environments and below —850 mV, the steel remained immune to cracking. These observations demonstrate the phenomenon of passivity and reactivity. The occurrence of SCC requires the existence of a critical balance between the two. If the balance is not maintained, there would be either pitting and general corrosion or complete passivity. The above observations are supported by electrochemical studies of steel in MgCk solution [87]. In Fig. 4.48, current vs time curves are shown. Three types of curves are observed. Curve 1 is indicative of the fact that the film breaks down by emergence of slip steps as discussed earlier. The surface re-passivates so rapidly that no time is given for crack propagation to occur. In curve 3, passivation occurs too slowly which is a favorable situation for pitting in curve 2; there is the possibility of crack propagation because of a passivation time which allows the crack propagation and a critical balance is maintained between reactivity and passivity. Propagation occurs because more fresh metal surface is generated than can be re-passivated. The above observations support the electrochemical mechanism of SCC. Electrochemical studies are also used to distinguish between SCC and HE (hydrogen embrittlement). 203 (b) Effect of Oxygen It is commonly believed that oxygen is essential for SCC to take place. This opinion has, however, been contradicted, and it has been suggested that oxygen only acts as a reducible species, a function which could well be served by Fe +3 and Cr 6+ rather than oxygen. It has been argued that the hydrogen bubbling only indicates a reduction process and it does not suggest hydrogen embrittlement. It is suggested that hydrogen enters steel when it is not escaping from its surface. (c) Role of Hydrogen The role of hydrogen in SCC is very crucial and yet more controversial. The effect of hydrogen has been mostly explained in terms of hydrogen embrittlement, which causes a loss of ductility in the metal. The following observations are in support of a hydrogen embrittlement mechanism: (1) Escape of hydrogen bubbles from cracks. (2) Fractographic studies. (3) Stimulation of cracking by hydrogen embrittlement. X ^ !__/" Time Figure 4.48 Schematic illustrations of the relationship between reaction current transient and the amount of reaction occurring at an emergent slip step. (From Staehle, R.W., Royuela, J.J., Raredon, T. et at. (1970). Corrosion, 26,451. By kind permission of NACE, Houston, Texas, USA) [87] 2 04 Principles of Corrosion Engineering and Corrosion Control and increases with anodic polarization. The reverse is true for the anodic dissolution mechanism - (active path mechanism) when SCC occurs (Fig. 4.49). In order to account for the failures of high strength steel by hydrogen two mechanisms have been suggested: (a) Active path corrosion (APC), anodic dissolution and (b) hydrogen embrittlement (HE). The active path mechanism which accounts for cracks by providing active paths for propagation of cracks to proceed in SCC differs from hydrogen embrittlement where adsorption of hydrogen at cathodic sites is followed by embrittlement of the material. The two processes, SCC and hydrogen embrittlement (HE), must be differentiated from each other, which for some systems is extremely difficult. To distinguish the two, the relationship between time to failure (TF) and potential is studied. In the case of hydrogen embrittlement, the time to failure decreases as cathodic current is applied 4.8.23 STRESS CORROSION CRACKING OF S T A I N L E S S STEEL In the following section, an attempt is being made to summarize the practical information regarding stress corrosion cracking (SCC) of various types of steels. Table 4.9a lists primary conditions which can cause SCC of stainless steels. Table 4.9b shows some typical results on the SCC of steels H Hydrogen embrittlement kPC Active path corrosion Tr Time to failure a c Anodic polarization Cathodic polarization Figure 4.49 Active path corrosion and hydrogen embrittlement mechanism. (From Wilde, B.E. (1971). Corrosion, 27, 326. By kind permission of NACE, Houston, Texas, USA) Types of corrosion: materials and environments Table 4.9a Primary conditions which cause stress corrosion cracking of stainless steels • • • • Susceptibility of the alloys Metallurgical conditions Damaging environment Time Table 4.11 Effect of various elements on SCC susceptibility of austenitic steels Element Ni(>8%) C Si Cb Ti V P Mo Cr(>16%) Cu Effect Harmful Beneficial Beneficial Beneficial Beneficial Beneficial Harmful Harmful Harmful Harmful 2 05 Table 4.9b Results of stress corrosion cracking on various stainless steels and high alloys exposed at 100°C to 100 ppm chloride Cracked 304, 304 L, 316, 316 L, 347,310,202 Resisted cracking 329, 430, 446, Fe-Ni-Cr 20, Fe-Cr-Fe Alloy 600, Fe-Ni-Cr 825 Several factors affect the SCC of austenitic steels in chloride solutions. The effect of some important factors are given below. (a) Alloy Composition and high alloys exposed to 100 ppm chloride. On top of the table is the 300 series stainless steels in annealed condition in chloride environment, as they represent the most common situation because of their wide application. Unfortunately, the environment-alloy interaction is least understood. The SCC of austenitic stainless steel in chloride has been the subject of considerable research. The factors responsible for the SCC of austenitic steels in chlorides are given in Table 4.10. The effect of various elements on SCC susceptibility is shown in Table 4.11. The nominal compositions of selected AISI standard grades of stainless steels are given in Table 4.4a. The highest susceptibility to SCC is shown by austenitic steels containing 8% nickel. Alloys containing less than 8% or more than 8% nickel exhibit a higher resistance (Fig. 4.50). The beneficial effect of a higher nickel content is shown by Incoloy, alloy 800, which fails in boiling 45% MgCi2, but which is resistant to the environments in which alloys of lower nickel content, steel 304 (10% Ni) and steel 316 (12% Ni), fail by stress corrosion cracking. Resistance to SCC in most environments is obtained by adding about 30% nickel. Nickel is useful in resisting corrosion in mineral acids. Nickel also stabilizes the austenitic phase [88]. Chromium is an essential element for forming a passive film, whereas other elements assist in stabilizing the film. The passive film is not very effective up to 10.5% chromium content, but as the chromium content is increased between 17 and 20%, the passivatingfilmbecomes very stable. Austenitic steels generally contain 18-20% chromium. Whereas the corrosion resistance of austenitic steel is increased by addition Table 4.10 Contributing factors to SCC in chloride solution No. 1. 2. 3. 4. 5. Contributing factors Concentration of chloride Elevated temperature pH>2 Oxygen Time 2 06 Principles of Corrosion Engineering and Corrosion Control wr TJ o In 1 "\ 1 /o % % o c ? p 00 | Crackingi] nOp O TT s n Tl o 1 / / O ° O QD oo o o o o o o o <3 o o o /f / 9 1 10 / /& /< /^ / |NoCrad Jng| D o o 1o o o o o o J ft // \ °O 11 < ) XW y-^Y i>0 40 Percent Nickel 61 3 80 Figure 4.50 Breaking time of iron-nickel-chromium wires in boiling 45% magnesium chloride solutions. (From Copson, H.R. (1959). Phy. Metallurgy of Stress Corrosion Fracture, Interscience Publishers ) [88] of chromium, its resistance, particularly to SCC, concentrated form, steels 316, 304 and other is slightly lowered at levels higher than 10%. 300 series suffer caustic cracking. The type of Steel 304 contains 18-20% of Ni and steel 316 cracking will be separately dealt with. The benhas 16-18% Cr. Chloride SCC occurs generally eficial and detrimental effects of various elements at pH levels above 2.0 provided the other con- on SCC of austenitic stainless steels are shown in ditions for cracking also exist. As the pH value Table 4.11. increases towards the alkaline side of the pH scale, It is to be noted that the harmful or benefithe tendency for SCC is reduced. At high tem- cial effect is only with regard to stress corrosion perature and when free caustic is present in a cracking and not to any other form of corrosion. Types of corrosion: materials and environments 2 07 only one day to crack at 100°C. Types 304 and 316 steels have been reported to be susceptible The susceptibility of austenitic steels to SCC to SCC in water containing 20-60 ppm chloride increases with temperature. Mostly failure occurs above 60°C. by concentration of chloride by evaporation. In normal testing of SCC in chloride solution, heat has to be applied to such a degree as to (c) Stress Levels cause evaporation of chloride or alternate wetting and drying to concentrate the chloride. It It is to be noted that SCC would never occur is to be remembered that stainless steels do not in the absence of either corrosive environment crack in strong chloride environments at ambient or stress. SCC can occur both by residual and temperature. tensile stress and never by surface compressive Dana and Warren [89] conducted SCC tests stress, although the later may be used to preon cracking of hot stainless steel pipe under vent SCC. SCC can occur from stress close to wet chloride-bearing thermal insulation. It was the yield point, down to as low as 0.1% of the shown by the experiment that at 90% the yield yield strength [90]. For many systems, a 'threshstrength, SCC occurs only at temperatures above old ' or limiting stress has been observed. Stresses 60°C (Fig. 4.51). It is observed from the figure as high as 4000-7000 psi are built up by corthat the specimen exposed to 1.00 ppm chloride rosion products as shown by the SCC tests on took 6 months to crack at 60° C, whereas it took steel 347. Whereas the role of stress is extremely (b) Effect of Temperature 100 / S O 100°C / I /n 1 I S 3 «o £ I so*e I 60 °C / f / / (6 mo) 8 40 6 20 1 (1 Day) ^ S Q^l (1 W k j X 1 (1 mo) 1000 to 100 Test time (hrs) 10000 Figure 4.51 Effect of temperature on the time for cracking of Type 304 stainless steel specimens exposed to water containing 100 ppm chloride. (From Dana, A.W. and Warren, D. (1957). Bulletin, DuPont Engg. Dept, ASTM Bulletin No. 225, October. Reproduced by kind permission of ASTM, Int., Ohio, USA) [89] 2 08 Principles of Corrosion Engineering and Corrosion Control concentrated by chemisorption, by leaching or by migration to pitting sites. If the temperature is ambient, no SCC would occur even in the presence of lOOOOppm chloride. SCC of austenitic steels would not occur in seawater if it is cathodically protected. The time for SCC of 304 steel in chloride-containing water at 100°C is reduced significantly on an increase in the concentration of chloride as shown in Fig. 4.52 [91]. important, it is to be considered that 'neither the residual nor the external stress are prerequisites for SCC because the corrosion process in chloride media can generate stresses of sufficient magnitude to cause S C C Stress corrosion cracking can occur in unnotched austenitic steel specimens which are stress-free when exposed to 42% MgCl2 at 135°C. Some of the stresses have been shown in Section 4.4 as intergranular cracking. (d) Chloride Concentration (e) Electrochemical Potential Intergranular SCC occurs over a wide range It is impossible to specify a chloride concentration at which SCC occurs. Stress corrosion of potentials. The process of film formation can occur at a wide range of concentration, and film breakdown occurs at a certain value of such as ranging from 10 000 to 0.02 ppm Cl~ potential termed 'criticalpotential' Transgranubecause chloride in small concentration can be lar type SCC occurs in zone 1 and zone 2, whereas £ 3 *Q I 5 10 100 Test time, hr 1000 10000 Figure 4.52 Effect of chloride concentration on the time for cracking of Type 304 stainless steel specimens exposed at 100°Cto chloride bearing water. (From Warren, D. (1969). Proc. 15th Annual Purdue Industrial Waste Conf.y Purdue Univ., May 1-19) [91] Types of corrosion: materials and environments intergranular corrosion occurs over a wider range of potential between the two zones [92]. It is to be noticed that in zone 1, there is a transition from active to passive state and hence, film formation takes place on the side of the crack and film rupture on the crack tip. Similar conditions are observed in zone 2, however, it is observed that zone 2 is above the critical pitting potential (£p) and hence, cracking is initiated by pitting in this zone. Because of chemical heterogeneity at the grain boundary and increased activity of the grain boundary with respect to the bulk material, intergranular SCC occurs over a wide range of potentials. The pitting potential of steels in 1000 ppm chloride solution is as shown in Table 4.12. 2 09 of austenitic steels. It has been reported that iodides inhibit the SCC of steel. Similarly, silicates, phosphates, carbonates, iodides and sulfites are effective inhibitiors. The sulfites inhibit corrosion of stainless steels by removing oxygen from the system [94]. 4.8.24 S T R E S S C O R R O S I O N C RACKING O F F ERRITIC STEELS The simplest stainless steel contains only iron and chromium. Chromium stabilizes the ferrite phase. Ferrite has a body centered cubic structure, it is magnetic, high in yield strength and low (f) Oxygen in ductility. Ferrite shows a very low solubility for carbon and nitrogen. The ferritic steel, AISI Type The role of oxygen is very important and yet 446, is used for high-temperature applications not fully understood. It is generally believed that and Types 430 and 434 for corrosion applications, oxygen is essential for SCC of austenitic steels such as automotive trim. in chloride medium. Oxygen reduction is the primary cathodic reaction in the SCC process. SCC can also occur in solutions where hydrogen undergoes the cathodic reduction. It is also (a) Stress Corrosion Cracking of argued that it is not the presence of oxygen Ferritic Steels in Chloride Media but the presence of a species which can be reduced which is important. The oxygen con- Ferritic steels, Types 430 and 434, are resistant tent in boiling MgCi2 test is estimated to be to SCC in MgCl at 140°C. High purity ferritic 2 0.3 ppm [93]. stainless steels are subjected to SCC in boiling 30% sodium hydroxide in tests exceeding 1000 h and in 42% MgCLz in a sensitized condition. (g) Effect of Anions and Cations Types 430 and 446 stainless steels are subject to It is known from the experimental work that chloride SCC in the welded conditions. In the chlorides, bromides and fluorides cause SCC presence of high residual levels of copper (0.37%) and nickel (1.5%), the alloys become susceptible to SCC in 42% MgCl2 solution. Conventional ferritic stainless steels containing the same amount of carbon and chromium Table 4.12 Pitting potential of steels as austenitic steels undergo sensitization more in 1000 ppm chloride solution rapidly than the austenitic steels. The ferritic steels generally become sensitized when quenched Steel type Potential (V) vs SCE from 927-1149 °C. Sensitization is eliminated or minimized by annealing between 732 and -0.05 430 843°C for a sufficient length of time. If the steel -0.22 304 is held for a sufficiently long period of time, -0.33 316 chromium diffuses back to the grain boundaries -0.95 216 and sensitization is eliminated. 2 10 Principles of Corrosion Engineering and Corrosion Control STRESS CORROSION that in aerated conditions, pitting occurs as the corrosion potential is noble to (higher than) the pitting potential of the steel. Pitting is facilitated as the anodic current density is increased. If a cathodic current is applied, it is observed that there is an increase in the time of failure. It is also observed from the figure that if the corrosion potential (Ec) is more active (lower) than pitting potential (£p), pitting is not likely to occur. The only cathodic reaction is the reduction of water which lowers the pH. At cathodic current densities higher than 25 |xA/cm2, however, hydrogen evolution is the major cathodic reaction and the time to failure decreases because of the absorption of hydrogen. There is a great deal of evidence to show that cathodically generated hydrogen can cause embrittlement of high strength steel. In a high strength, low-alloyed steel, with a martensitic structure, tempering at 400°C eliminates the susceptibility to SCC by hydrogen embrittlement. The best method to distinguish between SCC and hydrogen embrittlement or hydrogen induced cracking (HIC) is by cathodic polarization. 4.8.25 CRACKING OF MARTENSITIC STEELS It is possible to obtain austenite at elevated temperature with low chromium and relatively high chromium content. Fast cooling of austenite transforms it into martensite which has body centered tetragonal structure. Martensite is strong and brittle and it can be tempered to a desired level of toughness and strength. The corrosion resistance of martensite is, however, not as high as that of ferrite and austenitic steels. Martensitic steels are generally susceptible to SCC in a wide range of environments. The martensite steels, such as grade AISI 400 types, are susceptible to hydrogen embrittlement when stressed or exposed to sulfide or chloride environments. The specific anions to cause failure are not as important as hydrogen. Figure 4.53 shows the polarization curve for a modified 12% Cr martensitic steel in 3% NaCl at 25°C [95,96]. It is observed in the figure 0 -0.2 1 ! >Ecorr(AIr) 1 1 1 1 1 1 1 1 Air saturated n -0,4 to -0.6 4 S'"~ i E^r (Helium) ! ! 1 Current, pA/cm2 1 - - B> -0,8 •1.0 -L2 -1.4 1 1 1 1 1000 100 Anodic 10 1 10 Cathodic 100 1000 Figure 4.53 Polarization curves for a modified 12% Cr martensitic stainless steel in 3% NaCl at 25°C. (B.E. Wilde, The theory of stress corrosion cracking in alloys, NATO Scientific Committee Research Evaluation Conference, 1971. Reproduced by kind permission of British Corrosion Journal) Types of corrosion: materials and environments By increasing cathodic polarization, SCC i prevented but hydrogen cracking is accelerated. 2 11 100psi (7kg/cm2), atomic hydrogen reacts with the carbon component in the steel to form methane. Fe3C + 4H -> 3Fe + CH4 (4.50) 4 .8.26 H Y D R O G E N D A M A G E (HIGH TEMPERATURE H Y D R O G E N ATTACK) The removal of carbon as shown by equation (4.50) causes a loss of strength. Accumulation of methane inside builds up high internal The role of hydrogen in the embrittlement of pressure inside the steels and createsfissuresprefmartensitic steel has already been discussed above. erentially at the grain boundary or non-metallic However, embrittlement is not the only way in inclusions. Since neither molecular hydrogen nor which materials are damaged by hydrogen. Steels methane is capable of diffusion through the steel, are also damaged by hydrogen blistering at high so these gases accumulate. Therefore, hydrogen temperatures. Thus, there are three categories of attack at high temperature results mainly from hydrogen damage (Fig. 4.54) the formation of fissures by methane and by decarburization of steel. The steel after hydro(a) High temperature hydrogen attack (hydro- gen attack may also be found to contain blisters in addition to the fissures. These blisters, howgen damage) (b) Hydrogen blistering ever, differ from the low temperature blisters (c) Hydrogen embrittlement. in that they contain methane (CH4) instead of hydrogen. A comparison of three types of attack is shown Use of stabilized grades of steel is one prevenin Table 4.13. tive method. Prevention of blisters is difficult to avoid by coatings or linings because of the high permeability of hydrogen. (a) High Temperature Hydrogen Attack This type of attack requires the presence of atomic hydrogen because of the inability of the molecular hydrogen to permeate steel at atmospheric temperatures. At temperatures above 230°C and hydrogen partial pressure above (b) Hydrogen Blistering (Hydrogen Induced Cracking) This is caused by the atomic hydrogen diffusing into a steel and being trapped at a non-metallic inclusion or at a grain boundary to produce Cathode Figure 4.54 Illustration of Hydrogen diffusion 212 Principles of Corrosion Engineering and Corrosion Control CO *03 £ C/5 Ul <-M *G O ning uate teel), eatment ediu < u 4-» •S o *G bo G CO *G t* OH gen bli G C/5 03 3 fl J-H o 03 co 03 W> 0 3 CD S '5 'IJS co 3 C> L bG G G If « ^-J G . Prot ctive . Use (ant HIC . Cheimical corr sive poly ulfid ame a hyd « ^G U< 13 03 <+H ao O CD C/5 V-i CD 3 o3 CO o J -i o 13 O 13 03 <D . <*> 03 ' ^ fa £< D O • j-< 'Si" &2 < pH <—i o c/5 CO (N CO c/o G 13 *£ C <+-, D CD 13 G <« ^ 03 V-i 13 fl I M2 G£ ^ > U 11 ^ CO CO low and cture deep 83 CD O 03 .. '^ O Ui C D o3 ^ cj C ^ o3 CO CO <-l—1 C VM D O +-> 03 CO * £ CD C/5 CD Jl CO CD CO CD L> 2 GM X\ u *c3 13 G£ >s 13 to X! G +-> defec ;vere 1 ducti strain delay jvere 1 M O fl fl « -^ "**! P _* fr bG bG C D 03 03 <+H +-> 03 CD c/5 03 o CO CO J -i 13 CD O & CD *-i a S tJ *C a 03 co ,5 C T3 D -M JH ^ f l M r ^ <D H G CD 13 2 G G O 03 • i—( b X) O •3 o3 CD O 3 U CD O CO CO 3 o o CD o 3Kt 1+ 3 G O GO •fl <-M CD ^ 3 IS X 3^ 13 x> o > G CD bJD G •9 03 C/5 o S as 1 .3 CO V-i &° O <! m A <D G o o o CH •S 6 ^ <N CD G CD ^S (N CO CO PG O s G 3X> > ° ^3 ^D c/5 . 2C CU U-i CD O 3o fl C D "^ C -rt O CO >^ o u MO C DO CM co -£ C D 1 U u 3G £ K 3 " o G g 'fl > J ? 3 <J G fl ^ i3 .JH ^ c^ o O V CD SG H 03 6C CD V HG^ CU 03 P H > s^3 O W) fl C D co 03 8 ^ bG'S .3 .2 G o3 2 ^ *S ^ ^ ^ P^ e o3 5> CD ^ bD fl ^G G ^3 O O C D bO G w J3 •• c/5 O U G H ^4 03 G CD 03 bX) g <-> G CD bo c G b0i3 G e CD O CD o CD bJD O Ui G: G C -G D bX)^ °y o 55 Types of corrosion: materials and environments molecular hydrogen. As a result of formation Corrosive environment of molecular hydrogen inside the non-metallic inclusions or grain boundaries, a high pressure is localized at the inclusions or grain boundaries until the bulging occurs, producing blisters or cracks. These cracks are parallel to the surface along the original laminations generated at various depths. These are finally connected together. Stepwise cracking occurs when short blisters at Atmosphere varying depths within the steel link together to A Blistering form a series of steps. A schematic of hydrogen 8 Straight crack induced cracking in steel is shown in Fig. 4.55. C Stsep-wise crack Details are illustrated in Figs. 4.56 and 4.57. Blistering is a very common name and it is often confused with high temperature hydrogen Figure 4.55 Hydrogen induced cracking attack described earlier. It is, therefore, recommended not to use the name 'blistering.' Hydrogen induced cracking (HIC) can be used as an alternate name to avoid confusion. 2 13 Interior Sour acid electrolyte Corroding metal surface Figure 4.56 Hydrogen blistering in the wall of a container 2 14 Principles of Corrosion Engineering and Corrosion Control Bettrctyte' y~V. f (it) T ® 1 .. i ' .(B^ Figure 4.57 Blister formation Conditions Mechanism of Hydrogen Formation For the HIC to occur, the following conditions The formation of atomic hydrogen is exemplified, for instance, by the following reactions: must occur: (a) The presence of water phase. (4.51) H2S -> 2H + + S 2 " (b) The presence of atomic hydrogen. (c) An agent that retards the formation of molec(4.52) Fe + 2H + -> Fe + + + 2H° ular hydrogen at the surface. (d) Presence of grain boundaries or inclusions. (4.53) 2HU H2 (e) Maintenance of an active surface. (f) Discontinuity in metal, such as slag, inclusion Such a reaction takes place in the presence of and/or void. hydrogen sulfide. Free hydrogen ions are produced at the cathode during the formation of iron Sources of Hydrogen sulfide scale as well as during the dissolution of this scale. In order to protect the high strength non-stainless steels against corrosion, a good surface treatment must be provided. Incidently, surface treatment Prevention like pickling or electroplating, are likely to introduce hydrogen in the steel. In high strength steels, (a) Changing the corrosive environment. a small amount of hydrogen may cause serious (b) Coating or lining. cracking. Hydrogen absorption may also take (c) Using steel resistance to hydrogen induces place during manufacture, fabrication, welding cracking, such as steels containing Cu or and heat treatment. cobalt. Types of corrosion: materials and environments 215 of hydrogen adsorption unless the strength of the remaining material is less than the load applied. Once hydrogen has been absorbed by a material, Later, instantaneous final fracture occurs. The its effect, regardless of the source from where it failure by hydrogen embrittlement is mostly interhas been absorbed, is the same. Gaseous hydrogen granular. The fractured surface has, therefore, a and hydrogen released from a cathodic reaction crystalline appearance. differ from each other in the following respects: (1) Cathodic hydrogen is adsorbed on the surface as atomic hydrogen (reduced), whereas gaseous hydrogen is adsorbed in the molecular form and it then dissociates to form atomic hydrogen. (2) The internal pressure produced by the gaseous hydrogen is much lower than produced by cathodic hydrogen, due to the log term in the Nernst equation which converts an E value into an exponent on the hydrogen pressure. Hydrogen embrittlement occurs during the plastic deformation of alloys in contact with hydrogen gas and is strain rate dependent. Alloys, like ferritic steel, nickel-base alloys, and titanium, show highest degradation in properties when the hydrogen pressure is very high and strain rate is low. Hydrogen embrittlement is a phenomenon whereby hydrogen is absorbed in the metal (diffuses), exerts local stresses, and leads to embrittlement of material, such as high strength steels. (c) Hydrogen Embrittlement Mechanism No definite mechanism of hydrogen embrittlement has been suggested. It has been considered sufficient to identify hydrogen as a cause for cracking. It is a general opinion that impurity segregations at the grain boundary act as poisons, and increase the adsorption of cathodic hydrogen at these sites. It is widely believed that in BCC materials hydrogen embrittlement is caused by the interaction of hydrogen with defects in the structure. Such defects are vacancies, dislocations, grain boundaries, interface, voids, etc. Hydrogen is trapped in these defects and growth of a crack is facilitated. A large number of such defects interact with hydrogen and the combined trapping results in a significant loss of ductility. Difference between SCC and Hydrogen Embrittlement It is possible to distinguish between SCC and hydrogen embrittlement by applied currents. If Examples of Embrittlement on applying a current, a specimen becomes more (1) In plating operations. anodic and cracking is accelerated, the attack (2) In pickling operations. is SCC, whereas, if cracking is accelerated in (3) In cleaning of high strength steels in chloride the opposite direction and hydrogen evolution is or fluoride solution. observed, the attack is hydrogen embrittlement. (4) Manufacturing and fabrication processes. The statement is, however, oversimplified. It has been discussed earlier. The following are the differences between Materials Most Susceptible SCC and hydrogen embrittlement from a fracIron, titanium, zirconium, martensitic steels, tographic point of view: high strength aluminum alloys. (1) SCC begins at the surface, whereas hydrogen embrittlement begins internally. Identification (2) The magnitude of corrosion is higher at the Hydrogen embrittlement results in a brittle fracorigin of SCC than observed with hydrogen ture throughout the embrittled material as a result embrittlement. 2 16 Principles of Corrosion Engineering and Corrosion Control Prevention of Hydrogen Embrittlement (1) Select materials resistant to hydrogen embrittlement for elevated temperature applications. (2) In pickling or plating operations, submit the material to low temperature aging at temperature between 161 and 370°C to eliminate the effects of hydrogen embrittlement. Chlorides and fluorides are known to cause hydrogen embrittlement. There are recommended concentrations of such solutions for use in processing so that damage may be avoided. Hydrogen embrittlement of titanium can be avoided if the ratio of HNO3/HF exceeds 10. Cleaning and plating solutions should be carefully screened to avoid hydrogen embrittlement problem, special care being needed when cleaning welding tints off welds. (3) Use of inhibitors. Inhibitors may be added to pickling baths to minimize corrosion. (4) Baking. It is recommended that high strength steel be subjected to baking at low temperature after plating. Temperatures between 200 and 300°F are generally employed. (5) Change of design. Avoid sharp corners, as they act as stress raisers. Also eliminate sites for crevice corrosion. (6) Heat treatment. Temper at a higher temperature for a long period of time. This would lower the strength. (7) Surface preparation technique. Use techniques which introduce compressive stresses on the surface. (8) Substituting alloys. Alloys with nickel or molybdenum reduce susceptibility but do not eliminate it. (9) Use low hydrogen welding rods. 4 .8.27 C A U S T I C C O R R O S I O N Austenitic nickel-chromium stainless steels and mild steel are subject to stress corrosion cracking in caustic soda (caustic cracking) at elevated temperatures. The phenomenon, 'caustic cracking' is mostly encountered in boilers. Caustic is added as an additive to boiler water in order to preserve the thin film of magnetic iron oxide by raising the pH. Caustic addition creates problems only when it becomes concentrated. It can become concentrated by one of the mechanisms summarized below. Four instants in the life of a steam bubble are shown in Fig. 4.58. (a) m ** w «* * * m * * * * * * (J > ««*•«*««*« **»•** 4 i, . , , , , » • . . * • «x ,\»t»A,X,X,l,XX l,.,>.,«,„l;„li.,«,.*„t,l„y,„K,ul„ft«lMtll»„»,„»lM»l,»„^ (a) Bubble nucleates (b) Bubble becomes saturated with brine (c) Cone brine left after evaporation (d) Rinsing removes corrosion deposits Figure 4.58 Four instants in the life of a steam bubble Types of corrosion: materials and environments 2 17 (a) Departure from Nucleate Boiling (DNB) [97] This refers to the condition in which bubbles of steam nucleate at points on the metal surface. As these bubbles form, minute concentrations of boiler water develop at the metal surface at the bubble/water interface. As the bubbles separate, the water redissolves sodium hydroxide. The rate of bubble formation exceeds the rinsing rate at the onset of the departure from nucleate boiling (DNB) and, hence, caustic as well as other solids begin to concentrate. Also, the caustic soda may react with iron to produce hydrogen: Fe+2NaOH = Na 2 Fe0 2 + H 2 (4.56) If atomic hydrogen is produced it diffuses to the metal at inclusions and grain boundaries to form molecular hydrogen, or alternatively F e 3 C + 4 H ^ C H 4 + 3Fe (4.57) It forms CH4 as shown in the equation. Molecular hydrogen cannot diffuse into steel and, therefore, it accumulates on the grain boundary. Due to enormous pressures exerted by (b) Under Deposits hydrogen, discontinuous microcracks are formed in the grain boundaries. The metal strength conMany times steam is formed under the insulating tinues to decrease by the pressure by methane or deposits, which leaves a rich caustic residue. hydrogen until damage occurs. A steel tube in a boiler would burst under the pressure. (c) Evaporation at Waterline It is possible also for the caustic to concentrate at the waterline. Generally, the waterline area is always most sensitive to corrosion. Two types of failures are common in boilers and both are related to the effect of concentration of caustic. One which forms discontinuous microcracks and results in the bursting of tubes is called 'hydrogen damage,' and the other which results in the formation of continuous microcracks leading to intergranular corrosion is called 'caustic cracking.' Both are briefly discussed below: (2) Caustic Cracking In the presence of sufficient tensile stress and traces of silicon, hot caustic solutions can induce SCC of boiler steels. This phenomenon is not called 'caustic embrittlement,' as no loss of ductility occurs in caustic cracking. Tensile stress and caustic concentration cause the formation of continuous intergranular cracks in the metal. As the cracks progress, the strength of the metal is exceeded and fracture occurs. Improved water treatment practices and improved boiler designs have minimized the problem. (1) Mechanism of Hydrogen Damage by Concentrated Caustic Solutions Caustic solution acts by dissolving the magnetic iron oxide 4NaOH+Fe 3 0 4 = 2NaFe02 + Na 2 Fe0 2 + 2 H 2 0 (4.54) The protective coating of magnetic iron oxide is thus destroyed. After the coating is destroyed, water reacts with iron to evolve hydrogen: 3Fe+4H 2 0 = Fe 3 0 4 +4H 2 (4.55) 4 .8.28 SULFIDE STRESS CORROSION CRACKING This phenomenon is generally encountered in high strength steels of Rockwell hardness above 22 in a sour oil field environment. It is a special case of hydrogen stress cracking, also called hydrogen embrittlement. Factors responsible for sulfide stress corrosion cracking (SCC) are: (1) Notches, pits, irregularities, inclusions on the metal surface. 2 18 Principles of Corrosion Engineering and Corrosion Control (4) Microstructure. It varies from one alloy system to another but it has a direct influence on SCC. Mechanism The following steps lead to the formation of FeS (ferrous sulfide), which causes SCC (Fig. 4.59): Anode: Fe-> Fe 2+ + 2e (4.58) (2) Hydrogen diffusion. (3) Tensile stress. (4) Environment. H2S concentration. The partial pressure of H2S for a 80 ksi yield strength stress is 0.01 atm, and 0.0001 atm for a 130 ksi strength steel. (5) pH below 6.0. (6) Chloride has a significant effect on 12% Cr steels and little effect on low alloy steels. (7) Highest sensitivity is observed at 20°C. Metallurgical factors Cathode: H 2 S+H 2 0 UH+ + H S " + H 2 0 (4.59a) HS~ + 2 H 2 0 <> 2 H + + HS~ + H 2 0 = (4.59b) (1) Yield strength. Resistance to SCC decreases with increased strength. (2) Cold work. It decreases the resistance to SCC. (3) Hardness. Susceptibility to SCC increases with hardness. Combination: 2 e " + 2 H + + F e 2 + + S2~ ±=;2H°+FeS (4.60) Net: F e + H 2 S ^ l F e S + 2 H 0 (4.61) Figure 4.59 Mechanism for sulfide stress corrosion cracking Types of corrosion: materials and environments Prevention (1) (2) (3) (4) Use alloys less susceptible to SCC. Use steels of HRC below 22. Avoid the sensitizing temperature. Minimize tensile stresses in the system. 2 19 Environment Shell side - hot waste fuel. Tube side - cooling water. Material Type 316 stainless steel U-bend exchanger tube. Temperature of tube side. Service life is 5 years at 21°C. Cause of failure Failure analysis showed that 316 SS heat exchanger tubes suffered chloride stress corrosion cracking in straight sections as HISTORIES well as U-bend of heat exchanger tubes. The root cause was attributed to a blockage in cooling A. Case 1 water which activated the hot oil at 21 °C to raise the temperature of the water containing about In 1985, the concrete roof of a swimming pool 800 ppm chloride. The problem was resolved by collapsed. The roof was supported by stainless free circulation of the cooling water. steel rods. 4.8.29 F A I L U R E C A S E Environment Chlorine-based disinfectants were used in the pool. The temperature was maintained C. Case 3 at about 1 ° C above water temperature. High levels of humidity were observed. Chloride stress corrosion cracking of a 304L SS, 28" diam buried pipeline carrying CO2 feedstock, Cause Stress corrosion cracking was established to be a major cause. Chlorine-containing com- after seven years of service life. pounds were transferred via the pool atmosphere Environment The pipeline ran parallel to a roadto surface away from the pool itself resulting way which was frequently de-iced with salts durin the production of a highly corrosive film ing winter period. The adjacent soil of the pipe on the stainless steel rods which supported the contained 150-260 ppm of chloride. roof structure. The failure occurred around 30° C in highly stressed components which were not Cause A metallurgical history of the pipe showed frequently cleaned. Such components included that it was exposed to tensile stresses, thermal stresses and residual stresses from weldment. The brackets, rods, bars, wire ropes, fasteners, etc. temperature reached about 45° C. Surface topogPrevention The following remedies were recom- raphy suggested intergranular cracking. mended by the consultants: Prevention SCC occurred because of the leaching of chlorides from the adjacent soil which was (1) A better design to resist localized corrosion de-iced by salt in winters. The ingress of chloride and elimination of possible side of crevices. may be stopped either by some insulation or by (2) A rigorous inspection. changing the location of the pipe well away from (3) Effective management including mainte- the roadway. nance. (4) Reduction of temperature around highly stressed components. D. Case 4 (5) Regular cleaning of stressed components. (6) Recirculation of pool. Failure of landing gears. (Courtesy: S. /. Keccham, (7) Lower humidity and chloride buildup. Naval Development Center, Westminister, PA, USA.) B. Case 2 A main landing gear shock strut piston assembly failed catastrophically while on the Stress corrosion cracking of a U-bend waste fuel ground, separating both axle stubs and lower portion of the shock strut piston from the gear. heat exchanger tube. 2 20 Principles of Corrosion Engineering and Corrosion Control The aircraft had 559 flight hours and no records of 4 .8.30 P R E V E N T I O N O F S C C overweight. Material The steel used was air-melt 4340 and The following general methods are recomnot vacuum melt. mended to minimize stress corrosion cracking of austenitic steels: Observations The cracks near the origin indicated the failure to be caused by intergranular (1) Eliminate chloride salts as far as possible in cracking or hydrogen embrittlement followed by acid environments. fracture. The axle bore showed active pitting. (2) Remove oxygen or use oxygen scavengers The failure occurred by intergranular stress corin the case of chloride-containing environrosion cracking. ment. (3) Convert the tensile stresses to compressive Recommendations It was recommended to use stresses by shot peening. vacuum melt steel. Air-melt steel is more susceptible to stress corrosion cracking than vacuum melt (4) Avoid concentration of chloride or hydroxide by proper equipment design. Allow steel. To avoid hydrogen embrittlement, cadu complete drainage in equipment. Prevent mium plating in vacuum { ion plating' vacuum designs that would concentrate chlorides or equipment) was also recommended. caustics by vaporization. (5) Reduce stresses by stress relief of the steel. The stress relief of non-stabilized E. Case 5 grades or low carbon stainless steels renders them more susceptible to intergranFailure of 17-4 PH Bolts. {Courtesy: James A. ular cracking than if they are not heat Stanley, Proceedings of the First Joint Aerospace treated. and Marine Corrosion Technology Seminar, (6) Adopt a good design practice. Highly stresLos Angeles, CA, p. 30, 1960.) sed areas can be minimized by adopting a On a launch vehicle, high strength 17-4 PH good design practice. For instance, forged steel bolts were used on an aluminum body oxielbows are preferred over mitreed elbows, dizer valve. This martensite steel was heat treated and forged tees are preferred over direct (H-900 condition) to a minimum strength of stub-in joints. Avoid poor welding work190 000psi. On exposure to marine atmosphere manship. at Cape Kennedy, the bolts were found to be (7) Select a more crack-resistant alloy if other cracking. preventive measures fail to work. Use high nickel alloys or alloys containing very low Observations The failure was not caused by levels of nitrogen and other impurities, if SCC, shown by laboratory tests, but by hydropresent. For instance, purified 16 Ni-20 Cr gen induced cracking (HIC). The bolts were used is not susceptible to cracking. Choose one to hold down aluminum forging of alloy 7075-T6. of the lower nickel duplex stainless steels or Because of galvanic corrosion, due to contact of a ferritic steel. steel with aluminum, hydrogen was being released (8) Use coatings to provide a barrier between into steel, which caused HIC. the steel and the environment. Coatings Recommendations have been successfully used to prevent access of moisture on the outside of stain(1) Protection by over-aging. Over-aging at 51 ° C less steel lines, operating at 49-260° C for one hour was recommended to eliminate in the chloride-laden atmosphere of the HIC. Arabian Gulf. (2) Vacuum deposited coating to a thickness of (9) Use inhibitors. High concentration of phosnot less than 0.5 mil. phate have been successfully used. (3) Anodizing of aluminum alloy 7075-T6 which (10) Cathodic protection. Impressed current is in contact with steel. cathodic protection system has been (4) Application of lubricants, greases and paints. Types of corrosion: materials and environments successfully used to prevent SCC of steels. This system has been used to protect steel in seawater. (11) Lower the temperature if service conditions permit. For instance, keep below the critical temperature (85°C) for SCC to occur in chloride media. Whenever possible, maintain high velocity to prevent any deposition of debris. (12) Modify the environment, if possible, by changing pH or reducing oxygen content as in heat exchangers or fluids. In addition, the following measures may be taken to prevent SCC of martensitic, or ferritic steels. 2 21 of mechanical failure in aircraft have been investigated over the past fifty years [98] by Royal Aerospace Establishment. Fatigue is indicated to be the most frequent of the failure types compared to other modes. Fatigue failure is a major mode of failure of components in automobiles, transport, petroleum, petrochemical, shipping and construction industries. Corrosion fatigue is a process in which a metal fractures by fatigue prematurely under conditions of simultaneous corrosion and repeated cyclic loading at lower stress levels than would be otherwise required in the absence of a corrosive environment. Metals and alloys will crack in the absence of corrosion if they are subject to high cyclic stress for a number of cycles. The num(a) Do not allow hydrogen to be gener- ber of cycles for failure decreases as the stress is ated by any source as hydrogen in these increased. Below a certain stress the metal will steels cause hydrogen damage. last indefinitely. This level is termed as 'Endurance (b) If cathodic protection is being applied Limit' of the material. take care that no over-protection If, however, the material under cyclic stress is occurs, as a slight generation of hydro- subject to a corrosive environment, the endurance gen would lead to hydrogen damage. limit of the material is sharply reduced. The pre(c) Reduce the hardness of martensitic or mature failure of a material from the exposure to precipitation hardened steels to values the combined action of corrosion and cyclic stress lower than Rockwell C-400. is called 'Corrosion Fatigue.' (d) Take precautions against nitrogen or This type of failure is generally encouncarbon pick-up during mill process- tered in pump shafts, heat exchanger tubes, ing or during the welding of fabricated rotors, steam turbine blades, aircraft wheels, equipment. boiler and steel equipment. Some major examples where corrosion fatigue is encountered are given All of the methods given are not ideal below: and they must be used judiciously tak(1) Ships ing into consideration the past experience, (2) Offshore platforms materials compatibility, environments and (3) Drilling rigs the service life. (4) Navigation tuners (13) Reduce tensile stresses by shot peening. (5) Aircrafts Shot peening introduces surface compres(6) Submarine pipelines sive stresses. Shot peening counter-balances (7) Communication equipments tensile stresses. (8) Heat exchanger tubes (9) Pump shafts (10) Alternate dry and wet zones. 4 . 9 C O R R O S I O N FATIGUE INTRODUCTION 4 .9.1 4 .9.2 CHARACTERISTICS OF Fatigue failures are as old as the Industrial Revolution and they have been responsible for some of the major engineering catastrophes, for example the Comet aircraft in the past. Mechanisms C O R R O S I O N FATIGUE (a) The main distinguishing feature of corrosion fatigue cracks is the presence of several cracks 2 22 Principles of Corrosion Engineering and Corrosion Control near the fracture. In a normal fatigue failure, usually one crack is present. The corrosion fatigue cracks run parallel to each other and are aligned perpendicular to the direction of principal stress. Corrosion fatigue cracks in carbon steels often propagate generally from the base of corrosion pits. Generally, the fatigue corrosion cracks in certain steels are transgranular in nature (Fig. 4.60). The striations formed in corrosion fatigue failure are less pronounced than in normal fatigue failure, as shown by electron microscopy. Fatigue striations are shown in Fig. 4.61. The cracks generally initiate at the surface where the stress is maximum. The fractured surface is dull in appearance and may contain corrosion products. A good indication of the effect of corrosion on the fatigue strength of materials is obtained by the conventional S-N curves (stress vs number of cycles to failure). A corrosive environment promotes crack initiation and shortens the fatigue life. Figure 4.62 shows a typical S-N fatigue curve. The endurance limit is represented by the asymptotic portion of curve. It is defined as the stress below which no failure will occur for a given number of cycles. The effect of corrosive environment on the fatigue life is clearly shown in Fig. 4.63. It is observed from thefigurethat fatigue life is drastically reduced in the environments shown. It is also observed that the fatigue life in 0.1M NaCl is much less than in distilled water. Most important, no endurance limit is shown by the S-N curves in either distilled water or in NaCl. The variables affecting corrosion fatigue may be classified into three categories: (a) mechanical (b) (c) (d) (e) (f) (g) Figure 4.60 Transgranular cracks in corrosion fatigue Types of corrosion: materials and environments (b) metallurgical and (c) environmental, as shown below: 4 .9.3 223 EFFECT OF VARIABLES ON FATIGUE A N D CORROSION FATIGUE Variables in Corrosion Fatigue A. Mechanical Factors n Mechanical 1 Metallurgical ' Environmental 1. Stress Intensity This factor is of a fundamental importance in the prediction of brittle fracture using linear fracture mechanics (LEFM) principles. It is a function of both crack geometry and associated loading. The analytical expression given below can be used to express the relationship of the stress intensity factor to crack growth: -^- = C{AK)n dN I I Stress intensity ' I Cyclic load frequency ' I Stress ratio •—i Heat treatment ' Fracture toughness Microstructure Composition I Temperature 1 Species 1 pH Potential (4.62) where AK is stress intensity factor range (KmdLX - Km[n)y C and n are constants, the value of n varies between 2 and 4, depending on temperature, environment and frequency. The above equation is called Paris equation. The constants are called Paris constants. They can be found from Table 4.14. An expression similar to equation (4.63) can be applied to fatigue data for a range of materials in various environments. A schematic diagram of stress intensity, AK vs fatigue crack growth rate (da/dN) is shown in Fig. 4.64. The exponent varies with the environments. The corrosion fatigue crack propagates in both inert and aggressive environments as shown. (a) In some cases, the presence of air and environment causes a da/dN vs A Kplot to move to a higher da/dN value at a given AK (b) In some cases, corrosion fatigue increases da/dN more at low AKthan at high AX, so there is a reduction of exponent n in the Paris equation. (c) In the intermediate Arrange, the Paris law is approximated [98]. In this region, it is reported that the growth of fatigue cracks is Figure 4.61 Fatigue striation in a 7178 aluminum by ductile striation which is associated with alloy. (From Pelloux, M.N. (1965). In Metals Engineercorrosion fatigue. ing Quarterly, ASM, November) 2 24 Principles of Corrosion Engineering and Corrosion Control i £ For stress values above this line, failure occurs SD Number of cycles (A) Steel Number of cycles (B) Aluminum Figure 4.62 Typical S-N curves Types of corrosion: materials and environments 2 25 500 400 r^— ^ ^ ^C^N^ ~» : ENVIRONMENT DISTILLED WATER \ •....v..,,,.... \ 300 Stress Amplitude, S (MN/sqm) 200 i ^ \_ \ vYx 1 % AQUEOUS NaCI SOLUTION s\ A Y \A 100 y \ /\A,m^^mm^^ I ID3 I 105 I i 10T i i 1® 0 Number of cycles to failure, N Figure 4.63 S-N fatigue curves for Cr steel in distilled water and 1% aqueous NaCI solution A schematic representation of fatigue crack growth is shown in Fig. 4.64. It has basically three regions: Region A. The crack growth rates are very low. The region begins with a threshold value of stress intensity, below which crack propagation does not occur. This region continues until the slope becomes constant. Region B. Shows a linear relationship between log AK and log da/dN. Equation (4.62) shows a power law dependence. It can be expressed as On simplification it yields da log — = rclog AK+log C aN (4.64) i og^Uio g [C(A*:n (4.63) Equation (4.64) shows a straight line relationship between (da/dN) and AiC, with slope n and intercepts C. Region B shows the steady-state rate of crack growth. The value of n is 3 for steels and in the range of 3-4 for aluminum alloys. Region C. It exhibits a steep gradient and the crack growth rate accelerates. In Region C, the value of Kmax is attained. Region B is the most important region 2 26 Principles of Corrosion Engineering and Corrosion Control Table 4.14 Alloy 5086-H116 Paris constant and Paris exponent values Environment Air Seawater, free corrosion Seawater-0.75 V Seawater- 1.3 V Seawater- 1.4 V Air Seawater, free corrosion Seawater-0.75 V Seawater- 1.3 V Seawater- 1.3 V Seawater- 1.4 V Air Air Seawater, free corrosion Seawater - 0.75 V Seawater- 1.3 V Seawater- 1.3 V Seawater- 1.4 V Air Seawater, free corrosion Seawater-0.75 V Seawater- 1.3 V Seawater- 1.4 V Seawater- 1.5 V Air Seawater - 0.75 V logioC -10.75 -10.47 -10.09 -10.82 -11.13 -10.78 - 9.09 -9.55 - 10.50 -11.43 -11.47 -10.67 -10.90 -11.16 -10.95 -10.77 -10.95 -13.96 -11.01 - 9.85 -9.43 - 11.81 -15.02 -15.66 - 9.91 - 12.91 - 9.91 - 12.91 - 11.3 -9.91 - 12.91 -10.07 -10.00 -10.00 - 9.95 -9.70 -9.65 -9.60 n 3.53 3.41 3.16 3.69 3.76 3.53 2.26 3.05 3.30 3.95 3.95 3.53 3.68 4.20 3.79 4.25 3.70 5.10 3.69 2.97 2.74 4.11 6.31 6.86 3 .0(AK<28MNm" 3 - 2 ) 3.0(AK<28MNm- 3 ' 2 ) 3.0(AK<28MNm~ 3 - 2 ) 3.0(AK<28MNm- 3 - 2 ) 4.4 3.0 5.0 2.73 2.73 2.73 2.73 2.73 2.73 2.73 (Con 5086-H117 5456-H116 5456-H117 5456-H116 Seawater-0.95 V Seawater- 1.3 V 2219-T851 Argon H20 H20 H20 H20 H20 H20 Types of corrosion: materials and environments Table 4.14 Alloy H Y130,10 N i - C r - M o - V 10 N i - C r - M o - V 10 N i - C r - M o - V 12 Ni-5 Cr-3 Mo 4340 2 27 (Contd) Environment Air 3 % N aCl 3 % N aCl 3 % N aCl Air 3 % N aCl 3 % N aCl Air 3% 3% 3% 3% logioC -9.58 -9.38 -9.24 -9.11 -9.49 -9.10 -8.52 -9.58 -9.44 -9.39 -9.08 -9.59 -10.96 -9.51 -8.74 -9.82 -9.70 -9.56 -9.35 -9.12 -9.0 -8.35 -11.86 -11.42 -12.41 2 2 2 2 2.56 2.56 2.56 2 2 2 2 2 2.8 2.02 1.7 2 2 2 2 2 2 2 3.31 3.18 3.94 n 2 Ni-5 C r - 5 Mo (maraging steel) N aCl, N aCl, N aCl, N aCl, 10 Hz 1 Hz 0.1 Hz 1 Hz square wave AISI 4145 Cr-Mo steel Air 15 Hz 0.1NH2SO41.7Hz 0.1NH2SO40.7Hz Dry argon, 20 Hz Water vapor 10 Hz Water vapor 4 Hz Water vapor 2 Hz Water vapor 1 Hz Water vapor 0.5 Hz Water vapor 0.1 Hz Air H 2, 5 p si, R = 0 H 2, 5 p si, R = 0.4 AISI 4340 3 .5% N i - C r - M o - V for evaluating the life of engineering structures. Materials which are affected by environment are characterized by high crack growth rate. The effect of environment is generally to enhance crack growth rates. increase in the stress ratio (Fig. 4.65). R — ^ min l^rr (4.65) 2. Stress Ratio Stress ratio is defined as the ratio of minimum stress to maximum stress. The rate of corrosion fatigue crack propagation is increased by an The situation is rather complex. A higher load ratio results in a higher mean load, which causes an accelerated interaction between the environment and the crack tip. This would, however, vary from one material system to another. A uniaxial stress seriously shortens the corrosion fatigue life. The extension of corrosion fatigue relative to dry air increased from a factor of four at R (0.11 to 0.24) to as much as 20-30 fold at high 228 Principles of Corrosion Engineering and Corrosion Control REGION A Log dA 6U REGION C Log (AK>h Log(AK) Figure 4.64 A schematic diagram of fatigue crack growth Log- R (0.61 to 0.71). R is specified to quantify the amount by which a mean stress value is removed from zero. Aluminum alloys are sensitive to variations of R, w hereas its effect on steel is limited. A negative value of !Cmin i ndicates closure of the tip. If Km[n is negative, JR = AK = !Cmax> R = 0. 3. Cyclic Stress T he cyclic load frequency seriously affects corrosion fatigue. The service life of a component depends upon: (a) Number of stress cycles required to initiate cracks. (b) The number of stress cycles required for cracks to grow to a critical size before final fracture occurs. Local stress is not important in defining the conditions of crack growth. However, by defining Log(AK) Figure 4.65 R ratio Effect of crack growth rate on varying Types of corrosion: materials and environments it in terms of stress intensity factor, it is possible to predict how much time will be taken for a crack of a sub-critical size to reach a critical value before fracture occurs. Interaction between the material and environment depends on the rate of transfer of the environmental species to the metal. The time available for interaction is, therefore, important. The more is the time available, the greater the interaction. If sufficient time is available, the corrosion species will have sufficient time to reach the diffusion boundary layer, and cause failure by corrosion fatigue. If the time available is short, due to faster loading rates, sufficient time would not be available to the species to reach the boundary layer and, hence, the damage would be purely mechanical in nature. 229 increases because the crack tip remains in contact with the environment and also because of interaction with a larger volume of environment. 7. Stress Waveshapes The effect of environment is affected by the shape of the cyclic stress. Exposure of the metal to an environment at the peak stress accelerates corrosion fatigue failure. The effect of waveform is important on the crack growth. Certain waveforms have a more pronounced effect than others. The effect of triangular, square, saw tooth and trapezoidal waveforms on corrosion fatigue has been demonstrated. It has been generally observed that maximum effect occurs with sinusoidal, triangular and positive saw tooth waveforms. The square and negative saw tooth wave have no effect, because of very short rise time [99]. Different types of loading cycles are shown in Fig. 4.66. It is not possible to generalize that the effect of load waveform on corrosion fatigue at this stage, as the phenomenon is not clearly understood because of lack of corrosion fatigue data. In the case of stainless steels, the kinetics of reaction depends largely on the rupture of the oxide film on it, which is controlled by the strain rate. The effect of stress waveshapes is, however, not clear and more investigations are required to establish the effect of shape of the stress waves on corrosion fatigue. The effect of waveform is particular to a specific material environment system. 4. Stress Amplitude In general, a low amplitude of cyclic stress favors a greater contact of the environment with the metal and favors a longer fatigue life. In case of high amplitude, there is hardly an environmental interaction. In case of high frequency, there may not be an interaction between the metal and the environment. 5. Effect of Strain Rate New surface continues to be generated as the fatigue crack continues to grow, and the strain rate at the tip of the growing crack is high. The surface at the tip is influenced by passivation, adsorption and anodic dissolution. At low frequency, the corrosive environment is in contact for a long time period to allow the crack tip- 8. Stress Corrosion Cracking environment interaction. As discussed earlier, at Stress corrosion cracking can also occur in a high frequency, sufficient time is not available for system susceptible to stress corrosion cracking crack tip-environment interaction. above Kiscc- The adverse effect of environment takes place at loads below K\scc only during the increasing load portion of the stress 6. Mean Stress cycle. Crack growth is cycle dependent. Crack growth propagation occurs below the threshIf the mean tensile stress is high, the level of stress intensity at the crack tip increases and old for the time-dependent stress corrosion. The the crack remains open for a longer period of time-dependent crack growth can be attributed time. Also the crack has a wider opening under to the constant or the rising portion of load this condition. The corrosion fatigue damage cycle. 2 30 Principles of Corrosion Engineering and Corrosion Control Sinusoidal Positive saw-toothed load Negative saw-toothed load Triangular load Square load Reduced minimum load Reduced maximum load Figure 4.66 Different types of loading cycles. (Barsom, M.J. (1971). In Corrosion Fatigue, Int. Conf., June 14-18, NACE) 9. Environmental Factors The aggressivity of the environment is intimately related to fatigue life. Increased concentration of a corrodant generally decreases the corrosion fatigue resistance of metals and alloys. For instance, in seawater, chemical, physical and biological factors affect the resistance of materials to corrosion. The effect of environment is shown in Fig. 4.67. It is to be observed that the corrosion fatigue limit in salt water is lower (30 ksi) than in freshwater (40ksi). The corrosion fatigue resistance of high strength alloys is also affected by relative humidity and condensation conditions in the environment. 10. Temperature Temperature affects the rate of fatigue crack propagation and, hence, the fatigue life. Increase of temperature increases the rate of transport of the active species to the tip of the crack, and accelerates the propagation of the crack. It also increases the corrosion process by lowering the hydrogen over-voltage. The fatigue crack growth rate of a high strength steel in 3.5% NaCl increases with a rise in the temperature of sodium chloride. 11. Electrode Potential Application of a constant electrode potential can either decrease or increase corrosion fatigue. Types of corrosion: materials and environments 231 a series of constant potentials by a potentiostat. The fatigue life increased as the applied potential in the active direction (cathodic) increased as shown by failure tests (Fig. 4.68). At an applied stress of 53 x 103 psi below the fatigue limit failure did not occur within 107 cycles below -0.48 V(SHE). At 65 x 103 psi, maximum life was observed below -0.48 V(SHE). It shows that the protection potential in either case does not change with the applied stress. For specimens with a higher hardness value, Rc 52, tests at an applied stress of 141 ksi showed increased life for more active potentials, but the maximum life did not reach more than 107 cycles (Fig. 4.69). A maximum life of 7 x 106 cycles was observed at more active potentials from —0.55 to —0.85 V. ¥f 10* Kf The life is almost 100 times the life at corrosion CYCLES TO FAILURE potential. Hydrogen is, however, evolved at more Fatigue life of 4140 steel Re 5 2, i n moist and active value, which causes damage by hydrogen dry air and in aerated 3 % N ad, 2S*C embrittlement and shortens the life. In conclusion, cathodic protection is effective in 3% NaCl Figure 4.67 Effect of environment on corrosion for steels between hardness values of Rc 20 and fatigue. (From Lee, H.H. and Uhlig, H.H. (1972). Met. Rc 44 in 3% NaCl. Trans, 3, 351, November) Anodic polarization of steels, aluminum alloys and copper alloys, decrease their resistance to fatigue crack initiation, whereas cathodic polarization increases the resistance. Whether the corrosion fatigue would be accelerated or slowed down would depend on the METAL ENVIRONMENT environment, material (passive or active) and 4 .9.4 the direction in which the potential is applied I N T E R A C T I O N (active or noble). Generally, fatigue life increases with more active potential, however, the poten- When a metal is subjected to continuously applied tial applied to achieve cathodic protection is not cyclic stress, the extent to which fatigue crack sensitive to applied stress. The crack propagation initiation and propagation is influenced would rates are sensitive to changes in potential. The depend on: potential changes can either decrease or increase the crack propagation rates. The increase or (1) The thermodynamic tendency of the metal decrease would depend on the applied potential and the environment to react. in the negative or positive direction (cathodic or (2) The reaction kinetics, the type of reaction and anodic), the nature of reactive species in the envireaction produced. ronments, and the mechanism of interaction with the specimen. In aqueous systems, the following processes may Moderate cathodic protection can improve take place: a corrosion fatigue performance of low strength steels. In the case of high strength alloy, cathodic (a) Adsorption of atoms and molecules on the protection may accelerate fatigue crack initiasurface of the atom. tion. Steel specimens, AISI 4040 (Rc 20), in 3% (b) Film formation at the metal-electrolyte NaCl were stressed 10% above and below the interface. fatigue limit. The specimens were maintained at (c) Dissolution of the metal. 232 Principles of Corrosion Engineering and Corrosion Control -0.3 -0.4 Corrosion potential x tn -0.5 g -0,6 I -0.7 £ -0.8 65 ksi -0.9 -1,0 > I I M l ! J I 10s > i II Mi 10s J i i Ii W Number of cycles to failure Figure 4.68 Effect of applied potential on 414 steel Rc 20 stressed above and below the fatigue limit, on fatigue life in aerated 3% NaCl. (From Lee. H.H. and Uhlig, H.H. (1972). Met. Trans. ASM, 3, November, 2949-2957. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) A Aerated 3 %NaCI II -0,4 £ -0.6 - A. , A JL Aerated 0.5N Na 2 SoJ ^*^\ A § -0.8 m *4mJ 1 -i.o A -1.2 E c A JL t^^"" 1 l l1 Mil S A — •1.4 ~1T <& 1 1 JL M i l lJU 10s i i $ J I 1 II M 10 4 10 Cycles to failure 107 Figure 4.69 Effect of applied potential of 4140 steel Rc 52 stressed at 141 ksi (below the fatigue limit) on fatigue life in aerated 3% NaCl and aerated 0.5N Na 2 S0 4 ,25 0 C. (Lee, H.H. and Uhlig, H.H. (1972). Met. Trans., 2949, November. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) Types of corrosion: materials and environments However, as the crack progresses, the effect of environment also alters as forces such as mechanical, become very predominant. The effects of adsorption which cause embrittlement may, for instance, increase with an increase in the stress intensity. The effect of environment on crack initiation and crack growth varies from one metal to another. The fatigue life in the presence of an aggressive environment at a given stress value generally decreases. The corrosion fatigue is highly dependent on particular metal environment combination. The following can be the environment variables: (1) (2) (3) (4) (5) (6) (7) Temperature. Type of environment. Concentration of reactive species. Composition of electrolyte. pH. Viscosity of electrolyte. Coatings on metal surface or inhibitors added to electrolyte. of corrosion fatigue behavior of metals. 4.9.s 2 33 different INITIATION OF FATIGUE CRACKS The following are the possible sites for crack nucleation: (1) Discontinuities in metal, such as near the surface. (2) Inclusions or second-phase particles. (3) Scratches on a metal surface. (4) Sites of pitting or intergranular corrosion. (5) Twin boundaries. The cracks are initiated generally on the surface, however, initiation from subsurface is possible in the presence of surface defects. Intrusions may also develop into cracks. During the loading part of the cycle, slip occurs in a plane favorably oriented. The surface created by slip may oxidize during the unloading period. The first cyclic step may create either an intrusion or extrusion as shown in Fig. 4.70. By continued deformation A basic knowledge of the principles involving the mechanism of fatigue crack initiation and propagation is essential to the understanding t / * • ft Extrusion Intrusion i * Figure 4.70 Slip band intrusion and extrusion. (From Fuchs, H.O. and Stephans, R.I. (1982). In Metal Fatigue in Engineering, New York: lohn Wiley & Sons Inc., p. 28. Reproduced by kind permission of John Wiley, New York) 2 34 Principles of Corrosion Engineering and Corrosion Control represents one load cycle. At high AK values, striations become less significant. in subsequent cycles an intrusion may grow and form a crack [100-102]. 4 .9.6 FATIGUE CRACK 4 .9.7 CORROSION FATIGUE PROPAGATION The following is a summary of fatigue crack propagation: (a) A crack initiates in a crystallographic shear mode. It penetrates a few tenths of a millimeter. There is a large effect of microstructure, stress ratio and environment [103]. (b) The crack propagates in a direction normal to the stress axis (Stage 1). The stress concentration at the crack tip causes local deformation in a zone in front of the crack. The plastic zone increases in size as a result of crack growth. It continues to grow until it reaches the thickness of the specimen. When the size of the plastic zone becomes nearly equal to the thickness of the specimen, plain strain conditions do not exist any more. In the plain strain condition, the plane of fracture instability is normal to the axis of the principal tensile stress. There is zero strain in the direction normal to both the axis of applied tensile stress and the direction of crack growth. The crack propagates in a direction perpendicular to the tensile stress (Stage 2). A shear decohesion also contributes to crack propagation [104]. The crack undergoes rotation and the final rupture occurs in a plane stress mode (the plane of fracture instability is inclined to 45° to the axis of principal tensile stress, Stage 3). The crack propagates during each cycle. The crack growth during stage 3 is only a few nanometers per cycle. In the second stage, the rate of crack growth depends upon the square of fourth power of stress intensity range: da A — = AK4 dN , (4.66) CRACK GROWTH Corrosion fatigue crack growth for many systems occurs only above a certain threshold stress intensity factor AJCth- The growth increases linearly with increasing AiCand above AKth. It enters a region where da/dN is linearly dependent upon log AK. The rate becomes very rapid as Kmax approaches K\Q. Time-dependent stress corrosion cracking is shown in Fig. 4.71. In case of SCC, crack growth does not occur below a certain stress intensity factor (K\scc)- Cycle dependent fatigue crack growth is shown in Fig. 4.72. In an aggressive environment, corrosion fatigue crack growth may be quite different from the pure fatigue curve because of the metal-environment interaction. The fatigue crack growth response of materials in an environment depends on K\ at K levels approaching Kc or K\c (fracture toughness and plain strain fracture toughness, respectively) at high stresses and at levels approaching threshold at the lower end, with an intermediate region depending on some power of K The environment-assisted fatigue growth can be represented by three patterns of behavior. In general, three stages are encountered. In region (1), at low A lvalue, the crack growth rate is extremely dependent on stress intensity and the curves become almost parallel to the crack growth rate. The corresponding stress intensity appears to be environment-dependent and it is denoted by AXICF in analogy to i^iscc- Below Kiev, the corrosion fatigue crack growth is negligible. In region (2), the fatigue crack growth rate depends strongly on environment. Most results on corrosion fatigue studies have been reported in this region. It may be noted that acceleration of corrosion fatigue crack growth in region (2) may also occur around or below Kiscc> which shows that stress corrosion cracking and corrosion fatigue are two different phenomenon. Hence, a fatigue crack which accelerates in region (2) is called The striations observed on the fracture surface are related to the load cycle. Each striation Types of corrosion: materials and environments 235 Ktscc 10* - i Kie 4& -s c ur4 - 10* I m wm Cracks do not extend ""by stress corrosion cracking mechanism ! 10 I 100 1000 Static stress Intensity (AK), kr^sl JlirEh Figure 4.71 Time-dependent stress corrosion cracking. (From Gallaghar, J.P. and Wei, R.P. (1971). Int. Corrosion Fatigue Conf., June 14-18, NACE. Reproduced by kind permission of NACE, Houston, Texas, USA) 'True Corrosion Fatigue (TCF).y B oth processes may take place in region (2). Stress corrosion cracking may occur at a stress intensity threshold called Kiscc for a particular metal environment system. The effects of air fatigue and stress corrosion cracking on corrosion fatigue become apparent in dn/di vslog A K curve and their shape may change depending on which of the above is predominant. Cyclic-dependent fatigue crack propagation occurs below the threshold for stress corrosion. Type A (Fig. 4.73) shows the combined effect of fatigue and corrosion and it, therefore, represents true corrosion fatigue (TCF). This system is typified by AI-H2O system. As observed, there is a reduction in upper stress threshold for crack growth due to the corrosive environment. The crack growth is also accelerated. The environmental effect is diminished as K approaches Kc or K\c, because of mechanical-chemical interactions. There is a 2 36 Principles of Corrosion Engineering and Corrosion Control Kma* Fracture m2 JZ u B 9 u felO* e Cracks do hnot extend by fatigue mechanism u d10"* 10" 10 X 100 1000 Range of stress intensity ( AK), kpsi J inch Figure 4.72 Cycle-dependent fatigue crack growth. (From Gallagher, J.O. and Wei, R.P. (1971). In Proc. Int. Corrosion Conf., June 14-18, NACE. Reproduced by kind permission of NACE, Houston, Texas, USA) combined interaction of cyclic load and aggressive environment in the region where the rate of crack growth is not very rapid because the mechanical factors which lead to cracking are more dominant than chemical factors. The second type [B] represents stress corrosion fatigue. There is no environmental interaction below ^ ISCC- Here the sustained load contributes significantly to the cyclic crack growth. Above ^iscc> the stress intensity, cyclic frequency and waveform shape are major contributors. It is typified by the hydrogen/steel system. The environmental effects become very strong above a threshold for stress corrosion cracking and negligible below it. A mixed behavior is shown by type [C]. It is typified by a broad range of metalenvironment systems. In type [C],type [A] behavior is observed at K levels below the threshold K level, and type [B] behavior above the threshold K level. It shows a mixed pattern of SCC and CF. Types of c orrosion: materials a nd e nvironments 237 log(AK) (A) TCF (B) Log(K) SCF TCF and SCF Aggressive Aggressi 5 ? 3 Log(AK) (C) Log(AK) (D) Log(AK) (E) Figure 4.73 Schematic illustrations of basic types of corrosion fatigue behavior, SCC = stress corrosion cracking, TCF = true corrosion fatigue, SCF = stress corrosion fatigue. (From McEvily, A.J. and Wei, R.O. (1971). In Proc. Int. Fatigue Confi, June 14-18, NACE. Reproduced by kind permission of NACE, Houston, Texas, USA) 4 .9.8 THEORIES OF (1) Stress concentration a t the b ase of p its. T he c rack nucleation is r elated t o p its formed b y c orrosive attack. (2) Preferential dissolution of d eformed material which acts as a local anode with the undeformed material acting as a local cathode. C O R R O S I O N FATIGUE CRACK INITIATION A N D PROPAGATION T he theories of c orrosion fatigue a re b ased o n o ne o r m ore of the following principles. Extensive analytical studies have been made [105-108]. 2 38 Principles of Corrosion Engineering and Corrosion Control (3) Rupture of the protective film on the metal the plastic strain. The surface, after the film surface in the presence of aggressive ions by rupture, becomes very active and dissolves anodcyclic deformation. ically in the presence of corrosive media. The (4) Supply of reactants and removal of products crack advances by the conjoint action of film from crack tip. rupture and anodic dissolution. That potential (5) Reduction of surface energy due to adsorp- drop continues throughout the length of an altertion of species. nating stress experiment which suggests that the (6) Diffusion of a chemical species ahead of protective film is destroyed. crack tip. Several investigators do not support the film rupture theory. For instance, corrosion fatigue occurs in acid solutions, where there is no possi1. Stress Concentration at the Base bility for film formation. The observations given of Pits below do not support the film rupture theory: The role of pitting is not very clear. Materials (a) There is a critical rate of corrosion which is which develop pitting are found to be susceptible associated with corrosion fatigue. to corrosion fatigue, however, corrosion fatigue is (b) There is no fundamental shift of equilibrium also observed without pitting. Pitting is not a prepotential of steel, as shown by the effect of requisite to cracking as shown by low carbon steel cathodic protection on corrosion fatigue of in sodium chloride solutions. It is possible that steel. The potential is always shifted by film pits may have been developed after cracking. The formation. role of pitting was found to be crucial in fatigue (c) Emergent slip pattern is observed by metcrack initiation of martensitic 13% Cr stainless allographic studies for various grades of steels. There is sufficient evidence to suggest that steels. fatigue cracks are initiated by pits. Pitting plays a crucial role in corrosion fatigue. However, in view All these findings show that the film rupof the varying nature of results on various metal- ture theory cannot be accepted as a general environment systems, the effect of pitting on the mechanism. fatigue crack growth cannot be generalized. As shown by metallographic studies, slip bands are formed by emergence of slip steps or 2. Electrochemical-Mechanical Attack intrusion-extrusion mechanism (Fig. 4.74). The intrusion-extrusion pairs are larger in the presThere is a possibility that fatigue cracks advance ence of a corrosive media. These observations by an electrochemical mechanism. The distorted suggest that the emergent slip bands are attacked metal acts as the anode whereas the undistorted by a corrosive medium which causes local stress metal acts as cathode, and a galvanic cell is, there- intensification leading to premature failure. At fore, set up which provides the required driving preferentially attacked slip band steps, dislocaforce for the crack to advance. The active sites are tions are unlocked and it becomes easier for the provided by plastic deformation at the crack tip. metal to be deformed (Fig. 4.75). This has been observed for carbon steels in aqueous media. The slip bands produce numerous sites for crack initiation. The density of slip bands is increased by 3. Film Rupture Theory corrosion. Investigations on polycrystalline copper sugIt has been proposed that corrosion fatigue progest that copper fails in an intergranular manner ceeds by rupture of the surface film. There is considerable evidence to suggest that the rup- by corrosion fatigue. The dissolution of the grain ture of a protective film leads to initiation of boundaries is enhanced by deformation with fatigue crack. Figure 4.74 shows a schematic of respect to the slip bands. In the case of steels, a film rupture model. The film is ruptured at however, corrosion fatigue failures are always the crack tip by slip resulting in an increase in transgranular. Types of corrosion: materials and environments 239 —Source Extrusion Surface (B) Figure 4.74 Schematic example of mechanisms leading to a fatigue crack initiation (a) extrusion and intrusion in slip region due to cyclic stress (b) slip band intrusions and extrusions prior to crack initiation (Note the stress concentration is formed on the surface in this process). (From Fuchs, H.O. and Stephans, R.I. (1980). Metal Fatigue in Engineering, New York: John Wiley & Sons Inc., p. 28. By kind permission of John Wiley, New York, USA) In the case of aluminum alloys, failure occurs by hydrogen embrittlement according to mechanisms suggested under Hydrogen Damage. Hydrogen embrittles the regions in the vicinity of crack tip. The corrosion reaction continuously supplies hydrogen to the growing tip of the crack. It has been also suggested that the surface films on aluminum prevent hydrogen from escaping from the surface alloy and lead to the accumulation of dislocation debris. The situation may enhance the formation of cavities and lead to the propagation of cracking. 2 40 Principles of Corrosion Engineering and Corrosion Control smooth surface, whereas Tube 2 showed buildup of corrosion products. Tube 1 showed blunt transgranular cracking with minimum branching propagating from inside the tube. By EDX-ray analysis, the presence of copper and zinc ions, and some small amounts of chloride, sulfur, silicon and tin were observed. Conclusion The tubes failed by corrosion fatigue. Recommendation Admiralty brass tube was recommended because of its good corrosion performance. The tubes may be annealed to reduce residual stresses. Dislocation locking site Sy/~ Unlocked dislocations Figure 4.75 Schematic of simple model for corrosion enhanced crack nucleation at emerging slip step. B. Case 2 (From Duquette, D.J. and Uhlig, H.H. (1969). Trans. ASM, 62,839. Reproduced by kind permission of ASM, A number of cabin bolts from elevators had fractures. Metals Park, Ohio, USA) Examination Longitudinal sections from four bolts were examined. The bolts had a sudden transition in cross section and coarse grinding grooves, both of which are liable to decrease the fatigue strength. Longitudinal sections of bolts were prepared from the bolts and examined under microscope. A small crack (0.1 to 0.2 mm deep) was observed in one of the bolts. The other bolts were broken open at the point of cracking. In some bolts, narrow crescent-shaped cracks were observed. Some of the bolts had been normalized and some were quenched and tempered, therefore, the strength of the bolts fell into three groups: (a) 47, (b) 60-62 and (c) 65-71 kg/mm. Two bolts from group (a), two from the group (b) and four from group (c) had fractured. Conclusion The fatigue fracture was due to faulty material. 4.9.9 CASE HISTORIES 4. Adsorption Theory It is postulated that specific ions are absorbed and interact with strained bonds at the surface of the crack tip, thus reducing the bond strength, and permitting continued brittle fracture. This theory has been supported by observations in SCC. By chemisorption of the environmental species on the crack tip, the local fracture stress of the metal lattice is reduced. The theory has been applied to hydrogen embrittlement and liquid metal embrittlement. The adsorption phenomenon may be used to interpret the crack propagation mechanism of alloys which fail by hydrogen embrittlement, such as aluminum alloy 7075. A. Case 1 Two admiralty brass heat exchanger tubes from a cooler in a refinery unit showed cracks. 4.9.10 P REVENTION OF C O R R O S I O N FATIGUE The aim of the protective measures is to prevent Examination Both tubes showed cracks extend- corrosion fatigue cracks from initiating rather ing circumferentially above 180°C on the tension than stopping them from propagation once they side of the U-bend. Tube 1 showed a relatively have been initiated. Types of corrosion: materials and environments (1) The fatigue life in corrosive environments can be increased by introducing compressive stresses. The compressive stresses can be introduced by shot-peening, shot-blasting, nitriding and carburizing. (2) Cathodic protection has been used to prevent fatigue crack initiation. It can be achieved by attacking a more active metal than that being protected or by the impressed current technique. The use of electrodeposited coatings, such as zinc coating or cadmium coating, have been found to offer a better protection than the hot dipped zinc and cadmium coatings, because of the compressive stresses introduced during electrodeposition. (3) Application of nickel coating has been found to be effective in preventing corrosion fatigue. Use has been made of non-metallic coatings, like epoxy, rubber and enamel, to increase fatigue life, hence, once the coating is mechanically damaged, the fatigue resistance is adversely affected due to acceleration of underneath the coating. (4) The fatigue life of carbon steels and stainless steels is known to be improved by the formation of a passive layer on the surface. The passive layers can be formed by a technique known as *anodic protection.' In anodic protection, the steel is polarized by raising its potential to the passive region. The fatigue life of carbon steels and stainless steels maybe improved by this technique in oxidizing environment. This technique is generally used to improve the fatigue life of steels in an oxidizing environment. (5) The composition of an alloy can be modified to increase its resistance to corrosion fatigue. The resistance of stainless steels can be improved by increasing the Ni content. The resistance of mild steel can be improved by addition of 1-2% titanium. Such modifications in compositions are not the ideal solutions as the cost factor may sometimes be prohibitive for making such modifications. (6) The corrosion fatigue life can also be improved by environmental modification, such as adding inhibitors to the environment. The fatigue resistance of steel in a chloride-sulfate solution is increased by adding sodium dichromate. 2 41 Such inhibitors act by forming protective films on the surface of the metal being subjected to corrosion fatigue. The selection of a right inhibitor and its effective concentration is important for prevention of corrosion fatigue. The price is very expensive as large amounts of inhibitors are required. 4 .9.11 W H Y TO S T U D Y C O R R O S I O N FATIGUE There is a great demand for high static strength materials and a desire for higher structural performance from these materials. The demand for high strength-to-weight ratio of aerospace materials is increasing and resistance to fatigue is an important criteria. There is a need to improve the performance of aircraft alloys and of powder metallurgy products. Aggressive environments have a deleterious effect on the fatigue life, hence, an understanding of corrosion fatigue and of fatigue-environment interaction is extremely important for selection of materials resistant to corrosion fatigue. This is a localized form of corrosion, caused by the deposition of dirt, dust, mud and deposits on a metallic surface or by the existence of voids, gaps and cavities between adjoining surfaces. An important condition is the formation of a differential aeration cell for crevice corrosion to occur. This phenomenon limits the use, particularly of steels, in marine environment, chemical and petrochemical industries. 4.10 FRETTING CORROSION Fretting is a phenomenon of wear which occurs between two mating surfaces subjected to cyclic relative motion of extremely small amplitude of vibrations. Fretting appears as pits or grooves surrounded by corrosion products. The deterioration of material by the conjoint action of fretting and corrosion is called 'Fretting Corrosion.' Fretting is usually accompanied by 2 42 Principles of Corrosion Engineering and Corrosion Control The damage by fretting is, therefore, reduced at this temperature. The crucial factor is not the temperature by itself, but the effect of temperature on the formation of oxide on a metal surface. The nature and type of the oxide is the deciding factor. 5. Relative humidity The effect of humidity on fretting is opposite to the effect of general corrosion where an increase in humidity causes an increase in the rate of corrosion, and an increase in dryness causes a decrease in corrosion. Fretting corrosion is increased in dry air rather than decreased for metals which form rust in air. In case of fretting, in dry air, the debris which is formed as a consequence of wear on the metal surface is not removed from the surface and, therefore, prevents direct contact between two metallic surfaces. If the air is humid, debris becomes more mobile and it may escape from the metal surface, providing sites for metal-to-metal contact. corrosion in a corrosive environment. It occurs in bolted parts, engine components and other machineries. EXAMPLES OF DAMAGE FRETTING (1) Fretting of the blade roots of tube blades. (2) Loosening of the wheels from axles. For example, the coming out of railroad car wheel from the axle due to vibrations. (3) Fretting of electrical contacts. (4) Fretting damage of riveted joints. (5) Surgical implantations. 4 .10.2 F A C T O R S A F F E C T I N G FRETTING 1. Contact load Wear is a linear function of load and fretting would, therefore, increase with increased load as long as the amplitude is not reduced. 2. Amplitude No measurable threshold amplitude exists below which fretting does not occur. An upper threshold limit, however, exists above which a rapid increase in the rate of wear exists [109]. Amplitude oscillations as low as 3 or 4 nm are sufficient. 3. N umber of cycles The degree of fretting increases with the number of cycles. The appearance of surface changes with the number of cycles. An incubation period is reported to exist during which the damage is negligible. This period is accompanied by a steady-state period, during which the fretting rate is generally constant. In the final stage, the rate of fretting wear is increased. 4. Temperature The effect of temperature depends on the type of oxide that is produced. If a protective, adherent, compact oxide is formed which prevents the metal-to-metal contact, fretting wear is decreased. For example, a thick layer of oxide is formed at 650°C on titanium surface. 4.10.3 MECHANISM OF CORROSION FRETTING Although the mechanism of fretting corrosion is only partially understood, it is well-known that fretting proceeds in three stages: (a) The first stage is the metallic contact between two surfaces. The surfaces must be in close contact with each other. The contact occurs at few sites, called asperities (surface protrusions). Fretting can be produced by very small movements, as little as 10 - 8 cm. (b) The second stage is oxidation and debris generation. There is a considerable disagreement between the workers on whether the metal is oxidized prior to its removal or after its removal. It is possible that both processes may occur, each process being controlled by conditions which lead to fretting. In either case, the debris is produced as a result of oxidation. (c) Initiation of cracks at low stresses below the fatigue limit. Types of corrosion: materials and environments 2 43 4 .10.4 C H A R A C T E R I S T I C S EACH STAGE OF A. Adhesion In order for the metals to be in physical contact with each other, there must be no protective oxide layer. The breakdown of the protective layer is essential for the onset of fretting. The asperities C. Crack Initiation are bonded together at adhesion sites created by the relative slip of the surfaces. Fretting may occur Cracks grow in a direction perpendicular to the applied stress at the fretting area. Some at amplitude as small as 10~8 cm. If two metals in intimate contact are similar, of the cracks may not propagate at all at low protective films on both shall be disrupted, how- stresses because the impact of stress on a fretever, if one metal is soft and the other is hard, the ted surface extends to a shallow depth only. The layer on the soft metal will be destroyed, and on propagation of cracks is either restrained or prevented by the presence of favorable compressive the other metal it would not be disrupted. stresses. The stage of crack initiation is called fretting fatigue [110]. Crack propagation at higher stresses is of practical importance as it can lead B. Generation of Debris to failure of components, such as shafts and According to one school of thought, the metal axles. The crack originates at the boundary of is oxidized before it is removed from the sur- a fretted zone and propagates. During propagaface and according to other oxidation precedes tion, if a corrosion medium contacts the crack, the removal of the metal from the surface. This is corrosion fatigue also contributes to the crack no clear explanation for the two opposite views. propagation. Outside the sphere of the surface The material removed from the metal surface contact stress, the crack propagates as a fatigue due to fretting is called debris. The debris pro- crack, and upon fracture, a characteristic lip is duced by low carbon steel consists of mainly observed. ferric oxide, Fe2C>3. The debris can also contain The general mechanism of fretting corrosion unoxidized particles in the case of non-ferrous is shown in Figs 4.76 and 4.77. metals. The composition of the debris differs from one metal to another metal. If the particles of oxide becomes embedded in the softer material, the rate of wear is reduced and hence fretting is minimized. Loose particles increase the rate of wear and hence fretting proceeds at a high rate. Exposed metal {mechanical activities) a-feO (oxide debris powder) S tepl (Before) Step 2 (After) Figure 4.76 The mechanism of fretting corrosion at a steel surface (schematic view) 2 44 Principles of Corrosion Engineering and Corrosion Control : $^;mfim feGs arid FkaOf powter Figure 4.77 Sketch illustrating the mechanism of fretting corrosion 4 .10.5 C A S E H I S T O R Y Fretting failure of raceways on 52100 steel rings of an automotive front wheel bearing [111]. 4 .10.6 PREVENTION ( i) Increase the magnitude of load at the mating surfaces to minimize the occurrence of slip. Description of problem Fretting and pitting on (2) Keep the amplitude below the level at which the raceway of the ball bearing is shown in fretting occurs, if known for a particular Fig. 4.78. The inner and outer rings were made system. There is lower threshold limit below of cold drawn 52100 steel tubing. which fretting does not occur. Examination On physical examination, it was (3) Decrease the load at the bearing surfaces, discovered that serious fretting of the raceway however, it may not always work. in the ball contact area has occurred. Fretting (4) Use materials which develop a highly resisand pitting occurred at spacings equivalent to the tant oxide film, at high temperature to spacings of the balls in the retainer (Fig. 4.78). minimize the adverse effect of temperature Examination of the inner raceway showed lesser on fretting. attack. (5) Use gaskets to absorb vibration. (6) Increase the hardness of the two contactConclusion The failure was caused by fretting. ing metals, if possible, by shot-peening. During transportation of vehicle by sea, the body Compressing stresses are developed durof the vehicle was continuously vibrating without ing shot-peening, which resist and increase any rotation of the bearing. fretting resistance. Recommendations Sufficient preventive mea(7) Use low viscosity lubricating oils. sures were not taken during the transportation of (8) Use materials to resist fretting corrosion (Tables 4.15 and 4.16). the above vehicle. The rolling elements should be Use thick resin bonded coatings. taken off during packing and replaced by wooden (9) packing. Vibrations should be eliminated, as (10) Separate the surfaces, for instance, by interleaving sheets of wrapping paper between far as possible, during transportation of vehicles stacked sheets. by sea. Types of corrosion: materials and environments 2 45 5 2100 steel rm^s Slim steel at Roclcwell €60 to 64; 1008 steel Fretting and pilling on taceway I i Outer r *ig CS2I00 sieci) 1 9 4 <&*« Batt, 0.3?$ d am (SaiDO steel); I of 9 Retainer OOOSsteei) '' 07$ d&arm Frettirts and pitfenQ cm raceway Iimer iwg (S2I00 s t a f ) Figure 4.78 Fretting and pitting on raceway Table 4.15 Poor Aluminum on cast iron Aluminum on stainless steel Magnesium on cast iron Cast iron on chrome plate Laminated plastic on cast iron Bakelite on cast iron Hard tool steel on stainless steel Chrome plate on chrome plate Cast iron on tin plate Cast iron on cast iron with coating of shellac Fretting resistance of various materials Average Cast iron on cast iron Copper on cast iron Brass on cast iron Zinc on cast iron Cast iron on silver plate Cast iron on copper plate Cast iron on amalgamated copper plate Cast iron on cast iron with rough surfaces Magnesium on copper plate Zirconium on zirconium Good Laminated plastic on gold plate Hard tool steel on tool steel Cold-rolled steel on cold-rolled steel Cast iron on cast iron with phosphate coating Cast iron on cast iron with coating of rubber cement Cast iron on cast iron with coating of tungsten sulfide Cast iron on cast iron with rubber gasket Cast iron on cast iron with Molykote lubricant (M0S2) Cast iron on stainless steel with molykote lubricant Source: McDowell, J. R. (1952). STM Special Technical Publication, No. 144, American Society for Testing A Materials, Philadelphia, 24. 2 46 Principles of Corrosion Engineering and Corrosion Control Table 4.16 Fretting resistance of various materials Fretting resistance Poor Poor Poor Poor Poor Poor Poor Poor Poor Poor Average Average Average Average Average Average Average Average Average Average Good Good Good Good Good Good Good Good Good Combination Aluminum on cast iron Aluminum on stainless steel Bakelite on cast iron Cast iron on cast iron, with shellac coating Cast iron on chromium plating Cast iron on tin plating Chromium plating on chromium plating Hard tool steel on stainless steel Laminated plastic on cast iron Magnesium on cast iron Brass on cast iron Cast iron on amalgamated copper plate Cast iron on cast iron Cast iron on cast iron, rough surface Cast iron on copper plating Cast iron on silver plating Copper on cast iron Magnesium on copper plating Zinc on cast iron Zirconium on cast iron Cast iron on cast iron with coating of rubber cement Cast iron on cast iron with Molykote lubricant Cast iron on cast iron with phosphate conversion coating Cast iron on cast iron with rubber gasket Cast iron on cast iron with tungsten sulfide coating Cast iron on stainless steel with Molykote lubricant Cold-rolled steel on cold-rolled steel Hard tool steel on tool steel Laminated plastic on gold plating Source-. McDowell, J. R. (1952). In Symposium in Fretting Corrosion, STP 144, ASM. 4 .1 1 EROSION- CORROSION AND C AVITATION D A M A G E 4 . I i.i INTRODUCTION The problem of cavitation corrosion has plagued engineers continuously in a variety of disciplines, ranging from civil engineers to rocket engineers. Cavitation occurs wherever the local pressure in aflowfieldfalls below a certain critical value. Cavitation corrosion is a form of localized corrosion combined with mechanical damage, that occurs in a rapidly moving liquids and takes the form of areas or patches of pitted or roughened surface. Many of the terms related to cavitation damage need to be distinguished from each other: Cavitation corrosion It is a conjoint action of corrosion and cavitation. Types of corrosion: materials and environments Cavitation damage It is the degradation of a solid body caused by cavitation. The damage appears in the form of loss of material, change in appearance, surface deformation or changes in properties. It takes place when the velocity becomes so high that its static pressure is lower than the vapor pressure of liquid. Cavitation erosion It is a continuous loss of material by the impact of cavitation under the influence of erosion. Generally, cavitation includes the corrosion as well as the mechanical damage. 2 47 (6) Operation conditions. For instance, rotary pumps can be operated at the highest head pressure to avoid crevice corrosion. 4.11.4 MECHANISMS OF CAVITATION 4 .11.2 C H A R A C T E R I Z A T I O N CAVITATION DAMAGE OF The following are the characteristics of damage caused by cavitation corrosion: (a) Honeycombed surface of the metal. (b) Absence of corrosion product on the surface. (c) Occurrence of the attack in very sharply defined areas with a very sharp boundary between the affected and the unaffected metal. (d) Difference in the intensity of attack on two sides of the exposed surface. 4.11.3 ENVIRONMENTAL FACTORS AFFECTING C AVITATION D A M A G E (1) Amount of entrained air. (2) Presence of dust particles. The dust particles act as nuclei for cavity formation. (3) Temperature. There is a critical temperature above which the intensity of attack is decreased. (4) Corrosiveness of the media. For instance, the attack in salt water is at least 50% more than observed in distilled water. The intensity would, therefore, vary from one medium to another. (5) Selection of materials. Materials, like 18-8 steels and titanium, are resistant to cavitation damage. Under certain conditions, for example, in pumps and at the back of propellers, a partial vacuum is created. The water boils at ambient temperature at this reduced pressure, in the form of bubbles which collapse when they move to a position of increased pressure. The pressure exerted by the collapsing bubbles are very intense. Collapsing bubbles can produce pressure as high as 60 000 psi. If the process is repeated too often, the protective oxide layer is completely destroyed by the implosion of bubbles. The resistance of the metal to cavitation depends on its ability to form a compact, dense and adherent film. Brittle films are not normally protective and are subjected to cavitation damage. The collapse of bubbles is accompanied by vibrations and noise. The above phenomenon can be observed also during boiling, when vaporfilled bubbles are convected away from regions of high temperature into cooler regions, where condensation of vapors and bubble collapse is accompanied by a crackling sound. Many times large vapor-filled cavities develop which remain attached to the solid surface. The metal, which is subjected to a tensile force greater than its cohesive strength, is split open and forms such cavities. When such cavities form on surface, such as the blades of pumps, turbines, hydrofoils, etc. a marked change occurs in the flow field. Small bubbles are formed on the surface of these cavities and they collapse. The process is repeated continually. Physical damage always occurs if the collapse is adjacent to a solid boundary. Estimates have shown that 2 x 106 cavities may collapse over a small area in a matter of seconds. Shock waves with pressures as high as 200MN/m2 are produced. Such high forces are sufficient to cause plastic deformation. Failed pump parts show slip lines, which is the evidence of plastic deformation brought about by the collapse of cavities [112]. 2 48 Principles of Corrosion Engineering and Corrosion Control A plot of performance and noise in a turbomachine as a function of cavitation is shown in Fig. 4.80. Several mechanisms have been proposed. The various stages of bubble collapse mechanisms are shown in Fig. 4.79 [112]. The theory of bubble dynamics is well developed and impulsive pressures sufficiently high to cause metal failure can be predicted from various models of bubble collapse. The schematics of various mechanisms are shown in Fig. 4.79. The mechanism of spherical collapse, non-symmetrical collapse and for vital collapse are shown [113]. Readers interested in the above mechanism should refer to the references cited at the end of this section. 4 .11.6 C A S E H I S T O R Y A cylinder lining from a diesel motor had suffered extensive damage on cooling water side (Fig. 4.81) [115]. Visual examination Heavy pitting was observed on the cooling water side. The outside wall was also found to be coated with Fe3C>4, SiC>2 and CaO. Identification The attack was cavitation. This attack is generally encountered in diesel motor cylinders. The attack is observed when a body vibrating in a liquid attains high values of amplitude and frequency and the sluggish liquid cannot keep in pace with body. Implosion of bubbles on the point of maximum vibration occur and high shock wave pressures are produced on the metal surface. It maybe pointed out that at the moment of reversal, the cylinder is subjected to bending vibrations by the sideways, pressure of the piston and the cooling water may not be unison with the cylinder vibrations. 4 .11.5 EXAMPLES Cavitation occurs wherever irregular water flow takes place, particularly if high pressures and vacuum are involved in the flow systems. Cavitation is a serious problem in desalination plants. In multi-stage flash distillation units, cavitation occurs in: (a) suction of brine cycle, blowdown and distillate pumps, (b) the drain opening of the last heat recovery stage (HRS), (c) the brine recirculation control valve, Remedy (d) heat reject pipe. (1) Reduce the piston pressure. Cavitation is common in hydraulic equip- (2) Use a thicker wall lining to reduce the ment, ship propellers, impellers, inlets to heat amplitude. exchanger tubes and steam turbines. (3) Add a suitable oil to the cooling water so Cavitation can lead to the shutdown of a that a film is formed on the cylinder surdesalination plant. If there is cavitation in an ejecface, but note that an oil film may reduce heat tor condensate pump, it will fail to reach the transfer. discharge pressure and the required amount of condensate would not be extracted. One of the Some interesting applications The scales which requirements of pumping liquid is that the pres- develop on the surface of the pipes in geothermal sures in any point in the suction arm should plants are extremely hard and in many instances, never be reduced to the vapor pressure of the they approximate the strength of the pipe. These liquid as this causes boiling (at reduced pres- scales are extremely difficult to remove. Cavitasure). Too low a pressure at the pump suc- tion cleaning systems have been used to remove tion must always be avoided so that cavitation such scales. Cavitation nozzles at pressure drops is not caused. Cavitation in pumps is noticed up to 100 MPa have been tested. The focal length by a sudden increase in the noise level and of the bubble can be adjusted to cut the required its inability to reach discharge pressure [114]. scale thickness. Types of corrosion: materials and environments 249 1. Spherical collapse and rebound o 7777777777777T Cavitation bubble collapse commences. 77777777777777 Motion arrested at minumum volume. Bubble rebounds due to noncondensable gas. A pressure wave propagates outward. 2, Non-symmetrical collapse //////, A cavitation bubble near a boundary begins to collapse. Due to the presence of the boundary the bubble indents. 7777/ A high velocity jet is formed that strikes the boundary. 3. Final toroidal collapse oo 77777777777777 A jet strikes the boundary but causes little or no damage, ( CI *». * 77777777777777 The cavity collapses further in the form of a torust until minimum volume is reached. 77777777777777 The bubble rebounds due to noncondensabie gas, A pressure wave propagates outward. Figure 4.79 Schematic of various bubble collapse mechanisms. (From Arndt, R. E. (1977). Paper No. 91, Corrosion, NACE) [113] 250 Principles of Corrosion Engineering and Corrosion Control 1.10 0,05 0.10 0.15 0.20 025 030 Cavitation Index Stages of Cavitation (1) Limited (no performance loss) (2) Partial noise and vibration (3) Fully developed (large performance loss) Figure 4.80 Performance and noise in a turbo-machine as a function of cavitation index. (From Arndt, R.E. (1977). Paper No. 9 1, Corrosion. Reproduced by kind permission of NACE, Houston, Texas, USA) [113] Types of corrosion: materials and environments 251 Mfcrostructure of specimen takenfromthe cylinder liner Figure 4.81 Failure of cylinder liner from a diesel motor. (From Kauczor, E. (1979). Case Histories in Failure Analysis, p. 232, ASM. Reproduced by kind permission of ASM, Metals Park, Ohio, USA) In dental hygiene, extensive use has been made of cavitation cleaning for the removal of calcerous deposits. Water is fed to the tip of the tool at 35 cm3/min under pressure. The vibrating tip of the tool used for cleaning induces cavitation in the water flow, and removal of the scale is achieved by the cavitation cleaning technique. (3) (4) (5) 4 .11.7 P R E V E N T I O N C AVITATION D A M A G E OF Cavitation damage can be reduced by: (1) Using materials resistant to cavitation damage (Table 4.17) [116]. (2) Changes in design, for instance, by changing the diameter of the pipe, the flow geometry can be affected. The flow pattern can be changed from turbulent to lamellar (6) by increasing the pipe diameter. Smooth finishes on pump impellers and propellers reduce the cavitation damage. Reducing the amount of entrained air, if the process of protective film formation is not adversely affected. Avoiding dust particles and other impurities. Optimizing operation conditions, for instance, operate rotary pumps at highest pressures. Cavitation indicates that there is insufficient net positive suction head (NPSH) (NPSH is a measure of the suction head to prevent vaporization of the lowest pressure point in a pump). A good design can eliminate the cavitation in impeller vanes by providing sufficient NPSH and positive pressure at the suction inlet. Adding inhibitors, like chromates and nitrates. In order to prevent the cavitation damage of diesel engine cylinder liners, fNa2 &O4, is generally added to cooling 2 52 Principles of Corrosion Engineering and Corrosion Control Table 4.17 Group I Resistance of metals to cavitation corrosion [116] Most resistant. Subject to little or no damage. Useful under extremely severe conditions. Stellite hard-facing alloys Titanium alloys Austenitic and precipitation-hardening stainless steels Nickel-chromium alloys, such as Inconel alloys 625 and 718 Nickel-molybdenum-chromium alloys, such as Hastelloy C These metals are commonly used where a high order of resistance to cavitation damage is required. They are subject to some metal loss under the most severe conditions of cavitation. Nickel-copper-aluminum alloy Monel K-500 Nickel-copper alloy Monel 400 Copper alloy C95500 (nickel-aluminum bronze, cast) Copper alloy C95700 (nickel-aluminum-manganese bronze, cast) These metals have some degree of cavitation resistance. They are generally limited to low-speed low-performance applications. Copper alloy C71500 (copper-nickel, 30% Ni) Copper alloys C92200 and C92300 (leaded tin bronzes M and G, cast) Manganese bronze, cast Austenitic nickel cast irons These metals normally are not used in applications where cavitation damage may occur, except in cathodically inhibited solutions or when protected by elastomeric coatings. Carbon and low alloy steels Cast irons Aluminum and aluminum alloys Group II Group III Group IV Note: Applies to normal cavitation-erosion intensities, at which corrosion resistance has a substantial influence on the resistance to damage. water, but chromates are undesirable for health reasons. (7) Cathodicprotection application. Hydrogen bubbles produced by cathodic protection mitigate the effect of imploding cavitation bubbles. Sacrificial anodes, such as zinc or £ ' ' (8) Minimizing opportunities for cavities to be formed. It could be accomplished by keeping the liquid temperature and pressure high and by introducing air bubbles. (9) Making the surface very smooth. (10) Coating with resilient materials, like rubber. QUESTIONS UNIFORM CORROSION Multiple C h o k e Questions Mark one correct answer from the following: L The major component affecting corrosion in industrial atmosphere is a) carbon particles b) dust particles c) S0 2 Types of corrosion: materials and environments 2. In an industrial atmosphere with high humidity and SO2 content a) the corrosion product [Fe(OH)2] being formed is protective b) the corrosion product is formed away from the corroding site c) the corrosion product is formed at the corroding site 3. The following protective measures can be very effective in preventing uniform corrosion: a) Select a thinner material b) Prevent the material from atmosphere by covering c) Use an epoxy coating d) Give a corrosion allowance 4. If a metal undergoes uniform corrosion it becomes a) thinner b) thicker c) perforated 5. The following is the example of uniform corrosion: a) b) c) d) Cracking of a pipe in the soil Perforation of a water pipe Rusting of a steel tank in air Leakage of water pipe by corrosion 253 2. Referring to the galvanic series of some commercial metals and alloys in seawater, mark the condition which would lead to minimum corrosion by galvanic coupling. a) Coupling of 18-8 steel active with chromium stainless steel, 13% Cr (active) b) 70-30 brass with pure copper c) Silver and copper d) Magnesium and aluminum plates 3. Which of the following shows an unfavorable area ratio? a) b) c) d) Copper rivets in copper plate Steel rivets in copper plate Aluminum rivets in aluminum plate Brass rivets in copper plate 4. Which of the following would be most effective in preventing galvanic corrosion? a) Avoiding unfavorable area ratio b) Adding inhibitors c) Using welded joints of the same alloy as the alloy to be welded d) Increasing temperature at the point of contact of the two metals 5. Which of the following pairs of metals would show the highest rate of corrosion in seawater? a) b) c) d) Copper Copper Copper Copper and steel and zinc and aluminum and brass GALVANIC CORROSION Multiple Choice Questions Mark one correct answer from the following: 1. Galvanic corrosion is restricted when a) corrosion products form between the two metals b) pH of the electrolyte is reduced c) electrolytic conductivity is high d) polarization of less active metal takes place with difficulty How and Why Questions 1. In the following couples, which one would form the anode and which one the cathode? a) b) c) d) Copper and iron Stainless steel and brass Active and passive steel Zinc and aluminum 2 54 Principles of Corrosion Engineering and Corrosion Control PITTING CORROSION 2. In the galvanic series why active steel is placed far away from the passive steel and not bracketed together? 3. Explain why: Multiple Choice Questions Mark one correct answer from the following: a) The magnitude of galvanic corrosion is maximum at the junction of the two 1. The most important condition for pitting to take place is metals b) The magnitude of galvanic corrosion a) the metal must be pure decreases as the distance from the junction b) the metal must be in a passive state of the two metals increases c) the metal must be in an active state c) A new pipe joined to an old pipe corrodes d) the metal can be in any state d) A metallic couple with an unfavorable area ratio (small anode/large cathode) corrodes 2. Pitting can take place under the following faster than a couple with large anode and conditions: small cathode e) Galvanic attack is accelerated in the presa) Water in contact with stainless steel and ence of NaCl and humidity flowing at a high velocity b) Stagnant conditions in service 4. Explain why: c) Presence of NO^~ and CrO^ ions in water d) Deaerated water a) If steel rivets are fixed in a copper plate, the rivets are completely corroded after 3. The following reaction occurs in the pit cavity: sometime a) 2 H 2 0 + 0 2 + 4 e ^ 4 0 H b) Galvanized steel pipes are not suitable for b) F e + + + 2 e ^ F e hot water system above 80° C c) The polarization of the reduction reaction c) F e - > F e + + + 2 e d) 2H+ + 2e->H 2 predominates d) The aggressiveness of the environment determines the extent of galvanic corrosion 4. On the outside of the pit cavity, the main reaction is e) When dissimilar metals are to be coated, the more noble must be coated a) anodic polarization f) Water heaters with copper water inlet tube b) cathodic reduction of hydrogen corrode c) cathodic reduction of oxygen g) Welded joints are better than riveted joints 5. Which one of the following processes occurs mainly around the pit on aluminum? Review Questions 1. State the mechanism of galvanic corrosion for a steel and copper couple immersed in 1N solution of aerated NaCl. 2. State three main differences between the emf series and galvanic series. 3. What is the effect of area ratio and distance on the magnitude of galvanic corrosion? 4. State the mechanism of atmospheric corrosion of steel in an industrial atmosphere containing SO2, C and chloride particles. a) Cathodic protection by the anodic reaction b) Passivation by alkali formed in the cathode reaction c) Oxidation of aluminum d) Deposition of active metals 6. The following methods are used to prevent pitting: a) Adding inhibitors containing Fe + + ions orHg ++ ions b) Operations at the highest temperature c) Increase of oxygen concentration Types of corrosion: materials and environments d) Elimination of chloride from the environment, wherever possible 2 55 Review Questions 1. Explain the mechanism of pitting when it is initiated at sulfide inclusion in a carbon steel. 2. Explain the reactions which take place during the pitting of aluminum How and Why Questions 1. Explain why: a) at the mouth a) Stainless steels exposed to water containb) inside the pit mouth ing Fe 3+ , Cu 2+ or Hg 2+ develop pits c) outside the pit. within hours. b) Addition of 3% NaN0 3 to 10% FeCl3, 3. How can you distinguish between the detriinhibits the pitting of 18-8 steel. mental effect and beneficial effect of an c) A passive surface is necessary for the onset alloying element with respect to the shift of of pitting. corrosion potential either in the positive or in d) Oxygen is a prerequisite for pitting. the negative direction? e) A polished surface has a lesser tendency to 4. State briefly the main characteristics of adsorpdevelop pits than a rough surface. tion theory as applied to aluminum and steel. f) Addition of 2% Mo to steel increases its Mention one item of evidence against the resistance to pitting. theory. 2. Explain why: INTERGRANULAR CORROSION a) Acidity is developed within a pit. b) The outside of a pit cavity is cathodically Multiple Choice Questions protected. c) Pitting reaction is sustained even after Mark one correct answer from the following: oxygen is consumed. d) The pitting process is autocatalytic. 1. Addition of molybdenum is particularly e) Pitting takes place in contact with water made steels under stagnant conditions. a) to increase the tensile strength 3. Explain how: b) to increase its fluidity c) to increase the resistance of stainless a) The surface adjacent to a pit is protected steels (ferrite and austenite) to pitting and from corrosion. crevice corrosion b) The resistance of an alloy to pitting can be predicted. 2. Intergranular corrosion is c) Pitting can be differentiated from fretting corrosion. a) restricted to stainless steels d) Pitting can be differentiated from crevice b) not restricted to stainless steel but corrosion. includes nickel-base alloys also e) Pitting can be prevented by cathodic prec) restricted to alloys containing low carbon vention. content f) The resistance of steels to pitting can be increased. 3. Sensitization of austenitic steel would not g) Pitting damage can be evaluated microoccur on cooling fast through 580-850°C scopically. h) Pitting attack is decreased by an increase a) if the crosssection of the specimen is small of velocity of corroding fluid. b) the amount of carbon is increased 2 56 Principles of Corrosion Engineering and Corrosion Control c) if it is reheated after quenching between 550 and 850°C temperature range d) if it is slowly cooled through 800° C, quenched and reheated to 235° C c) Heat treatment does not influence distribution and precipitation of carbide and, hence, the degree of sensitization 7. Sensitization in ferritic stainless steels can be eliminated by a) lowering the carbon content to 50 ppm b) lowering chromium to less than 12% c) lowering the amount of interstitial carbon and nitrogen to 50 ppm and 150 ppm, respectively 8. Which of the two stainless steels, ferritic or austenitic, would sensitize more under the given conditions? a) Ferritic steel when quenched from 800° C or higher b) Ferritic steels on slow cooling from 800°C c) Austenitic steels when slowly cooled through 500°C 9. Knife-line attack in stainless steels is caused by one of the following: a) The base metal (18-8 steel) adjacent to the fusion line is heated in the temperature range in which the steel is sensitized b) Subsequent welding passes during welding pre-heat the area adjacent to the fusion line and the temperature range to cause chromium carbide to be formed c) Heating stabilized steel in the temperature range 510-760° C 10. Which of the methods given below can be used for prevention of intergranular corrosion of ferritic stainless steels? a) Reduction of interstitial carbon and nitrogen content b) Heating at 925°C and rapid quenching c) Increasing the nickel content 4. Titanium and niobium carbides are formed at temperatures above 1219°C, whereas chromium carbides are formed in the range 787-1219°C. Steels 321 and 347 contain either chromium or titanium carbides as received from the mill. Intergranular corrosion would develop in these steels if a) material is heated below 400° C b) metal is heated to 1310°C, quenched in water and reheated for 30min to 500° C and cooled c) the material is heated after welding to above 1055°C and cooled rapidly d) metal is heated between 505 and 760° C. 5. Which of the following is true about carbides? a) They are present in the grain boundaries in the form of a continuous film b) They utilize all the carbon present in the alloy for their formation c) The carbides are generally represented by M23C6, where M = Cr and Fe d) The composition of carbides is fixed 6. Which of the following assumptions regarding the chromium depletion theory for stainless steels has more evidence in support of it? a) The local chromium-carbon-carbide equilibrium is reached at the grain boundary and the chromium content at the grain boundary is thermodynamically determined by the equilibrium b) The chromium content in the vicinity of the carbide particles is negligible and the variation in the degree of sensitivity results from changes in the morphology of distribution of particle along the grain boundary How and Why Questions 1. Why the precipitation of carbide results in the depletion of chromium in the grain boundary, Types of corrosion: materials and environments 2 57 and why it becomes sensitive to the attack of S T R E S S C O R R O S I O N corrosion? C RACKING 2. Are the carbides stoichiometric or they can be represented by M23C6? Multiple Choice Questions 3. For 304 stainless steel, the maximum carbon allowed for 700°C anneal is 0.08%, and at 1. Which of the following is true about stress 600°C it is 0.02%. Why the alloy would be corrosion cracking? susceptible to intergranular attack if the above carbon amounts are exceeded? Assume that a) The stress required to cause cracking may 12% chromium is required to form a passive be below the macroscopic yield stress film. b) It is not necessary to apply a tensile stress 4. Why ferritic stainless steel resists corrosion if it for SCC to occur is held at 788° C for a sufficient period of time? c) It can take place in vacuum 5. Why intergranular attack is not prevented in d) It cannot take place in the absence of a ferritic stainless steel even if carbon is reduced thermal stress to 0.3%? 6. Although carbon reacts with titanium and 2. Select one correct statement: niobium which are added as stabilizers, yet intergranular corrosion is not observed in a) Specific ions are necessary for metals and their presence. Explain. alloys 7. What is the major difference between knifeb) An environment that causes SCC of one line attack and weld decay as far as the location alloy would also causes SCC of another of the attack with respect to the weld is alloy concerned? c) As specific ions form a film on the metal 8. Why stainless steels which are stabilized surface, hence, those ions which form undergo knife-line attack? If the zone on films may be adequate to cause SCC which knife-line attack has occurred is reheated between 950-1450°C, would you 3. State which of the following is characteristic expect the steel to be sensitive to intergranular of stress corrosion cracking: attack? a) The failure by SCC is always ductile b) The failure by SCC is dependent on the metallurgical condition of the alloy Conceptual Questions c) Cracks generally become visible very soon (in hours) and take a very long time to 1. Differentiate clearly between the austenitic, propagate ferritic, martensitic, duplex and precipitation hardened stainless steel. List two examples 4. The following variables may affect SCC of each. ( two): 2. State four salient characteristics of the model of grain boundary attack based on chromium a) Temperature depletion theory. b) Impurity 3. What would happen if austenitic stainless steel c) Mechanism of crack propagation is held at the sensitizing temperature for, say 2000 h? Would it be still susceptible to inter5. In the film rupture mechanism, it is implied granular attack? If yes, why. If no, explain the that reason. 4. Differentiate clearly between knife-line attack a) the stress acts to open the crack and ruptures the protective film and weld decay in terms of mechanism. 2 58 Principles of Corrosion Engineering and Corrosion Control b) the crack advances by preferential dissolution of the grain boundaries c) the crack tip remains bare, because the rate of film rupture at the crack tip is less than the rate of repassivation 11. Addition of a small amount of Ni to ferritic stainless steeL a) increases its resistance to corrosion b) decreases its resistance to corrosion to SCC c) has no effect 12. Which of the following similarities are common between stress corrosion cracking and hydrogen embrittlement? (Mark two correct answers.) a) A general increase in sensitivity to the degradation with the increased strength of the alloy b) Discontinuous crack propagation c) Liberation of hydrogen upon cathodic polarization d) Appearance of blisters on the surface of material which has cracked 13. The following are the reasons for concentration of caustic which lead to caustic cracking: a) Departure from nucleate boiling in which case the rate of bubble formations exceed the rinsing rate b) Concentration of sodium hydroxide at the water level c) Formation of steam over the insulating deposits d) Adding a higher concentration of caustic than the concentration prescribed 14. By the term caustic embrittlement, it is understood that a) there is a loss of ductility b) there is no loss of ductility c) the failure occurs by weakening of the grain boundaries leading to cracking without any loss of ductility 6. By the adsorption of specific active species on the metal surface a) the interatomic bond strength is increased b) the stress required for cleavage fracture is decreased c) the cracks propagates in a discontinuous manner 7. The process of stress corrosion cracking can be adequately explained by a) mechanistic chemical model b) a broad spectrum of mechanism, such as the theories based on active path, strain assisted and adsorption related phenomenon c) the film rupture mechanism 8. In the stress corrosion cracking of austenitic stainless steels a) the greatest susceptibility to SCC is exhibited at a nickel content of 8% b) the least resistance is shown by alloys of a higher nickel content c) nickel content has no particular effect on SCC of austenitic stainless steels 9. The stress levels required for SCC of steels is a) 0.1 % of the yield stress b) vary from one metal to another c) close to the yield stress 10. Ferritic stainless steels like austenitic stainless steels are a) subjected to SCC in chloride media, such as boiling MgCl2 at 140°C b) only subjected to cracking in 5% NaOH at 25°C c) do not crack in chloride media Conceptual Questions 1. Briefly mention at least six important characteristics of stress corrosion cracking. 2. List three different sources of residual stresses. Types of corrosion: materials and environments 259 3. Explain the main difference between the prec) chlorides, bromides and iodides existing active path, strained generated active d) deposits, debris on a metallic surface paths and specific adsorption theories. 4. State three major factors which cause the 2. The main reason for crevice corrosion is concentration of caustic. Explain briefly the mechanism of caustic cracking in boilers. a) the formation of differential, aeration cells 5. Why localized corrosion is initiated by the b) the difference of potential between two rupture of a protective film on steels, and contacting surfaces how it assists SCC? c) development of stress cells on the metal 6. Specific anions are not important for SCC d) polarization of the anodic and cathodic of martensitic steels, as long as hydrogen is areas evolved. Substantiate this statement. 7. State briefly how the rupture of a protective 3. When the ratio of the crevice solution to film assists in SCC of steels. crevice area is small 8. How a local galvanic cell is formed by precipitation of carbides at the grain boundary? a) the critical value for initiation of crevice 9. Why low carbon stainless steels are not as corrosion is rapidly achieved much susceptible to SCC as high carbon steel? b) the critical value for initiation of crevice 10. Outline the main difference between hydrocorrosion is delayed gen attack at 200° C and blistering caused by c) the critical value has no affect on the hydrogen. initiation of crevice corrosion 11. What would happen if hydrogen is adsorbed on an active metal surface? 4. The acid produced in the crevice is by 12. If by applying a cathodic current, the time to failure decreases and increases by anodic a) Oxidation of Fe + + polarization, which mechanism of SCC is b) Reduction of F e + + suggested? Active path mechanism or hydroc) Hydrolysis of M + + gen embrittlement? d) Hydrolysis of M+Cl 13. Explain the main difference between the pre-existing active path, strained gener- 5. To prevent the crevice corrosion of 18-8 steel, ated active paths and specific adsorption the following measure is very important: theories. 14. State three major factors which cause the a) Decreasing the velocity of water concentration of caustic. Explain briefly b) Painting the anodic surface the mechanism of caustic cracking in c) Avoiding contact with non-hygroscopic boilers. material d) Using welded joints in preference to bolted joints CREVICE CORROSION Multiple Choice Questions Mark one correct answer from the following: 1. Crevice corrosion is caused by a) metallic contact between two different metals b) high water velocity Review Questions 1. What are the three important metallurgical factors which affect crevice corrosion? Explain the reasons. 2. Explain the mechanism of crevice corrosion based on concentrations of metal-ion cell and differential aeration cell. 3. Explain the mechanism of crevice corrosion. 2 60 Principles of Corrosion Engineering and Corrosion Control 5. Which of the following is true about the oxidation-wear mechanism? a) A thin layer of a protective film which is formed on the metal surface is ruptured and oxide debris is produced b) When metals are placed in contact, the oxide layer is broken at all points c) No frictional forces are involved in the theory based on oxidation-wear concept d) Cold welding or fusion occurs at the interface between the contacting surface How and Why Questions 1. How is oxygen depleted in the crevice? 2. Why does the surface adjacent to the crevice not become the anode? 3. Why hydrolysis causes a fall in pH in the crevice solution? 4. How is acidity produced in a crevice? FRETTING CORROSION Multiple Choice Questions Mark one correct answer from the following: 1. Select the best fretting resistant material from the following: a) Hard tool steel on tool steel b) Cast iron on cast iron with phosphate coating c) Aluminum on cast iron d) Chrome plate on chrome plate 2. Fretting corrosion occurs at contact areas between material under load and subjected to a) b) c) d) attack by environment static load vibration and slip cavitation damage How and ^^ Q uestions 1. What are the four important factors which affect fretting corrosion? 2. Why film rupture causes fretting corrosion? 3. What is the essential difference between the wear-oxidation and oxidation-wear mechanism? 4. How is the debris which is formed as a result of oxidation affects fretting? Review Questions State the main points of disagreement and agreement between the theory of wearoxidation and oxidation-wear. What is the essential difference between fretting wear, fretting fatigue and fretting corrosion? State two examples of each. State four important factors which affect fretting. 4. State four methods of prevention of fretting corrosion. C AVITATION D A M A G E 3. Which one of the following is not a requirement for fretting corrosion to occur? a) The interface must be under load b) Relative motion between contacting surface c) Sufficient relative motion to produce slip d) Presence of a fatigue crack 4. Which one of the following is not an example of fretting corrosion? a) b) c) d) Fretting of blade roots of turbine blades Bolted tie plates on railroad rails Fretting of electrical contacts Failure of a ship propeller Multiple Choice Questions Mark one correct answer from the following: 1. Cavitation corrosion damage on metals may a) be the result of chemical changes in the environment b) occur in slowly moving liquids Types of corrosion: materials and environments c) be limited sometimes by inhibitor d) be easily eliminated by cathodic protection 2. Cavitation damage is caused by a) degradation of a solid body resulting from exposure to cavitation b) loss of original material from a solid surface c) collapse of cavities within a liquid when subjected to rapid pressure changes d) loss of mechanical strength of the metal 3. Which of the following are the examples of cavitation corrosion? a) Cracking of 18-8 steel pipe transporting brackish water b) Vibration in the blades of turbines c) Loosening offlywheels from shafts d) Failure of a ship propellers 4. Cavitation corrosion may be prevented by a) coating of materials b) allowing dust particles to interact with the surface c) applying cathodic protection d) reducing the size of cavities 5. Cavitation in a pump can be observed by a) failure to reach the discharge pressure b) quiet operation c) regular power consumption 6. A classical example of cavitation damage is shown by a) a bolted tie plate on railroad rails b) interface between press fitted ball bearing shaft c) pump impeller d) boiler tubes 2 61 2. Why is the magnitude of cavitation corrosion in salt water greater than in distilled water? 3. What is the most important factor which leads to the cavitation damage of pumps? 4. How does cathodic protection prevents cavitation damage? Review Questions 1. State clearly the difference between cavitation corrosion and cavitation damage. 2. Briefly describe the mechanism of cavitation damage. D EZINCIFICATION Multiple Choice Questions Mark one correct answer from the following: 1. In dezincification of brass a) copper is selectively attacked b) zinc is selectively attacked in preference to copper c) the attack is confined to both zinc and copper d) zinc is selectively attacked only if its concentration is more than 35% 2. Plug type of dezincification is observed a) in admiralty brass when exposed to water (pH = 8.0) b) in brasses with a high zinc content in alkaline environment c) in heat exchanger tube of 90-10 Cu-Ni alloys d) in tooth of a gear wheel 3. Which one of the following steps does not represent the mechanism of dezincification of a 70-30 brass? a) b) c) d) Oxidation of copper Plating of copper ions Reduction of H2O Anodic dissolution of zinc How, Why and What Questions 1. How are extremely high pressures created on the surface of metals by cavitation? 2 62 Principles of Corrosion Engineering and Corrosion Control c) Composition of alloy d) Cyclic loading frequency 3. A plot of log (AK) vs log (da/dN) a) can be integrated in conjunction with stress intensity solution to predict life of a material b) cannot be used for corrosion fatigue studies as it is purely used for design and evaluation of engineering structures c) cannot be used to mitigate corrosion fatigue d) region II of the plot is not affected by environment 4. One of the following factors does not influence corrosion fatigue: a) b) c) d) e) Load frequency Stress ratio Environmental composition Alloy composition Loading parameter 4. Dezincification can be minimized by a) b) c) d) addition of oxygen addition of 0.2% titanium addition of 0.1 % arsenic using brasses with a zinc content higher than 45% How, Why and What Questions 1. What is the main cause of dezincification of a-brass? 2. What is the essential difference between the plug type and layer type of dezincification? 3. Why zinc is always leached in seawater in preference to copper? Review Questions 1. State the essential difference between the two main theories of dezincification. 2. What is de-alloying? Is it appropriate to call dezincification as de-alloying? 3. State four important factors which affect dezincification. What alloys you would suggest to be used for seawater services? 5. Corrosion fatigue proceeds more rapidly a) b) c) d) at higher frequencies at low frequencies at low stress ratios at very high frequencies C O R R O S I O N FATIGUE Multiple Choice Questions 6. Low cycle fatigue region is characterized by Mark one correct answer from the following: 1. In corrosion fatigue a) the number of cycles for failure increase as the stress is increased b) there is always a greater effect of environmental factors than mechanical factors c) the endurance limit of a material is sharply reduced d) the surface remains bright after fracture 2. The following are the metallurgical factors which affect corrosion fatigue: a) Stress intensity b) Potential a) b) c) d) N< N< N> N> 104 cycles 103 cycles 102 cycles 105 cycles 7. According to film rupture theory a) it is not necessary that a film be present as a metal surface b) the surface becomes active only after the protective film is ruptured c) there is no fall of potential while the film is being destroyed because all parts of the film are not ruptured d) there is a fall of surface potential when the film is being destroyed Types of corrosion: materials and environments 8. Which of the following elements responsible for corrosion fatigue? (Mark YES or NO) a) b) c) d) The basic material Mechanical deformation pH Passivity 2 63 8. How is crack growth related to the kinetics of mass transport? 9. Why is the rate of corrosion fatigue crack propagation increased by high stress ratios? Review Questions 1. State the effect of stress ratio and stress intensity range on corrosion fatigue. 2. Describe the characteristics of a log(AX) vs log( da/dN) curve. Specify the characteristics a) Potential of each region. b) Current density 3. Does the above curve take into consideration c) Film formation the effect of varying stress ratio? d) Passivity 4. State the effect of electrode potential on corrosion fatigue crack growth. 10. In an aggressive environment, cyclic fre5. State the difference between a SCC crack and quency a corrosion fatigue crack. 6. What is the basis of film rupture mechanism? a) affects the fatigue strength significantly Give one piece of evidence in support of the b) does not affect fatigue strength above theory. c) affects the fatigue strength only in highly 7. Differentiate clearly between fatigue, corrosion acidic medium fatigue and stress corrosion cracking. 8. State clearly the effect of stress ratio, stress 11. The fatigue life in corrosive environments amplitude and mean stress on fatigue crack can be increased by propagation. 9. State the factors which play an important role a) introducing tensile stresses in corrosion fatigue crack propagation. b) use of coatings, such as zinc and cadmium coatings c) application of anodic current d) increasing the stress ratio 9. The following are the electrochemical factors which contribute to corrosion fatigue: (Mark YES or NO) How, Why and What Questions 1. How does corrosion affect fatigue life? 2. How can corrosion of certain steels be prevented by cathodic protection? 3. How does film rupture accelerate the propagation of fatigue crack growth? 4. Why is fatigue life of a material drastically reduced in corrosive environment? 5. Why is the rate of corrosion fatigue not appreciably accelerated at a high frequency? 6. What is the difference between stress corrosion cracking and corrosion fatigue? 7. How is the rate of corrosion fatigue increased by a higher stress ratio? REFERENCES Uniform Corrosion [1] Ahmad, Z. and Afzal, M. (1995). Brackish water corrosion of glass lined water heaters in sea-coastal environment. AJSE, 20, (4 A), October. Galvanic Corrosion [2] Ashbaugh, W.G. (1986). Corrosion Failures in Metals - Handbook Vol.11, 9th ed. ASM. [3] Ahmad, Z. (1973). A technical report submitted to Sakaii Housing Corporation. Ministry of Works, Iran. 264 Principles of Corrosion Engineering and Corrosion Control [28] [29] [30] [31] Strehblow, H.H. (1984). Werkstoffe und Korrosion, 35,437. Strehblow, H.H. (1976). Werkstoffe und Korrosion, 27, 729. Chao, C.Y., Lin, L.F. and McDonald, D.D. (1981). /. Electrochem. Sc. and Tech., 8,1187. Pickering, H.W. and Frankenthal, R.P. (1972). /. Electrochem. Soc, 119,129. Kowaka, M., Tsuge, H., Akashi, M., Masamura, K. and Ishimota, H. (1984). Applications of statistics of extremes, Japan Society of Corrosion Engineers, Marzen Pub. Co. Makabe, H. (1978). How to use Weibull probability, papers, JIS, Tokyo (in Japanese). Dezincification [4] [5] Vernick, E.D. and Jr. Heldersback, R.H. (1972). In ASTM-STP, 516, 303, ASTM. Ashbough, W.G. (1986). Corrosion, 11, ASM, 178. Crevice Corrosion [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] Kain, R.M. (1980). Technical Paper No. F-P-37, Laque Center, Wrightsville, March. Jackson, R.P. and Van Rooyen, D. (1974). Localized Corrosion, ASTM-STP, 516, 210. Brigham, R.J. (1974). Corrosion, 396, November. Anderson, D.B. (1978).Statistical aspects of crevice corrosion. Paper TL-26, T-O.P. (Rand, D.) INCO. Kain, R.M. (1979). Corrosion, Paper No. 230, March. McKay, R.J. (1922). Transactions TESOA, 4 1,201. Fontana, M.G. and Greene, N.D. (1987). Corrosion Engineering, 3rd ed. New York: McGraw-Hill. Oldfield, J.W. and Sutton, W.H. (1978). B. Corros. /., 13, 3. Nauman, F.K. and Spies, F. (1978). Eisenforschung, Max Planck Institute, Dusseldorf. ASM(1988). Metals Handbook, 13, 632. [32] [33] Intergranular Corrosion [34 [35 [36 [37 [38 [39 [40 [41 Pitting Corrosion [16 [17 [18 [19 [20; [42 [43 [21 [22; [23 [24; [25 [26; [27; Wranglen, G. (1969). Con. Sc, 9, 585. Foroulis, Z.A. and Thubrikar, M.J. (1975). /. Electrochem. Sc. and Tech., 122,1296. Engell, H.J. and Stolica, N.K. (1959). Con. Sc, 9, 455. Engell, H.J. and Stolica, N.D. (1959). Z. Phys. Chem.N.F., 20,113. Broli, Aa., Holton, H. and Andreassen, J.B. (1976). Werkstoffe und Korrosion, 27, 497-504. Broli, A., Holton, H. and Midjo, M. (1973). Br. Corros. J., 8, 173. Hoar, T.P., Mears, D.C. and Rothwall, G. D. (1965). Con. Sci., 5,279. Kolotyrkin, Y.M. (1964). Corrosion, 19, 216t. Hoar, T.P. and Jacobs, W.R. (1967). Nature, 216,1299. Vetter, K.J. and Strehblow, H.H. (1970). Ber. Bunsenges. Phys. Chem., 74,1024. Sato, N., Kudo, K. and Noda, T. (1971). Electrochem. Acta., 16, 1909. Sato, N. (1971). Electrochem. Acta., 16, 1683. [44; [45 [46; [47; [48 [49 [50 [51: [52 [53 [54 [55. Bain, E.C., Aborn, R.H. and Rutherford, J.B. (1933). Trans. Am. Soc. Steel Treating, 2 1, 481. Tedmon, C.S. Jr. and Vermilyea, D.A. (1971). Corrosion, 9, 376. Strawstrom, C. and Hillert, M. (1969). /. Iron and Steel Institute, 77, 207. Baumel, A., Buhler, H., Schuller, E., Schwaab, H., Schwenk, P., Ternes, W. and Zitter, H. (1964). Corros. Sc, 4, 89. Stickler, R. and Vinckier, A. (1961). Trans. Am. Soc. Metals, 54, 362. Kinzell, A.B. (1952). /. Metals, 4, 469. Streicher, M.A. (1959). /. Electrochem. Soc, 161, 106. Stickler, R. and Vinckier, A. (1963). Corros. Sc, 3 ,1. Aust, K.T., Armaijo, J.S. and Westbrook, J.H. (1966). Trans. Am. Soc. Metals, 59, 544. Aust, K.T., Armaijo, J.S. and Westbrook, J.H. (1967). Trans. Am. Soc Metals, 60, 360. Aust, K.T., Armaijo, J.S. and Westbrook, J.H. (1968). Trans. Am. Soc. Metals, 61, 270. Hanneman, R.F. (1968). Ser. Metall, 2, 235. Houdermont, E. and Tofante, W. (1952). Stahl U. Eisen, 72, 539. Columbier, L. and Hochmann, J. (1965). Stainless and Heat Resisting Steels, New York: St. Martins Press, p. 168. Bond, A.P. (1969). Trans. Met. Soc. AIME, 245,2127. McGuiire, M.F. and Trolano, A.R. (1973). Corrosion, 29, 7, 268. Hodges, R.J. (1971). Corrosion, 27,119. Armaijo, J.S. (1968). Corrosion, 24. Armaijo, J.S. (1968). Corrosion, 2 1. Edeleanu, C. and Snowdenn, P.P. (1985). /. Iron and Steel Institute, 186,406. Wulpi, D.J. (1990). Understanding How Components Fail, ASM. ASM(1986). Failure Analysis and Prevention Metals Handbook, Vol.11, p . 180. Types of corrosion: materials and environments 265 Stress Corrosion Cracking Hoar, T.P. and Hines, J.G. (1956). In Stress Corrosion Cracking and Embrittlement. New York: John Wiley, p. 107. Champion, F.A. (1948). In Symposium on Internal Stresses in Metals and Alloys. Inst. Of Metals, London, 469. Logan, H.L. (1950). /. Res. N. B. S., 48, 99. Forty, A.J. (1959). In Physical Metallurgy of Stress Corrosion Cracking, New York: Interscience, p. 99. Keating, F.W. (1948). Symp. On Internal Stresses in Metals and Alloys, I. O. M. London, 311. Sandoz, G. (1972). Stress corrosion cracking in high strength steels and in titanium and aluminum alloys, Naval Research Lab, 131. Petch, N.J. (1956). Phil. Mag, 1, 331. Ulhig, H.H. (1952). /. Res. N. B. S., 48, 99. Jones, R.H. (1987). Metals Handbook, 13, ASM, 145. Wearmouth, N.R., Dean, C.P. and Parkins, R.N. (1973). Corrosion, 29, 251. Parkins, R.N. (1974). 5th Symposium, American Gas Association, Catalog No. L-30174. Dix, E.P. (1940). Trans. Alml, 11,137. Mear, R.B., Brown, R.H. and Dix, H. (1944). Symposium on SCC of Metals, ASTM/AIME, Phil, 329. Dix, E.H. (1940). Trans. AIME, 11,137. Parkins, R.N. (1980). Corros. Sc, 20,197. Humpries, M.J. and Parkins, R.N. (1969). Proc. 4th Int. Cong. Met. Corros., Australia: Sydney. Bakish, R. and Robertson, W.D. (1956). /. Electrochem. Soc, 103, 320. Sprowls, D.O. and Browny, R.H. (1969). In Fundamental Aspects of SCC Conference, Houston: NACE. Parr, S.W. and Straub, F.G. (1928). Univ. III. Bull, 177. McEviley, A.J. and Bond, A.P. (1965). /. Electrochem. Soc, 112,131. Pugh, E.N. (1971). Theory of Stress Corrosion Cracking, NATO. Green, J.A.S., Mengelberg, H.D. and Yollan, H.T. (1970). /. Electrochem. Soc, 117,433. Robertson, W.D. and Bakish, R. (1956). Stress Corrosion Cracking and Embrittlement. New York: John Wiley, p. 32. Hoar, T.P. and West, J.M. (1962). Proc. Royal Soc, A.268, 304. Orwan, E. (1944). Nature, 154, 341. Paleh, N.J. and Stables, P. (1952). Nature, 169, 842. [82] Paleh, N.J. (1956). Phil. Mag, 1, 331. [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] Coleman, E.G., Weinstein, D. and Kostokar, W. (1961). Acta Metallurgia, 9,491. Heady, R.B. (1977). Corrosion, 3 3,441. Scully, J.C. and Powell, D.T. (1970). Corros. Sci, 10, 719. Perkins, R.N. and Fessler, R.R. (1978). Mater. Eng. Appl, 80. Staehle, R.W., Royuela, J.J., Raredon, T.L., Morin Serrate, E. and Farrar, R.V. (1970). Corrosion, 2 6,451. Copson, H.R. (1959). Physical Metallurgy of Stress Corrosion Cracking, Rhodin, T.N., ed. New York: Interscience, 247-272. Dana, A.W. and Warren, D. (1957). DuPont Engineering Department, ASTM Bulletin, 225, October. Hoar, T.P. and Hines, J.G. (1956). /. Iron and Steel, 184, 166-172. Warren, D. (1960). Proc. 15th Annual Purdue Industrial Waste Conference, Purdue University Mag., 1-19. Bruce Craig, (1988). In Corrosion, Metals Handbook, 13, ASM, 151. Engle, J.P., Floyd, G.L. and Rosenes, R.B. (1959). Corrosion, 15, (2), 69-72. Snowden, P.P. (1961). Nuclear Engineering, 6, 409-411. Wilde, B.E. (1971). Theory of Stress Corrosion Cracking of Alloys, NATO, SCREC. Parkins, R.N. (1972). Br. Corros. J., 7, 15. Port, R.D. (1984). Corrosion-84, New Orleans: NACE. Corrosion Fatigue [98] [99] [100] [101] [102] Griffith, A. (1920). Phil Trans. Royal Soc Lond., 163. Meyers, M. and Chawla, K.K. (1984). Mechanical Metallurgy, Prentice Hall. Inglis, C.E. (1913). Proc. Inst. Naval Arch., 55, 163. Irwin, G.R. (1957). /. App. Meek, 24, 361. Mitchell, M.R. Fundamentals of modern fatigue analysis for design, Fatigue and Micro-structure, ASM, Ohio: Metals Park, p. 385. Feltner, C.E. and Landgraf, R.W. (1969). ASMF, Paper No. 69, DE-59, New York: ASME. Coffin, L.F. (1954). Trans. ASME, 76, 31. Sander, B.I. (1972). Fundamentals of cyclic stress and strain. University of WisconsinMadison, Wisconsin. Paris, P.C. and Erdogan, F. (1963). /. Basic Eng. Trans. ASME, 85, 528. Erdogan, F. (1967). Crack propagation theories, NASA-CR-901. Forman, R.G., Kearney, V.E. and Engle, R.N. (1967). /. BasicEng.Trans. ASME, 89,459. [103] [104] [105] [106] [107] [108] 266 Principles of Corrosion Engineering and Corrosion Control [124] Theories of Fretting Processes in Fretting and Fatigue (1981). Applied Science, pp. 217-211. Fretting Corrosion [109] [110] [Ill] Pugh, E.N., Bertocci, U. and Ricker, R.E. (1987). N. B. S. Lindley, T.C., and Nix, KJ. (1982). ASTM, Std. 853, 340. Widner, R.L. (1990). In ASM Metals Handbook, Vol. 11, 420, Ohio: Metals Park, USA. K EYWORDS Galvanic Corrosion Cavitation Corrosion Anode/cathode ratio It is the ratio of the anodic area to the cathodic area. Bimetallic corrosion Corrosion of two metallics in [112] Hammilt, F.O. (1987). In Metals Handbook electrical contact. The term galvanic corrosion is used generally in place of bimetallic. Corrosion, 13, 9th ed., p. 163. [113] Arndt, R.E. (1977). Corrosion, Paper No. 91. Cathode The electrode of a galvanic cell where positive [114] Khan, A.H. (1956). Desalination Processes and current flows from the solution to the electrode. Multistage Flash Distillation Plants, Elsevier, Concentration cell A galvanic cell in which the emf is due to the difference in the concentration of reactive Oxford, UK. [115] Unterweiser, P.M., ed. (1979). Case Histories constituents of the electrolyte. in Failure Analysis, ASM, 232. Differential aeration cell A corrosion cell formed by [116] Tuthill, A.H. and Schoollmaker, CM. the difference in the concentration of oxygen in the (1965). Ocean Science and Ocean Engineering electrolyte in contact with a metal. Conference, Marine Tech., Washington: SOC. Galvanic series It is a list of metal and alloys based on their relative potentials in a specified environment. The environment generally used is seawater. Galvanic couple A pair of dissimilar metals in contact S U G G E S T E D READING FOR with an electrolyte. Galvanic current The current which flows between GALVANIC CORROSION two dissimilar metals in contact with an electrolyte. Galvanic corrosion The corrosion caused by a gal[117] Dillon, C.P., ed. (1982). Forms of Corrosion, vanic cell. Recognition and Prevention, Chapter 3, p. 45, Galvanic potential The potential arising from two or NACE. more electrochemical reactions preceding simultane[118] Hack, H.P. (1988). Galvanic Corrosion, STP, ously on the metal surface. It is also called 'galvanic cell 978, ASM. potential.' [119] Bradford, S.A. (1998). Practical Self-study Guide to Corrosion Control, Casti Publishers. [120] Fontana, M.G. (1986). Corrosion Engineering, Pitting Corrosion 3rd ed. McGraw-Hill, pp. 4 1-51. Active-passive metal A metal which shows transition [121] Jones, D.A. (1996). Principles and Prevention of from an active state to a passive state, e.g. stainless steel, Corrosion, 2nd ed. Prentice Hall, p. 168. titanium. Passive A metal characterized by low corrosion rate in a certain potential range whose oxidation is the predominant reaction. W E B S I T E FOR GALVANIC Passivity It is the reduction in the anodic rate of reacCORROSION tion because of the formation of a protective film of oxide. [ 122] Pits It is the result of localized corrosion confined to a small area in the metal surface, in the form of a small cavity. Pitting factor The ratio of the depth of the deepest pit S U G G E S T E D READING FOR divided by the average penetration of pits on the metal FRETTING CORROSION surface. Pitting frequency The number of pits per unit area. A n excellent review of fretting corrosion is given Pitting probability P= Np/N x 100, where iVp is the number of specimens in the pit, and N is the total by W aterhouse, R.B. in: number of specimens. [123] Fretting Corrosion (1972). Pergamon Press, Pitting potential (Ep). Also called critical pitting pp. 111-113. potential. It is the most negative potential required to T ypes of corrosion: materials and environments initiate pitting on a metal surface which is in the region of the passive potential. At this potential, the current density shoots up sharply during anodic polarization. Passive-active cell (Active-Passive) A corrosion cell in which the area of the metal surface in the active state is the anode and the area in the passive state is the cathode. Protection potential (£pP). It is the potential below which no pitting is expected to initiate or propagate. 267 martensitic structure is produced by quenching from the austenite phase region. A martensitic structure is body-centered tetragonal. Non-sensitized steels Steels which are specially heat treated so that they are not susceptible to intergranular corrosion are called non-sensitized steels. Precipitation hardening Hardening caused by the precipitation of a constituent from a supersaturated solution. Precipitation hardening steels These are steels with high toughness, high strength and having optimum Intergranular Corrosion creep properties. Aluminum, molybdenum, copper Austenitic A solid solution of one or more elements in and titanium additions are made for intermetallic a face centered cubic iron. strengthening. Austenitizing Heating a steel into the austenitic tem- Residual Stress Stress that is within a metal as a result perature range so that its structure becomes austenitic. of plastic deformation. Carbide A compound of iron and carbon Fe3C Precipitation of carbides At the grain boundaries of (cementite) is a hard and brittle substance. stainless steels, chromium reacts with carbon to precipConcentration cell A corrosion cell formed by a differ- itate chromium carbides. Carbides of Ni, molybdenum, ence in the concentration of some constituents in the titanium and niobium, are well-known. Chromium electrolyte. carbide is represented by 0*23 C6Chromium depletion Sensitization of 18-8 stainless Sensitization When austenitic steels, such as AISI steel results in the precipitation of chromium carbide Type 304 are cooled slowly through the range at grain boundaries with depletion of chromium in 550-580° C, they are frequently susceptible to severe regions adjacent to the grain boundary. intergranular attack when exposed to a corrosive Diffusion (Self) It is the migration of atoms in a environment. This phenomenon has been termed material by thermal random-walk. 'sensitization which means the metal is sensitive to Diffusion (Interstitial) The migration of interstitial grain boundary corrosion. atoms in a matrix lattice. Sigma-phase In stainless steels containing Mo, stainless Duplex steels Alloys containing both the body centered steel types, such as 316 and 317, and in high nickel alloys cubic alpha-phase and the face centered gamma-phase containing Mo, heating to temperatures in the range are known as duplex steels. They may be low car- 700-850° C causes the precipitation of sigma-phase. bon (less than 0.03%) low chromium alloys (less than Solution-heat treatment Heating an alloy to a temper20% Cr) with a continuous austenitic matrix or high ature which causes one of the constituents or more to chromium (>20%) alloys with a ferrite matrix. enter into solid solution. A solid solution is a single, Dislocation A crystalline imperfection in which a lattice solid crystalline phase containing two or more species. distortion is centered around a line. Streicher test This is a test to determine the susceptibilFerritic A solid solution of one or more elements in a ity of Fe-Ni-Cr alloys to intergranular corrosion. The BCC iron (a-phase). specimen is exposed in a solution of 50 wt% H2SO4 4Ferritic steel Binary iron chromium alloys contain- 25g/literCuSO 4 forl20h. ing 12-30% Cr. Their structure remains ferritic under Stabilized steels Steels containing elements which form normal heat treatment condition. carbides that are more stable than chromium carbides. Grain boundary A surface imperfection which sepa- Niobium and titanium are used as stabilizing elements. rates grains of different orientation in a polycrystalline Steel AISI Types 321 and 347 are stabilized grades of aggregate. steels. Heat Affected Zone (HAZ) The portion of a metal Transpassive region The region of an anodic polarwhose microstructure is altered by heat, such as during ization curve above the passive potential range which is welding or heat treating. characterized by a significant increase in current density Intergranular corrosion Preferential corrosion occur- with an increase in the potential to the noble range. ring at grain boundaries or at regions adjacent to the Weld decay This is the intergranular attack which grain boundaries. occurs in stainless steels or certain nickel-base alloys Inter-crystalline corrosion Same as intergranular as a result of sensitization of the steel in the sensitizing corrosion. temperature range. It occurs a short distance away from Knife-line attack Intergranular corrosion of the weld bead. an austenitic steel (stabilized grade) extending only a few grains away from the fusion line of a weld Stress Corrosion Cracking during sensitization of steel. Martensitic stainless steel Essentially Fe-Cr alloys Active path (pre-existing) Structure features, such as containing 12-17% Cr with 0.15-1.0% C so that a impurity, segregates, precipitates, which exist on the 268 Principles of Corrosion Engineering and Corrosion Control Stress intensity It is denoted by symbol K and it denotes the intensity of the applied stress at the tip of the crack. K is a function of nominal stress, crack length and component geometry. Stress intensity factor (K\) grain boundaries and cause a local galvanic cell to be set up. These cells provide active paths for the propagation of cracks. Active path (strain generated) If the protective film is disrupted, the exposed metal is attacked until the disrupted film is again formed. As a result of disruption of the protective films, the metal becomes active. The plastic strain in the underlying metal is responsible for disrupting the film. This type of disruption provides a strain generated active path. Adsorption This is the surface retention of solid, liquid or gas molecules, atoms or ions by a solid or liquid. Anodic zone Anodic zones are produced as a result of microstructural differences, compositional differences, grain boundaries and sub-grain boundary or heterogeneous effects of stress or prior cold work. Brittle fracture Brittle fracture proceeds along characteristic crystallographic planes called cleavage planes and has a rapid crack propagation. Most brittle fractures are transgranular. They proceed with very little plastic deformation. Cleavage. This is the splitting of a crystal in a crystallographic plane. A cleavage fracture is marked by characteristic pattern, such as river markings, herring bones, tongues, etc. Ductile fracture A fracture or tearing of metal associated with gross plastic deformation. Film (breakdown and repassivation) This is the rupture of a film and the reformation of afilmon the metal surface. An emergent slip step breaks the thin protective film. The film may reform again on the surface except a small area. The latter is called 'repassivation.' Film rupture This is the breakdown of a protective film on a metal surface. Plane strain This is a stress condition in which the strain is zero in a direction normal to both the axis of the applied tensile stress and direction of crack growth (parallel to the crack). This condition is achieved with thick specimens. Plane strain fracture toughness (K\c) The minimum stress intensity required to cause catastrophic failure. Plane stress This is a stress condition in which the stress in the thickness direction is zero. This stress arises as a result of strain gradient. This condition is achieved in thin specimens. Residual stress The stress that is confined in a material as a result of plastic deformation. Slow strain rate technique The specimen is pulled in uniaxial tension at a controlled very slow strain rate in a given test environment to detect its susceptibility to SCC. Slip Plastic deformation by irreversible shear displacement of one part of crystal with respect to another in a definite crystallographic direction and on a specific crystallographic plane. Strain rate This is determined from the specimen gauge length during tensile test. In evaluation of SCC, a constant strain rate of lo-6 s e c -l is applied to a tensile specimen in a given environment. It represents the rate of deformation of a metal. Ki=KxdxVc where K\ is stress intensity factor derived from crack length C is crack length M is stress across the narrow section of the specimen k is constant. It is also reported as K\ = d^/ay, where a is the crack length and y is the correction factor. Stress intensity factor for SCC (Ki_scc)- The value of stress intensity above which cracks grow because of stress corrosion cracking. Stress raisers Discontinuities in the structure which produce local increases in the stress. Stress relief This occurs when the alloys after welding are subjected to stress relief by appropriate heat treatment in order to reduce the residual stresses. This is also called 'post-weld heat treatment.' Stress level The magnitude of stress required to cause stress corrosion cracking. Threshold stress intensity factor AKt^. The value of AK[ below which crack propagation does not occur. Transgranular cracking This is the cracking which takes place across the grains. Hydrogen Damage Caustic embrittlement It is the phenomenon of stress corrosion cracking of steels in caustic soda at elevated temperatures. Hydrogen blistering The formation of blisters on the metal surface due to excessive internal pressure of hydrogen. Hydrogen embrittlement The ductility of a metal is reduced and it cracks due to absorption of hydrogen. Hydride formation The degradation of mechanical properties and cracking of metals, such as titanium, resulting from hydride formation due to pick-up of hydrogen. Hydrogen Induced Cracking (HIC) This is characterized by the brittle fracture of a ductile alloy normally in the presence of hydrogen (same as hydrogen embrittlement). Hydrogen sulfide cracking This refers to cracking of high strength steels in a sour gas environment encountered generally in oil drilling. Hydrogen trapping This is the binding of hydrogen atoms to impurities and structural defects. Hydrogen may be trapped at mobile dislocations or grain boundaries. Types of corrosion: materials and environments Seasonal cracking This refers to the cracking of brass in ammoniacal environment. 269 Cavitation Damage Cavitation The formation and collapse of innumerable tiny cavities within a liquid subjected to intense pressure changes. Cavitation corrosion This is the deterioration of a metal subjected to a conjoint action of cavitation and corrosion. Cavitation damage This is the damage to a material caused by cavitation. Cavitation erosion This is the gradual loss of material from a solid surface as a result of continuous exposure of a material to cavitation. Erosion Removal of surface material by individual impact of solid and liquid particles. Protective film This is generally a continuous film of oxide on a metal surface which acts as a barrier between the metal and the environment. Shockwave A wave of the same nature as a sound wave but of much greater intensity. Cyclic loading This is a periodic or non-periodic fluctuating loading applied to a test specimen (also called fatigue loading). Ductile fracture Tearing away of a metal accompanied by appreciable plastic deformation. Constant amplitude loading A type of loading in fatigue in which all peak and valley loads are equal. Fatigue life The number of loading cycles that a specimen can sustain before failure. Fatigue limit This is the limiting value of medium fatigue strength. Fatigue crack growth The rate of crack growth under constant amplitude fatigue loading (da/dN). Fracture toughness This is a measure of the resistance of a material to crack extension (MPaVrn). Fracture toughness (plane strain) K\c The crack extension resistance under conditions of crack tip plane strain. Loading amplitude (or alternate loading). This is onehalf of the range of a cycle. Loading ratio Dezincification De-alloying It is the selective leaching of one or more component of an alloy. De-aluminification Selective leaching of aluminum in aluminum containing alloys. De-nickelification Selective dissolution of nickel from nickel containing alloys. Dezincification Selective leaching of zinc from copperzinc alloys. Graphitic corrosion This is the deterioration of cast iron in which the iron-rich matrix is selectively leached or converted to corrosion products and a weak residue of graphite flakes is left behind. Leaching Dissolution of a component of an alloy or dissolution of a metal by leaching agents, such as sulfuric acid or sodium hydroxide, etc. R= Minimum load a m i n . . —- = (a is given in MPa) Maximum load crmax Mean load 8m = ( a is given in MPa) Corrosion Fatigue Anodic polarization Polarizing a metal in a positive direction by applying an external EMF. Axial strain A linear strain parallel to the longitudinal axis of the specimen. Brittle fracture The separation of a solid accompanied by either little or no plastic deformation. Cathodic polarization Polarizing a metal surface in an active (positive) direction by application of an external EMF. Corrosion fatigue Cracking produced by the conjoint action of repeated or fluctuating stress and a corrosive environment. Crack size A linear measure of a principal plane dimension of a crack. Crack extension An increase of crack length. Crack opening displacement This is the displacement at the tip of the crack (CTOD). Plane strain A stress condition in linear elastic fracture mechanics in which there is zero strain in the direction normal to the axis of applied tensile stress and direction of crack growth. It is achieved in thick plate, along a direction parallel to the plate. Plane stress A stress condition in linear elastic fracture mechanics in which a stress in the thickness direction is zero. It is achieved in loading thin steel along a direction parallel to the surface of a sheet. Plane stress fracture toughness, KQ The value of stress intensity at which crack propagation becomes rapid in sections thinner than those in which plane strain conditions occur. Stress intensity factor is given as iCa = a y / 7ra/(a/w), whereas a/w—• 0, f(a/w) -> 1.0. In the above expression f(a/w) is the correction factor which depends on the width (iv) and flow size (a). The critical value of stress intensity, K\Q, is commonly known as plane stress fracture toughness. It is expressed in units of MPa. S-N diagram A plot of stress amplitude against the number of cycles to failure. Stress corrosion cracking A process of cracking which occurs by the action of a corrodent and a stress. Stress concentration factor It is the ratio of greatest stress in the region of a notch or discontinuity to the corresponding nominal stress. 270 Principles of Corrosion Engineering and Corrosion Control Subscripts for loading conditions: KQ = plane stress fracture toughness. K\ = stress intensity factor. K\c = plane strain fracture toughness. Kiscc = threshold intensity for stress corrosion cracking. iCth = threshold intensity for SCC. AK = the range of stress intensity factor during a fatigue cycle. Stress intensity (Kj) It represents the intensification of applied stress of the tip of a crack of known size and shape. The range of stress intensity factor is given by AK = K max - K min (MPa Vrn~). Stress ratio (A or R) It is the ratio of minimum stress (5 m j n ) to maximum stress (S max )- CATHODIC PROTECTION in the direction of local action anode potential, thus reducing further the potential difference athodic protection is a proven corrosion between the anodes and cathodes. Complete control method for protection of under- cathodic protection is achieved when the metallic ground and undersea metallic structures, such structure becomes cathode (more negative). The as oil and gas pipelines, cables, utility lines severity of corrosion is directly proportional to the and structural foundations. Cathodic protection magnitude of the difference of potential between is now widely applied in the protection of oil the anode and the cathode, hence by eliminating drilling platforms, dockyards, jetties, ships, sub- this difference, corrosion may be eliminated. marines, condenser tubes in heat exchangers, bridges and decks, civil and military aircraft and 5.2 B A S I S OF CATHODIC ground transportation systems. The designing of cathodic protection systems P R O T E C T I O N is rather complex, however, it is based on simple electrochemical principles described earlier Figure 5.3 illustrates the simple principle of in Chapter 2. Corrosion current flows between cathodic protection. On application of an external the local action anodes and cathodes due to the current, the difference of potential between the existence of a potential difference between the cathodes and anodes on the structure decreases. two (Fig. 5.1). As shown in Fig. 5.2, electrons Corrosion stops when potential of cathode released in an anodic reaction are consumed in becomes equal to the potential of anode. The the cathodic reaction. If we supply additional anode would become more negative and the electrons to a metallic structure, more elec- cathode more positive. Cathodic protection is, trons would be available for a cathodic reaction therefore, achieved by supplying an external negwhich would cause the rate of cathodic reac- ative current to the corroding metal to make the tion to increase and that of anodic reaction to surface acquire the same potential to eliminate decrease, which would eventually minimize or the anodic areas. The anodic areas are eliminated eliminate corrosion. This is basically the objective by transfer of electrons. After a sufficient current of cathodic protection. The additional electrons flow, the potential of anodic areas would become are supplied by direct electric current. On applica- negative enough for corrosion to stop. tion of direct current, the potential of the cathode shifts to the potential of the anodic area. If suf- (a) There must be an anode, a cathode, an elecficient direct current is applied, the potential trolyte and a metallic path for the transfer of difference between the anode and cathode is elimelectrons. inated and corrosion would eventually cease to (b) A source of DC current to supply electrons. occur. (c) Sufficient direct current should be applied As the cathodic current increases (more transto eliminate the potential difference between fer of electrons), the cathodic reaction polarizes the anode and the cathode. 5 ,1 INTRODUCTION C 272 Principles of Corrosion Engineering and Corrosion Control Figure 5.1 An electrochemical cell. Corrosion cell between two areas on single metal surface. Current flows because of a potential difference that exists between anode and cathode. Anions (e.g. F e + + ) leave at the anode which corrodes and are accepted at the cathode where corrosion is prevented. Electrons are insoluble in aqueous solutions and move only in the metal - /Seetrotyte " Figure 5.2 Simple electrochemical cell 5.3 WORKING OF 5.4 FACTORS LEADING CATHODIC PROTECTION SYSTEM Figures 5.4a and b show how, in principle, a cathodic protection system works. Figure 5.4a shows a buried pipeline with anodic and cathodic areas prior to the application of the cathodic current. Figure 5.4b shows the same pipeline after the cathodic protection. TO C O R R O S I O N O F UNDERGROUND METALLIC S T R U C T U R E S The corrosion encountered by metals in aqueous solutions is always electrochemical in nature. It occurs because of the formation of anodic and cathodic areas and the flow of electrons through Potential Difference (a) Anode ] Cathode <t>) Anode p Cathode More negative More positive a: Potential difference before direct current is applied b: Potential difference after the application of current Figure 5.3 Effect of external current in the metallic structure Sectrolyte © © © ©©© y ® © © TO TO ©v ® © O ® © Q MM Arte die reactions ® © .0 Figure 5.4a Anodic and cathodic reactions on a metal surface Electrolyte © f ©f t iP) t (£) t (F) © t @ t ® f ©f t © t ® t ® @®©®©©@© Electrons from external source Decreased anodic reactions Figure 5.4b Increased cathodic reaction and decreased anodic reaction (insignificant) caused by introducing electrons from an external source 2 74 Principles of Corrosion Engineering and Corrosion Control would form a galvanic cell. Several examples of galvanic cells are given in Chapter 4. A galvanic cell is basically a corrosion cell. the metallic path. Structures, such as pipelines buried in the ground are affected by the presence of concentration cells, galvanic cells and stray currents. Some typical cells are described below. (a) Concentration cell A concentration cell is formed by the differences in concentration of salts, degree of aeration, soil resistivity and degree of stress to which the metals are subjected. The above differences cause the formation of anodic and cathodic areas on the surface which lead to corrosion. (b) Galvanic cell A galvanic cell is formed when two metals differing in potential are joined together. For instance, if copper is joined to aluminum, aluminum would corrode because it has a more negative potential (—1.66 V) than copper (+0.521 V). Copper being less active becomes the cathode and aluminum becomes the anode. But if iron is joined to aluminum, the iron corrodes (in seawater), due to the passive film on aluminum which causes it to behave like a nobler metal than iron (but not nobler than copper). The formation of such galvanic cells often leads to the corrosion of underground buried structures. A steel plate with copper rivets 5 .4.1 EXAMPLES OF C ONCENTRATION C E L L S (a) Differential Aeration Cell Suppose, a pipeline is buried in a completely uniform soil and that some areas of the line have a free supply of oxygen and other areas have a restricted supply. The part of the pipe buried in the soil with a free supply of oxygen (high oxygen content) would form the cathode and the part with a less supply of oxygen or poorly aerated forms that anode (Fig. 5.5). The current (Fe + + ions) would flow in the soil from anode to cathode resulting in the corrosion of the pipe end buried in poorly aerated soil. Such a cell is commonly called a differential aeration cell. (b) Dissimilar Electrode Cells Joining of a new pipe to an old pipe results in the corrosion of the new pipe, as the new pipe becomes anodic to the old pipe. The old pipe by virtue of film formation has a less active potential Loose sandy sod Hard packed earth roadway loose sandy soil High oxygen content Low to normal oxygen content High oxygen content Cathode Anode Cathode '••"•"•'A ^X-M^f^^mz^^ Figure 5.5 Formation of differential oxygen cells. (From CORRINTEC, USA, Presentation at KFUPM Dhahran, Saudi Arabia, March 5,1982) Cathodic protection 275 Old pipe (Cathode) New pipe (Anode) > Old pipe (Cathode) Figure 5.6 Dissimilar electrode cell formation by joining of an old pipe with a new pipe than the new pipe and it, therefore, becomes the cathode (Fig. 5.6). Corrosion also occurs at the coating flaws when the new pipe is connected to the main with a bonded coupling. and areas of stress at welds form the anode and corrosion occurs at these areas. (d) Different Types of Soils Suppose a pipeline is buried in two types of soils, one being a clay soil and the other sandy soil. Anodic areas may be formed in regions of high The area of pipe buried in the clay soil would stresses and cathodic areas in regions of low form the anode, whereas the area buried in the stresses as shown in Fig. 5.7. Bright steel at threads sandy soil would form the cathode, because of (c) Stress Cells /T\f^\ /r%^ f C athode Cathode Cathode /r\^ Cathode Ky^/ Area o f stress at weld Anodic area Bright steel a t thread Anodic area ^%J/ Bright steel at thread Anodic area Figure 5.7 Stress cell. Corrosion caused by areas of stress and areas of bright steel on an underground pipeline 2 76 Principles of Corrosion Engineering and Corrosion Control excess of oxygen and higher porosity. The pipe contact with wet soil (high moisture) would corin the clay soil, therefore, corrodes. Corrosion rode, whereas the pipe in contact with the dry soil caused by different types of soils is shown in would not corrode. The area of the pipe in conFig. 5.8. tact with the dry soil becomes the cathode and the area in contact with the wet soil, the anode. Differential concentration cell is shown in Fig. 5.9. (e) Different Moisture Contents If a pipe is buried in two types of soil having different moisture contents, the area of the pipe in At the anode: Fe —• F e + + + 2e~ At the cathode: 0 2 + 2 H 2 0 + 4e~ -> 4 0 H " Clay soil Sandy soil Figure 5.8 Corrosion caused by different types of soils ^ 0 \ l Wet soil >> > \*\ X V, \ \\\\;\\\\\\\x\\\\\\^ w ;/..•• \\\\\\\\\\\v\\A\\\v\v.\\\vl^ Figure 5.9 Differential concentration cell Cathodic protection 2 77 5.4.2 GALVANIC C E L L S 5,5 ELECTRICAL B A S I S OF This type of corrosion arises by two dissimilar metals in contact with each other. In pipe lines it can occur when replacing an old section of the pipe by a new section. The same problem arises when the new lines are laid adjacent to the existing lines and connected across by means of branches. Any dissimilarity within the two metals would bring about this type of corrosion. The more active metal in this situation becomes the anode and the less active, the cathode. For example, if zinc is connected with steel, zinc would corrode as it would be anodic (more active potential), and by sending electrons to the cathode (steel) it would protect the steel. Any metal with a more active potential would always form the anode and the one with less active potential the cathode. The conventional current in the soil always flows from anode to cathode. Figure 5.10 shows a cast iron box containing copper tubes. Electrons would flow in the metallic path from the cast iron box to the copper tubes as the cast iron has a more negative potential and so would be the anode, whereas copper has a noble potential would be the cathode. Eventually the box would be corroded and the copper tubes protected. By installing a magnesium anode, a hot water steel tank may be protected against corrosion as shown in Fig. 5.11. In this case magnesium being very active would become the anode and corrode. CATHODIC PROTECTION Let us suppose a pipe is buried in a soil. A difference of potential exists between the anode and the cathode, i.e. AE = (£a - Ec) A £ is the difference in potential. If R& + Re is controlling resistance, and £a — Ec has a finite value of circuit resistance, the corrosion current (Jc) would flow. If complete freedom from corrosion is desired, Ic must be made zero. £a = ~£c = o (i = - \ Ohms law where 7C = corrosion current (A) £a = cathode potential Ec = anode potential jRa = anode resistance (ohms) Re = cathode resistance. The corrosion current can be made zero by making £ a + i ^ equal to infinity, for example by pointing. This can also be achieved by equalizing the potential difference between Ec and £a or making (Ec — £ a ) equal to zero. The structure can be made cathode (negative) by supplying an electric current from outside until its potential becomes equal to the potential of the anode, and the difference between Ec and £a completely disappears. To further illustrate the principle, the equivalent circuit (a) of a cathodically protected metal is shown in Fig. 5.12A. Complete cathodically protected metal is shown in Fig. 5.12B. Here EA-Ec = hRA + Rc(h+h) i 1 = ( -L„—: I ' Tubes V "I " Galvanic corrosion EA-(Ec + RCh) # A + Re where £A = open circuit anode potential EQ = open circuit cathode potential #A = effective anode resistance Figure 5.10 Currentflowsin the metallic path from case iron box into copper based tubes 2 78 Principles of Corrosion Engineering and Corrosion Control Mild steel tank Pitting Figure 5.11 Galvanic corrosion in a hot water tank Re = effective cathode resistance I\ = current from anodic area Ii = current from the external anode. If corrosion is not to occur, I\ must be zero, therefore, (EQ + Re h) = £A- It means that sufficient current must flow through RQ for the potential of the anode to be equal to the potential of the cathode (open circuit potential). The above relationship suggests that enough current must be provided by an external anode to flow through the cathode resistance RQ to make the cathode potential (EQ) equal to the anode potential (£A)« When this condition, £A = ( £c + Re h)y is achieved, no corrosion would occur and the structure would be cathodically protected. Figure 5.12 illustrates the complete and incomplete protection. In part(b) current is not flowing to all anodic areas, whereas in part(c) current is supplied to all anodic areas. Figure 5.12A shows partial protection and Fig. 5.12B shows complete protection, E\ = {EQ + RQ h). 5 .5.1 S O U R C E S OF CURRENT It has been shown above that there must be a source of current to supply electrons to the areas of the metal which is corroding. In a metal buried in ground, anodic areas corrode by release of electrons and if an equal number of electrons are not introduced from an external source, the metal would continue to corrode. An external anode which supplies such current is called auxiliary anode in the electrochemical cell and referred to as anode in a cathodic protection system. Electrodes of graphite, cast iron, platinum and titanium act as conductors of electricity and supply the desired current to the structure to be protected. The conductors are energized by a DC source. The rate of consumption of anode electrodes C athodic protection 279 *U External anode f*c External DC Source I|-H (a) Equivalent circuit (li>0) (A) Partial protection ll—ll (b) !*hysicaf ctrcyit (c) Physical circuit (8) Complete protection Figure 5.12 Equivalent circuit of cathodically protected metals. (From Spencer, K.A. (1960). The Protection of Gas Plant and Equipment from Corrosion. Joint Symp. Corrosion Group of the Society of Chemical Industry, Sept 22-23) 2 80 Principles of Corrosion Engineering and Corrosion Control A structure without cathodic protection is analogous to an electrochemical cell, and is shown in Fig. 5.13a. In a cathodic protection system, the current (electrons) leaves the anode and enters both the anodic and cathodic areas of the buried metal or by analogy to the electrochemical cell, the current from the auxiliary anode (AA) enters both the anodic and cathodic areas of corrosion, A and C, where A = anode and C = cathode, returning to the source of DC current (Fig. 5.13b). Naturally, when the cathodic areas are polarized to the open circuit potential of the anode, all the metal surface is at the same potential and the local active current would no longer flow. The metal does not corrode as long as the external current is maintained. The electrodes A and C would achieve the same potential if they are polarized by the external current supplied by AA. The potential of the cathode would shift close to the potential of the anode, and after some time both A and C would acquire the same potential, thus, by supplying an external current it is possible to prevent corrosion. As the two electrodes, A and C, become more negative, the method is called cathodic protection. Cathodic protection is, therefore, a method of preventing the corrosion of a metallic structure by making it a cathode with respect to the auxiliary anode (anode supplying the current). depends on the material used. The more expensive the anode, such as titanium, the lesser the rate of consumption. The above anodes are called impressed current anodes. The other types are sacrificial anodes which corrode in the soil and generate electric current required for protection of the structure. These anodes are not energized by an outside DC source. Details of galvanic and impressed current anodes are provided later in this chapter. •~~~~~~^^^^— *\ r I \' ( 1 v^. (a) Structure without cathodlc protection A « anodic area of structure C = cathodic area of structure Auxia^anode ^ 5 6 ELECTROCHEMICAL ^ THEORY OF CATHODIC P ROTECTION *M (b) Structure witfi cathodic protection Figure 5.13 Diagram illustrating the principle of cathodic protection In Fig. 5.14a, £0,c represents the reversible potential of the cathode and £0>a represents the reversible potential of the anode. /?a represents the Tafel slope of the linear portion of anodic curve (slope = 0.12V/decade), and /3C the cathodic slope. Econ represents the corrosion potential, where the rate of oxidation equals the rate of reduction. 7corr is the corrosion current corresponding to corrosion potential (100 |xA/cm2) (Figs 5.14a and b). Extrapolation of the linear portion of the anodic curve to the equilibrium potential of the anode, gives the exchange current density for the anodic reaction (/0,c). Io is defined Cathodic protection 2 81 noble QA 03 02 A Simulated experimental anodic polarization curve © Emulated experimental cathodic polarization curve 2 04 0 I m B 1 active ~0J 10* ,i . jjiiiil i uuffll muwl 1.1.1ii Atl i ami i uiiiiil i ujwiL.i yjuiil 12 1 a 3 4 s 10" lO" 10° 10* 10 10 10 10 Cyrrent density, A/cm2 Ui U Figure 5.14 a Anodic and cathodic Tafel curves for corrosion of metal, M, showing the principle of cathodic protection. (From Jones, D.A. (1981). Principles of Measurement and Prevention of Buried Metal Corrosion by Electrochemical Polarization. In Underground Corrosion, STP 741, ASTM. Reproduced by the kind permission of ASTM, Philadelphia, PA, USA) as equal and opposite ('exchange') current per unit area on a reversible electrode at equilibrium potential (reversible potential). I0 is a kinetic factor and has a significant influence on the rate of corrosion. For example, pure zinc has no reaction in reducing acids (e.g. H2SO4), whereas impure zinc evolves hydrogen. The IQ ( H + /H2) on pure zinc is much lower than IQ ( H + /H2) on commercial zinc. The values of I0,a and IQyC in the diagram are 10~2 and 1.0 |xA/cm2, respectively. The relation between the current and overvoltage is given by the Tafel equation discussed in Chapter 3. and the equilibrium potential E0). If the cathode, for instance, is polarized by applying an external current, the cathodic process would be accelerated (M + + e -> M) and the anodic process (M —> M + + e) would be retarded. Similarly, an anodic process could be accelerated by supplying an external current to the anode and the cathodic process suppressed. In both cases, however, charge would be conserved and the current applied would be equal to the difference between the ic and ia- ^applied = k - h (where ic is current density at the cathode and ia is current density at the anode). If ia is made smaller than ic, /applied would i be equal to ic and only the cathode would be h polarized. Under such conditions, a linear Tafel behavior would be observed. On applying an elecMg —>• M g 2+ + 2e (Tafel equation) trical current (/applied) equal to 104 |xA/cm2, and continuously polarizing the cathode, the corroM+ + e -* M sion current would be reduced from 100 |xA/cm2 2 Here rj denotes the over-voltage of an anode at Icon to 1 |xA/cm at Ja in Fig. 5.14a which (the difference between the actual potential of is a 99% reduction in the rate of corrosion of the electrode at certain value of applied current M (Fig. 5.14a). By continuously increasing the 2 82 Principles of Corrosion Engineering and Corrosion Control noble Oh | o > r-*rte- 2r L% 5 - (Oh^ *- —3- N ^ ^ ^ 1 .^-"^ ^\ -0.4 r * + e —» X" M — » M* + e s 1 P a * 0 .15 VOltS l/\. y* P = 0.10 VOltS a m \^^ L Eeorr - . — — — — — — - - - « - — . ^ ^ ^ ^ " % = 0*05 volts •*corr^ss^^ -0.6 h m c ***S^ [ -*^^—~~~ ^ < ^ ^^^ /^ JS I am Jr>\ * i c ^^^^i £ -0.8 h r -1.0 io-3 re 1 UJIilll 104 1.1.1 llllll 10s active i i i u i i i l 1 1 1 iiinl i i l l mi! i U4.Mll 1 11 llllll . 101 10° 10s 10* 103 Current density, jWcm2 Figure 5.14 b Effect of anodic Tafel slope on polarization and current necessary for cathodic protection of three different metals with (hypothetically) identical £Corr and ICOrr> but different Tafel f$ slopes. (From Jones, D.A. (1981). Principles of Measurement and Prevention of Buried Metal Corrosion by Electrochemical Polarization. In Underground Corrosion, STP 741, ASTM. Jones, D.A. (1971). Corrosion Science. 11, 439. Reproduced by the kind permission of ASTM, Philadelphia, PA, USA) applied current, the value of JCOrr can be made negligible and corrosion can be stopped. In order to reduce the current density to zero, an applied current density of 106 |xA/cm2 would be required (point P in Fig. 5.14a). Figure 5.14b (for three different metals) shows how the amount of corrosion is controlled by the anodic Tafel slope, /?a. The lower the value of /?a, the greater is the reduction in the rate of corrosion for a given cathodic polarization. Lower /3a means lower polarization required to achieve the required degree of protection. The Tafel slope fia gives the measure of the amount of cathodic polarization required to reduce the anodic reaction by one order of magnitude. In order to reduce Icorr, from 102 |xA/cm2 to 1 |xA/cm2, a potential change of 100 mV (0.1 V) is required for Pz= 0.05. To achieve the same amount of reduction in 7Corr a potential change of 0.2 V and 0.3 V would be required, respectively, for /*a= 0.1V and /3a= 0.15 V. Thus, lesser polarization is needed with a lower value of /3a. It is, however, not necessary to reduce the rate of corrosion to zero at £0,a (Fig- 5.14a) as this would require a large amount of current which might not be economically justified. It is only sufficient to reduce the corrosion to a negligible amount which would depend on the number of years the cathodic protection structure is to be designed for. By cathodic polarization the potential of the cathode (£o,c) becomes nearly equal to the equilibrium potential of anode (£0,a) and the Cathodic protection metal surface attains a uniform potential, hence, corrosion is prevented. As long as the value of £COrr is brought very close to the value of £G,a by applying an external current, corrosion is prevented, hence, it is not necessary to reduce the corrosion rate to zero completely. galvanic anodes: Al -> A l + + + + 3e Mg -> M g + + + 2e Zn -> Z n + + + 2e 283 5.7 ANODIC POLARIZATION Imagine what would happen if the structure is now polarized in the opposite direction. It would amount to polarizing the potential of the anode to that of cathode in the positive direction. Theoretically, such a practice should result in creating corrosion rather than protection. But for some metals, positive polarization forms a protective oxide/hydroxide surface film and this phenomenon of passivation for a limited number of metals results in retardation of corrosion. By this method called anodic protection, it is possible to passivate active-passive metals. Metals, such as iron, chromium and nickel are passivated by anodic polarization, which leads to retardation of corrosion. The potential of this must, however, be maintained in the region of passivity by a potentiostat. Anodic protection is widely applied in transport of acids and corrosives in containers and other applications. 5.8 CATHODIC PROTECTION SYSTEMS Two types of cathodic protection systems exist: (a) Galvanic anode system or Sacrificial anode system. (b) Impressed current anode system. In the galvanic or impressed current system, the metallic structure is made the cathode (negative) by connecting it to galvanic anodes, which are more negative than the metallic structure to be protected. In this system, the current is generated by the corrosion of active metals, such as magnesium, zinc and also aluminum, which are The anodes of the above materials are utilized as sources of electrons which are released when the anodes are buried in the soil corrode. The electrons released pass through the metallic connection between anode and steel, and thus enter the structure to be protected. A suitable anode is buried adjacent to and level with the invert (lowest part) of a pipeline. A connection is made between the anode and the pipeline. The anode, generally magnesium or zinc, is connected to the pipeline or any buried metallic structure by an insulated cable. A schematic diagram of a galvanic anode cathodic protection system is shown in Fig. 5.15a. The figure shows a carbon steel pipe (A), magnesium anode (B), chemical backfill (C) surrounding the anode, wires connecting the carbon steel pipe to the anode, the soil (E) and test station (F). The details of test station are shown in Fig. 5.15b. The resistive component of a galvanic circuit is shown in Fig. 5.16. The copper wire connection provides a passage for flow of electrons to the pipe to be protected. The electrons are released by the consumption of Mg anode in accordance with the anodic reaction, Mg — M g + + -j- 2e. The outer > circuit is completed by the passage of electrons from the pipe (cathode) to the anode (Mg anode) through the copper wire (D). The pipe continues to be protected as long as it receives a regular supply of electrons from the anode. A typical anode installation in detail is shown in Fig. 5.17. The figure shows galvanic anodes (A) connected by a test station (F) and separated from each other by a distance of 8 feet. The test station provides a connection between the anode lead wire and the structure via the test panel. The details of the surface box housing test station are shown in the figure. The surface box is sometimes buried below the ground level. The anodes are connected to the pipe via a central control test panel. For measurement of pipe-to-soil potential and currents from the magnesium anode ground-bed, test stations 284 Principles of Corrosion Engineering and Corrosion Control Test station junction box Galvanic anode in chemical backfill Figure 5.15 a current flow) Typical vertical galvanic anode in soil (arrows show the direction of convention positive Pipe cap Plan view Rr. \ / 30* Insulated terminal block Socket head set screw 2 " dia galvanized pipe L^ -jfr^mm^ mmim & Wis >*' I ttilStttlgjwiI rani s / Concrete base, U^W ± Lead wire / JUO 18" Ok Ik Insulated Ashing Underground structure 1 Figure 5.15b Details of a test station Cathodic protection 2 85 - AAAGrade ~7~m //////////////////// TTTT Backfill Bare ferrous metal (cathode) Earth or water (electrolyte) E* = potential of anode (half cell potential) Ep » potential of pipe (half cell potential) Rw «resistance of connecting wire (metallic resistance) Rai ~ anode film resistance Rp » backfill resistance Rpg = resistance of anode backfill to ground R g « resistance of cathode to ground (this value can become the major e component if pipe is coated fccf« cathodic film resistance la = anode current - Sacrificial type anode Figure 5.16 Resistive components of a galvanic circuit are utilized. A special backfill, such as hydrated gypsum, bentonite and clay is placed around the anode, to ensure low resistance contact to the local soil. The anodic installation is often designed for ten years but may last much longer if current demand is low. The potential of the pipe must be continuously monitored and the value should not be allowed to fall below -0.85 V ( CuS0 4 reference electrode used). A 70 lb Mg anode practically gives a current of more than 300 mA in a soil of average resistivity of 2000 ohms-cm. Bare steel sometimes requires about 15mA/ft2. A single anode can protect about 2 square feet of the pipe. By applying a coating, the current requirement is reduced to 0.5 |xA/cm2, hence one Mg anode can protect up to 6000 square feet of the pipe surface. A potential value of 1.11 V is obtainable from the magnesium anode. This subject will be discussed in details later in this chapter. The following are the advantages and the disadvantages of the galvanic anode system: (a) Advantages 1. It requires no external source, which might fail. 2. It is economical. 3. It can be easily installed. 4. It can be easily maintained. 5. It can be used in areas where the soil resistivity is low. 2 86 Principles of Corrosion Engineering and Corrosion Control Test post Surface box housing junction box Plan view of junction box Anode connection to pipe via control test panel A-E Anodes F Junction box Galvanic anodes Control test panel Figure 5.17 Typical galvanic anode installation layout and test points. (By kind permission of Cheveron Corp., USA) 6. Lesser interference with the other metallic The following are the advantages and disadstructures is caused because of a relatively vantages of impressed current anode systems: low current output. (a) Advantages 7. The current is evenly distributed, 1. Rectifiers available in unlimited current output. (b) Disadvantages 2. May be designed for long lives. 1. It has limited applications compared to 3. More economical. impressed current. 4. Possibility of variation of current to suit the changes in the system. 2. Driving voltage is fixed and cannot be manipulated, except by selecting Mg (b) Disadvantages instead of Zn for example. 3. The cost of protection is high for bare systems (uncoated structures). 4. As no above-ground equipment is used, it is difficult to trace the protected system, unless contact posts are provided. 1. 2. 3. 4. External power is essential. More complicated system for installation. Less economical for smaller jobs. Limited to use below a soil resistivity of 3000 ohms-cm. Cathodic protection 2 87 5 .9 C O M P O N E N T S OF GALVANIC S Y S T E M S 5 .9.1 ANODES Wire Potting compound As mentioned earlier, magnesium and zinc anodes are customarily used in this system. The magnesium anodes are most popular because of their high current output. They have the following advantages: (a) Favorable position in the emf series (very active). (b) Non-polarizing. (c) High current output. An average current output of 500-600 A-h/lb can be obtained assuming the efficiency to be 65%. Magnesium anodes are usually applied in soils of resistivity lower than 3000 ohms-cm. A standard 17 lb magnesium anode will produce 1A of current for one year or 0.1 A for ten years, and so on. The current output, however, depends on the soil resistivity. With a resistivity of 1000 ohm-cm, the current produced would be 0.1 A. If the resistance is very high, the produced current would be lower. To obtain a maximum efficiency the anodes are surrounded with a mixture of gypsum and bentonite called as backfill The composition of a typical backfill is given in Table 5.1. * E u in Silver solder connection Hagnesium alloy Galvanised steel core Figure 5.18 Typicl 14.5 kg magnesium galvanic anode draw) of a standard alloy magnesium anode is —1.55 V with respect to CU-Q1SO4 reference electrode. The open circuit potential The mixture must be 100% capable of passing of a high manganese magnesium anode is through a 20 mesh screen and 50% through a 10 — 1.75 V with respect to a copper-copper sulmesh screen. fate half cell. If iron is polarized to —0.85 V, the driving potential of standard alloy magnesium would be (1.55 - 0.85 = 0.70 V) and 5 .9.2 MAGNESIUM ANODES that of a high manganese magnesium anode is The details of a magnesium anode are shown in (1.75 - 0.85 = 0.90 V). The composition of Fig. 5.18. The open circuit potential (no current some magnesium anodes is shown in Tables 5.2a and 5.2b. At 100% efficiency, the output is limited by Table 5.1 Composition of backfill local corrosion cells. To avoid this, magnesium is alloyed with Al and Zn. Practically, an efficiency of Element Percentage 60-70% can only be obtained. Galvomag anodes are used for seawater service. A typical compoGround hydrated gypsum 75 sition is shown in Table 5.3a. The consumption Powdered Wyoming 20 rate is 17.52 lb/A-year. The compositions of high bentonite (local brand) purity and Galvomag Mg anodes are also shown Anhydrous sodium sulfate in Table 5.3a. 2 88 Principles of Corrosion Engineering and Corrosion Control Table 5.2a Composition of standard magnesium anodes Element Aluminum Manganese Zinc Silicon Copper Nickel Iron Other impurities Magnesium Percentage 5.30-6.7 0.15-min 2.50-3.5 0.30-max 0.05-max 0.003-max 0.003-max 0.300-max Balance 5 .9.3 ZINC ANODES Table 5.2b High manganese magnesium anodes Element Aluminum Manganese Copper Iron Nickel Other impurities Magnesium Table 5.3a anodes Specification Specifications Cu Al Si Fe Mn Ni Zn Sn Pb Mg 0.02 0.01 max 0.03 0.5-1.3 0.001 0.01 max 0.01 max Remainder Percentage 0.010-max 0.50-1.31 0.02-max 0.03-max 0.001-max 0.3-max Balance Zinc anodes are frequently used for protection of submarine pipelines. They are commercially available in weights from 5 to 601b. Prepared backfill should be used for anodes if they are to be installed in the earth. They have a driving potential of-l.lOV compared to a Q1-CUSO4 reference electrode. The details of zinc anodes are shown in Fig. 5.19. Specifications of some zinc anodes are given below. Suppliers specifications must, however, be consulted before application. Some commercial specifications and details of zinc anodes are appended to this chapter. The composition and characteristics of zinc anodes are shown in Tables 5.3b and c. (a) Characteristics of zinc anodes (1) Corrosion products insulate the anodes and the anodes are, therefore, installed below the water table in soils with no free carbonate or phosphate so that passivity does not occur. (2) Zinc is not generally polarized anodically and the theoretical current output can be as high as 372 A-h/lb for a 99.99% pure zinc. Specifications of galvomag Mg Galvomag Wire Silver solder connection {insulated with rubber and tape) - Zinc 50% Efficiency -1.70VAg/AgCl Potential Capacity - Ampere hours 1230 per kg 560 per lb AgCl is a reference electrode M: - Galvanized steel core Figure 5.19 Typical 13.6 kg (30 lb) zinc anode Cathodic protection Table 5.3b Zinc anode for soil or fresh water use Element Aluminum Cadmium Iron Lead Copper Impurities (T oral) Zinc Percentage None 0.004 min 0.0015 max 0.006 max 0.005 max 0.010 99.99 2 89 Table 5.4a Elements added in various classes of aluminum anodes Type I II III IV V VI VII Additives 3.0-7.0% Zn 3.5-9.0% Zn, 0.10-0.5% Sn 7.0% Zn, 1% Sn 0.5% Sn 0.45% Zn, 0.045% Hg 0.1-0.40% Zn, 6.0-8.0% Mg and 0.08-0.15% Hg 1-5% Zn, 0.1-0.05% Mn and 0.06-0.15% Hg Table 5.3c water use Zinc anode for seaTable 5.4b anodes Percentage Solution potential of aluminum Aluminum Cadmium Iron Lead Copper Impurities (Tbtal) Zinc 0.3 max 0.06 max 0.003 max 0.006 max 0.005 max 0.014 Remainder Type Open circuit potential (£cu-CuS04) I II III VI VII -1.06 -1.10 -1.35 -1.43 -1.10 Working potential (Driving potential) -1.00 -1.15-1.25 -1.10 -0.93-1.30 -0.93-1.13 Consumption rate (kg/A-year) 4.7-3.6 4.3-3.6 3.3-2.9 - (3) Zinc can operate up to 95% efficiency and a current output of 335 A-h/lb can be obtained (a) Characteristics of aluminum (0.90 x 372 = 335 A-h/lb). (4) The open circuit potential of zinc is gener- anodes ally — 1.10 V with respect to a copper-copper sulfate reference electrode. (1) The cost is low and they are light in weight. (5) Based on a polarization potential of —0.85 V (2) The corrosion products do not contamifor steel, the driving potential of zinc is nate the water. 0.25 V. (3) The rate of consumption varies between 7 and 9lb/A-year. The efficiency varies between 87 and 95%. 5 .9.4 ALUMINUM ANODES (4) The anodes are easily passivated and must be rinsed with NaCl to reactivate. Backfill must These are mostly employed for seawater applibe used with aluminum anodes. cations. The base metal contains 98-99% of (5) The consumption of Galvalum-Aluminum aluminum. Elements commonly added in difanodes (Fe = 0.08, Si = 0.11-0.21, Zn = ferent types of aluminum anodes are shown 0.35-0.50 and Hg = 0.035-0.40) is in Table 5.4a. Table 5.4b shows the solution 1285 A/h/lb (Table 5.4c). It is a very popular potential of various classes of anodes. anode. 2 90 Principles of Corrosion Engineering and Corrosion Control of galvalum- s . 9 . s INSTALLATION OF Table 5.4c Composition aluminum anodes Galvalum* -aluminum GALVANIC A N O D E S Galvanic anode installations are simple compared to impressed current installations. Figure 5.20 Specification: shows a single packaged anode installation. The 0.08 max Fe common 17 pound packaged magnesium anodes Si 0.1 max are commonly used for a packaged installation. Zn 0.35-0.50 When several magnesium or zinc anodes are to Al Remainder be installed at a single location, the anodes are connected to a header wire which is directly 90% Efficiency Potential -1.05V,Ag/AgCl connected to the pipeline. Packaged anodes in multiple may be installed as shown in Fig. 5.20. 2830 per kg Capacity - Ampere hours The anode line may be parallel or perpendicular 1285 per lb to the pipeline. Zinc anodes are generally kept at * There are three types; Galvalum I, Galvalum II and a distance of 5 ft from the pipelines, if they are in parallel lines. Magnesium anodes are generally Galvalum III 15-20 ft from the pipelines. If space is not available and soil resistivity conditions are very low, galvanic anodes are Connection to pipe Where necessary, hole may be angled |siightly to place anode directly under the pipe 3" minimum for Mg anodes No minimum for 2n anodes 71 Packaged anode Figure 5.20 Galvanic anode installation for packaged anodes. (From Source: Peabody, A.W. Control of Pipeline Corrosion, NACE) Cathodic protection Table 5.5 Anode spacing factors No. of anodes in p arallel 5 2 3 4 5 6 7 8 9 10 .Adjusting ; factors Anode spacing (feet) 10 15 20 1.920 2.705 3.455 4.188 4.902 5.598 6.277 6.964 7.643 1.946 2.795 3.625 4.429 5.223 6.000 6.768 7.536 8.304 JMB 2 91 The current output of a magnesium alloy can be determined by the following relationship = 150 000 FY (for an uncoated pipeline) (5.1) 1.839 2.455 3.036 3.589 4.125 4.652 5.152 5.670 6.161 1.964 2.848 120 000 FY (for a coated pipeline) JMR 3.714 (5.2) 4.563 5.411 where: 6.232 4ig = current output of magnesium anode 7.036 p = soil resistivity, ohm-cm 7.875 F = factor from Table 5.6 8.679 Y= correction factor for pipe-to-soil potential value (Table 5.7) Example 1 Calculate the current output for a 32 pound packaged magnesium anode buried in a 1750 ohm-cm soil for the protection of a bare pipeline where the expected resultant pipe-to-soil potential is -0.85 V. Solution: The given data for a 32 pound anode is: F = 1.06 from Table 5.6. From Table 5.7, the Yfactor for a pipe-to-soil potential of—0.85-1.00. Pipe-to-soil potential is the potential between the pipe and the surrounding soil. Putting the data in equation (5.1), one obtains 7 installed below the pipe. Very deep holes are required to place multiple anodes. The anodes are safer in deep soils as they are not affected by variation in the moisture contents and soil conditions. Due to seasonal changes, non-packaged anodes are frequently used. In this case, the anode and the backfill is not installed as single unit, but separately. Galvanic anode may also be installed horizontally. Table 5.5 shows the anode spacing factors. 5.9.6 CALCULATIONS OF C U R R E N T OUTPUT OF MAGNESIUM A N O D E S The current output of galvanic anodes is affected by the resistivity of the soil. A higher current would be required for a low resistivity soil. The geometry of the anode is also important, for instance, the longer the length of the anode, the higher the current output would be. Lastly, the higher the potential of the alloy, the higher would be the current output. The efficiency of anodes is a major factor even under the best conditions. The Mg anodes, for instance, does not have an efficiency higher than 50%. The current output is an important information which must be known. Mg 150 000 x 1.06 x 1 = 90.9 mA 1750 For multiple anodes, the current output obtained for a single anode should be multiplied by an adjusting factor (Table 5.5). Extending the above example further for a 4 parallel anodes, 10ft apart, one obtains: 3.455 (adjustment factor - Table 5.5) x 90.9 mA = 310.50 mA Similar calculations can be made for the output of magnesium anodes used with the coated pipelines. Equation (5.2) would be used under such conditions. 2 92 Principles of Corrosion Engineering and Corrosion Control Table 5.6 ' F' factor for different anode weights Details Factor'F' A node weight (pounds) Standard Anodes: 3 5 9 17 32 50 50 Long Anodes: 9 10 18 20 40 42 Extra Long Anodes: 15 20 25 (Packaged) (Packaged) (Packaged) (Packaged) (Packaged) (Packaged - anode dimensions 8" diam x 16") (Packaged - anode dimensions 5" x 5" x 3.1") 0.53 0.60 0.71 1.00 1.06 1.09 1.29 ( 2.75" x 2 .75" x 26" backfill 6" x 31") ( 1.5" x 1.5" x 72" backfill 4" x 78") ( 2" x 2 " x 72" backfill 5" x 78") ( 2.5" x 2 .5" x 60" backfill 5" x 66") ( 3.75" x 3 .75" x 60" backfill 6.5" x 66") ( 3" x 3 " x 72" backfill 6" x 78") 1.01 1.71 1.81 1.60 1.72 1.90 ( 1.6" d iam x 10' backfilled to 6" diam) ( 1.3" x 2 0' backfilled to 6" diam) ( 2" d iam x 10' backfilled to 8" diam) 2.61 4.28 2.81 Table 5.7 P/S -0.70 -0.80 -0.85 -0.90 -1.00 -1.10 -1.20 ' Y' correction factors Magnesium 1.14 1.07 1.00 0.93 0.79 0.64 0.50 Zinc 1.60 1.20 1.00 0.80 0.40 0.00 0.00 an expected pipe-to-soil potential of —0.85 V. Determine the current output. hn 50000 x f x Y Putting the data from the tables, as in the case of Mg anodes: 50000 x 1.06 x 1 = 70.7 mA 700 For a four parallel anodes, 10ft apart, the estimated output would be: (The factor for 4 parallel anodes, 10ft apart is 3.445) (Table 5.5) / = 70.7 x 3.455 = 243.92 mA IZn = Example 2 - Current Output of Zn Anodes A bare pipeline is to be protected by zinc anode. A 32 pound packaged is to be used for protection. The anode is buried in a 750 ohm-cm soil with Cathodic protection 5 .9.7 CALCULATIONS OF 2 93 N U M B E R OF A N O D E S R E Q U I R E D A N D S PACING The number of anodes required for a pipe or any bare structure can be conveniently estimated as per example given below. Suppose 10 000ft of a bare 4 inch pipeline is to be protected and if the resistance of soil is 1000 ohms per cubic centimeter, the pipe requires 1 mA/ft2 for protection. The anode output curve shows 100 mA per anode in this type of soil. The number of anodes can be easily calculated as shown below. Calculations (a) First calculate the area of the pipe: Area 4 x 3.14 x 10 000 10 500fV 12 Total current requirement = 10 500 x 1.0 = 10 500 mA Number of anodes required Total current required Current output of a single anode 10 500 mA = 95 anodes lOOmA/anode (b) Total current requirement = Protected area (m2) x Current density (A/m2) or Protected area (ft2)x Current density (mA/ft2) (c) Total weight of anode material required: Current (A) x Design life (years) x 8760 Capacity A-h/kg 5.10 IMPRESSED C U R R E N T SYSTEM In contrast to the galvanic anode system, the flow of current from the anode to the cathode is forced from a DC source in the impressed current system. Thus, whereas the current is provided by the corrosion of the electrode in the anodic galvanic system, the electrode acts as a conductor and hardly corrodes in the impressed system and the AC input is transformed and rectified to a varying DC voltage. A transformer rectifier is the most important component of the system. Figure 5.21 shows a schematic diagram of an impressed current cathodic protection system. The direct terminal supply is obtained from a transformer rectifier (T/R unit) designed to step down normal alternating mains voltage and then rectify this to the direct current. This output is adjustable over a wide range to suit requirements. An AC power supply with an ammeter is connected to the rectifier. A switch box is installed in the AC circuit which contains a magnetic or thermal circuit breaker to protect the rectifier against overloads. The AC goes to the rectifier unit. It passes through the primary coil (P), the resulting magnetic field continuously expands, contracts and reverses direction. The changing field induces AC in the secondary coil, that is proportional to the turns ratio between the two coils. Primary turns Secondary turns Primary volts Secondary volts The rectifier cells are shown from B to E. Their function is to allow the current to flow in one direction and to block it in the opposite direction. When AC is applied to the AC terminal of the rectifier, DC appears at the output. The negative terminal of the rectifier assembly is connected to the pipeline (G) and the positive to the anode bed. The five anodes are represented by numbers 1 to 5. The anodes are surrounded by a mixture of coke breeze. The anodes are energized by the rectifier and they are made of silicon cast iron, steel scrap, platinized titanium, graphite or lead silver. 2 94 Principles of Corrosion Engineering and Corrosion Control Ground bed anodes Use rectifies solar units, or generator to supply external source of direct current Negative wire Pipeline (cathodes) z Figure 5.21 Schematic diagram of impressed current cathodic protection system Cathodically protected / 2 The electrons from the rectifier enter the pipe 5 .10.1 A D V A N T A G E S A N D or any structure to be protected. The anodes are D ISADVANTAGES OF placed in a trench, which is termed a ground-bed. As a matter of practice, the anodes are placed in I M P R E S S E D C U R R E N T SYSTEM the lowest resistivity soil. Details of a typical impressed current instal- (a) Advantages lation is shown in Fig. 5.22. The AC power unit (a) One installation can protect a large area is connected to the rectifier (R). The negative terof metal. minal is attached to the structure (S) and positive terminal to the ground-bed. The anodes are sil(b) The system can be used for a wide variety icon cast iron and represented in the figure by of voltage and current requirement. Ai to A4. The individual anode leads are brought (c) Schemes can be designed for life in into a junction box (J) and a shunt is inserted in excess of 20 years, if required. each anode lead. The junction box allows the mea(d) Current requirements and potentials surement of current output of individual anodes can be easily adjusted to the varying by determining the IR drop. The anodes can be needs of protection. connected directly to the positive terminal of the (e) Can be applied to a wide range of rectifier by a ring main cable system shown in structures. the figure or via a control box (direct connecting system) to monitor the current output above (f) Requires generally a small total number ground. of anodes. Cathodic protection 2 95 Oil cooled transformer rectifier box Direct connection system {via control box or ring main cable system) Silicon Iron anode Figure 5.22 Details of typical power impressed ground-bed installation (g) Requires simple controls which can be automated. (h) The use is less restricted by soil resistivity. (i) A large area can be protected by one installation. (b) Disadvantages 5 .1 1 C O M P O N E N T S O F IMPRESSED CURRENT S Y S T E M S A ND AC/DC SYSTEMS 5 .11.1 POWER SOURCE Rectifiers are commonly used as the source of (a) Possibility of interference effects on power. Although rectifiers are most commonly other buried structures. used as a source of DC power for the impressed (b) Regular maintenance is essential. current systems, other external power sources, (c) Electrical power cost is high. such as wind generators, thermoelectric gener(d) Power failures can cause serious prob- ators and wind-driven generator units can also lems, and faults may go unnoticed for be employed if conventional AC power is not long times. available. 2 96 Principles of Corrosion Engineering and Corrosion Control RECTIFIERS flows through direction Di, follows the path (3) and through diode D4 it enters the negative terThe rectifier units consist of a transformer, rec- minal T2. In the next half-cycle (1/120th) of a tifier stacks, meter, switch and the transformer second later, polarities at Ti and T2 are reversed tap connections. The rectifiers convert AC to DC (see diagram B). The current is blocked by diode current. The transformer is used to step down D4 and flows through D2, follows the path (3) the supply of voltage to that required for opera- through D3 in the same direction as before. The tion of the rectifier stacks. There are two types of load Ri thus receives energy in the form of pulses rectifiers: at 120 per second. Although three-phase rectifiers are used as mentioned before, each single bridge shares a pair (a) Selenium (Oil and Air-cooled) of diodes with one of the other bridges. The threeRectifier phase bridge is like three single-phase bridges, with each bridge sharing a pair of diodes with one The selenium air-cooled rectifiers have lower iniof the other bridges. tial cost than the oil immersed units, but have A rectifier consists of three important compopoor ventilation. The oil immersed rectifiers are nents: circuit breaker, transformer and rectifying less vulnerable to air and dust than the air-cooled type. In either case, the structure to be pro- elements (stacks) (Fig. 5.24). Brief details are tected is connected to the (-) negative terminal given below. of the rectifier. The installation is normally at 8 miles interval for 36" pipe and 12 miles for (a) Circuit breaker These are basically switches with an inter24" pipe. nal mechanism which opens the switch when the current exceeds a prescribed designed (b) Silicon (Oil and Air-cooled) limit. They also serve as 'on and off switches. Rectifier There are two types of switches: (1) magnetic and (2) thermal. The circuit breaker protects The silicon rectifiers (oil and water-cooled) have equipment from over loading. longer lives and higher efficiency compared to In the magnetic type, a coil is woven selenium rectifiers. The oil-cooled types are not around a brass tube and a magnetic field is susceptible to damage by dust and dirt. They set up by a current flowing in the coil. The are 20-50% smaller in size than the selenium magnetic slug is held at one end of a tube by rectifiers. a spring. The magnetic field attracts the slug, but at or below the rated current the slug does not move. At overload, the magnetic field 5.H.3 COMPONENTS pulls the slug into the coil. When the slug is drawn to the opposite end of the tube, the OF A RECTIFIER circuit is completed for the trip mechanism and the breaker switch trips. The movement The three-phase bridge is the most common cirof the magnetic flux is slowed down and a cuit for rectifiers operated from a three-phase AC time delay is provided. The breaker can trip power line. Each phase of a three-phase AC curon to 101-125% of the rated current. Overrent is spaced 120 electrical degrees apart and loads of ten times the rated currents can be therefore the voltage of each secondary winding sustained. The dropping is very fast when the reaches its peak at different times. overload is ten times. Figure 5.23 shows the operation of a singlephase bridge rectifier. The direction of flow In thermal magnetic breakers, the thermal tripping is caused by the flowing current reverses 60 times per second for 60 cycles AC. In through the resistor close to the bimetala positive half-cycle (diagram A), current origilic strip. When the current exceeds the nates at Ti on the secondary winding. It is blocked rated value, the bimetallic element trips the by D3 (silicon diode). The current, therefore, S.11.2 Cathodic protection 2 97 (A) First half-cycle Input i i^ i i_ i h hi n Output (8) Second half-cycle n n r\ Output (C) Final result fYYYYY\ Figure 5.23 Operation of a single phase bridge rectifier. Arrows show conventional (postitive) current flow direction breaker and a long time delay is involved before the breaker can be closed, (b) Transformer This consists of two coils of wire wound around an iron core. The coils are not connected electrically, but the core provides a magnetic link between them. AC voltage is applied to one coil (primary), the changing magnetic field crosses to the other coil (secondary) and induces a voltage in it. The changing field induces the AC voltage in the secondary coil that is proportional to the 2 98 Principles of Corrosion Engineering and Corrosion Control 5.11.4 A,C Input RECTIFIER EFFICIENCY <\\Bj Input selection <—>*> lightening arrester l^ * p Circuit breaker . Primary winding Secondary winding Rn Fine taps pqpryYyvpq Coarse taps T 1 This is the ratio between the DC power output and AC power input. Rectifiers are used as a source of DC power. Rectifiers convert the AC current (60 cycles) to DC current through rectifier operated at maximum efficiency at the full rated loads. Overall rectifier efficiency DC power output x 100% AC power input An efficiency filter can be used to minimize the ripples. ¥' H t Stack Change-over switch —& (~)i D.C Input lightening arrester o(+) 5 .11.5 C A B L E S Cable conductors to the anodes are made from copper or aluminum. These conductors must, therefore, be insulated. High-density polyethylene has good properties with respect to abrasion and high-temperature while maintaining excellent dielectric properties. Technical information about cables is provided by the manufacturers. Figure 5.24 Components of a rectifier turns ratio between the two coils. Primary turns Secondary turns (c) Rectifier cells Primary volts Secondary volts 5 .11.6 I N S U L A T I O N A N D ELECTRICAL CONTINUITY The structure to be protected must be free from all interconnected metal work in order to limit The change of AC power to DC is done by the flow of electric current on the structure. This rectifying elements. They act like check valves is made possible by electrical insulation. It is the by offering low resistance to current flow in condition of being electrically isolated from other one direction and high resistance in the other metallic structures. The term insulation is used direction. The function of the rectifying ele- interchangeably with isolation. The use of isolated ment is to allow the current to flow readily flanges and joints is common in pipelines. in one direction and to block current flow in the opposite direction. The Selenium cell is INSULATION OF the most common rectifier cell. Selenium is 5 . H . 7 applied to one side of an aluminum base plate I N S U L A T I N G F L A N G E S which has been nickel plated. A thin metallic layer is applied over the selenium layer. This Insulating flanges are installed to isolate the layer acts as counter electrode. It collects the cathodically protected buried pipelines from current and provides low resistance to the above-ground pipelines. An insulating kit concontact surface. These cells may be arranged sists of a non-conducting gasket separating the in stacks or parallel to produce the desired flange faces, an insulating sleeve and a washer for the bolt (Fig. 5.25a). Details of an insulated flange voltage and current rating. Cathodic protection 299 Hate; Sleeve extends partially into each insulating w a s t e Note: All studs insulated as shown Insulating gasket , Insu'a«n9 washer y / / S t e e l washer —EQ m f~) I aximum current output is 50A/m2 and the rate of consumption is between 90 and 250 g/A/year. Anodes containing Mo are used in 5,13 MAJOR I M P R E S S E D high-temperature media. A typical analysis of high silicon cast iron CURRENT ANODES anode is shown below in Table 5.8. It is The following are the major impressed current generally used for onshore cathodic protection anodes: applications. 3 04 Principles of Corrosion Engineering and Corrosion Control Table 5.8 Analysis of a typical silicon cast iron anode develops an adherent oxide layer of high electrical resistance. The oxide layer prevents corrosion by acting as a barrier. Titanium acts as an inert Element Percentage support for the platinum. Platinum can withstand very high current density and it is generally Silicon 14.35 min applied to a small area only. The platinum layer Carbon 0.85 max is normally 2.5 microns in thickness and it has an Manganese 0.65 max estimated life expectancy of 10 years. Titanium Iron Remainder sheets, 1-2 mm thick with a platinum coating of 2.5-5.0 |xm, can be loaded to 10 A/dm2 or over a period of years. Rod anodes of 10-25 mm diam5 . 1 3 . 2 M ETAL S C R A P eter are used frequently for protection of vessels, pipes, condensers, heat oil terminals, etc. CurIt has the advantage of being cheap and abun- rent densities up to 50 A/ft2 (540 A/m2) can be dantly available. The rate of consumption of mild obtained. The anode should, however, be used at steel scrap pipes, rails and cast iron scrap castings a low voltage. varies. For mild steel scrap the rate of consumption is 6.6-9.0 kg/A/year, and for cast iron the rate is 0.9-9.0 kg/A/year. Steel in the form of 5 .13.5 L E A D A N O D E S old railroad line, pipes and structural sections is used. The rate of consumption of steel scrap is Lead anodes are made of various lead alloys, such generally uniform. The material is mostly avail- as Pb-lAg-6Sb and Pb-lAg-5Sb-lSn. The den3 able in the form of long and thin section and sity of a lead anode is around 11.0 to 11.2 g/cm . Pb-lAg-6Sb has a capacity of 160-220 A/m2 depending whether these sections are installed horizontally or vertically, they may encounter soil and a consumption rate of 90 g/A/year or 2 strata with different resistivities resulting in non- 0.009 kg/A/year at a current density of 10 A/ft 2 uniform corrosion. Cast iron has the advantage of (108 A /m ). The other anode containing 10% Sn 2 being thick in section and of such form that any and 5% antimony has a capacity of 500 A/dm one piece will be in soil of more or less uniform and the rate of consumption 0.3 to 0.8 kg/year. resistivity. Also, a graphite surface is left exposed This alloy has good mechanical properties, and as the outer iron is consumed. The remaining can be extended to any shape. Lead-silver or leadiron in the form of a graphite, therefore, acts as a platinum anode with a diameter of 7.5 cm, length 75 cm and weighing 36 kg or with a diameter of graphite anode. 5 cm, length 180 cm and weighing about 45 kg are used in the form of round anodes to protect corrosion of marine structures. These are also used 5 .13.3 G R A P H I T E A N O D E S for protection of ships. These have the advantages of long-life corrosion protection, low maintenance cost and high efficiency. The typical anode current density 5 . 1 4 PROTECTION OF is between 10.8 and 40.0 A/m2 (1.4 A/ft2). The rate of consumption is between 0.225 and 0.45 kg S U B M A R I N E P I P E L I N E S (0.5 and 1.0 lb) per year. These are generally cylindrical in shape, although other forms are Pipelines in seawater are protected by so-called bracelets (annular anodes) as shown in Fig. 5.30. available. In marine structures, corrosion is at maximum at a small distance below the water line and decreases 5 .13.4 P L A T I N I Z E D T I T A N I U M with depth. Corrosion is less severe in mud. For protection of bare steel in seawater, an iniThese anodes are used for salt water or fresh water tial current density of 15mA/ft2 (161mA/m2) where the conductivity is very low. Titanium is required, and this decreases after some time Cathodic protection 3 05 Table 5.9b Degree of hardness vs the protective current density Carbonate, hardness (Degree of hardness) <2 2-10 >10 Protective current density (mA/m2) 250-320 100-150 70-120 Table 5.9c Recommended current densities Environment Current density Current density (mA/m2) (mA/ft2) Figure 5.30 Bracelet type of anode for marine application. (From B.K.L. Cathodic Protection Division, Birmingham, England) to 5mA/ft2 or (43mA/m 2 ). In the impressed current system non-consumable graphite anodes are required, whereas in the galvanic system a magnesium anode is the best material. Zinc and aluminum anodes are also used as galvanic anodes, but the cost is high. The protective current densities for steel structures in various seawaters are given in Table 5.9a. The current requirement of steel is affected by the degree of hardness as shown in Table 5.9b. The current required for protection of steels in various environments is shown in Table 5.9c. Complete protection of buried steel or iron may require 0.75-5.0 mA of current per square foot of the surface. On a well-coated line, the current may be as low as 0.01-0.2 mA/ft2. Table 5.9a Protective current requirements in different seawater Area Gulf of Mexico Nigeria Alaska Arabian Gulf North Sea U.S. West Coast Protective current requirements (A/m2) 80-150 85 250 65-85 90-150 24 Soil Fresh water Seawater Moving seawater Sea mud 0.75-5.0 1-3 4-5 1-3 1-3 40-58 11-32 43-64 11-32 11-32 The potential necessary to protect buried steel is —0.85 V, however, in the presence of sulfates, reducing bacteria a minimum potential of-0.95 V with respect to copper sulfate electrode would be necessary. Approximately 15-100 mA/ft2 current is needed for protection of bare steel in sluggish water. In rapidly moving water, 1-10 mA/ft2 for bare steel in a soil would be necessary. Current requirements in various environments can be found abundantly in the literature as well as cathodic protection specifications. For submarine pipeline, a current density of 5 mA/ft2 is required. 5.15 DESIGN P A R A M E T E R S IN CATHODIC PROTECTION The basic design requirement for cathodic protection is the choice of current density per square foot or square meter of the surface area to be protected. The choice of current density can vary from something in the order of 100 mA/ft2 for a bare structure in water to as low as 0.0001 mA/ft2 for well-coated pipes or 3 06 Principles of Corrosion Engineering and Corrosion Control structures of high resistivity. To estimate the curIt is necessary to experimentally conduct a rent requirements, knowledge of soil resistivity is current requirement test for the structures to of a primary importance. The following are the establish the magnitude of power requirement for major characteristics of soils: the system to be installed. As a rough guide, the following current den(1) Sandy-type soils are less corrosive than non- sities would be required to ensure anode life for homogeneous soils. 10 years in soils of different resistivities. (2) Homogeneous soils are less corrosive than 800 ohm-cm and below 300 mA heterogeneous soils. 800-200 ohm-cm, 180 mA (3) Well aerated soils are less corrosive than 2000-3000 ohm-cm, 90 mA sparsely aerated soils. The more aerated soils tend to be brown in color. (4) Soils low in organic matter are less corrosive than soils with high amount of organic S . 1 7 BACKFILL matter. (5) High acid and high alkaline soils (high pH) Galvanic anodes are surrounded by a backfill which is usually a mixture of gypsum, bentonite, are more corrosive. (6) Soils containing sulfate reducing bacteria are and clay. Table 5.11 shows a typical composition more corrosive than soils free from this of a backfill. The backfill for galvanic anodes serves the bacteria. following purposes: (7) Soils having low electrical resistivity are more corrosive than soils having high electrical (1) It isolates the anode from the surrounding resistivity. soil and protects the anode from the effect of chemicals contained in the soil. The soils can be classified as below with respect to (2) It provides a lower anode-to-earth resistance the corrosivity of steel and iron (Table 5.10). because of its low resistivity. (3) It provides a higher current output because of the low resistivity of the surrounding soil. 5 .1 6 C U R R E N T REQUIREMENTS Generally, the quantity of current varies with the type of soil and the quality of coating. The current requirement for an uncoated structure may be seventy times the current requirement for a coated structure. For instance, 10 miles of wellcoated pipelines (24") may require about 8 A for protection compared to 560 A required by a non-coated line. Table 5.10 Classification of soils according to corrosivity of steel and iron Soil Resistivity 0-1000 ohm-cm 1000-3000 ohm-cm 3000-5000 ohm-cm 10 000 and above Tendency to Corrode Very corrosive Corrosive Mildly corrosive Non-corrosive For the impressed current anodes, the standard material is coke. The physical and chemical analysis of the coke breeze is given in Tables 5.12a and 5.12b, respectively. Backfill should be installed very dry around the anodes except in desert conditions. The coke breeze provides a low resistance between anode and earth and a longer life for the impressed current anode. Because the greater part of the current passes from the anode to the backfill particles, the anode is consumed at a slower rate. Table 5.13 shows the amount of backfill required. Table 5.11 Composition of backfill Gypsum (%) 50 75 Bentonite (%) 50 20 Soil Resistivity (ohm-cm) Below 3000 3000 and above Cathodic protection Table 5.12a Physical analysis of coke breeze (0.525 in) (0.263 in) (0.131 in) 307 available to determine the soil resistivity and a few important methods are given below. 1% maximum to remain in No. 2 mesh screen 85% minimum to remain in No. 3 mesh screen 12% maximum to remain in No. 6 mesh screen (a) Four-pin Direct Current Method (Wenner's Method) On the site, four stainless steel pins are buried in the ground and spaced in a straight line. The spacing between the pins represents the depth to which the resistivity is measured. By increasing pin spacing, the resistivity to a greater depth is measured. The instruments used for measurement are easily available. Instruments, such as a Vibroground, are often employed for the measurement of soil resistivity (Fig. 5.31a). This method is based on voltage drop. Figure 5.31b shows two outside pins Q and C2, and the two inside pins Pi and P2. The outer pins are connected to a DC source (12 V). In the meter all instruments, such as the power supply, the current meters, the switch, etc. are contained in one single case. A desired amount of current is made to flow in the earth between the two Table 5.12b Constituent Chemical analysis of coke breeze Percentage 14.70 3.14 76.66 1.50 4.00 Moisture Volatile matter Fixed carbon Ash Sulfur Table 5.13 Amount of backfill required Hole size (in) 6x80 6x84 8 x84 10 x 84 Backfill (lb) 62 61 108 180 Anode size (in) 1 x60 1.5 x 60 2 x60 3 x 60 Vibrogroimd instrument s* +~~—" Current / electrode / ^><r^ Ci I j^ n* 5.18 M E A S U R E M E N T S IN Potential Pi 4 ^electrode / /p2 / 1 Current I electrode CATHODIC PROTECTION 5.18.1 ML SOIL RESISTIVITY MEASUREMENTS he i Soil resistivity measurement is the first impor/h tant step in the design of a cathodic protection system as the current requirement would differ from one soil resistivity to another for the system. It may, however, be pointed out that there AC Current is no single method available to determine precisely the degree of corrosivity caused by soils. Soil resistivity only provides a rough guide to the Figure 5.31a Vibroground method for determinacorrosivity of the soils. There are several methods tion of soil resistivity 3 08 Principles of Corrosion Engineering and Corrosion Control Galvanometer Potentiometer adjuster Switch Figure 5.31b Four-pin (Wenner) resistivity measurement outside pins Q and C2, and the voltage drop is measured between the two inner pins Pi and P2. The millivoltmeter is used to measure the voltage drop in the earth between the inner pins Pi and P2, located along the current path between C\ and C2. The existing voltage between the two potential pins Pi and P2 is recorded before any current is applied. The test current is turned on and the voltage drop between the two potential pins is recorded. Resistivity is calculated by using the relationship Current reading - 8 mA Voltage across the inside pins - 50 mV with no current flowing Voltage across the inside pins - 500 mV with current flowing. Solution: 191.5 x Dx I AV 191 x 3 x (500 - 50) = 32 231 ohm-cm p= — x Spacing of pins (ft) x 191.5 (Factor) (Voltage in mV) where D = distance between pins A V= voltage drop Illustrative problem Calculate soil resistivity from the following data given: Pin spacing - 3 ft Impressed voltage - 10 V Similarly, if D=10'6" AV = 412mV 7 = 267 mA Cathodic protection 191 x 10.5 x 412 267 3094 ohm-cm 3 09 P= (b) Soil Box Method In the soil box method, no multiplying factor is necessary for calculation of resistivity as the dimensions are made such. A typical soil box has dimensions of 1"(D) x 1.5" (W) x 8.5"(L). It is constructed from a non-conducting material like plastic. The box contains the current terminals and potential terminals. The end plate of the box acts as current terminal and the inside contact plates as potential terminals. The box is filled with required soil which is packed firmly. The potential change is divided by the current to obtain the desired resistivity quickly. The resistivity of soil or water can also be measured by a soil box which has four terminals. The measurements can be conveniently made in the laboratory. The soil box method is a replica of a four pin method mentioned earlier. A typical soil box is shown in Fig. 5.32. In the figure shown, Ci and C2 are current terminals and Pi and P2 potential terminals. A DC source is connected to C\ and C2 terminals and the voltmeter is connected to the terminals Pi and P2 (potential terminals). A measured current is passed through the soil sample and the voltage drop is read across the Pi and P2 pins. Knowing the voltage drop and the current introduced, the resistivity of the soil can be conveniently measured. The box can be used either with AC instrument, such as Nilssons or Vibroground or with Miller meter. When using Vibroground, the Ci and C2 terminals are connected to the end plates of the box and the Pi and P2 terminals to the intermediate terminals of the box. 5 .18.2 S I G N I F I C A N C E SOIL RESISTIVITY OF Soil resistivity gives an indication of the corrosivity of soil. Some typical approximations are given as below: Soil resistivity (ohms-cm) 0-900 901-2300 2301-5000 5000-10 000 10 000 and above Corrosion Severe corrosion Severe corrosion Moderate corrosion Mild corrosion Very mild corrosion The above is only a rough guide to predict corrosion. A soil resistivity survey is required to determine the current requirement for a given pipeline. Soil resistivity may be very high in cold areas, such as Alaska, and virtually no cathodic protection may be required for coated pipes. On the contrary, in tropical areas near the sea shores, Current electrodes Potential electrodes Figure 5.32 Typical soil box 3 10 Principles of Corrosion Engineering and Corrosion Control Table 5.14 Characteristics of corrosive and non-corrosive soils Corrosive (1) Brackish water (2) Poorly aerated (3) High acidity or alkaline pH (4) Sulfides present (5) Black or gray color (6) Anaerobic microorganism (7) Soil with high salt contents Non-corrosive Low moisture content Well aerated Low acidity or neutral pH or slightly acidic Sulfates present Red or brown color Aerobic microorganism Dry soil 5.19 PIPE-TO-SOIL POTENTIAL In order to determine the extent of corrosion of a buried pipe in soil, it is essential to determine its pipe-to-soil potential. It is to be realized that a potential difference existing on a pipe surface can cause corrosion, similar to the situation in a dry battery cell in which zinc is the anode and corrodes. The pipe-to-soil potential would indicate whether the pipe is corroding or it is fully protected. The normal procedure for measurement of the structure-to-soil potential is to connect the negative terminal of a high resistance voltmeter (cm/V) to the pipe and the positive terminal to a standard copper/copper sulfate half cell (reference electrode) which is placed near the pipe (Fig. 5.33a). The Cu-CuS04 half cell acts as the cathode and the steel pipe as the anode. The difference of potential between the Cu-CuS04 half cell and the steel pipe is generally —0.55 V (pipe negative), when all connections have been made. The reading of the voltmeter indicates the structure-to-soil potential. The reference Cu-CuS04 is shown in Fig. 5.33b and connections are shown in Fig. 5.33c. in particular, the soil resistivity may be low and cathodic protection for the pipes would be essential. Soil resistivity measurements will, however, depend upon dampness of soils and the prevalent weather conditions. The following are the characteristics of corrosive and non-corrosive soils (Table 5.14). Potential test station High resistance voltmeter Copper - copper sulfate half cell (pushed into soil) Figure 5.33a Typical pipe-to-soil potential measurement C athodic protection 311 A* Half-cell components C^'Tr-Q-**— Thumb screw OOD*-™ Copper nut __ c=*«i—~ Washer __ _ -Plastic top B. Half-ceil assembly Plastic washer - Copper nut iopper rod Electrolyte (Distilled water) Copper rod u ~ Plastic tube Eartti contact 1 ^^ Figure 5.33b -(f ^ Porous wood plug L . ^ Rubber guard saftssMssss Copper-Copper sulfate half cell components +0.55 V High resistance voltmeter I s ^ ^ \ Copper reference Current flew Common electrolyte * Earth Figure 5.33c Copper-Copper sulfate half cell used as a reference electrode 3 12 Principles of Corrosion Engineering and Corrosion Control 5.20 FACTORS AFFECTING P I P E - T O - S O I L POTENTIAL 5 .20.1 PLACEMENT OF REFERENCE ELECTRODE OF FOR M E A S U R E M E N T PIPE-TO-SOIL POTENTIAL is no further increase in the negative reading, the remote position is reached which implies that the electrode is outside the influence of the structure-to-earth resistance. This is the desired location for placing the reference electrode. At this position, there is no further change in IR drop (potential drop). The remote earth does not offer any resistance to current flow. A remote location is outside the potential gradient of the anode bed and the pipe. One of the important points in the measurement of pipe-to-soil potential is the placement of the reference CU-C11SO4 electrode. Two locations are normally used, one near the structure and the other electrically remote. As a matter of principle, the reference electrode should be placed in such a way that the circuit resistance remains at the minimum. If a pipe is fully coated, the reference electrode may be placed in either remote or on the pipe position and it would not make any difference as the coating resistance makes up most of the resistance between the pipe and the soil. The potential of a cathodically protected pipe is measured by placing the reference electrode at least five or six pipe diameters away from the structure along the surface of the earth. In the case of an unprotected system, it may be essential to take measurements of pipe-to-soil potential with the reference electrode directly over the pipe and also at a position remote to the pipe. At the close position the potential of a small segment of the pipe is only indicated. Suppose an electrode is placed at a distance of 3 feet from a 12 inches diameter uncoated pipe, it would only survey a segment of six feet long as the electrode surveys a distance equal to twice its distance from the pipe. For a pipe length of 100 feet, the electrode must be placed at a distance of at least 50 feet. As a rule of thumb, the circuit resistance must be kept to minimum. For a bare structure, the reference electrode must be placed at a position electrically remote from the test section. This can be done by continuously increasing the distance and observing the readings in the voltmeter which becomes more and more negative. When the successive reading becomes more negative, the electrode is not outside the structure-to-earth resistance. Once, a point is reached at which there 5 .20.2 ELECTRICAL CONTACTS Extreme caution must be taken to maintain a good electrical contact between the probe and the structure, and between the soil and the reference electrode. 5 .20.3 INSTRUMENTS Only high resistance voltmeters must be used to avoid errors in reading of potential. The voltmeter should have a minimum sensitivity of 50000ohms/V. 5 .21 POTENTIAL S U R V E Y It has been stated earlier that the tendency for a metal to corrode can be predicted by its potential in a particular environment. The potential surveys are made to assess the magnitude of corrosion of pipelines and detect special areas and spots where the degree of corrosion is severe; the later being termed as a hot spot. A general idea of corrosion of a pipeline can be obtained from the average pipe/soil potential. Newer pipelines show generally a lower negative potential than the older and coated pipelines. For example, a new pipeline may show an average potential of —1.65 V when compared to a potential of —1.2V shown by an older pipeline. It is, however, important to detect the areas on the pipelines which are subjected to severe corrosion (hot spots). Cathodic protection 3 13 -1.0 | -0.8 Remote electrode reading ft S § -0-6 I% ^1-0.4 I JL 1000 2000 Pipeline length, meters X 3000 4000 Figure 5.34 Potential survey - remote electrode Figure 5.34 shows a plot of pipeline length vs pipe-to-soil potentials taken with the electrode in the remote position. The peak in the curve indicates the points towards which current is flowing to the pipe and discharging to the soil. It should be noted that the higher the value of P/S potential, the more would be the magnitude of corrosion. The peak, therefore, represent the hot spot or the areas where corrosion is active. 5.2f.i POTENTIAL SURVEY MEASUREMENTS The distribution of potential along the survey of earth above the pipeline indicates the location of corroding areas. Different types of soil encountered by the pipe affect the potential of the pipe. The changes in soil resistivity also induce potential differences. The surface potential surveys are made to determine the anodic and cathodic areas on the pipe. The structure-to-soil potentials do not give a qualitative measurement of corrosion, however, they are very useful in prediction of corrosion when used in conjunction with other data, such as soil resistivity. Several methods are used for potential measurement, such as the one electrode and the two electrodes methods. The two electrodes method is described below. The two electrodes method for measuring the surface potential is shown in Fig. 5.35. Two Cu/CuS04 electrodes (a) and (b) are placed on a wooden clipboard and connected to each other through a resistance voltmeter. One of the electrodes is in rear position and the other in the forward. The figure also shows the positions along which the electrodes are marked 1, 2 and 3. The anodic and cathodic areas are also shown in the figure. The IR (potential) drops at each location is measured between the two electrodes. The two electrodes leap-frog method is shown in Fig. 5.36. The survey is started at position 1 and the potential is recorded. Electrode CA' is left at position 1 and electrode 'B' is moved at position 2. The potential difference is recorded and also the polarity of electrode is noted. This procedure is continued along the whole length of the pipeline. For instance, consider the potential at position 1 to be —0.72 V. If the potential drop at position 2 is 0.065 V 314 Principles of Corrosion Engineering and Corrosion Control (3) (2) (1) 4^W/W/M>/sW>W/A^^^ Current flow I Cathode % X^ *C7 C athode© -* Anode 0 «- Current flow © f u^ Figure 5.35 Surface potential method for determining corrosion involves electrodes (a) and (b) connected through a voltmeter (V). Potential is measured between the two electrodes Rear electrode is leap-frogged to forward position for each subsequent reading High resistance voltmeter Survey direction First reading R».l4 Second reading \ - Third reading ! . Fourth reading .*-' ! etc P05.4« ! I .'" */ - v ; : <: 1U ."Known-V; *.'»;' -•':' .'spacing/-, * - Y _ / -\ /, ( ^&oiJ.v; - 'J';-''*'*"<'*»' '.* , •/•'; • , v \ ' ., i '.' *' ' «V -V - * : - ; v-> "•'.•%'. V Pipeline A v^r.v. • r ' : ^ ' - ^ ' : W ; ^ - / : . : ' • v-'C;;{;>;;-;:-/ Figure 5.36 Two electrodes method of measurement of potential. (By kind permission of Cheveron, USA) A*U i ••, • v ,u A • • i 5 - 2 2 MEASUREMENT OF and the polarity is positive, then the pipe-to-soil potential would be -0.785V. If the polarity of CURRENT FLOW forward electrode is negative, the potential drop would be subtracted to get the final pipe-soil It is possible to measure the current flow in a potential. pipeline or any buried structure by determining Cathodic protection 3 15 1.5(2.0 + 2.0) the voltage drop for a given length, if the resis3.0 mV 10 tance is known. Take, for instance, the current flow in a 6" OD steel pipeline. Assume the length to be 500 feet. As the voltage drops are Thus the true voltage is 3 mV (CU-Q1SO4). The next step is to determine the resistances of often small, low resistance voltmeters may be the pipe. The resistance of a standard weight pipe used. For example, a voltmeter with an interis obtained by substituting in a simple equation: nal resistance of 100 ohms/V on a 2 mV scale and external resistance of 2 ohm would be suitable 215.8 L for this measurement. Refer to Fig. 5.37 for the R= 1000 000 W measurement technique. Before determination of current flow, the true where voltage between the tests points A and B and the R = resistance of the pipe per given length resistance of the steel must be known. The latin ohms ter information can be determined from standard W= weight of pipe per linear foot (consult tables for steel pipes. tables) L = length of the test pipe in feet (given) Voltage drop The weight of a 6 inch steel pipe is 19 pounds per linear feet. The test length is 500 feet. The = Voltmeter reading between A and B resistance of the pipe is, therefore, = 1.5 mV 215.8 x 500 True voltage between A and B = R= 19 x 1000 000 Voltmeter reading 0.000216 x 500 Voltmeter resistance + External resistance Voltmeter resistance 19 0.0005684 ohms Ik Voltage drop with low resistance voltmeter Bare pipe Figure 5.37 Measurement of current flow 3 16 Principles of Corrosion Engineering and Corrosion Control anode when bridged is then shown by the ammeter. The modern method employs using AC audio frequency energy for measurement. By applying Ohm's law, I = E/R, the magnitude of current flow can be determined: As shown in the figure, the polarity indicates that the flow of current is from left to right. In the previous example, the magnitude of the current already flowing in a pipeline is determined. The amount of current flow may also be determined by passing a known amount of battery current and measuring the voltage drop between two points. The current is determined from Ohm's law (E = VR). 5,23 REFERENCE E LECTRODE The emf of a cell is comprised of its two half cells. For instance, the emf of a Daniel cell is the sum of the two half cells, i.e. Zn in ZnS04 and Cu in Q1SO4. The resulting emf is 1.1 V. In order to determine the half cell potential of zinc in any media, zinc is connected to the negative end of a voltmeter and a reference electrode of a known potential to the positive end. The difference of potential is directly read from the voltmeter. The protective potentials for steel in seawater at 25°C and soils with respect to commonly used reference electrodes are shown in Table 5.15. Several reference electrodes can be used for measurement of half cell potential as shown in Chapter 2. The most common electrode used for cathodic protection measurements is the coppercopper sulfate electrode. It contains a copper rod, a saturated solution of CUSO4 containing the copper rod and a porous plug for contact with the electrolyte. The container is non-conducting. The space within the tube isfilledwith a saturated solution of copper sulfate and should contain an excess of crystals to insure that the solution would be saturated all the time. A rubber cap may be placed over the wooden plug when the electrode is not in use. An increase in the temperature causes an increase in concentration of the solution and a decrease in temperature causes a decrease in concentration of the solution. The temperature correction is 0.5 millivolts per degree change (°F). Temperature potential correction charts may be used for correction. Example Determine the true half cell potential of a steel structure to a copper-copper sulfate reference electrode when the outside temperature is 90°F. 5 .22.1 Z E R O R E S I S T A N C E AMMETER The zero resistance ammeter can be used to determine the exact amount of current that would flow between two points of different voltages in a given circuit. The zero resistance ammeter is useful for measurement of currentflowin low resistance and low current circuits. If an ordinary ammeter is used to read current in such a circuit, an appreciable margin of error would be introduced by its internal resistance, the true voltage would not be shown. The basic advantage of this instrument is its ability to read test current accurately without introducing any resistance. For instance, if the circuit has a total resistance of 0.1 ohm, and an ammeter of 0.2 ohm resistance is introduced, the current flow would be reduced by 67% by a 0.2 ohm ammeter. Thus, the margin of error would be 67%. By not introducing any resistance in the circuit the meter is able to read true current. It is designed to work on a null principle. The arrangement for measurement of current by zero resistance ammeter (ZRA) is shown in principle in Fig. 5.38. According to the arrangement, the open circuit potential between the pipe and the anode balanced by a potentiometer circuit. Sufficient current is allowed to flow from a battery through a variable resistor, until the IR drop balances the open circuit potential. At this point, the galvanometer registers zero current. The true current between the pipe and Cathodic protection 3 17 Battery »«^^VVSAAAAAA/"^^^^^^'^^^'' Variable resistor Ammeter A ^WWWWVResistor Galvanometer ^s> Q«Open circuit voltage Grade /4S Magnesium anode Figure 5.38 Measurement of current flow by the zero resistance ammeter method The pipe-to-soil resistance potential reading is-0.70 V. 5.24 ^ C O AT IN G _ RESISTANCE TESTS (a) Temperature correction +0.316 + (+0.007) = +0.323 V (Correction factor from the Suppose the coating resistance is to be measured on coated 24" diameter steel pipe, the procedure chart). is simple. Pass a known amount of current in (b) Half cell potential of the steel structure the given section of the test pipe and note the without correction: voltage drop. By applying Ohm's law, determine the resistance (R) of coating. An experimental = -0.70 - (+0.316) = 0.384 V setup is shown in Fig. 5.39. Between the insulated flanges, three voltmeters are fixed at a predeter(0.316 V is the potential of Cu-CuS0 4 mined spacing. A is the test section of a pipe, electrode.) B is the on and off switch, C is the power source, And with correction: D is an ammeter and E, F, G are the voltmeters. H is a copper-copper sulfate reference electrode. = -0.70 - (+0.323) = -1.023 V. 3 18 Principles of Corrosion Engineering and Corrosion Control The soil resistivity affects the coating resistance significantly. The pipeline resistance is the combined resistance of coating and the resistance of the pipeline to remote earth. In the example shown above, the resistance was determined in 1000 ohm-cm soil. The resistance can be changed from one soil resistivity to another soil resistivity. For instance, if a resistance of 0.4 ohms obtained in a soil resistivity of 10 000 ohms, the resistance is 10 000 ohm, would be 10 000/1000 x 0.4 = 4 ohms. (b) If now it is assumed that the test section was bare and in 1000 ohm-cm soil, the resistances to earth of the above line would be lower. Let us assume it is found to be 0.0040 ohm in a soil having a resistivity of 1000 ohm-cm. If the average soil resistivity is 100 000 ohm, the resistance would be 100 000 1000 (This is the resistance of the bare pipe) Table 5.15 Protective potentials for steel in seawater and soils (aerated) Reference electrode Value for Eprot Fe in aerated soil or seawater -0.85 V - 0.80 V Copper/Copper sulfate (CSE) C11/C11SO4 Saturated calomel (SCE) Hg/Hg2Cl2(s)/KCl (Saturated) Silver chloride Ag/AgCl/Seawater Standard hydrogen electrode (SHE) Pt,H 2 /H+(a+ = l) Zinc Zn/Seawater -0.80 V -0.53 V +0.20 V A known amount of current is passed and the values of potential 'on' and 'off' are obtained by opening and closing the switch which controls the circuitry. The value of A V is obtained from V(on) and V(off). The amount of current passed is already known. The A Vvalues of the voltmeters are averaged. By applying Ohm's law, the coating resistance is determined. Here is an example. Example (a) Length of the pipe = 2 miles, area = nDL, OD = 24", soil resistivity = 10 000 ohms The amount of current passed = 0.096 A, A VE = 0.77 V, A VF = 0.75, A VG = 0.77 Average voltage drop (A VE + A Vp + A VG) A = 0.75 V. E The resistance by Ohm's law is R = — = 0.75 V = 7.8125 ohms 0.096 A The area of the above pipe = 24 x ix x 2 x 5280 = 66 316 ft2 The effective coating resistance would, therefore, be 0.0040 x = 0.40 ohms The difference between the resistance of the bare and coated pipe is 0.40 ohm. The difference is added to the resistance obtained above, i.e. 7.8125 and multiplied by the area of the pipe to obtain the effective coating resistance of the pipe per average square foot. In the example above, the effective average coating resistance would be 7.8125 + 3.60 = 11.412 x 66 316 = 756 002 ohm per average square foot. Area of pipe = 518 093 ohms per average square foot. The resistance of the steel pipes of various dimensions can be estimated from Table 5.16. It is based on the relationship R= 7.8125 ohms x 66 316 = 51 792.7 ohms per average square foot (approximately) 16.061 x Resistivity in micro-ohms Weight per foot = Resistance of one foot of pipe in micro-ohms C athodic protection 319 \w \w Nx^ >fr^ Groundbed \W A ) D, Ammeter Insulated joint Pipe Insulated joint © 0 © Reference electrode - 1 Reference electrode - 2 Reference electrode * 3 I Figure 5.39 Determination of coating resistance. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Cheveron, Houston Research Center, USA) 5.2S CURRENT REQUIREMENT T E S T S Basically sufficient current is supplied from a power source to lower the potential of a structure to —0.85 V. At this potential all cathodic areas are polarized to the open circuit potential of the anodic areas, hence no corrosion would occur. A temporary ground-bed is made and a known amount of current is forced in the structure to determine the total current required for cathodic protection. Protection of steel in an aggressive soil is achieved if the metal-soil potential is more negative than —0.85 V or -0.95 V for anaerobic soil conditions. In a bare pipeline or a structure, sufficient current is introduced in the system until a protective potential is achieved. To protect a structure from corrosion, say, a pipeline, a current requirement test is crucial, because if the required current to achieve a pipe-to-soil potential of—0.85 V is not provided, the pipe would not be protected as desired. If a bare pipe is buried in a soil of known resistivity and its characteristics are well-known, a selected current density can be given to the structure, without the necessity of current requirement test. In such instances experience is the best guide. In areas where experience lacks, t he situation is different and current density required for protection can be determined experimentally by applying a range of current densities and selecting the most desirable current density to achieve cathodic protection. T he ' current drain* t est is c ommonly employed to determine the current requirement for a coated pipeline. A temporary drain point is set to 3 20 Principles of Corrosion Engineering and Corrosion Control Steel pipe resistance^l ^ Outside diameter (inches) Wall thickness (inches) Weight per foot (pounds) Resistance of one foot ^ in (ohms x 106) (millionths of an ohm) 79.2 26.8 15.2 10.1 7.13 5.82 5.29 4.61 4.09 3.68 3.34 3.06 2.82 2.62 2.44 2.28 2.15 2.03 Table 5.16 Pipe size (inches) 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 2.375 45 . 6.625 8.625 10.75 12.75 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 0.154 0.237 0.280 0.322 0.365 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 0.375 3.65 10.8 19.0 28.6 40.5 49.6 54.6 62.6 70.6 78.6 86.6 94.6 102.6 110.6 118.7 126.6 134.6 142.6 (1) Based on steel density of 489 pounds per cubic foot and steel resistivity of 18 micro-ohm-cm as stated in the text 16.061 x Resistivity in micro-ohm-cm resistance of one foot of pipe in micro-ohms (2)R = Weight per foot determine how much current would be required to protect the line. The site must be selected where the pipeline is exposed or a valve box is located so that a connection may be made. The test current can be supplied either by a storage battery (100 A) or a welding machine. A machine with an output voltage of 40 V is sufficient for the test. The temporary ground-bed usually consists of scrap length of pipes, driven steel rods or aluminum foil. If desired, actual anode could be used instead of scrap length of pipes and left on the location. They can be later utilized in the permanent ground-bed. For final installation, currents from 2-3 A up to full ratings of the machine can be utilized. The procedure is continued until the quantity of current required to protect the entire line of test section is determined. The ground-bed must be between 100 and 50 ft of the pipeline. If the temporary installation (pipeline + ground-bed) is operated, for instance, at 25 A and a sufficient protection is obtained, it signifies that a rectifier, when set at this value, would also protect the pipeline. It is to be remembered that the temporary ground-bed described above has a high resistance, therefore, a high voltage is required from the welding machine. When a rectifier is installed with a proper ground-bed (low resistance), the voltage used would not be as high. The temporary ground-bed also would be consumed in a few days under test conditions because of high power consumption, however, it has no bearing on the protective current density determined for the protection of the pipe by the above test method. The method is illustrated in Fig. 5.40. Current is drained at point A by the welding machine. Two points, such as B and C are located on the pipeline test section after a polarization run of 3 or 4 hours. The two points should Cathodic protection 3 21 P/S = ~ L 0 Volt P{S**~ 0,8 Volt P/S a p\pe to soil potential s Figure 5.40 Current drainage method have potentials of 1.0 and 0.8 V, respectively. The section between the two points will be at a potential of about 0.9 V, which is the potential required for protecting the line. The line current is measured between each of these points. The difference between the line currents at the two points would be the current picked up. If two points having P/S of —1.0 V and —0.8 V are not located, readings are taken which approximate the above values. Potential and current are adjusted to obtain the desired quantity. A static potential of —0.6 V is assumed. On the above basis, P/S and current curves are drawn as straight lines, and the points where the values are —1.0 and —0.8 are located in Fig. 5.41. If the distance between the two points is L miles, the average current required to bring them to 0.9 V line JB - fc> and (JB - IQ)IL would be average current, in amperes per mile. Once the two points are located, the current value at those points is determined. The total current requirement to raise the potential in a range —1.0 to —0.8 V can, therefore, be estimated. It also possible to obtain data that would give information on the current requirement while the coating resistance tests are being made. Such a setup is shown in Fig. 5.42. In the setup a current interrupter is provided (automatic switch). The switch is 'on' position for 30 seconds and in 'off' position for 15 seconds. The data is taken at every specified length's interval for calculation of coating resistance. The values of A V (voltage 'on' and voltage 'off') and AI (current 'on' and current 'off') are measured. The average pipe potential change, which is mostly the voltage drop across the pipe coating due to the current increment (arithmetic difference between AI value) is determined. From the data, the current required to protect the coated pipe as well as the resistance of the coating is estimated. The estimate is preliminary only. S .26 CURRENT DENSITY R E Q U I R E M E N T FOR A BARE S T R U C T U R E In case of a bare structure, or a pipe, a current requirement test is made by placing the electrode at a position remote to the structure. The setup for a current requirement test of a bare structure is shown in Fig. 5.43. The most important condition is that the structure must be fully polarized to a potential —0.85 V for complete cathodic protection. It is to be noted on the current off position, the potential must show —0.85 V immediately. This value should not be reached gradually on 3 22 Principles of Corrosion Engineering and Corrosion Control Figure 5.41 Measurement of pipe-to-soil P/S potential and line current. The difference in the current values for the two points gives the current demand for the distance L Temporary ground connection T High resistance voltmeter rnV oo -=» DC Current source High resistance voltmeter fk j DC Ammeter mV oo Automatic Switch Well coated pipe Insulating flange Figure 5.42 Current requirement test using close Cu/CuS04 electrodes. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Cheveron, USA) opening the 'current on switch. It may take a very long time in several instances to achieve polarization. To shorten the waiting period for achievement of complete polarization (—0.85 V), the curve can be extrapolated. Figure 5.44 shows a polarization curve. It is a plot of pipe potential to remote earth vs time. The immediate objective is to determine the amount of current for the test section to achieve a potential of -0.85 V. Temporary grounded O —O—I—O i Power O C& J Ammeter \ Switch Insulated flange Pipe Bare or poorly coated pipeline Insulated flange High resistance voltmeter i I €u/€u$04 reference electrode (remote) Figure 5.43 Cathodic protection of a bare structure. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Cheveron, USA) -0.90 Test current ON -0.50 L 10 20 30 Hours 40 50 60 Figure 5.44 Polarization curve. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Cheveron, USA) 3 24 Principles of Corrosion Engineering and Corrosion Control (d) The total current required would, therefore, be 25 + 0.822 = 25.822, or say 26.0 A. The field test data can be modified by this method. (e) The current approximation for a temporary DC power source on the basis of 2.50 mA/ft is calculated as below for a bare line = r6-5/8in~l L 12 r ft 1/ A\ 7 5280 — T 0.0025 - r2 _T 1 niileJV ft / Referring to the polarization plot shown in Fig. 5.44, the following are the salient characteristics: (1) The broken line in the plot shows the pipeto-soil potential before the application of cathodic protection, in the diagram it is shown as 0.560 V. (2) Variation of pipe potential with time read in current 'off' position (curve II). (3) Variation of pipe potential with time read in current 'on' position (curve III). The curve extrapolated to read maximum potential, -0.838 V. (4) The value of current recorded to achieve maximum potential is recorded. Assume the current value as 25 A (JA)« Example Assume that the test section of bare pipe, 1 mile long, 6-5/8" OD, is to be protected, and the approximate current requirement is 25 A. An approximate quantity of current is introduced in this pipe or buried structure. This approximate value is based on the experience of cathodic protection engineer. Later this approximate quantity is modified on the basis of the results obtained from the polarization plot. The following steps show clearly, how the exact current requirement is worked out: (a) Determine the total change of voltage (A£) from the original value of pipe potential (-0.560 V) = T ft 1/ A\ 3.417T 5280 — 0.0025 - r ) L 12 _T 1 mileJV ft2/ = 0.25A r 0.5521 s2. .61 C U R R E N T DENSITY REQUIREMENTS Table 5.17 shows typical current requirements of uncoated steel in various environments. Table 5.17 Typical current density requirements for cathodic protection of uncoated steel Environment Current density, (mA/ft2) AFM 88-9 0.4 to 1.5 2 to 3 lto6 3 to 15 6 to 42 Gerrard 0.4 to 1.5 2 to 3 2.5 to 6 5 to 15 up to 42 Neutral soil Well-aerated A£ = -0.838 - (-0.560) = -0.278 V neutral soil Wet soil (b) Calculate the average potential change per Highly acidic soil ampere for the test section Soil supporting active sulfate AE 0.278 V reducing bacteria = =0.01127 25 A Heated soil Stationary fresh water (c) The additional current required to raise the Moving fresh water potential from -0.838 to -0.850 is Turbulent fresh - 0.850 V - (-0.838) = -0.01 V water containing dissolved oxygen 0.01 V Seawater = 0.822 0.0112 A 3 to 25 1 to 6 9 to 25 5 to 15 5 to 25 5 5 to 6 5 to 15 3 to 10 5 to 25 Cathodic protection 3 25 5.27 STRAY C U R R E N T CORROSION s .27.1 INTRODUCTION The part of the pipe receiving the current from the rails becomes the cathode and the part of the pipe from which the current leaves the pipe becomes the anode. It has been shown earlier that anode is the area of exit of F e + + ions (corrosion site). In a cathodic protection system, the conventional current (e.g. Z n + + ions) flows from the 5.27.2 M A J O R S O U R C E S O F ground-bed through the earth towards the metallic structure. If a current encounters a metallic S T R A Y C U R R E N T S structure on its way it is picked up by the metallic structure, transmitted to other parts of the (a) DC Transmit Systems, such as structure and finally discharged from the struc- Electric Trains ture through the earth and returns to the cathode. The following are the two routes: Stray currents create a great source of difficulty in DC transit systems. In such transit systems, (a) Ground-bed (Anode) - Earth - Metallic the overhead feeder is connected to the positive structure - Earth - Cathode bus of DC substations. The load current which is (b) Ground-bed (Anode) - Earth - Cathode. required to operate the trains is expected to return via the tracks connected to the negative bus. For If an underground metallic structure is present, illustration, one train and a substation of a transit the point at which the current is discharged system is shown in Fig. 5.45. A part of the load from the metallic structure to the ground becomes current may enter the earth as the tracks are not the anode and, therefore, corrodes (forming Fe + + completely insulated from the earth. Load curions). The undesired current which enters the rent would enter where the tracks are positive and metallic structure on its way to the cathode is take a path back to the substation. In the current called stray current and the corrosion caused by pick areas, any underground bare pipes or other stray current is called stray current corrosion. This metallic structures would pick up the current and type of corrosion is associated with DC tran- they would be cathodically protected at no cost sit systems, such as in the case of electric trains to the other. The current after being picked up on super-grid contacts which cross the electrical by a pipeline would be carried in the neighborlines. There are numerous other sources of stray hood of the DC substation where it would be currents, such as welding machines, elevators, discharged as Fe + + ions to the earth. The diselectroplating machines, etc. Speaking simply, charge area of the pipe would, therefore, become stray currents are uncontrolled currents which the anode and the pipe would seriously corrode. originate mostly from DC systems and cause Thus, in a current pickup area the pipe would corrosion at the point of leakage from the sys- be protected and in a current discharge area it tem (site of exit of Fe + + ions). There are two would corrode. If all negative return was to be main categories into which these currents can be carried by the tracks completely insulated, divided: this problem would not exist. In DC systems, the negative side of the generator is grounded because of (A) Static type the contact of rails or structures with the earth. In the example given of an electric railway system (1) Cathodic protection rectifier the positive side of the generator is connected to the trolley contact wire through the feeder cables (2) Railroad signal batteries and the negative side to the rail. The return path of the current is shown in (B) Dynamic type Fig. 5.45. The current flows back to the generator, either through the rails or an earth path parallel (1) DC equipment in mines to rails. There is strong possibility of such current (2) Electric railway generating equipment. 3 26 Principles of Corrosion Engineering and Corrosion Control DC - *— DDDD1DD J&LW.,.. Protected pipe Stray currents Eart& return Pipe corrosion "' , W CEE Figure 5.45 Showing how a current flowing between a train and the distant generating station tends to split itself between the track and a pipeline being picked up by a nearby metallic structures. From the structure the current would discharge to the earth, returning to the rails and finally to the negative terminal of the generator. The underground metallic structure would corrode at the point where the current discharges back to the ground before its entry to the rail and its return to negative terminal of the generator. Any underground metallic structure would corrode at the point of exit of Fe + + ions. To prevent this undesirable stray current corrosion a metallic bond, such as a bond cable between the pipeline and the negative bus of the DC substation, is installed as shown in Fig. 5.46. The current is then drained off by the metallic bond and all the surface of the secondary pipes becomes completely cathodic. The situation here is rather over-simplified, as there may be hundreds of substations serving the system depending on the traffic load and the load may vary during the 24 hours period. In certain instances, a bond connected may not be useful as the direction of flow of current in the bond may reverse and the current may flow to the pipelines rather than to the negative return. In order to handle this problem, rectifier discs may be inserted in the circuit so as to prevent the reversal of the current flow. involved and they escape to earth where they are picked up by the buried metallic structures. While the ships are being repaired in the dock basin, welding generators are placed on shore and the ground DC lines are taken to the ship. Any current returning from welding electrode through the water to the shore would destroy paint work and cause damage to the hull of the ship. This happens particularly where the earth circuit is closed. The stray current on the ground would be picked up by underground metallic structures and cause corrosion at the point of exit. In such instances, the DC generator should be placed on ship rather than on the shore to minimize the magnitude of stray currents. 5.28 CATHODIC PROTECTION INTERFERENCES For cathodic protection of an underground structure, current is directly injected into the soil in the neighborhood of the pipe or structure to be protected. High current density, therefore, exists close to the ground-bed. If another metallic structure, such as a water pipe or a cable, happens to be present in the immediate neighborhood of the (b) Welding Generators cathodically protected structure, it would pick up During the electric arc welding of ships and fab- some current at one point and discharge it via rication of large structures, large currents are the soil at other point on the metallic structure. Cathodic protection 3 27 DC Substation Overhead positive reeoer #»«•< Load current required to operate train Pipe £ ^3 D DPPIDD Stray Tracks negative return _. / ^ v ~ JJJM dr Discharge area subject to corrosion Moving current pickup area Current flowing around high resistance or insulating joint Figure 5.46 Stray current corrosion caused by DC transit system. (From Peabody, A.W. (1967). Control of Pipeline Corrosion, NACE. Reproduced by kind permission of NACE, Int., USA) Any water pipe or cable in the vicinity would corrode as a result of discharge of positive current from it (Fe + + ions). This is an unwanted corrosion and is referred to as cathodic protection interference. It refers to an undesirable current discharge from a pipeline or structure in the vicinity of a cathodically protected pipeline or structure. s .28.1 EXAMPLES OF INTERFERENCE A foreign (unprotected) pipeline may either cross a protected pipeline or passes close to the ground-bed of the protected line. Such situations cause varying degrees of interference. Suppose a foreign pipeline does not cross the cathodically protected line but passes very close to a groundbed of cathodically protected line. This is an example of interference caused by radial current flow. When a foreign pipeline lies across the normal flow lines of current, there would be a little difference in the potential between the path nearest to the anode and nearest to the cathode as only small amount of current would be interrupted. If, however, the foreign pipeline runs parallel to flow lines of current, there would be a large amount of current pickup which would be discharged to the protected structure. This is because current prefers a low resistance path (metal path) rather than higher resistance (via soil). A tank close to an anode bed may pickup current at one point and discharges it as F e + + ions from another point (Fig. 5.47). The discharge area would undergo heavy damage and corrode due to the consequent loss of Fe. In areas of positive soil potential, the foreign pipeline would pickup the current. This current would leave the foreign pipeline in remote areas in order to reach the protected pipeline and flow back to the rectifier to complete the circuit. Because of several areas of Fe + + current discharge, the foreign pipeline would corrode. However, if a foreign pipeline is coated, it would not be affected to the same degree as a bare pipeline or structure, although it may be in the region of potential gradients. To ascertain the damage, potential readings can be taken at the points of crossing of the pipes. If the potential of the foreign pipeline becomes positive, it would indicate corrosion caused by Fe + + current discharge. In order to prevent this, the bare pipelines maybe coated in the region of crossing or a metallic bonding may be introduced between the two as illustrated in Fig. 5.48. 328 Principles of Corrosion Engineering and Corrosion Control -Soil Protected Pipe •RQEgefiflr ^•A^- . Ariose Jbed /" Soil Soil - Figure 5.47 Damage caused by current flow. Corrosion occurs at position 2 due to discharge of F e + + ions 4r / / */ i i x * Anodes (scrape iron or graphite) / i i / / * i i / Rectifier DC power AC source wi /// /// / * i t / / t \ Sdii i i l i i * i » Soil i \ \ \ \ Bonding \\\ v\s \\\ \\\ \ \ * \ \ k k^-iuJi Protected pipeline ft Figure 5.48 Showing current impressed by a DC source being picked up by an unprotected line and returning to the protected line via a bonding link Cathodic protection The following is a summary of preventive methods which could be used to minimize the interference: 3 29 involving a foreign line, the foreign line must be coated. (1) The current output of the main rectifier may S .29 D E S I G N CHARTS be reduced. (2) The ground-bed maybe re-sited, if necessary. Figure 5.49 shows a typical horizontal anode This is applicable if a foreign pipeline passes design chart where the number of anodes is shown close to the ground-bed. on the abscissa and the resistance of the anode (3) Installation of a crossing bond between the at different spacings on the ordinate axis. The pipes. A bond between the two points of following basic information is required for the crossing is installed and the amount of cur- construction of a design chart: rent flow is controlled by a resistor. (a) Anode size (4) Installation of magnesium anodes on the (b) Size of coke breeze column corroding structure. (c) Resistivity of coke breeze (5) Isolation of the anodic section of the struc(d) Anode spacing tures and installation of continuity bonds (e) Soil resistivity across the anodic section. (6) Coating the metal/electrolyte interface, or The resistance of a given number of anodes in the contact surfaces. In cases of corrosion a soil of 100 ohm-cm at a specified spacing can Anode spacing (centerline to centsrline) 10 15 Number of anodes Figure 5.49 Typical horizontal anode design chart for impressed current ground beds. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Cheveron, USA) 3 30 Principles of Corrosion Engineering and Corrosion Control Once the current requirement of a structure is established, the ground-bed is designed to keep the resistances as low as possible. Such an arrangement allows a small voltage for the purpose of driving the required protective current. By lowering the resistances, the cost of power consumption is also reduced. The simplified relation given below can be used to determine the resistance of a single or vertical anode to earth. be directly noted from the chart. It is to be noted that separate charts need preparing for each anode size, size of coke breeze column and resistivity of the coke breeze. Resistance can be conveniently converted from one soil resistivity to another soil resistivity. Suppose it is desired to obtain resistance of 20 anodes in parallel at 20 ft spacing in a soil of 3000 ohm-cm, the following is a step-wise procedure. 8L 1 RY = 0 .00521- x 2.3 log (1) Obtain the parallel resistance of 20 anodes at 20 ft spacing in 1000 ohm soil directly from the chart. It is known as D'Wights equation for single (2) Determine the resistance in a soil of vertical anode, where 3000 ohm by conversion. Suppose the anode Ry = resistance of vertical anode (ohms) resistance determined is 0.50 ohm. It to earth would be L = anode length in feet p = soil resistivity (ohm-cm) /3000\ 0.50 x = 1.5 ohm-cm d = anode diameter (in feet) -H Vioooy A more common form of equation for multiple (3) Add to this resistance (1.5 ohm-cm) the vertical anodes is: internal resistance of the anode and divide by the number of anodes in parallel (twenty 81 2.303 log RY = 0.0052 in this case). m '2L The above procedure can be used to determine x 2.303 log 656N +i the anode to earth resistance. Design charts can be constructed for vertical anodes and horizontal anode ground-beds for both impressed and gal- It is called Sundae equation. vanic anode systems. The development of design H. B. D'Wights equation can also be used curves is discussed in Section 5.31.2. to determine the resistance to earth for a single horizontal anode. Rh = 0.00521p (2303 log — 5,30 GROUND-BED DESIGN + 2.303 l o g - + — - 2 J The ground-bed design is an essential component of a cathodic protection system. The following are where the salient points which need consideration while L = anode length in feet selecting a site for the ground-bed installation: Rh = resistance of horizontal anode d = anode diameter (including the backfill) (1) Presence of metallic structures in the vicinity h = depth from surface to center of the of the ground-bed. anode (ft). (2) Location of the site with respect to the For multiple anodes, multiply R^ by the pipeline. adjusting factors for the parallel anodes given in (3) Source of power. standard tables. (4) Ready access. Cathodic protection 3 31 With the help of the above formula, it can also The following relationship is widely used to be shown that a long slender anode has a lower determine the resistances of the anode backfill: resistance than a short square anode, given the 0.0171p same weight. RV = ( 2303 log ^ - l ) Illustrative Examples Calculate the resistance to earth of a 3 in x 60 in long vertical anode in 1000 ohm soil. The anode length is 5ftand diameter is 0.25 in. What would be its resistance in a 5000 ohm-cm soil? Solution: Using the formula mentioned above (D'Wights equation) R= 0.00521/? S .31 S OIL RESISTIVITY AND P I P E - T O - S O I L POTENTIAL S U R V E Y 5 .31.1 SOIL RESISTIVITY SURVEY ( 2 .303logf-l) and inserting the appropriate values R= 0.00521 x 1000 (* 303 log 8 x 5 0 25 4.24 ohms In a soil of 5000 ohm-cm, the resistance would be 4.24 x 5000 = 21.20 ohms 1000 Conducting soil resistivity surveys is a primary step in designing of a cathodic protection system for pipelines. The methods of determination of soil resistivity have been described in an entire section. Pipelines in low resistivity soils would require a greater amount of current for protection than pipelines in a high resistivity soils, because of a higher magnitude of corrosion in the former. Hence, low soil resistivity areas are selected to install the anode ground-bed. Figure 5.50 shows a typical plot of soil resistivity survey. The peaks in the plot indicate the areas of high soil resistivity and the valleys, the i S O 6000 Kilometres Figure 5.50 Soil resistivity data 3 32 Principles of Corrosion Engineering and Corrosion Control areas of low soil resistivity, the preferred areas for along wet coated pipelines. The peaks in the installation of anode ground-beds. potential profile represent locations liable to corrode. The basic idea of the survey is to detect anodic and cathodic areas on the pipe for cathodic proDetermining the degree of cathodic protection: tection. The potential readings differ for coated A reading less negative than -1.2 V vs Cu/CuS04 and uncoated pipelines and are influenced by galwould indicate adequate C. P. and a readvanic corrosion and external current. Therefore, ing more positive than -1.2 V would indicate no generalized statement can be made on the inadequate C. P. value of potentials above which the areas are anodic and below which are cathodic. More negative potentials indicate the anodic areas in general and these areas need to be protected. The fol- 5 .31.2 C O M B I N E D P L O T S lowing are the guidelines for determining the ( C O M P O S I T E P L O T S ) pipe-soil potential: Composite plots of pipe-to-soil potential and soil (1) Potentials for newer pipelines are more neg- resistivity surveys can be constructed to idenative than the other ones. tify the hot spots (corroding areas determined (2) Potentials are more negative in neutral and by both soil resistivity and pipe-to-soil potential acidic soils than those in alkaline environ- surveys). The corrosive areas in the composite figure are shown by shaded areas where the ments. (3) Regions of more negative potential corre- peaks in the pipe-to-soil correspond with the valspond to locations of low soil resistivity leys in the soil resistivity survey (Figs 5.50-5.52). A low soil resistivity (say, 1000 ohm-cm) and (anodic areas) in uncoated pipelines. (4) Typical values of potentials along uncoated a high negative value of pipe-to-soil potenpipelines are, in general, in the range of tial (say, —1.3 V) would constitute a corro-0.5 V 0.6 V and -0.65 V 0.75 V sion area. -1.7 -L6h Adequate protection Inadequate protection ~&8 0 JL JL 3000 6000 9G00 12000 Pipeline length (meters) Figure 5.51 Pipe-to-soil potential survey. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Texaco, Houston, USA) Cathodic protection 3 33 100000 3000 4000 Pipeline length, ft Figure 5.52 Composite data - pipe-to-soil potential and soil resistivity surveys. Shaded areas represent hot areas which need protection. (From TEXACO Cathodic Protection - Design and application school, Texaco Houston Research Center, Training Manual. Reproduced by kind permission of Texaco, Houston, USA) 5.32 C A L C U L A T I O N S IN (c) Dimensions of the backfill Suppose the following data is given: a = 40 ohm-cm (backfill resistivity) b = 3" x 5' (anode dimensions) c = 6" x 1' (backfill dimensions) CATHODIC PROTECTION DESIGN 5.32.1 DESIGN CURVE CALCULATIONS It has been stated earlier that the resistance of a solution: number of anodes can be determined from design Applying Sundae equation, determine first the charts. The design charts are made for definite resistance of vertical anode to outer edges of the sizes of anodes, in a known soil resistivity. The backfill \R\)following calculations show how the design curves 0.00521 / , 8L \ are constructed. *va = ^ p ( 2 . 3 l o g T - l J (A) Determination of resistance of the anodes to outer edge of the backfill (internal resistance) In order to determine the internal resistance, the following data would be required: (a) Resistivity of the backfill (b) Dimensions of anode #v = (0.00521)(40) r , 4° 2 . 3 1 o0g — - 1 .25 = 0.166 ohms (use anode dimensions and backfill resistivity) 3 34 Principles of Corrosion Engineering and Corrosion Control (0.00521) (40) Rvhf = = 0.107 ohms RX = £ Va ~ RVM 2.3 log --'1 0.5 J (C) A design curve is to be developed for 10, 15, 25 and 30 anode beds at a 10 ft spacing in 1000 ohm-cm soil resistivity. Required data: (1) p = 1000 ohm-cm (soil resistivity) (2) N = 10, 15, 20, 25, 30 (number of anodes) (3) 5 = 10 ft (anode spacing) (4) d = 6" x 1' (backfill dimensions) = 0 .166-0.107 ohms = 0.059 ohms (B) Calculation of total resistance of one anode to earth Data required: i?i = resistance of anode to outer edge of backfill (internal resistance) p = resistivity of soil (1000 ohm-cm) d — dimension of backfill column (6" x 7') as in (A) R2 = resistance of backfill column to soil Solution: ^v(total) = R\ + Ri (internal resistance -h backfill to soil) R\ — has already been determined in (A) to be 0.059 ohms #2 — resistance of backfill to soil is now to be determined, which is (0.00521)(1000) 0.00521 p{(2.3logf-l) + ( — 2.3 log 0.656 N ] | Workout R for 10,15,20,25 and 30 anodes from the given data. (1) F o r N = 1 0 (O00521X1000)}/ R= 8x7 i31og \ io(7) |( a5o"!) + (^2.3log0.656(10)M = 0.47216 R2 ( 2 .31ogg-l) (2) F o r N = 1 5 ^0.00521X1000) j / 15(7) IV 3 = (0.74)(2.3 log 1 1 2 - 1 ) = (0.74)(2.3x2.04-l) = (0.74)(3.692) = 2.76 ohms 8x7 0.50 X / + p^2.3log0.656(15)H = 0.343 (3) F o r N = 2 0 R= (a00521)(1000)|^3lo88x7_^ 20(7) # i — has already been determined in (A) to be 0.059 ohms AS jRv(total) =Rl+R2 = RT RT = ^v(total) = 0.059 + 0.76 = 2.822 ohms + f^2.3log0.656(20)M = 0.272 Cathodic protection (4) ForW=25 \ (0.00521X1000) \U a 1 8 x 7 R = (25)(7) t^OSO-V (2) i ? i 0 i b r S = 1 5 ( 0.00521X1000)1/ 10(7) IV h 335 8X7_\ o.5 ; +Y ^ 2 . 3 l o g 0 . 6 5 6 ( 2 5 ) M = 0.227 + (^2.31og0.656(10)M = 0.471 ft (3) £ 1 0 forS = 20 (0.00521)(1000) ( / (5) F o r N = 3 0 R= (0.00521)(1000) [/, „ _, 8x7 1 \ io(7) ll2-310^-1; 8x7 \ {{'^Tf- ) (30X7) +f^2.3log0.656(10)M = 0.3742 ft (4) i? 1 0 forS=25 (0.00521)(1000) \( 10(7) \\ +( = 0.355 £2 (5) jRi 5 for5=10 (0.00521)(1000) 10(7) /2x7 , ,x + ( —2.3log0.656(15) ) ! = 0.343 £2 (6) #i5 for 5 = 1 5 (0.00521X1000) J / . 23l /2x: +( 2.3log0.656(30) = 0.1955 )l 8x7 °^- 1 \ ) From the above calculations tabulate the results as below: Number of anodes 10 15 20 25 30 R (ohms) 0.47216 0.343 0.272 0.227 0.1955 2.3log0.656(10) Repeat the calculations for the following anode spacing 5=10 5=15 5 = 20 5 = 25 For 5 = 15, N= 10 (1) £ 1 0 f o r 5 = 1 0 (0.00521)(1000) 15(7) K' 2 310 8x7 ^- ) 1 \ + f^2.3log0.656(15)M = 0.290 £2 (7) £ 1 5 for5 = 20 , 8x7 (0.00521)(1000) {(2.31og^-l) 15(7) IV ""° 0.! + f^2.3log0.656(15)M = 0.2635 £2 K 2.31og — - : +(^2.3log0.656(loAj = 0.471 £2 3 36 Principles of Corrosion Engineering and Corrosion Control (8) Rl5 for S = 25 (0.00521)(1000) f ^ a | For R = 25 and S = 10,15, 20 and 25: (13) £25 for S = 10 (0.00521)(1000) 25(7) + f^2.3log0.656(25) = 0.248 Q = 0.229 ft (9) J? 2 0 forS=10 ( 0.00521)(1000)f^ a | (14) ^ 2 5 f o r S = 1 5 )) 15(7) I^-OJ-1) 8x7 \ + ( 2 ^2.31og0.656(15)M — m — \{ - iJ- ) + ( 2 -^2.3log0.656(2oAj = 0.272 £2 2 3H 8x7 1 \ (0.00521)(1000) 25(7) K - ^- ) 2 310 1 + ( ^2.3log0.656(25)M = 0.19ft (15) i^ 2 5forS=20 (10) i?2oforS=15 (0.00521X1000) \( 8x7 \ (0.00521X1000) j / 8x7 g \ —20(7)—K2-3108-^-1; + ( - ^ - 2 . 3 log 0.656(20) ) 1 25(7) IV 0.5 J + ( ^2.3log0.656(25AJ = 0 .17^ = 0.2275 ft (16) ^25forS=25 (11) i^20forS = 20 (0.00521)(1000) \( 20(7) (0.00521)(1000)j/ 2 3 8 X7 6 \ 25(7) 8x7 \ IV 0.5 J + ^ ^2.3log0.656(25AJ = 0.158Q From the above results tabulate the calculations in the following form: No. of anodes Rio Ris Rio R25 + 1 - ^-2.3log0.656(20))| 20 = 0.205 ft C (12) £20 for S = 25 (0.00521)(1000) } / 20(7) 8x7 \ Anode spacing 10 0.472 0.343 0.272 0.229 15 0.407 0.29 0.2275 0.19 20 0.37428 0.2635 0.205 0.17 25 0.355 0.248 0.192 0.158 + f ^2.3log0.656(20)M = 0.192ft Cathodic protection Plot the data and obtain the number of anodes (N) vs resistance curves for a soil of 1000 ohm-cm resistivity and backfill dimensions of 0.5 ft diameter and 7 ft long. The following example shows how the design curve can be used to obtain the resistance of impressed current and galvanic anodes. Separate charts are constructed for impressed current ground-beds and horizontal ground-beds. With each chart the following design information is provided: (1) (2) (3) (4) (5) (6) (7) Anode dimensions Backfill dimensions Anode resistance Soil resistivity Spacing of anodes Internal resistance of anode Type of anode 337 Apply a current of approximately 74.2 A to the pipe. In order to ensure the accuracy of the estimation, a pipe potential vs time polarization curve is plotted as shown in Fig. 5.52. By extrapolation of the polarization curve, the maximum potential achieved by the predetermined current is noted. If, for instance, by applying 74.2 A of current, the potential achieved is —0.8386, the additional current needed to achieve —0.85 is worked out and added to the approximated current to obtain an accurate value of current requirements. For example, the amount of current approximated raised the potential of the structure to —0.838 rather than -0.85 V, which is the required potential to achieve cathodic protection. If the voltage before cathodic protection is applied is —0.6 V, then the total voltage change is AE = - 0.838 - ( -0.6) = - 0.238 V The average potential change per ampere is A£ 0.238 V = = 0.0032AI 74.2 A Amount of current required to raise the potential from -0.85 - (-0.838) = 0.01 V. Therefore, additional amount of current required: 0.01V 0.0032V/A 3.125A The charts can be prepared for special applications by utilizing the data and working out the resistance according to the method given above. 5 .32.2 CURRENT REQUIREMENTS Example to illustrate the total current requirements for a bare 1 mile section of a 10(3/4") OD pipe. The current density required to protect the pipeis50mA/ft 2 . Total current = 74.2 + 3.125 = 77.325 A. Solution: Additional current needed to raise the potenFor a bare steel structure, a potential of —0.85 V tial from -0.838 V to -0.85 V is 0.892 A. The total on it provides a reasonable degree of cathodic current required is, therefore, protection. The structure is polarized by a known amount of current until a potential of 74.2 A + 0.892 A = 75.092 A -0.85 Vcu-CuS04 is acquired by the pipe. Step 1 - Estimation of current requirement 10.751 , ft / A\ x (3.14) x 5280 - — 0.005 - T which is not too far away from the approximation (74.2 A). 12 J mile \ ft2/ 5.32.3 DETERMINATION = 0.895 x 3.14 x 5280 x 0.005 A = 74.2 A O F C OATING R E S I S T A N C E OF A P I P E (An approximate current density on the basis of Determine the coating resistance of 4(6/8)' experience is first selected.) OD pipe, 4 miles long. 3 38 Principles of Corrosion Engineering and Corrosion Control Illustrative Problem 2 Design an impressed current system to protect a coated pipeline 4 mile long, 6(5/8)" OD in a soil of 2000 ohm-cm resistivity. Graphite anodes 3 " x 5' are to be used. The back voltage between the pipeline and ground-bed is 3.0 V. Data (1) In order to cause a potential shift (AV) of 0.2 V, ( -0.65 to -0.85), 0.13 A of current is applied as required by the current requirement test (extrapolation method). (2) The coating has 2% holidays. (3) A current density of 3 mA/ft2 is to be applied. (4) Resistivity of anode bed = 2000 ohm-cm. The following values are to be calculated: (1) (2) (3) (4) (5) (6) (7) Current requirement. Pipe-soil resistance. Maximum allowed circuit resistance. Wire resistance. Anode bed resistance. Anode bed size. Weight of the backfill and its volume. Solution: Data provided: Location 1 2 3 Volts (ON) -1.10 -1.45 -1.30 Volts (OFF) -0.83 -1.19 -1.00 DF 0.27 0.26 0.30 The current measured is 0.05 A. (A) Calculate the average value of AE AE = 0.27 + 0.26 + 0.30 3 _ 0.83 = 0.276 V (B) Calculate the area of the pipe Area = TTDL = 4.625 12 x ( TT)(4)(5280) = 25559.6 ft2 (C) Calculate the resistance, R I _ 0.276 V " 0.05 A = 5.52 ohms The resistance of the coating for one average square feet is (5.52 ohms)(25559.6) = 141088.99 ohms-ft2. Illustrative Problem 1 From the design chart for 37/ diam x 5' long anode, determine the resistance of 25 graphite anodes at 10 ft spacing in a 3000 ohms-cm soil. Solution: (1) Suppose the resistance of 25 anodes at 10ft spacing in 1000 ohm-cm soil is 0.26 ohms from a design chart. (2) The resistance in 3000 ohm-cm soil 3000 ^V3000) = 0.26 x 1 } Solution: Current required (1) From the data given above, surface area of the pipe ffl* )(4)(5280) = 36474 ft2 (2) Percentage of uncoated pipeline = 2%. (3) Current density required, 0.003 A/ft2. Total current needed = 36474 x 0.02 x 0.003 = 2.188 A. A rectifier, possibly 12 V, 4 A may be installed. However, an 18 V, 6 A rectifier is recommended to make allowances for future requirements. (4) Pipe-to-soil resistance from the given data: J = current applied for voltage drop = 0.13 A AE = voltage drop = 0.2 V AE 0.2 = 1.5 ohms. jR = I 0.13 = 0.78 ohms 1000 Cathodic protection Calculations of maximum allowable circuit resistance in this case Back voltage = 3.0 V (provided in the data) Rectifier voltage = 18 V (available from the rectifier) A£=18-3 = 15 V (maximum rectifier output) 7 = 6 A (maximum current output) Maximum allowable circuit resistance: A£ 15 = — = 2 . 5 ohms. I 6 Calculation of resistance of wire Wire selected (gauge of wire) = 6 Resistance of the wire = 0.410 x 10" 3 ohms (from tables) Length = 60 ft Additional length for safety = 10% of the original = 6 ft Total length of wire = 66 ft Resistance of the wire = 0.259 x 1 0" 3 ohms (0.259 x 10~3)66ft = 0.017 ohms. Anode bed resistance Maximum circuit resistance = 2.5 ohms Pipeline-soil resistance =1.5 ohms Wire resistance = 0.017 ohms ^(anode bed) = maximum circuit resistance — pipe to soil resistance — wire resistance = 2.5 - 1.5 - 0.017 = 0.983 ohm in a soil of 2000 ohm-cm. The resistance in a soil of 1000 ohm-cm shall be 0.983 x 1000 2000 0.491 ohms. 339 (8) Calculation of the anode bed size Suppose 25 anodes are placed at 10 feet spacing. The resistance of 25 anodes from the design chart (suppose) = 0.26 ohms in 1000 ohm-cm soil. The resistance of 25 anodes (anode bed) in a soil of 2000 ohm-cm = 0.26 x 2000 = 0.520 ohms. 1000 Internal resistance (from data) = 0.520 ohm/25 = 0.020 ohm. The header cable resistance (assume) = 0.017 ohm Total anode bed resistance = Resistance of 25 anodes bed + internal resistance + header wire resistance. Total anode bed resistance: 0.020 + 0.017 (internal (header-wire resistance) resistance) = 0.557 ohms (9) Total resistance of circuit Pipe-to-soil resistance = 1.5 ohms (given in data) Anode bed resistance = 0.557 (Step 8) Lead wire resistance = 0.017 (Step 6) Total = 1.5 + 0.557 + 0.017 = 2.077 ohms It is obviously less than the maximum permissible resistance of 2.5 ohms (as shown in Step 5). (10) Weight of backfill Anode dimension (given) = 3" x 5' = 0.245 ft3 Backfill dimension = 8" x 1' = 2.446 ft3 Number of anodes is 25 at a spacing of 15 ft = 0 .52+ 3 40 Principles of Corrosion Engineering and Corrosion Control Backfill volume = 25 (2.45 - 0.246) = 5 5.1+ 20% for over-design = 66.12 ft3 Weight of backfill at 70 lb/ft3 = 7 0x66.12 = 4628 lb/ft3 __ (48)(0.116)(0.50)(0.85) ~" 0.103 = 22.97 years * consumption rate = A/year/lb (inverse of theoretical consumption rate) ** efficiency of anode Illustrative Problem 3 Calculate the expected current output of a single 48 lb Galvamog magnesium alloy anode. The size of the backfill package is 10" x 40". The steel has been polarized to a potential of —0.85 V. The resistivity of soil is 2000 ohm-cm. The solution potential of Galvamog is —1.75 V. Calculate the life of 48 lb magnesium anode. Solution: (a) First determine the resistance of the anode from the given data: d = 10" = — ft = 0.83 ft 12 // 40 r L = 40" = — ft = 3.33 ft 12 R= T 0.00521 x Illustrative Example 4 The interior of a tank is to be protected. The tank contains 5000 barrels of salt water. The following data is provided: (1) (2) (3) (4) (5) (6) (7) Water level = 25.0 ft Height of tank = 30.0 ft Diameter of tank = 40 ft Water level is maintained at 25 ft Current density = 4 mA/ft2 Resistivity of water = 1 0 ohm-cm Length of 2 AWG wire = 130 ft Resistance = 0.162 x 10~3 ohms/ft From the above data, calculate the following: (1) Current requirement. (2) Current output per anode if 3 " x 5' graphite anodes are used and the anode current density is 2.0 A/ft2. (3) Number of anodes required. (4) Anode to electrolyte resistance. (5) Resistance of lead wire (# w ) 130 ft long. (6) Resistance of anode. (7) Potential drop in the circuit. (8) Size of the rectifier to be selected. Solution: (1) Firstly, the area of the tank in contact with water is calculated (wetted area, Aw) Aw = nd2 = ndL 4 T T(40) 2 + T T(40)(25) L = L 'K - ] L 0.00521 r 8x3.33 (2000) In 3.33 = 7.72 ohms °-83 -] (b) The driving potential = -1.75 - ( -0.85) - ( -0.10) = 0.8 V. (c) According to Ohm's law: I=I= E R 0.8 = 0.103A 7.72 = 1256 + 3140 = 4396 ft2 The current requirement is: The life expectancy is determined by (Weight of anode) (0.116) (0.50*) (0.85**) 0.103 7 = 4396x0.004= 17.6 A. Cathodic protection (2) Current output of each anode. From the previous calculation we have observed that the current requirement is 13.2 A. The area of the anode surface is (Aa) Aa = dnL 3n = — x5 12 12 2 2 3 41 The resistance of the circuit is given by 1 R Here Rw = resistance of the wire which is calculated to be 0.02 ohms £ a = resistance of the anode to electrolyte which has been determined to be 0.0425 ohms Therefore, 1 4 0.021 + 0.0425 4 0.0635 = 62.992 R = 0.016 ohms (6) The potential drop is determined by Ohms' law: E = Ix R = 17.6 x 0.016 = 0.282 V Rw + -Ra = 3.92ft ( - 4 ft ) The current output is, therefore, Q = Cda x Aa(Ca = current output of anode, Aa = area of anode) Cda = 2 . 0 - r X 3.92 = —8.0 — (Cda = current density anode of anode) ~R As the total current requirement is 17.6 A, the number of anodes required is: 17.6 ~ 2 anodes (3) Anode to electrolyte resistance use D'Wight equation as before: „ 0.00521 0.00521 I" 8L "I (10) Lii-il L 0.25 J = 0.0425 ohms The size of the rectifier to be used must be a 30 V-30 A circuit rectifier would be quite suitable. (4) Calculation of resistance of lead wires Illustrative Example 5 The resistance of the lead wires is the estiCalculate the current output and life expectancy mated average length x resistance of 2 AWG of a Galvomag anode from the following data: wire. The known resistance is 0.162 x 3 10~ ohms/ft. Backfill package size 3 = 8" x 32.5" (0.65' x 2.710 Resistance = 130 x 0.162 x 10~ ohms Potential of the polarized structure = - 0.95 V Soil resistivity = 2000 ohm-cm = 0.021 ohms. Solution potential of Galvomag anode =-1.75 V (5) The resistance of the anode. We have now Weight = 48 lbs all the necessary information to calculate the Polarization potential = — 0.10 V resistance of the two anodes. 3 42 Principles of Corrosion Engineering and Corrosion Control Solution: Apply the following relationship to obtain the resistance: R= 0.00521 pr (3)P= where T 8L i p= £= L= A= (4)P= T resistivity in ohm-cm resistance length between two points area Inserting the value 0.00521, r 21.7 1 2000 In 1 2.71 L 0.67 J R= = 9.53 ohms The driving potential (ED) ED = - 1.75 - ( -0.95) - ( -0.1) £ D = 0.7V Determine the current J, by using E= IR I= 0.700 = 0.073 A 9.53 where L = length between two points in the soil or metal surface. (5) Four Pin, DC Method for soil resistivity p= ITVAR A = distance between two points, in cm R = resistance between two points (6) Four Pin, AC Method for soil resistivity p= D= £= p= 191DR pin spacing, feet instrument (potentiometer reading) resistivity (7) Life of Mg anode (years) _ 57.08 x W _ 38.24 x W Life expectancy = (48)(0.116)(0.50)(0.85) 0.73 where LMg Lzn W J = life of magnesium anode = life of zinc anode = anode weight (lb) = current (mA) = 32 years 5.33 IMPORTANT (8) Driving potential of an anode = Solution potential — Potential of polarized structure — Polarization potential (-0.100) (9) Current output of Mg anode and Zinc anode (coated) F O R M U L A E IN C A T H O D I C PROTECTION CALCULATIONS (1) Series Circuit J = Rx + R2 R JMg = (2) Parallel Circuit where 1_ 1 I_ Ri+R2 150 0 0 0 / 7 , JZn = 50 0 0 0 / 7 R~Y1^Y2~TRM) IM% = current output of Mg anode (mA) IZn = current output of Zn anode (mA) Cathodic protection p = soil resistivity, ohm-cm / = factor from Table I Y = factor from Table II (Condition: The above equations apply if the soil resistivity is above 500 ohm-cm and the distance between the anode and structure is not more than 10ft.) (10) E. D. Sundae equation for resistance to earth. (a) Using multiple vertical anodes Ry 4L r i 191.5 NL | 2.303log 10 — - 1 343 (e) Ry (Total resistance of vertical anode) = #i (Resistance to vertical anode-tobackfill) + #2 (Resistance of vertical anode to earth) (f) Working diameter of magnesium anode Area = nd2 (g) Internal resistance of the anode to backfiU R\ = ^V(anode) ~ ^backfiU column 2L + — x 2.303 log10 656 N (b) H. E. D'Wight equation for resistance to earth for a single anode (horizontal) 0.00521 x 4L Rh = ——(p) 2.303 log10 (h) Resistance of a single graphite high silicon cast iron anode installed vertically (no backfill) Ri = 1000 To where ~\ K = constant p = soil resistivity (i) Resistance of graphite of high silicon cast iron anode with and without backfill + 2.303 log10 where RY Rh L S h = = = = = L 2h k+ T-2\ resistance of vertical anode resistance of horizontal anode length (ft) anode spacing, ft depth from surface to center R where IR. x NY 1000 (c) D'Wight equation for single vertical anode Rv = 0.00521 x r i 8 I 1 ; (p) [ 23Iog T -lj p = soil resistivity N = number of anodes Y = spacing factor from the curve £ = resistance in ohms (11) Current requirement on a coated pipe (d) Sundae's equation for multiple vertical anodes R= 0.00521 N (p) NL H ( — 2.3log0.656N ) where Nis the number of anodes As x % uncoated pipe x current density As = surface area of pipe % uncoated pipeline current density (12) Wire resistance Length of wires + 10% safety factor + resistance of wire ohms/ft 3 44 Principles of Corrosion Engineering and Corrosion Control Y = impressed current system design life, years J = total current required in A C = anode consumption rate in kg/A-yr W = weight of anode in kg (b) Number of anodes based on weight (lb) Number of anodes (lb) Total weight of anode material Weight of one anode (c) Number of anodes based on current output, N N= Total current output (A) Output of one anode (A) (use manufacturer's data) (d) Current output of one anode (A) (13) Total volume of backfill needed = (Number of anodes) (Volume of backfill — Anode volume) + 20% (14) (a) Induced emf Secondary winding x Applied voltage Primary winding (b) Efficiency of rectifier DC volts x DC amps x Seconds x 100 Kh x 3600 x Revolutions where Kh = meter constants DCVxDCA or Input watts xlOO Input Watts = Kx Nx 12 (N= number of dial revolutions in a 5 min period) (15) Number of anodes (N) Total current requirement Current output of one anode (A) (16) Rectifier current rating AC Driving potential of the anode (V) Resistance of the anode (ohms) (18) Structure to electrolyte resistance R,= where Von ~ Vqff he = where lac = £ dc = Idc = F= £ ac = £dc X *dc F x £ ac Von = potential (on) VQft — potential (off) Ion = current applied to give Von (19) Maximum circuit resistance R -* AC current (A) DC volts DC current (A) rectifier efficiency AC volts (17) (a) Number of anodes required based on the anode consumption rate N": where N = number of impressed current anodes /YxlxC\ where ED = driving potential of anode J = current requirement (20) Allowable ground-bed resistance -Ragb = ^max(-Rs + -^Lw) where Rs = structure to electrolyte resistance £ Lw = lead wire resistance Cathodic protection 3 45 QUESTIONS A. M ULTIPLE C HOICE QUESTIONS Select one correct answer for the following questions: 1. A bare structure requires more current than a coated structure because [ ] the coated structure corrodes rapidly [ ] the bare structure has a more negative potential than a coated structure [ ] a coated structure is more rapidly polarized than a non-coated structure [ ] the bare structure takes a very long time to polarize than a coated structure 2. A backfill is used around an anode [ ] to provide a uniform environment around the anode [ ] to accelerate the rate of consumption of the galvanic anode [ ] to increase the anode-to-earth resistance [ ] to increase the magnitude of the current which is to be provided to the structure 3. Cathodic protection in a metallic structure is achieved by [ ] polarizing the cathode to the open circuit potential of the anode [ ] polarizing the anode to the open circuit potential of cathode [ ] shifting the potential of the structure to less negative values [ ] producing a film of oxide on the surface of the metallic structure to be protected 4. The following is the criteria for cathodic protection: [ ] A shift in the pipe-to-soil potential in the negative direction by 0.40 to 0.50 V from the initial potential for bare structures [ ] To achieve a pipe-to-soil potential of 0.85 V with respect to Ag-AgCl electrode [ ] To achieve a potential of —0.85 V with respect to a copper sulfate electrode [ ] To polarize the whole structure to the cathodic potential of the structure 5. In the impressed current cathodic protection system [ ] the pulsating direct current goes from the positive (+) terminal of the rectifier to the ground-bed [ ] the pipeline is the positive return or external circuit of the electrolytic cell [ ] AC current is directed to a rectifier where a step-down transformer increases the voltage [ ] it is not necessary to install a magnetic circuit breaker in the AC circuit 6. The following are the disadvantages of the rectifier ground-beds: [ ] Larger driving voltages [ ] Higher current outputs [ ] Protection of larger and more expensive structures [ ] High installation costs 7. The following is the H. B. D'Wight equation for resistance to earth of a single vertical anode: [ ] Ry = L 1 2.303 log10 — - 1 10 191.5 NL |_ d 2L , 1 + — 2.3log100.656N 0.00522 p[ ,41 ~ I 32.303 log10 — L Ih 1 + 2.303 l o g 1 0 - + — - 2 r n [ ] £v = r n 0.00521 f , 8L 1 [ ]Kv = - ^ p { 2 . 3 l o g T - l J [ ] None of the above 3 46 Principles of Corrosion Engineering and Corrosion Control ] place the electrode 100ft from the structure ] place the electrode contacts far enough until there is no increase in the negative reading ] place the electrode always 50 ft away from the structure 13. The driving potential is the systems can be ] difference between the 'on' value and the 'off' value of potentials ] difference between the 'on potential' and 'static potential' ] difference between the solution potential and polarization voltage 1 None of the above 14. The total resistance of the vertical anode to earth is resistance of the anode to backfill + resistance of the backfill to the soil resistance of the anode to backfill — resistance of the backfill to soil resistance of the backfill to soil + resistance of the pipe None of the above 15. A low resistance backfill is specified in anode bed design, because it ] increases the area of contact between anode and the soil ] decreases the area of contact between the anode and the soil 8. The pipe-to-soil potential of a pipe is — 1.35 V and the potential of the polarized structure —0.90 V. The resistance of the anode-toearth has been found to be 0.150 ohm. The anode output would be [ [ [ [ ] ] ] ] 5A 4A 3A 1.15A 9. Cathodic protection described as [ ] transferring corrosion from the protected structure to the anodes [ ] transferring corrosion from the protected structure to the soil [ ] effective when all current has stopped flowing in the system [ ] effective when all piping is made sufficiently negative 10. The surface potential survey requires [ ] the use of a high resistance voltmeter or potentiometer connected to the copper-copper sulfate reference electrode [ ] the use of a rectifier [ ] the use of a low resistance voltmeter [ ] a strip-chart recorder 11. Which one of the following is a good indicator to ensure that the structure is receiving current? [ ] The achievement of —850 mVcu-CuS04 ] provides a lower anode to earth resiswith C. P. applied tance [ ] A change of 200 mV from the original ] has a higher current output potential [ ] The structure become passive A deep well anode bed design is selected [ ] Polarized potential of—600 mVcu-CuS04 ^ ' because 12. While determining the current requirements ] the soil has a higher resistance at a for a bare structure, it is essential to greater depth [ ] place the electrode over the structure directly ] the soil is more aerated at a higher depth compared to a lower depth Cathodic protection [ ] the soil has a high resistance near the 21. surface and a low resistance at a greater depth [ ] it can be useful in congested areas and will 17. Which of the following is true? 347 The best remedy to minimize stray current corrosion is to [ ] relocate the metallic structure [ ] coat the metallic structure [ ] bury the metallic structure deeper in the ground [ ] install a metallic bond between the structure and the source of stray current [ ] A remote ground-bed protects a smaller area of pipe than a close ground-bed [ ] A remote ground-bed protects a larger 22. In measurement of soil resistivity by four-pin area of the pipe method [ ] Soil nearest to the anode is most negatively charged [ ] an alternate current is passed between [ ] The further is the soil away from the the outer electrode and the resulting anode, the more positively charged it voltage drop is measured between these becomes two electrodes [ ] the galvanometer is adjusted to read 18. Which of the following is true for current zero by means of a potentiometer which originating from a rectifier? is calibrated to read directly in ohms [ ] a direct current is passed between the [ ] Return to the rectifier through the two outer electrodes and the voltage pipeline drop is measured between the inner [ ] Travel to the metallic structure and to electrode get discharged from the metallic struc[ ] the depth of pins is made to be the same ture to the ground as the distance between two electrodes [ ] The current flow reverses 240 times per second [ ] None of the above 19. When the current enters the remote earth [ [ [ [ ] ] ] ] the resistance of the soil increases the resistance of the soil decreases the soil offers no resistance at all there is an increase in the potential of the pipe B . How AND W H Y Q U E S T I O N S Explain why (very briefly) the following: 1) A bare structure requires approximately 10 000 times more current than a coated structure. 2) The resistivity of soil decreases with higher salt content. 3) The reference Cu-CuS04 electrode is placed directly over a coated pipe and away from the pipe if it is not coated. 4) Stray currents cause the uncontrolled corrosion of underground pipes. 5) Well-coated structures polarize more rapidly than bare structures, when cathodic protection is applied. 6) Long slender anodes are preferable over short squat anodes for high resistivity soils. 7) Cathodic current protects the outside of the pipes only. 20. A stray current is indicated if [ ] a positive potential is indicated at the point of corrosion [ ] a positive potential is indicated at the point of protection [ ] the current flows from the electrolyte, like soil, into the metal [ ] very small magnitude of current flows between the anode and the protected structure 3 48 Principles of Corrosion Engineering and Corrosion Control Estimate anode radius from c = 3.14 x d Hint c=nd Working diameter = Core diameter Anode diameter 8) The lowest resistivity soil is often the best location for placement of galvanic anode. 9) High resistance voltmeter must be used to make accurate measurement of pipe-to-soil potential. 10) Over-protection causes damage to the pipeline. +R= 0.00521 p 2.3 log 81 1 C. PROBLEMS 7 -' 1. Calculate the current output of a single anode bed and its life. Assume a pipe-to-soil potential of 0.4 ohm. The following data is given: Driving potential = 0.5 V Resistance = 7.18 ohms Weight of Mg per bed = 128 lbs Consumption rate = 0.116A-year/lb Efficiency = 0.5 Utilization factor = 0.81 2. The open circuit potential of a magnesium anode is — 1.55Vcu-CuS04- It protects a steel tank polarized to a potential of —0.950 VQI-CUSCV Estimate the driving potential of magnesium anode. 3. The open circuit potential of a single 481b Galvomag magnesium alloy anode is —1.75 V. The surrounding backfill has dimensions of 8" x 30". The anode has polarized the steel in a soil of resistivity 3000 ohms-cm to —0.85 V. Estimate the current output of the anode. 4. Calculate the life of the anode given in Problem (3), if it has an efficiency of 50% and a utilization factor of 0.85. 5. Calculate the expected current output of a 725 lb Galvalum anode, from the data given below: Resistivity of seawater = 30 ohms-cm Anode length = 96" Anode width = 10" x 10" Anode core = 4"d Steel polarized to -0.900 V Open circuit potential of anode -1.15Vcu-CuS04 6. From the following data calculate the life of a magnesium anode: Weight of magnesium anode = 50 lb Number of available ampere hour per pound = 500 Current output = 35 mA (8760 h in one year) 7. Find the number of 17 lb magnesium anodes and the spacing between the anode to protect a 10 000 ft of a bare 4" diameter pipeline in a corrosive soil having a resistivity of 800 ohms-cm. Assume a current requirement of 2 mA/ft2 and an anode output of 100 mA per anode. 8. Determine the internal resistance from the anode to the outer edge of the backfill column (R\) from the following data: Resistivity of backfill = 8000 ft-cm Anode dimension = 3" x 5' Backfill column dimension = 8" x 11' 9. In Problem (8), R\ is to be determined. The total resistance of a vertical anode-toearth Ry = R\ + R.2, where R\ is the resistance of the backfill column to earth. If the soil resistivity is 2000 ohms-cm and the dimension of the backfill column are 0.50' x 8', determine the total resistance of the anode-to-earth. 10. Calculate the total resistivity of the anode bed from the following data: Soil resistivity = 1000 ohms-cm Backfill resistivity = 70 ohms-cm Backfill dimensions = 5" x 7' = C athodic protection A node weight = 16 lbs Anode dimensions = 15" x 5' Hint: Use D'Wight's equation to determine the resistance of a single anode. 1 1. D etermine the amount of current needed to protect a bare pipe, 3 miles long, 6" O D. The estimated current density required for protection is 2mA/ft 2 . 12. F rom the data provided below, determine the coating resistance of a 5 mile, 10% OD pipe: a) Current output as recorded by ammeter = 0 . 5 0 A b) The following volts (on) and volts (off) reading were taken at three different locations: Location A B C Volts (on) -1.10 -1.45 -1.28 Volts (off) -0.83 -1.17 -0.99 349 17. A n appropriate rectifier is to be selected for designing an impressed current system. The minimum current required has been estimated to be 2.36 A and the total circuit resistance is determined to be 2.54 ohms. Specify the nearest commercial size of the rectifier. (Hint: E— I x R) 18. C alculate the minimum potential vs CU-CUSO4 reference electrode to which cadmium must be polarized for complete protection. (£sp;Cd(OH)2 = 2 X 1 0" , where Ks? is solubility product) 19. I ron corrodes at a rate of 3 mdd in Arabian Gulf water. Calculate the minimum initial current density (A/m 2 ) necessary for complete cathodic protection. 13. C alculate the surface area of a 10 mile long, 4-5 ft diameter, 2% uncoated pipeline, if the current density required for protection is 2mA/ft 2 . 14. C alculate the life of an anode bed and its current output from the following data: Driving potential of Mg anode = 6 V Resistance = 5.18 ohms Weight of Mg per anode = 5 x 32 = 120 lbs (5 anodes to a bed) Consumption rate = 0.116 A-year/lb Efficiency of Mg anode = 0.6 Utilization factor = 0.85 15. I n DC four-pin method, a direct current of 221 mA is passed between the two outer electrodes and a voltage drop of 191.8 mV is observed between the two inner electrodes. The pin spacing and anode depth is 7.5 inch. Determine the soil resistivity. 16. If p = 2 n AR a nd A = 30.48 D, prove that p = 191 DE/L D — d istance between electrode, E = v olts, J = mA RECOMMENDED L I T E R A T U R E ON CATHODIC PROTECTION [1] [2] Peabody, A.N. (2001). Control of Pipeline Corrosion. Blanchetti, R. L. ed. NACE Int., 2nd ed. Texas: Houston, USA. Von Beckmann, W.> Schwenk, W. and Prinz, W. eds (1998). Cathodic Corrosion Protection. Houston: Gulf Publishing Company, Houston, Texas, USA. Edward Pope, J. (1996). Rule of Thumb for Mechanical Engineers. Gulf Publishing Company, Texas: Houston, USA. Kent Muhlabaner, W. (1966). Pipeline Risk Management Manual, 2nd ed. Gulf Publishing Company, Houston, Texas, USA. Parker, M.E. (1954). Pipeline Corrosion for Cathodic Protection. Gulf Publishing Company, Texas: Houston, USA. West, L.H. and Lewicki, T.F. (1974). Cathodic protection design. Civil Engineering Corrosion Control, Vol. 35, 61-73, AFCEC Technical Repot No. 74-76, Tyndall, Florida: AFB, USA. Uhlig, H.H. (2000). Corrosion Handbook. (Revised by Revie, R.W.) New York: John Wiley. Schultz, M. ed. (2000). Corrosion and Environmental Degradation. Vol. 1, Weinheim: Wiley VCH. [3] [4] [5] [6] [7] [8] 350 Principles of Corrosion Engineering and Corrosion Control Depolarization The reduction of a counter emf by removing or diminishing the cause of polarization. Diode An electrical device with two electrodes, which allows electrons to pass through it in one direction only, hence converting AC to DC. Drainage Conduction of positive electricity from an underground metallic structure by means of a metallic conductor. Drainage (forced) Drainage applied to an underground metallic structure by impressed current or by sacrificial anode. Drain point The point of connection between a cable and protected structure. Efficiency of rectifier DC volts x DC ampere Input watts Electronic current Charge flow by electrons. Energy content Maximum capability of current output of an anode expressed in either Ah/lb or lbs/A-year. For instance, the energy content of a standard magnesium anode is 1230 Ah/kg or 559 Ah/lb. Faraday Faraday = 96 400 Coulombs per gram equivalent. Forward bias The current proceeds in the forward direction and is blocked in the backward direction. Full-wave rectification Rectification producing both AC waves in the DC output. Galvanic anode A sacrificial anode that cause a spontaneous current flow. Galvanic cell A corrosion cell formed by combination of metals differing in potential. Ground-bed Anodes (impressed current or galvanic) buried in a soil with special backfill and connected to the positive terminal of a current source. Impressed current anode Electrodes, such as scrap iron, titanium, lead-silver and silicon cast iron which provide current to an underground structure under cathodic protection. Impressed current system A cathodic protection system which receives the required current for protection from a transformer-rectifier. Insulation flanges Flanges employed to electrically isolate the over-ground pipeline from the underground pipeline. Internal resistance of anode Resistance of anode to backfill — Resistance of backfill to earth. Interrupter-current It is a device which momentarily stops current. Ion current Charge transfer taking place by ions. IR drop Voltage drop caused by a current flow (1) through the conductor of resistance R. Isolating joint A joint or coupling between two lengths of pipe inserted to provide electrical discontinuity (insulation). Junction boxes Connect electrical cables used in cathodic protection system. Microampere 0.000001 ampere. KEYWORDS Anode It is the electrode in a corrosion cell which corrodes by passage of electrical current into the electrolyte. Anode bed (ground-bed) The specific area where anodes are buried in soil and a backfill is placed around them. Anode life The number of years taken by an anode to be consumed at a certain current output. Anodic polarization That portion of polarization which takes place at the anode. The potential becomes more noble as anodic polarization proceeds. Attenuation curves Curves obtained by plotting driving voltage (AJB) and polarization potential (A Vp) on semi-log paper against distance are called attenuation curves. Backfill The special soil placed around the anodes to provide uniform resistivity. The material used as backfill in an impressed current system is generally coke breeze whereas in the galvanic system the backfill is composed of a mixture of gypsum, bentonite and clay, the composition being dependent on soil resistivity. Bond An electrical connection between two metallic structures. Casing voltage profile A plot of the voltage vs the depth of an oil well casing. Cathode The electrode in a corrosion cell through which conventional (positive) direct current leaves the electrolyte. Reduction takes place at the cathode processes, for example, oxygen reduction and hydrogen reduction takes place at the cathode. Electrons from the cathode are consumed at the cathode surface by O2 + H2O forming O H - ions, or by H + ions forming H2. Cathodic polarization That portion of polarization which takes place at the cathode. The potential becomes more negative as cathodic polarization proceeds. Cathodic protection Elimination or reduction of corrosion of a metal surface by making it the cathode (negative), using either a galvanic or an impressed current. Cell An anode and a cathode is an electrolyte. Close ground-bed In this arrangement series of anodes are used. The length of the pipeline protected by a single closed anode depends on changing the potential of earth around the pipeline with respect to earth. The earth is made more positive in areas where protection is needed. Coating resistance The electrical resistance of a coating to the flow of current. Concentration cell An electrolyte cell in which the emf is the result of a difference in concentration of the electrolyte or active metal at the anode and the cathode. Copper sulfate half-cell A reference electrode consisting of copper rod in a tube containing a saturated solution of copper sulfate. It is used for measurement of potential of buried structures. Current density The amount of current required to protect a metallic structure. The magnitude of current varies with the environment. Cathodic protection Mote earth A position in earth which offers no resistance to the flow of electrical current. Overload protection It is a device which protects an electronic component from destruction by excess current. Over-protection Current in excess of that required for protection. Pipe-to-soil potential It is the potential of a pipe measured in a soil which acts as an electrolyte. Also, the corrosion potential of a metal in soil. Polarization A shift in the potential of a metal in an electrolyte by passage of current flow. Positive current Current flow of cations (e.g. F e + + ions) or hypothetical positron (electron holes). Potential criteria Attainment of a potential of—0.85 V vs Cu-CuS04 for steel structure in soil, such as steel pipe. Potential decay (attenuation) The drop in the potential of a pipeline with increasing distance. For instance, the pipe-to-soil potential is at maximum near the area of influence where the soil is positive and decreases with the distance. Potential gradient Potential difference per unit distance. Potential 'on' Pipe-to-soil potential with switch on. It includes also IR drop. Potential 'off* Pipe-to-soil potential without any IR drop. Varies with time. Potential shift criterion Shift of potential required to completely protect a structure. A shift of 200-300 mV negative from the original value of potential brought about by external current to the structure to be protected is considered safe for protection. Potential survey Survey of the potential of a pipeline with respect to soil over a defined distance. A plot of potential with respect of Cu-CuSC>4 electrode vs the distance is called a potential profile. Rectifier An electrical device which converts AC to DC. A diode. 351 Resistance bond A metallic connection between the point of drainage of current from a structure to the origin of the current to avoid interferences. Resistivity (soil) Resistance in ohms of a cm3 of a material, measured across opposite faces [p = R {II a), where / = length; a = a rea]. The unit is ohms-cm. Reverse bias diode It has extremely high resistance and blocks all current flows. Silicon diode Silicon anode is positive with respect to cathode. Stray current Current flowing in the soil or water environment of a structure and arising mainly from electric power or traction installation. It is the uncontrolled current. Structure-to-soil potential Potential in a buried structure and a non-polarizable electrode placed in soil. Surface potential survey Survey of pipe-to-soil potential by two copper sulfate electrodes. Tap transformer. A connection brought out of a winding at some point between its extremities to permit changing voltage or current ratio. Test stations Special devices installed above in cathodic protection systems. They are used to measure pipe-to-soil potential, line current and current flow of a bond, to monitor potential measurements and also to measure stray current corrosion. Transformer It is a device which makes it possible to transfer power from one circuit to another by mutual induction. In a step-up transformer, the energy transferred is from a lower voltage circuit to a higher voltage circuit. In a step-down transformer, the transfer of energy is the reverse of the step-up transformer. CORROSION CONTROL BY I N H I B I T I O N 6 .1 INTRODUCTION 6.2 S C O P E OF INHIBITOR orrosion phenomena are well-known in the petroleum industry and cause a maximum damage to oilfield equipment. Significant corrosion protection efforts have been made by petrochemical industries to prevent corrosion damage. The practice of corrosion prevention by adding substances which can significantly retard corrosion when added in small amounts is called inhibition. Inhibition is used internally with carbon steel pipes and vessels as an economic control alternative to stainless steels and alloys, and to coatings on non-metallic components. One unique advantage is that adding inhibitor can be implemented without disruption of a process. The addition of an inhibitor (any reagent capable of converting an active corrosion process into a passive process) results in significant suppression of corrosion. Corrosion inhibitors are selected on the basis of solubility or dispersibility in the fluids which are to be inhibited. For instance, in a hydrocarbon system, a corrosion inhibitor soluble in hydrocarbon is used. Twophase systems composed of both hydrocarbons and water, utilize oil soluble water-dispersible inhibitors. Corrosion inhibitors are used in oil and gas exploration and production, petroleum refineries, chemical manufacturing, heavy manufacturing, water treatment and product additive industries. The total consumption of inhibitors in USA alone costs over one billion dollars annually. C Corrosion control by use of inhibitors is extremely useful in many environments, however, there are certain exceptions, such as: (a) equipment and components subjected to turbulent flow. (b) systems operating above the stability limits of inhibitor. (c) equipment subjected to high velocity, beyond 4 m/s. However, adding inhibitors can raise the value of the critical flow rate above which erosioncorrosion starts. 6.3 EXAMPLES OF APPLICATION OF INHIBITORS (1) Corrosion is a serious problem in all cooling water systems. The cooling water may be salt water (35 000 ppm TDS), brackish water (3000-5000 ppm) or fresh water (<300ppm TDS). Inhibitor treatment is required for heat exchanger and distribution lines. (2) Corrosion may be caused in the feedwater and boiler sections, if dissolved oxygen and Corrosion control by inhibition CO2 is not removed by water treatment. Scales and deposits may also be formed by dissolved and suspended solids. Excessive alkalinity in boilers can lead to caustic cracking. High alkalinity is caused by high TDS (total dissolved solids) and alkalinity. External treatment includes demineralization and reduction of alkalinity, corrosion inhibition and biological control. Morpholine inhibitor is added as inhibitor for treatment of condensate corrosion. Petroleum Industry. Corrosion phenomena in the petroleum industry occur in a twophase medium of water and hydrocarbon. It is the presence of a thin layer of water which leads to corrosion, and rigorous elimination of water reduces the corrosion rate to a negligible value. The inhibitors used in petroleum industry, both in production and refining are either oil soluble-water insoluble types or oil soluble-water dispersible compounds. New inhibitors are being developed. For instance, traditional filming amines are being replaced by several others, such as propylenedramine, and they work by adsorption on the surface. Sour Gas Systems. A major problem is encountered in steel pipelines in various sour gas environments. Chemical inhibition is one of the effective methods used to mitigate sulfide induced corrosion. Inhibitors containing alkylammonium ions are found to suppress corrosion effectively. Potable Water Systems. Corrosion is experienced in potable water transportation pipes of steels and cast iron. Inhibitors, such as Ca(HCC>3)2 and polyphosphates are commonly used to combat corrosion. Engine Coolants. Inhibitors, such as NaCr04 (sodium chromate), borates and nitrites (NaN02) and mercaptabenzothiazole are widely used for protection of automobile engines. Chromates are a health hazard. Packaging Industry. For transportation of machinery, components and equipment by sea, vapor phase cyclohexylamine and hexamethylamine are used. Construction Industry. Corrosion of rebars in concrete poses a serious threat to building structures. Inhibitors, such as chromates, 3 53 phosphates, nitrates and sodium metasilicates are used to suppress corrosion. Addition of sodium tetraborate and zinc borate has shown promising results. Commercial inhibitors, such as MCI 2022, MCI 2000 and Rheocrete 222, have shown good promise. Baker Petrolite, Du-Pont and NALCC are some of the known names in inhibitor manufacturing. The above summary shows the wide range of applications of inhibitors in wide spectrum of environments. (3) 6.4 IMPORTANT C O N S I D E R A T I O N IN S E L E C T I O N OF INHIBITORS (1) The magnitude of suppression of uniform and localized corrosion. (2) Long range effectiveness. (3) Effect on bimetallic coupling to other metals joined to the main system. (4) Effect of temperature and concentration on the performance of inhibitors. (5) Effect on the existing condition of the system to be protected. For instance, a metallic structure maybe partly corroded; the important point would be to observe the effect of inhibitor on the corroded areas. (6) Effect of inhibitor on heat transfer characteristics. (7) Toxicity and pollution problems. (8) Economically and technically competitive with other considered inhibitors. (4) (5) (6) 6.5 IMPORTANT T E R M S R E L A T E D TO I N H I B I T O R S (1) Additives. A typical reagent for treatment of a corrosive fluid contains one main active gradient and one or more additives which assist in achieving the purpose of the reagent. (2) Solvent. Keeps the active reagents in the liquid form and controls their viscosity. (3) Solubility. The active reagent should have the ability to dissolve in the solute. (7) (8) 354 Principles of Corrosion Engineering and Corrosion Control Inhibitors can also be classified on the basis of their functions. For instance, chromates and nitrates are called passivating inhibitors because of their tendency to passivate the metal surface. S ome inhibitors, such as silicates, inhibit both the anodic and cathodic reactions. They also remove undesirable suspended particles from the system, such as iron particles, by precipitation. Certain types of inhibitors make the surrounding environment alkaline to prevent corrosion. Such inhibitors in the gas phase are called 'vapor phase inhibitors^ a nd they consist of heterocyclic compounds, such as cyclohexylamine. These inhibitors are used within packing crates during transportation by sea. (4) Dispersibility. It is a measure of the reagent's ability to be transported by fluids or gases. The treating reagents must exhibit a high dispersibility. (5) Emulsion. A heterogeneous system consisting of an immiscible liquid dispersed in another liquid in the form of droplets. (6) Surfactants. It is a molecule with two components, each having different chemical properties, one end is polar (hydrophillic), and the other end non-polar (hydrophobic). 6 .6 CLASSIFICATION O F I NHIBITORS I nhibitors may be classified as shown in Fig. 6.1. There are two major classes: inorganic and organic. The anodic type of inorganic inhibitors includes chromates, nitrites, molybdates and phosphates, and the cathodic type includes zinc and polyphosphate inhibitors. The film forming class is the major class of organic inhibitor as it includes amines, amine salts and imidazolines - sodium benzoate mercaptans, e sters, a mines and ammonia derivatives. 6.7 DESCRIPTION O F I NHIBITORS 6.7.1 ANODIC PROCESS AND ANODIC INHIBITORS C onsider an anodic dissolution process: (1) F e ^ Fe 2 + + 2e (6.1) Classificiation of Inhibitors Inorganic Organic Environmental Conditions Film ft>cmer Anodic Chromates Phosphates Nitrates Molybdates (Also called passivation inhibitors) Cathodic Phosphate Zinc Poisons Arsenic Precipitators Scavengers Sodium sulfite Btockles Ozone Calcium Carbonate F i g u r e 6 .1 C lassification of inhibitors Corrosion control by inhibition During the dissolution process in an aqueous media, the phenomena of adsorption and disruption of species is very predominant (equation (6.1)). It can be represented in a stepwise manner as: 355 hydroxide, is observed in steel. Both the processes, passive film formation and stifling of anodic sites, lead to suppression of corrosion. However, the kinetics of film formation is different from anodic stifling, and the mechanisms of the process are beyond the scope of this text. Identification of (2) Fe + OH~ = FeOH" (ad) (6.2) thin film can be conveniently performed by trans(3) FeOH" = FeOH + e (6.3) mission electron microscope. The number of metal ions dissolving is effectively reduced by (4) FeOH = FeOH+ + e~ (6.4) anodic inhibitors. The effect of adding anodic (5) FeOH+ = FeOH+ (6.5) and cathodic inhibitors on potential difference is shown in Fig. 6.2. [underline indicates that the species are adsorbed In terms of potential, addition of anodic on the surface] inhibitors reduces the difference of potential How corrosion slows down would depend between the anodic and cathodic sites, and conon the electron transfer rate limiting step and sequently reduce the driving force for corrosion the environment. Corrosion could be increased reaction to occur. The potential of the anode shifts or decreased depending on the electrolyte con- to the potential of the cathode. The number of stituents, such as halogen ions and benzoate, for metal ions dissolving as a result of anodic reacinstance, other ions. If, for example, the reac- tion is reduced, and the potential shifts in a more tion Fe + O H - —• FeOH" is suppressed by noble direction. Figure 6.2 illustrates the effect of electrolyte constituents, corrosion is decreased, anodic and cathodic inhibitors on the potential and if the reaction FeOH - /ads = FeOH + is and the driving voltage. promoted, corrosion is accelerated. Species of FeOH + are at the kink sites at the edge of lattice and they are liable to dislodge easily, or at the dissolving edge of a terrace and are ready for 6 .7.2 CATHODIC P R O C E S S E S active dissolution. All reactions shown above are subject to interaction with environment and A N D C A T H O D I C I N H I B I T O R S the adsorption or deadsorption would depend on the environmental species and the effect of As the name suggests, the class of inhibitors which coordinating groups. Consider once again reac- decrease the rate of cathodic reaction in a metal tion (6.5) and suppose it is rate controlling or the surface are called ' cathodic inhibitors.'* To underslowest of all (reactions (6.1-6.4)). On increas- stand the mechanism, consider the two major ing the concentration of the O H - ion, ferrous cathodic reactions: hydroxide would be precipitated which would (1) 2H 2 0 + 0 2 + 4e -> 4 0 H " (oxygen retard the rate of desorption and lead to suppresreduction) sion of corrosion. In anodic control, passivation (2) 2H+ + 2e «> H 2 (hydrogen reduction) of the surface is the controlling factor, such as in stainless steels. Inhibitors broaden the range of It may be pointed out that both the cathodic passivation. reactions take place in several steps and the step A passive film formed on the metallic sur- with the slowest rate of reaction is generally the face block will suppress corrosion. They are very rate controlling step. thin films (50 to 100 A). In aluminum, thin films Consider a cathodic site where oxygen is difof boehmite (AlOOH) and bayerite (Al(OH)3) fusing to the metal/electrolyte interface. If an are formed which affect the corrosion process. inhibitor, like zinc and magnesium, is added to Thick films are liable to breakdown and acceler- the metal/electrolyte system, it would react with ate corrosion. Copper and nickel also have good the hydroxyl ion and precipitate insoluble compassive film forming properties. The stifling of pounds which would, in turn, stifle the cathodic anodic sites by corrosion products, such as iron sites on the metal. In oxygen-induced corrosion, 3 56 Principles of Corrosion Engineering and Corrosion Control Nore positive Potentia difference Nore negative Anode Anode Figure 6.2 Effect of adding anodic and cathodic inhibitors on potential difference the controlling step is the mass transfer of oxygen to the metal (cathode/electrolyte interface). Consider now that the metal (cathode/electrolyte interface) is in a stagnant condition. The oxygen is rapidly depleted in this condition and the reaction rate is slowed down. On the contrary, in a flowing system, a high rate of reaction would be maintained because of the continuous supply of fresh oxygen to the system. All factors, such as temperature, pressure and salt contents, which affect the solubility of oxygen would also affect the reaction rate (reduction of rate) of oxygen. On arrival of oxygen molecule to the metal (cathode/electrolyte interface), it must be absorbed. Consider: 2H -$- H2 (underline represents adsorption) A process of reduction follows the adsorption process: (a) Oxygen molecule is reduced in two oneelectron step to hydrogen peroxide which is reduced in a one-step reducing step to produce O H - ions. The slowest step is rate controlling. (b) An oxygen-oxygen bond is ruptured producing two chemisorbed oxygen atoms which transform to hydroxyl ion after each atom picks two electrons and a proton. In either case, (a) or (b), the reaction is completed after desorption of hydroxide ion. The surface of an anode changes continuously to make fresh surface available for the anodic reactions to continue. Consider now hydrogen evolution on the metal (cathode/electrolyte interface) in acid solution. Similar to oxygen reduction, hydrogen reduction also involves several steps, such as (1) H+<±H+ (2) H + + e -> H (ads) (ads) A s 2 0 3 + 6Hads -> 2As I + 3 H 2 0 Despite several steps leading to the formation of molecular hydrogen, two steps are common: (a) Adsorption of hydrogen ions (b) Reduction of adsorbed hydrogen ion to atomic hydrogen Corrosion control by inhibition Hydrogen may then be formed either by 357 achieve the same objectives have been used with success. There is, however, one inherent danger in adding poisons. A good example is H2S, 2H+>H 2 which poisons (prevents) the recombination of atomic hydrogen. H has an inherent tendency or H + H + + 0 2 «± H 2 to diffuse into steel surface and cause blistering The reaction is completed by desorption of of the steel structure, if it is not converted to the hydrogen molecule molecular hydrogen (Fig. 6.3). Another example is hydrogen sulfide cracking of steel equipment H2 H2 • in oil drilling. It must be remembered that at (De-adsorbed) (Adsorbed) the cathode, hydrogen ions and oxygen molecules It is primarily the adsorption character which consume electrons and the alkalinity of the elecdetermines the rate of cathodic reaction and the trolyte at the metal (cathode/electrolyte interface) degree of inhibition. The evolution of hydrogen is is increased which leads to the precipitation of affected by an increase in the over-voltage. Salts of cathodic inhibitor on the surface and formation of bismuth and antimony are added to obtain a layer a protective layer. Contrary to the situation shown of adsorbed hydrogen on the cathode surface. The for anodic inhibitors, the open circuit potential of over-voltage for hydrogen evolution is increased the cathode shifts to the potential of the anode in and the cathodic reaction is suppressed. In acids, the more negative direction (Fig. 6.2). the formation of hydrogen on the cathodic sites is retarded by the addition of arsenic, bismuth and antimony. For instance, if arsenic trioxide E F F E C T OF INHIBITORS is added, it plates out to form arsenic on the 6 .7.3 cathodic sites. ON POLARIZATION BEHAVIOR Arsenic thus suppresses the cathodic reaction. Elements, like P, As, Sb and Bi, are called The effect of inhibitor on the metal/electrolyte poisons and they retard the cathodic reduction of system can be successfully evaluated from the hydrogen. To avoid toxicity, organic inhibitors to polarization diagrams discussed in Chapter 2. Figure 6.3 An example of hydrogen blistering 3 58 Principles of Corrosion Engineering and Corrosion Control 430 S.S. t»* = -520 mV 3D3 5 5, £»« « -446 mV 304 $.$. &»* * -345 mV 316 $5. EMIT - .311 mv sofutro: in MaSCfc Sweep rate: 1 mV/min 3 MP Current; jiA/cm* 103 Figure 6.4 Polarization plots of ASTM standard type stainless steels showing the passive region and initiation of film formation Consider, for example, a polarization diagram shown in Fig. 6.4 which shows the polarization plots of three different types of steels in IN H2SO4. The electrochemical parameters, /critical (critical current density), /passive (passive current density) and £pP (passivation potential), are sensitive to changes in composition of electrolyte. Clearly, Steel 303 is very active as it shows the highest /critical* highest /passive and lowest passivation range. The effect of chloride ions and the behavior of the anodic polarization curve is shown in Figs 6.5 and 6.6. The critical current density increases from /a (no chloride) to 1^ (with Cl~). The three points, a, b and c, represent an anodic dissolution current corresponding to a reduction current (cathodic current). These points represent corrosion potential. At location a, active dissolution would proceed, whereas at point c, the passivity is retained and, hence, corrosion would not occur. Now consider Fig. 6.6 which shows a polarization curve in a higher concentration of chloride. Passivation is not observed. The intersection of the cathodic curve with the anodic curve shows at point V a larger corrosion current, and hence a higher corrosion rate. To clarify the effect of electrolyte composition on the anodic polarization behaviors, consider the three cases shown in Fig. 6.7. Case I. The cathodic curve intersects the anodic polarization curve in the active region. The corrosion potential is in the active region, £1. A high rate of metal dissolution is expected. Case II. The cathodic curve intersects the anodic polarization curve at points £2 and E'2. C orrosion control by inhibition 359 no Q • -^^^''^ - -* ' '""LJl---^^ with a s UJ Jc W & > 8. 1 L - - - ~ _ ^ r^> V a _ ^vb I N od « ! Applied current density, A/cm' (a) Applied current density, A/cm* (b) Figure 6.5 Schematic representation of the effect of a small chloride concentration on the polarization curve N od tons - -"•" Si Crthodlc cyrvt AppHecf wrmt 1$®®®% Afatf hm Figure 6.6 Schematic polarization curve in the presence of a large concentration of chlorides 3 60 Principles of Corrosion Engineering and Corrosion Control z^> Current density, A/cm2 CASE 2 Current density, A/cm1 H*fr .J. Current density, A/cm2 CASE 3 Current density, A/cm* i Current density, A/cm2 Currant density, A/cm1 Figure 6.7 Active-passive polarization behavior Corrosion control by inhibition This shows an unstable state of potential which oscillates between £2 and E'2. It represents a case of serious corrosion. Case III. The cathodic curve intersects at £3 in the passivation range. The potential is stable. It represents a case of low corrosion. The above three cases make it convenient to understand and interpret the effect of inhibitors. Consider inorganic inhibitors, like chromates, which passivate the surface (also called passivators). The effect of such inhibitors can be evaluated from the diagrams shown above. Examine Fig. 6.8, which shows the effect of concentration of a hypothetical inhibitor. For reasons stated earlier, Case I represents a situation where the concentration of inhibitor is sufficient to passivate the metal surface. The effect of adding anodic and cathodic inhibitors is shown in Figs 6.8a and b. It is to be remembered that: (a) a cathodic inhibitor shifts the corrosion potential in the negative direction; and 3 61 (b) an anodic inhibitor displaces the potential in the positive direction. Figure 6.8a shows that Icorr of the uninhibited electrolyte is higher than the I'coxx of the electrolyte to which an anodic inhibitor is added. On adding a cathodic inhibitor (Fig. 6.8b), the ICOTT is lowered compared to the I'con of the uninhibited electrolyte (Icon < Ic0n) and £corr is displaced in a more negative direction. The effect of addition of mixed inhibitors (anodic and cathodic both) is shown in Fig. 6.9. Mixed inhibitors show the characteristics of both the types of inhibitors seen in Figs 6.8a and b. As stated earlier, the corrosion potential is displaced in the positive direction on addition of anodic inhibitor and in the negative direction on addition of cathodic inhibitor. The mixed inhibitors protect the metal in three possible ways: (a) Physical adsorption. (b) Chemisorption. (c) Film formation. W (inhibited} < Um (uninhibited) EW (inhibited) > Em (uninhibited) ^ (Inhibited) £m (uninhibited) W (inhibited) low (uninhibited) d inart density A/o# Figure 6.8a Effect of addition of anodic inhibitors 362 Principles of Corrosion Engineering and Corrosion Control 8 * W (inhibited) > W (uninhibited} Eoort* (inhibited) < Bom (uninhibited) Anodic curve Emt (uninhibited) EW (Inhibited} U (uninhibited) W (inhibited!) Current density, A/am* Figure 6.8b Effect of addition of cathodic inhibitors Anodic curve (anodic inhibitor) ' Anodic .curve { to Inhibitor) y Anodttcurve (Catltodic inhibitor} Cathodic curve (anodic inhibitor) ^ Qthodic curve v (No iftomw) Current density, A/am2 Figure 6.9 Effect of addition of inhibitors on potential Corrosion control by inhibition 3 63 Physical adsorption is caused by electrostatic forces which exist between the inhibitor and © the metal surface. The metal surface can be Metal surface Electrolyte either positively charged or negatively charged. For instance, during cathodic polarization, metals get negatively charged due to the discharge of cations on the metal surface. The reverse n ©© happens on anodic polarization. When a metal _ © _ ©. • 0 surface is positively charged, negative charged ©© (anionic) inhibitor is adsorbed on the metal sur©© "~© face (Fig. 6.10a). On the other hand, catonic species would not be adsorbed on a positively © ©•<&•• charged surface(Fig. 6.10b). Similarly anionic species would not be adsorbed on a negatively charged surface (Fig. 6.10c). Physically adsorbed species can be removed Figure 6.10b No adsorption of positively charged from the surface by physical force such as species with a positively charged metal surface increased temperature and increased velocity. On the other hand, chemisorption results in a strong binding of the inhibitors with the metal surface. Positively charged ions in the presence Metal sytface of negatively charged ions can be adsorbed on a Electrolyte positively charged metal surface in the presence of negative ions, such as Cl~ ions, which act as ©© ^ ©. - © ~ a bridge between the two. The negative ions are ©© adsorbed on the positively charged metal surface *©•«©« and the positive ions are attached synergistically to © © %<*%*% the dipole (a molecule with an unique distribution ©© of charges, such as water). The adsorption of positively charged inhibitor on a negatively charged ©© surface in the presence of negatively charged ions 1© © _© © © ® ^ © ~ ©. « ©^ ©^© ^ © ^© 1© © e ©^ 0 ^© 0. 0 0 0 0 Figure 6.10c No adsorption of negatively charged species with a negatively charged metal surface is called 'synergistic adsorption (Fig. 6.10d). This phenomena is caused by sharing of charges or charge transfer between the inhibitor species and ^ © ^ © ^ €> the metal surface. The process of chemisorp© © © tion is accelerated with time and temperature. CS C^i Ci © While in the process of physical adsorption, de-adsorption may take place under adverse con© r\ © © ©^© ditions. Chemisorption is not reversible, however, it is a more effective process. © ® 0® © ® The process of film formation is highly complex and the properties of films are dependent upon its thickness, composition, solubility, Figure 6.10a Adsorption of negatively charged species temperature and other physical forces. For instance, films of AI2O3 produced by anodizing on a positively charged metal surface Metal surface Electrolyte © 364 Principles of Corrosion Engineering and Corrosion Control do not require oxygen for passivation, whereas non-oxidizing types, like phosphates, tungstates and molybdates, passivate the surface only in the presence of oxygen. The action of anodic, cathodic and mixed inhibitors is summarized in Figs6.11(a-c). 0 Nelal surface Electrolyte I© © © © © © 0 © 0 © ^ © _ 0' 0 0 © © © 0© © ©© ©© © ^ ^ ^ 6.7.4 CHARACTERISTICS OF CATHODIC INHIBITORS X0 ^ 0 ^© 0 0 0 (A) Polyphosphates T he structure of a sodium molecule is O O polyphosphate Figure 6.10d Adsorption of positively charged species with a negatively charged metal surface are highly resistant to corrosion and they produce a very high resistance. Thick films may loose their adhesion due to mechanical damage, like air formed films on steel surface, if the thickness reaches beyond a critical point. Inhibitors of passivating type, like chromates and molybdates, p roduce passive films and resist corrosion. Passivating inhibitors, like NaCrC>4 and NaN02 N aO P O - ONa O Na O O = orthophosphate n = 1 p yrophosphate n = 2 &# E«,o \V Mm®$ v v InNbltar r \, y 6*J £«,« L og! (b) (c) Logl Logl m Figure 6.11 Action of corrosion inhibitors: (a) anodic inhibitors examples: chromate, nitrite, molybdate, tungstate, orthophosphate, silicate, benzoate, (b) cathodic inhibitors examples: Cu(HC03)2> ZnS04, Cr4(S04)3, NiS04, polyphosphate, aminoethylene phosphate and (c) mixed inhibitors examples: organic inhibitors containing nitrogen and/or sulfur, like amines, triazoles, alkythiourea Corrosion control by inhibition They are also referred to as condensed or polymer phosphates. The chain length is determined by repetition of the portion of the structure denoted by n. The characteristic glassy structure is observed with longer chain lengths of the polymer. One good example of a glassy structure is sodium hexametaphosphate which is extensively used as an inhibitor. The polyphosphate molecule bonds with divalent calcium and other ions to form positively charged colloidal particles which are attracted to the cathode and form a protective film. As some metal ions, such as iron, may also be adsorbed on the film, polyphosphate also shows a partial anodic behavior although basically they are cathodic inhibitors. The mechanism of corrosion prevention by polyphosphates is shown in Fig. 6.12. Metallic ions, such as iron and copper affect polyphosphate. If an iron-polyphosphate complex is formed, dissolution rather than inhibition would proceed. If copper ions are present, a galvanic couple would be formed by the passage of copper ions from the polyphosphate film to the iron substrate, and corrosion would progress. 365 One major disadvantage of using polyphosphate is the hydrolysis of the phosphorus oxygen bond which converts the polyphosphate to orthophosphate, which is a proven weak inhibitor. A pH range of 6.5-7.5 is considered generally suitable to avoid reversion to orthophosphate which allows undesirable algae growth. Polyphosphates are blended with silicates and ferrocyanide to overcome this limitation. In actual operation, two or three inhibitors are blended to maximize the advantage of each other and to minimize their limitations. Frequently anodic and cathodic inhibitors are combined to maximize metal protection. The process is called 'synergistic blending.' (B) Zinc Zinc salts are well-known cathodic inhibitors in cooling water systems. They are, however, not used alone as the films formed by them are unstable. They are, however, used very effectively with polyphosphate as a synergistic blend to maximize the effect of inhibition. These synergistic blends m) + 2r © 000 © € C t" C jC; Sttft. complex Iii / © @©@ © - . • © © © © © © @ J ' ' :— c ':< e c cc Metal Figure 6.12 Formation of polyphosphates and their reaction with divalent ions in water 366 Principles of Corrosion Engineering and Corrosion Control minimize the inhibitor concentration. Chromates can also be used in synergistic blends with polyphosphates. Zinc and chromate inhibitors are toxic and not environmentally friendly. They also need a careful pH control as they have a tendency to form scales above a pH of 8.00. 6.8 INORGANIC ANODIC INHIBITORS The addition of inorganic inhibitors causes suppression of electrochemical reaction at anodiccathodic areas. Most of the times, inhibitors are used in a blended form. These inhibitors only react at an adequate level of concentration. and is not advised. In industrial water, the threshold concentration is 120mg/L. A high concentration is required if the systems to be inhibited contain bi-metal junctions or a high chloride concentration. They are oxidizers and raise the anodic current density above the limiting value needed for passivity. Chromate inhibitors contain either Na2CrC>4 or Na2Cr20y. The protective passive film which is formed contains iron oxide and chromium oxide which makes the chromate inhibitors very effective (Fig. 6.13). Chromates are reduced to form chromium (III) according to the following reaction: Fe -> Fe 2+ + 2e (Oxidation of iron) CrO~ + 8H + 3e~ -> Cr 3+ + 4 H 2 0 (Formation of Cr 3+ ) (A) Chromate Inhibitors A mixed potential is created by the oxidaThey are most effective inhibitors, but they are tion of iron and reduction of chromium and toxic and, hence, their application is restricted this potential lies somewhere in the passivation x/ ® -^© + * 0>Cb + feOs film (mixed axkte film) Heta! surface Figure 6.13 Formation of a mixed iron oxide and chromium oxide film Corrosion control by inhibition range, which accounts for the passivating effect of chromate inhibitor. 367 The passive film can only be formed in the presence of oxygen. They are very expensive and used with other inhibitors in synergistic blends. (B) Nitrites They are effective inhibitors for iron and a number of metals in a wide variety of waters. Like chromates, nitrites are anodic inhibitors and they inhibit the system by forming a passive film with ferric oxide. These are environmentally-friendly inhibitors. Besides steel, nitrites also inhibit the corrosion of copper, tin and nickel alloys at pH levels 9-10. Chromate is an extremely effective inhibitor for corrosion prevention of aluminum alloys. Nitrites should not be used in open systems as they would oxidize to nitrates in the presence of oxygen. (F) Silicates They have been used with success for years in potable water systems. The complex silicon ion has a tendency to form negatively charged colloidal particles which migrate to anodic areas and form passive films. Silicates are strong anodic inhibitors and passive films can be formed even on the corroded surface. The monomeric silica does not provide any protection. In waters below pH levels of 6.0, the silicate used is Na20-2SiC>3 and with a pH greater than 6.0, it is Na2C>3 -3Si03. Silicate inhibitors are also useful to prevent red water formation in plumbing systems by oxidation of ferrous carbonate in natural water or steel 2NO2 + 0 2 = 2NO3 pipes encountering soft water. Red water formaNitrites are not effective inhibitors. The pres- tion seriously affects plumbing fixtures and the ence of chloride and sulfate ions can damage the problem appears particularly in galvanized pipes protective film formed by nitrites. They are often if the temperature exceeds 65°C, due to reversal blended with borax in closed recirculating system. of polarity. Silicate treatment also prevents dezincification in brass and corrosion of copper. Mixtures of silicates and phosphates have been (C) Nitrates effectively used as inhibitors. They protect solder and aluminum. They are not very effective and limited to use only in closes recirculating systems. 6.9 ORGANIC INHIBITORS Organic inhibitors are abundantly used in the oil industry to control oil and gas well corrosion. Most common types are long chain (Cis) Phosphate retards corrosion by promoting the hydrogen and nitrogen containing compounds. growth of protective iron oxide films and by Organic inhibitors are neither anodic nor healing the defects in protective films. The effeccathodic, but they inhibit both the anodic and tiveness of phosphate inhibitor is reduced by cathodic areas to varying degrees depending on chloride ions which damage the protective film the type of inhibition. The most common types formed by phosphate. of organic inhibitors are shown below: (D) Phosphate I nhibitors (1) Monoamine: Primary amine, RNH2 Molybdenum is an alloying element which is Secondary amine, R2NH known to increase passivation of stainless steels. Tertiary amine, R-N(CH3)2 Steels of type 316 contain molybdenum as a minor (2) Diamines constituent and promote passivation. Sodium R-NHCH2CH2CH2NH2 molybdate forms a complex passivation film at the (3) Amides iron anode of ferrous-ferric molybdenum oxide. R-CONH2 (E) Molybdates 3 68 Principles of Corrosion Engineering and Corrosion Control amine nitrogen group at the end of a hydrocarbon chain. The active (-NH2) group contains a pair of unshared electrons which it donates to the metal surface. A chemisorption bond is, therefore, formed which impedes the electrochemical reaction. The polar amine group displaces water molecules from the surface. On adsorption, most of the metal surface is covered by the adsorbed water molecules. The inhibitors react by replacing water molecules by organic inhibitor molecules. Org.molecule(aq) + nfyOiads) -> Org.molecule(ads) + nfyOisoln) Here n represents the number of molecules which are replaced to accommodate the organic molecule. The hydrocarbon part of the organic inhibitor is oil soluble, hence, it repels water from the metallic surface. It, therefore, provides a barrier which keeps water away and thus prevents corrosion. The hydrocarbon chain attracts the organic molecules and forms an oily layer which prevents corrosion by acting as a barrier against fluids. For instance, diethanolamine effectively inhibits corrosion of carbon steel in petroleum/ water mixtures. The organic inhibitors are physically adsorbed on the surface. A scientific description of the mechanism is beyond the scope of this book. Interested readers should look into the suggested literature. To summarize, the following are the main features of corrosion inhibitions by organic inhibitors. The polar nitrogen groups attached to a hydrocarbon chain donate electrons to the metal surface and form a strong chemiadsorbtive bond. The strength of protection is dependent on this bond. The hydrocarbon portion of the inhibitor is oil soluble and it is water repellent. The large hydrocarbon chain orients towards the solution and forms a hydrophobic network (repels water from the metal surface). The water molecules are desorbed and replaced by organic molecules [Org(soln) + nH20(ads) -> Org{ads) + nH20(soln)]. Water molecules, which are the main source of corrosion, are thus eliminated. Figure 6.14 shows a simplified mechanism. (4) Polyethoxylated compounds (a) Amines (CH 2 -CH 2 0) ;c H R-N (CH 2 -CH 2 0) y H (x and y vary between 2 and 50) (b) Diamines (CH.-CH.O^H R-NCH 2 CH 2 N (CH 2 -CH 2 0) r H (CH 2 -CH 2 0) z H ( + y + z varies between 3-10) % (5) Quaternaries cr ( C 12 - C 18 chains) R\ and JR2 are Q 2 to C\$ chains. Organic inhibitors react by adsorption on a • metallic surface. Cationic inhibitors (+), like amines, or anionic inhibitors (—), like sulfonates, are preferentially adsorbed depending on the charge of the metal surface (+) or (—). At zero point of charge, there is no particular preference • for an anodic or cathodic inhibitor. In such a situation, a combination of an inhibitor which would • be strongly adsorbed at more negative potentials along with cathodic protection would provide a greater degree of inhibition than either applying cathodic protection or using inhibitor separately. The formation of a bond between the metal substrate and the organic inhibitor (chemisorption) bonds impedes the anodic and cathodic process and protects the metal surface. Consider an inhibitor molecule, e.g. containing a polar Corrosion control by inhibition 3 69 rr • • *m.—*m—*® 4 • • 66 • • • • HOOC © © © © © © CHa.HOOC Ntorgraup A Oil droplets {§) Oiemisorbed layer Figure 6.14 Mechanism of corrosion inhibition by polyphosphates 6.10 SYNERGISTIC INHIBITORS It is very rare that a single inhibitor is used in systems such as cooling water systems. More often, a combination of inhibitors (anodic and cathodic) is used to obtain better corrosion protection properties. The blends which are produced by mixing of multi-inhibitors are called synergistic blends. Examples include chromate-phosphates, polyphosphate-silicate, zinc-tannins, zinc-phosphates. Phosphonates have been used to cathodically protect ferrous materials. Following are the major applications of synergistic blends of inhibitors. Chromatepolyphosphate Chromateorthophosphate Polyphosphatesilicates Metal surface cleaning More effective corrosion control in oilfield Cooling water system Polyphosphateferrocyanide Zinc-tannins Amino-alcoholsodium nitrite Protection of ferrous and non-ferrous constructional materials Protection of copper and many ferrous materials Combines the precipitation effect of nitrite with the film forming properties of hydroxylklamine 6 .1 l SCAVENGERS Oxygen, even in very small amounts, may cause serious corrosion in feedwater lines, stage heaters, economizers, boiler metal, steam operated equipment and condensable piping. It must, therefore, be removed from the closed system. The solubility of oxygen varies with both pressure 3 70 Principles of Corrosion Engineering and Corrosion Control Table 6.1 Advantages and disadvantages of sodium sulfite and hydrazine Chemical Advantages Sodium sulfite • Rapid reaction • Non-toxic Disadvantages • Does not reduce ferric oxide to magnetite • May decompose to form corrosive gases and temperature. Oxygen is the main cause of corrosion. It reacts by consuming electrons at the cathode causing cathodic depolarization and enhancing the rate of corrosion. Chemicals which eliminate oxygen from the closed systems are called scavengers. Ammonium sulfite (NH 4 )2S0 3 , and hydrazine (N2H4) have been successfully used over the years to eliminate oxygen. Oxygen scavengers remove oxygen as shown below: (NH 4 ) 2 S0 3 + - 0 2 -> (NH 4 ) 2 S0 4 (1) Org. molecule(aq) -f nHiO^ads) — > Org. molecule(ads) + nfyO^soln) • Contributes no solids • Reduces ferric oxide to magnetite Hydrazine • Less dosage for scavenging compared to sodium sulfite required • Reacts less rapidly compared to sodium sulfite • More expensive than sodium sulfite (2) N 2 H 4 + 0 2 -> N 2 f + 2H 2 0 (3) Na 2 S0 3 + - 0 2 -> Na 2 S0 4 (4) N H 4 HS0 3 + - 0 2 -* NH 4 HS0 4 (Ammonium hydrogen sulfate) The oxygen scavenger reaction rate changes by a factor of 2 for every change in temperature by 10°C as shown by reaction (1). Sodium sulfite reacts to form sodium sulfate and increases the total dissolved solid contents of the boiler water. Generally, eight parts by weight of Na2S03 are needed to scavenge one part of O2, but dosage depends on the purity. Hydrazine can react directly with dissolved oxygen as shown in reaction (1), but the rate of reaction is slow at temperatures below 150°C. The indirect reactions of hydrazine with oxygen, however, proceed rapidly at temperatures as low as 70°C. The indirect reaction proceed as below: (a) 4Fe 3 0 4 + 0 2 -> 6Fe 2 0 3 (b) 6Fe 2 0 3 + N 2 H 4 -> 4Fe 3 0 4 + 2 H 2 0 + N 2 t • Toxic and flammable hydrogen. At normal dosages, the quantity of such products is not significant. At times a combination of both may be required. Hydrazine is considered to be carcinogen by (OSHA). As a result, new scavengers have appeared in the market. The following are new oxygen scavengers which are being promoted in the market: • • • • Carbohydrazide Diethyehydroxlamine (DEHA) Ammonium isocascorbate Hydroquinone For the above reactions, lppm of hydrazine is needed for lppm of oxygen. It is the most Experience with the above inhibitors is presently not sufficient. Factors, such as pH, type economical and controllable scavenger. Advantages and disadvantages of sodium sulfite and of catalyst, temperature, presence of H2S (such hydrazine are shown in Table 6.1. At higher as in oilfields) and biocide must be considered steam temperatures, excess hydrazine decom- while using oxygen scavengers, as they effect their poses to form ammonia, nitrogen and sometimes properties. Corrosion control by inhibition 3 71 6.12 NEUTRALIZERS form relatively insoluble CaCC>3 scale as shown below. Neutralizing inhibitors reduce corrosion by reducing the concentration of H + ions in Ca + + + CO^~ = CaCC>3 (precipitates) solution. Neutralizers control a small amount of HC1, CO2, SO2, H2S, organic acids and other Unfortunately, the scaling tendency of CaC03 similar species that react with water to produce increases as temperature increases, because H + ions. The most commonly used neutralizing CaCC>3 exhibits an inverse temperature depeninhibitors are morpholine, cyclohexylamine and dence, the solubility decreases with increased diethylamine-ethanol. The formation of H + by temperature. With an extended period of time, CO2 is shown below: the scale (CaC03) becomes harder and harder. Sodium chloride may be added to decrease the C 0 2 + H 2 0 -+ H2CO3 scaling tendency, but it begins to increase when the salt concentration reaches 150 g/L. The sol-f- HCO3"" H2CO3 -> H+ ubility of CO2 also increases with increasing (Carbonic acids) (Hydrogen ions) (Bicarbonate ions) pressure. On further heating (beyond 70°C) the carbonate decomposes. CO3— + H 2 0 = C0 2 t + 2(OH)" The O H - ions combine with Mg + + ions to form a precipitate of Mg(OH)2 which is 6.13 S C A L E INHIBITORS Scales are the precipitators which are formed on surfaces in contact with water as a result of physM g + + + 2 (OH)" = Mg(OH)2 ical or chemical changes. Calcium carbonate and calcium sulfate are the major types of scales. The formation of scale creates operating problems, Mg(OH)2 scaling is predominant above 82°C and such as blockage of the flow lines, and reduc- C aC0 3 scaling below 80°C. Calcium sulfate scaling is of three types: tion in the diameter of tubings. Scales are formed gypsum (CaS04 • 2H2O), anhydrite (CaSC^) and when the solubility product of scale forming hemihydrate (CaSC^ • | H2O). The gypsum is the constituents is exceeded. Seawater contains dissolved chemicals includ- more common form. Gypsum exhibits a maxiing carbonates, bicarbonates and hydroxyl com- mum solubility at 40° C followed by a decrease, pounds. These are identified as soft scale whereas anhydrite becomes decreasingly soluble compounds. At ambient temperature, seawater on increasing the temperature. Calcium sulfate scale is harder and less porous than CaC03 scale. is saturated with CaC03 and Mg(OH)2. Calcium carbonate is soluble in acids, whereas The mechanism of scale formation of each CaSC>4 is insoluble in acids. The effect of temperdeposit is different. Calcium carbonate is the ature on the solubility of CaCC>3 and gypsum is most common scale deposit. The solubility of shown in Fig. 6.15. CO2 decreases with increasing temperature and the equilibrium between CO2, CO^~ and HCO3 is upset. Bicarbonate ions break down (Tem6 .13.1 S C A L E R E M O V A L perature >70°C) causing liberation of CO2 and BY I N H I B I T O R increase of COJ~ concentration. A large number of compounds are known to inhibit scale formation by removing the scale As CO2 is evolved, the pH increases and forming ions and suspended solids from the H CO^ converts to the less soluble form CO^~. water. The common inorganic inhibitors used The CO^~ ion combines with Ca + + ions to are sodium hexametaphosphate and sodium 2HCO~ = C0 2 t + CO~" + H 2 0 3 72 Principles of Corrosion Engineering and Corrosion Control D Temperature f C) Figure 6.15 Variation of solubility of gypsum and CaC03 with temperature tripolyphosphate. They are effective at low concentration (2-5 ppm for CaC03 and 10-12 ppm for CaS04). The hexametaphosphate is, however, subject to reversal to the undesirable orthophosphate form above 140°F (59.5°C). Amongst the organic scale inhibitors, aminotrimethylene phosphoric acid (ATMP) is extensively used. It is stable up to 250°F (120°C) and at all pH values. In desalination plants CaSC>4 scaling can be controlled by allowing normal seawater to concentrate not more than 1.8 times at a time. The maximum temperature attainable in desalination plants without CaSC>4 • 2H2O formation is 120°C. Scale removal is an extensive subject and specialized literature must be consulted for further information. H O H H H 6 .13.2 S C A L E O REMOVAL BY C H E M I C A L A N D MECHANICAL METHODS O — P — C —N — C — P — O O H H HCH HOOPOH H O H Scales can also be removed by chemicals. Calcium carbonate is soluble in hydrochloric acid, formic acid, acetic acid and sulfamic acid. Hydrochloric acid can be used to remove CaCC>3 scale, however, it must contain one of the sequestering agents, like acetic acid, oxalic acid or gluconic acid, to prevent the undesirable precipitation of iron. Structure of ATMP Corrosion control by inhibition Gypsum (CaS04 • 2H2O) is insoluble in acids. It should, therefore, be converted to another scale, such as calcium carbonate, which is soluble in acid. Ammonium bicarbonate or sodium carbonate (Na2C03) convert gypsum to soluble CaCC>3. Gypsum can also be converted to Ca(OH)2 by KOH. Both the carbonates and hydroxide are soluble in acids. Ethylenediamine tetramine acid (EDTA) dissolves gypsum directly, but it is not economical. Mechanical softeners, known as 'pigs,' are sent through the pipe for removal of the hard scale. A proprietary class of inhibitors, 'threshold inhibitors,' are also added in small concentration to inhibit crystal growth. 3 73 6 .14 REBAR INHIBITORS concentrate (pH > 12.0) because of the development of a passive layer on steel surface under alkaline environment. The key to success is to restrict the permeability of the concrete. Mitigating corrosion inhibitors can reduce the mobility of corrosive ions and neutralize the corrosion species. The corrosion process in concrete is shown in Fig. 6.16. The passivation of reinforcement of concrete is provided by the high alkaline environment (pH = 12.8) from lime continuously produced from the hydration reaction of cement. This passive layer may, however, be destroyed due to ingress of chloride and carbon dioxide in certain environments and serious corrosion is caused. Concrete is permeable and allows the ingress of atmosphere. The carbon dioxide reacts with alkalies and forms carbonate and thus reduces the pH values to 9. Ca(OH)2 + C 0 2 + H 2 0 -> CaC0 3 + 2 H 2 0 If the carbonated front penetrates deeply into concrete and reaches the reinforcement surface, A serious problem which is faced by building industry is rebar corrosion in concrete. In steel reinforced concrete structure, corrosion attack is prevented by high alkaline environment in Penetration of oaygen and chloride kms 6 Impermeable dense concrete T$ &einforcin9 rod Cathodic area Cathode area LOSS or passivation by cfcloride, rusting $nti eventual spaing of concrete Figure 6.16 Schematic representation of the electrolytic microcell structure in reinforced concrete causing spalling 3 74 Principles of Corrosion Engineering and Corrosion Control protection is lost and steel begins to corrode as biofilms has occurred in many water treatment both moisture and oxygen are available to initiate plants during water pressure testing, even before the corrosion reaction of steel. the plants are put into use. This is due to stagnant water, even for short periods. Once started, it is extremely difficult or impossible to Fe —• F e 2+ + 2e stop such MIC. These include sulfate-reducers, slime formers, iron oxidizing bacteria, sulfur - 0 2 + H 2 0 + 2e -> 20H~ bacteria, algae, yeast and molds. Corrosion attack on steel is caused by one of the following mechanisms: F e 2+ + 2 0 H " -> Fe(OH)2 During re-alkalization treatment hydroxyl ions are regenerated by applying a negative voltage and corrosion is brought under control. Inhibitors, such as chromates, phosphates, sodium benzoate and sodium nitrate, have been applied to suppress corrosion. Sodium nitrate is known to preserve passivity on steel rebars and prevents the ingress of chloride. Proprietary blends, such as MCI 2022 (a blend of surfactants and amine salts in water) and MCI 2000 (alkanolamine), manufactured by CorrTech Corporation, USA, have been used in inhibition of rebars in 3.5% NaCl solution. The observed corrosion rate of rebar, after inhibition, was equal to that of steel 304 and 316 which are highly resistant to corrosion. Experience has shown calcium nitrite to be promising. It competes with chloride to react with Fe 2+ to maintain passivity by formation of oxide layers as shown below. 4 Fe 2+ + 4 0 H " + 2NO" -> 2NOFe 2 0 3 + 2H2 Inhibitors, such as sodium tetraborate (Na2B2Oy) and zinc borate (2ZnO3B 2 03) have proved effective inhibitors. ci- (a) Generation of iron sulfide cathodic to steel by bacterial action on a steel surface. (b) Formation of concentration cells by deposition of slime masses on the metal surface (Fig. 6.17). (c) Corrosion caused by H2S and carbonic acid generated by aerated alkali and bacteria. Two types of chemicals are employed to control bacterial growth: bacteriostats or bacteriocides. Bacteriostats put the bacteria in a dormant state without killing bacteria, whereas bactericides kill bacteria. Both types are jointly called biocides. The following is a list of important biocides. (1) Cocoamine acetate - i^ • N H 2 • HOOCH (2) Dialkyl-benzyl ammonium chloride. C fi H fi 6.15 BlOCIDES Bacterial corrosion (MIC) is extremely damaging and causes serious corrosion problems including plugging due to accumulation of solids and slime. Bacteria are unicellular, microscopic organisms (3) Acrolein, (CH2 = CHCHO) found in fresh water and in brine (pH 5-9). (4) Chlorine dioxide, (C102) The organisms are responsible for plugging and contamination problems in cooling towers and The objective of the biocide is not to kill the filters. Serious pitting of stainless steels under bacteria, but to control their counts and minimize Corrosion control by inhibition 375 Momacygefi Dirt/Slime deW$ Less oxygen OH * ' 1 OH 1 Q^hocte-' \ ^J"%sU Metal /»Cathode \ L_ 6.16 INHIBITOR / Figure 6.17 Differential oxygen concentration cell formation in slime deposition their deleterious effect on corrosion and water quality. APPLICATION TECHNIQUES Basically, there are three well-known inhibitor application techniques: (1) Continuous injection (2) Batch treatment (3) Squeeze treatment In the continuous injection method, a constant supply of chemicals is maintained at a controlled rate. The chemical injection pumps, however, require constant monitoring to check their performance. It is a cost-effective system and widely practiced in oil industry. A diagram of crude oil unit overhead showing NH3 injection and automated pH control is shown in Fig. 6.18. 6 .16.2 BATCH TREATMENT 6.16.1 C O N T I N U O U S INJECTION As the name suggests, inhibitors are injected in the system to achieve inhibition objectives through the system. Normally the inhibitor is injected into the system by means of an electric or gas driven chemical pump. The inhibitor is added at the point of turbulence to achieve uniform mixing. This method is used for municipal water supplies, cooling towers and oil wells, to minimize scaling and corrosion problems. This is a periodic treatment in which a large quantity of chemicals is used for an extended period of time. It is commonly used to treat flowing oil wells. Batch treatment is also called slug treatment. For batch treating, the tube displacement method is employed. Several barrels of inhibitor are pumped into the tubing at the top. The inhibitor is displaced to the bottom of the tubing with the fluids in the oil well. The well is closed for a specific period before operation. The batch is used mainly to treat water with biocides and not to supply inhibitors or scavengers. It is applied in areas where continuous injection is not practical. The process is not economical as substantial amount of chemicals may be wasted. Also, during the period of shutdown, 3 76 Principles of Corrosion Engineering and Corrosion Control Water recirculation N HJ«#| (T Sour water to safe disposal Figure 6.18 Crude oil unit overhead showing NH3 injection and automated pH control. (From Alley, D.W and Coble, N.D. (2003). Metal Perf., 44, May 03) there is no way that corrosion could be controlled. the success of this method. This method allows continuous treatment, however, it can damage the oil producing formation. It is also not cost effective. 6 .16.3 SQUEEZE TREATMENT Continuous treatment of oil wells by inhibitors is achieved by this method. The liquid inhibitor is pumped down through the tubing into the oil producing geological formation under low pressure which acts as a chemical reserve. In oil wells, 1-2 drums of inhibitor is mixed with 10-20 bbl of water (lbbl (British barrel) = 36 gallons), and is pumped into the well followed by pumping in over-flush fluid (50-75 bbl). The inhibitor is absorbed by the formation. It slowly escapes from the formation over a period of time to inhibit the corrosive fluids. A continuous slow release of the inhibitor from this producing formation into the corrosive fluids is the key to 6.17 VAPOR PHASE INHIBITORS By definition vapor phase inhibitors (VPI) are volatile by nature and they are transported to the desired site by volatilization from a given source. They also extend the protection offered by VPI-impregnated wrapping papers to areas out of contact with the wrapping papers so that protection may be achieved by the areas where wrap cannot be applied to complex shapes of the component. Two most popular VPIs Corrosion control by inhibition 377 are cyclohexylamine carbonate and dicyclohexywhere lamine nitrite (DCHN) with vapor pressures of 0.4 mm and 0.0001 mmHg, respectively. DCHN Qinh = quantity of inhibitor, kg may be hostile to magnesium, cadmium, zinc and Vfluid = volume of fluid to be inhibited lead. Cyclohexyl carbonate (CHC) may attack copper and its alloys and discolor plastics. The Qnh = inhibitor concentration, ppm mechanism of inhibition of VPIs is not clearly understood. It is assumed that undissociated molecules of the above inhibitors migrate to the Example metal surface and undergo hydrolysis by mois- Calculate the dosage of sodium chromate ture to liberate nitrite, bonzoate or bicarbonate. required to be added to 500 000 liters of In the presence of oxygen, they passivate the water, if the concentration of sodium chromate steel surface. Both DCHN and CHC are impreg- is 5 ppm. nated in kraft wrapping papers to effectively Solution: protect the steel surface. It is, however, difficult to achieve protection in tropical humid ~ V Na 2 Cr0 4 n environments. Vapor phase inhibitors are used Q Na 2 Cr0 4 = ~ TZZ X <-Na2Cr04 I n P P m frequently in storing equipment for a long period of time and for shipping of machinery and 500 000(kg) components. = 1.0 x 106 > 5 ( P P - ) < = 2.5 kg 6.18 INHIBITOR QUESTIONS A. M U L T I P L E C H O I C E MUtbTlONb Mark one correct statement in each of the following questions: 1. A corrosion inhibitor protects a metal surface by a) eliminating the corrosion from the metal surface b) dissolving the corrosion products formed on the metal surface c) increasing the thickness of the metal surface d) being adsorbed on the metal surface 2. Cathodic inhibitors react by a) increasing the rate of cathodic reduction by hydrogen EFFICIENCY AND INHIBITOR CONCENTRATION (1) The efficiency of corrosion inhibition can be expressed as CR0 - CRi -Rnh — CR^ where £inh = efficiency of a corrosion inhibitor CRo = corrosion rate with zero inhibitor CRi = corrosion rate in the presence of an inhibitor (2) The quantity of inhibitor required for a fluid to be inhibited can be obtained by the relationship: Qinh = Vfluid 1.0 x 10 6 x Q n h (ppm) 3 78 Principles of Corrosion Engineering and Corrosion Control b) increasing the rate of oxygen reduction at the metal/electrolyte interface c) shifting the potential of the cathode in a more negative direction d) shifting the potential of the cathode in a more positive direction 7. When present in sufficient concentration, organic inhibitors affect a) only the anodic area b) only the cathodic area c) the entire surface 8. The efficiency of certain organic molecules, such as organic amines, improve in the presence of certain halogen ions, such as chloride, bromide or iodides. A combination of amine and chloride ions gives a greater degree of efficiency than either of the two alone. This process is called a) physical adsorption b) chemisorption c) synergism d) additive effect 9. Vapor phase inhibitors, like dicyclohexylamine, inhibit the corrosion most importantly by a) b) c) d) liberating carbon dioxide scavenging oxygen making the environment alkaline passivating the surface of the condenser tube 3. Which one of the following is not an important factor in the adsorption of organic inhibitors on the metal surface? a) Nature of the inhibitor b) Surface charge of the metal c) Position of the metal in galvanic series d) Nature of electrolyte 4. Which one of the following is not characteristic of organic inhibitors? a) The inhibitors have no effect on surface potential b) The inhibitor increase metal activity by the process of desorption c) The inhibitor does not form a complex at the electrode d) It forms a physical barrier and decreases the diffusion of the reactant 5. Which one of the following inhibitors does not require the presence of oxygen to cause passivation of the surface? a) Phosphate b) Tungstate c) Molybdate d) Nitrite 6. Which one of the following classes of inhibitors may be termed as dangerous inhibitors? a) Cathodic inhibitors b) Anodic inhibitors c) Organic inhibitor d) Mixed inhibitor 10. Hydrocarbon chains play an important role in the corrosion inhibition process by a) forming a thin layer of oil and water on the metal surface b) dispersing the oil layer in the water layer c) forming an oily layer with a few molecules thick on the surface and creates a barrier to diffuse d) decreasing the film life of the inhibitor 11. Chromate inhibitors are effective because a) they form a chromium oxide film that combines with iron oxide fully b) they remove oxygen from the cooling water system Corrosion control by inhibition c) they are not cost effective d) they are initially applied at low rate 12. Which one of the following kills bacteria? a) Bacteriostats b) Bactericides 13. Which one of the following is not an advantage for the continuous treatment process? a) b) c) d) It supplies chemicals all times It is cost effective The injection rate may be optimized The injection pumps need proper maintenance 379 14. Which one of the following is not an advantage for a batch treatment process? a) Allows treatment in an area where continuous treatment is difficult b) A significant portion of the chemicals in water c) Corrosion and bacteria are not continuously controlled between treatments d) Wells are to be closed for a specific period 3) Why are inhibitors bonded by chemisorption more effective than inhibitors bonded by physical adsorption? 4) As the cathodic reactions consume electrons in the reduction reaction, how is alkalinity affected by these reactions? 5) Why do hydrocarbon molecules keep the water away from the metal surface? 6) How does sodium sulfite scavenge oxygen from the water? 7) What is the major difference between batch treatment and continuous treatment? 8) What is the difference between synergistic adsorption and chemisorption? 9) What is the difference between cathodic protection and corrosion inhibition? 10) What is the effect of increase and decrease of temperature and pressure on the scale forming tendency of calcium carbonate and calcium sulfate? C. C O N C E P T U A L Q U E S T I O N S 1) State the basic difference between the anodic and cathodic inhibitors in terms of shift of anodic and cathodic potentials. 2) State briefly the mechanism of prevention of corrosion treatment with sodium phosphate and how does it differ from polyphosphate 15. Which one of the following inhibitors is treatment. considered effective for inhibition of rebar 3) Describe the essential difference between procorrosion? cesses of chemisorption and physical adsorption with examples of inhibitors which react a) Zinc borate on the metal surface by the above two processes. b) Calcium nitrite 4) State the mechanism of cathodic inhibition by c) Sodium tetraborate addition of poisons of Group VA elements. d) MCI 2000 5) Summarize the squeeze treatment process in five important steps. B . How AND WHY QUESTIONS 1) Why are inhibitors needed to be injected in S U G G E S T E D READING pipelines transporting oil and gas? 2) The bulk of inhibitors in use in oil and gas [1] Alley, D.W. and Coble, N.D. (2003). Corroindustries are long chain hydrocarbons, such sion inhibitors for crude distillation columns. as aliphatic amines. They are very successMaterials Performance, July, 44,44-50. fully used as corrosion inhibitor because they [2] Byars, H.G. (1999). Corrosion Metal in Petroleum significantly suppress corrosion. Explain Production, TCE publication 5, 2nd ed. NACE, how the protection is achieved. Texas: Houston, USA. 380 Principles of Corrosion Engineering and Corrosion Control Bavarian, B. (2000). Corrosion inhibitors: STE rebar in concrete. Newsletter, Cortec Corporation, 1-4, USA. Bergman, J.I. (1963). Corrosion Inhibitors, New York: Macmillan. Bockris, J.O.M. etal (1967). /. Electrochem Soc, 114, 994, USA. Bradford, S.A. (2001). Corrosion Control Canada: Casti Publishing Company. Davenport, P.D. (1986). Protection of metals from corrosion. In Storage and Transit. Ellis Horwood, 171-192, UK. Drew Chemical Corporation (1980). Principle of Industrial Water Treatment, 3rd ed. Dunlop, A.K. (1978). Theory and use of inhibitors, NWS Corrosion Seminars. NACE, 147-152, Texas: Houston, USA. Greene, N.D. (1982). Mechanics and application of oxidizing inhibitors. Mat. Pref, 21,(3), 20. Hackerman, N. and Snavely, E.S. (1984). In Corrosion Basics, NACE, 127-146, Texas, USA. Kemmer, F.H. ed. (1988). The NALCO Water Handbook, 2nd ed. McGraw Hill. Munteanu, V.F. and Kinney, F.D. (2000). Inhibition properties of a complex inhibitor mechanism of inhibition. CANMET, Canada: Hawa, 255-269. Mohsen, E. (1978). Overhead corrosion control in crude distillation unit. Corrosion/78, Paper No. 132, NACE, Texas: Houston, USA. Oakes, B.D. (1981). Historical review of inhibitor mechanisms. Corrosion/81, Paper No. 248, NACE, Texas: Houston, USA. Pettus, P.L. and Strickland, L.N. (1974). Water soluble corrosion inhibitors: A different approach to internal pipeline corrosion control. Corrosion, 7 4,1-11, Texas: Houston, USA. Port, R.D. and Herro, H.M. (1991). TheNALCO Guide to Boiler Failure Analysis. New York: McGraw Hill, USA. Revie, W. ed. (2000). Uhligs Corrosion Handbook, 2nd ed. New York: John Wiley, USA, 89-1105. Riggs, O.L. Jr. (1973). In Corrosion Inhibitors. Nathan, C. C. ed. Texas: NACE 11. Simon-Thomas, M.J.J. (2000). Corrosion inhibitors selection: Feedback from the field, Paper No. 56, Corrosion, 56, NACE, Texas: Houston, USA. Wood, W.G. (Coordinator) (1999). Metals Handbook, 9th ed. Vol. 5, ASTM, Ohio: Metals Park, USA. Wilken, G. (1992). Optimization of corrosion inhibition of sour gas gathering pipelines, Report, Saudi Aramco, AER 5436. Ash, M. and Ash, I. (Compilers) (2001). Handbook of Corrosion Inhibitors, NACE, Texas: Houston, USA. [24] Davies, M. and Scott, P.J.B. (2000). Guide to Use of Materials in Water, NACE, Texas: Houston, USA. Sastri, V.S. (1998). Corrosion Inhibitors: Principles and Applications. New York: Wiley. Nathan, C.C. (1973). Corrosion Inhibitors, NACE, Texas: Houston, USA. [25] [26] KEYWORDS Adsorption Attractive interaction of an inhibitor with a metal surface due to ionic charge of the inhibitor and charge on the metal surface. It is a physical process and does not result in bond formation. Anodic inhibitors They suppress the anodic reaction, displace the corrosion potential in the positive direction, and reduce the current density. Batch treatment Addition of an inhibitor at one time to provide protection for an extended period of time. Additional quantities may be added later depending on the need. Cathodic inhibitors Inhibitors which decrease the corrosion rate by suppression of cathodic reduction reaction, such as oxygen and hydrogen reduction. Cathodic precipitates Inhibitors, like calcium carbonate and magnesium carbonate, which precipitate on cathodic areas. Chemisorption Formation of a bond between the inhibitor and the metal substrate by donation of a pair of electrons, for example, chemisorption of amines by a metal surface. Critical concentration The concentration of an inhibitor below which it would not inhibit corrosion. De-passivation Destruction of a passive film formed on a metal surface. De-passivation increases the rate of corrosion contrary to passivation. Inhibition Suppression of corrosion by use of inhibitors. Inhibitor efficiency Degree of inhibition offered by an inhibitor. It is given by (CR(unjn) — CR (inh)/ CR (unin)) x 1000> w h e r e CR(inh) i s t h e corrosion rate in the presence of inhibitor, and CR(unjn) the corrosion rate in the absence of inhibitor. Mixed inhibitors A mixture of anodic and cathodic inhibitors. They inhibit both the anodic and cathodic corrosion reactions. Neutralizing inhibitors Inhibitors which reduce corrosion by reducing the concentration of hydrogen ions in solution. Non-oxidizing inhibitors Inhibitors which do not need oxygen to passivate a surface. Organic inhibitors Long chain nitrogen containing molecules which suppress corrosion by being adsorbed on the metal surface and forming a chemisorptive bond. C orrosion control by inhibition Passivating inhibitors Inhibitors which shift the potential of a metal in the noble direction. They stabilize passive films, and also repassivate the metal surface if the passive film formed is damaged. Poisons Elements from Group VA, like As, Bi, etc. which retard the process of cathodic reduction. Scaling Formation of scale by precipitation of an ionic material from water. Examples are calcium carbonates as calcium sulfate scale in tubing and flow lines. 381 Scavengers Chemicals added to reverse oxygen from feed-water, e.g. sodium sulfite and hydrogen. Squeeze treatment A method of continuous feeding an inhibitor in an oil well by pumping a quantity of inhibitor into the well. The inhibitor is absorbed by the formation from which it slowly escapes to inhibit the produced fluids. Synergistic inhibitors An increase in the inhibitor efficiency of organic inhibitor in the presence of halogen ions. COATINGS 7 .1 INTRODUCTION A lthough coatings have been used for over thousands of years for decorative and identification purposes, the industrial importance of coatings has only been recognized after World War II. The total amount of paint sold annually amounts to billions of gallons. In 2000, USA alone manufactured about 3.5 billion gallons of paint. About one-third of the production of paint is used to protect and decorate metal surfaces. All forms of transport, such as trains, ships, automobiles, aeroplanes, underground buried structures, such as tanks, oil and gas pipelines, offshore structures, iron and structures and all metallic equipment require the use of coatings. The coating industry has, therefore, turned out to be one of the largest in terms of production. The importance of coating can be judged from the fact that coating can hardly be ignored in any corrosion protective scheme. Corrosion protection of over-ground and underground structures by protective coatings is one of the most proven methods. Other methods include cathodic protection, environmental modification, material selection and design. In contrast to the behavior of rust on steel, the formation of an oxide affords protection against corrosion. If the resistivity of electrolyte is increased and the electron flux is retarded, the rate of corrosion is decreased. By applying coatings of high resistivity, such as epoxies, vinyls, chlorinated rubbers, etc. the flow of electric current to the metal surface is impeded. Also, the higher the thickness of the coating, the higher would be the electrical resistance of the coating. A much higher resistance to the current flow would, therefore, be offered. Thus increasing the electrical resistance of metals by coating offers an excellent method of corrosion prevention. Another method to prevent corrosion is by the use of inhibitors. This can be achieved by using inhibitive pigments, like zinc chromate, red lead and zinc phosphate in coatings. An alternative method is to use a metal more anodic than iron, such as zinc. This is done by using zinc-rich paints. The zinc metal prevents the corrosion of iron by releasing electrons into the iron surface. Thus, coating is an effective method to control corrosion. Coatings must have the following characteristics for good corrosion resistance: (a) (b) (c) (d) a high degree of adhesion to the substrate minimum discontinuity in coating (porosity) a high resistance to the flow of electrons a sufficient thickness (the greater the thickness, the more the resistance) (e) a low diffusion rate for ions such as Cl~ and for H 2 0. Coating and paint technology is adapting to the environmental requirements. The development of water-borne coatings and solvent-free coatings signify new health and safety trends in coating technology. The description of high performance coatings is beyond the scope of this book. Coatings 3 83 7.2 OBJECTIVES (1) Anodic Oxides The following are the objectives of coatings: A layer of AI2O3 is produced on aluminum surface by electrolysis. As the oxides are porous, they are (1) Protection of equipment and structures from sealed by a solution of potassium dichromate. The the environment by acting as a barrier object of sealing is to minimize porosity. Howbetween the substrate and the aggressive ever, chromates have health hazards and are not environment, such as the marine and indus- allowed in some countries. trial environments. (2) Control of solvent losses. (3) Control of marine fouling; certain con- (2) Inorganic Coatings stituents in coating control the growth of These include coatings like ceramics and glass. mildew and marine fouling in seawater. (4) Reduction in friction; coating reduces fric- Glass coatings are virtually impervious to water. Cement coatings are impervious as long as they tion between two contacting surfaces. (5) Pleasant appearance; certain types of coatings are not mechanically damaged. provide a pleasant appearance and produce attractive surroundings. (3) Inhibitive Coatings (6) Change in light intensity; by selection of appropriate coatings the light intensity in rooms and buildings can be varied as desired. In several instances, inhibitors are added to form (7) Visibility; many combinations of colors surface layers which serve as barriers to the envibecause of their visibility from large distances ronment. Inhibitors, like cinnamic acid, are are used on TV and radio towers to warn added to paint coatings to prevent the corrosion of steel in neutral or alkaline media. aircraft. (8) Modification of chemical, mechanical, thermal, electronic and optical properties of (4) Organic Coatings materials. (9) Application of thin coatings on low-cost subEpoxy, polyurethane, chlorinated rubber and strates results in increased efficiency and cost polyvinyl chloride coatings are extensively used in savings. industry. They serve as a barrier to water, oxygen, and prevent the occurrence of a cathodic reaction beneath the coating. The barrier properties are further increased by addition of an inhibitor, like 7 . 3 CLASSIFICATION OF chromate in the primer. COATINGS Coatings can be classified in the following categories according to corrosion resistance: (a) (b) (c) (d) Barrier coatings Conversion coatings Anodic coatings Cathodic coatings. 7 .3.2 CONVERSION COATINGS Phosphate and chromate coatings are examples of conversion coatings. Conversion coatings are so-called because the surface metal is converted into a compound having the desired porosity to act as a good base for a paint. If iron phosphate is used, the following reaction takes place: 2Fe+3NaH2P04^ 2FeHP0 4 -+Na3P04+tH2 Coating 7 .3.1 BARRIER COATINGS Barrier coatings are of four types - anodic oxides, inorganic coatings, inhibitive coatings and organic coatings. The corrosion resistance is enhanced by phosphating. 3 84 7 .3.3 Principles of Corrosion Engineering and Corrosion Control ANODIC COATINGS Consider nickel coating on a steel substrate. A layer of bright nickel is laid on the dull layer of By anodic coating, it is meant that a coating which nickel. Over the bright nickel a layer of chromium is anodic to the substrate, such as zinc aluminum is laid. The bright nickel (high sulfur content) is or cadmium coatings. On steel such coatings are more noble than the steel substrate. Such a coating generally called sacrificial coatings. They protect system is called duplex coating. the substrate at the expense of the metallic coating applied. The zinc coatings protect the substrate by MISCELLANEOUS acting as a sacrificial anode for the steel which is 7 .3.5 cathodic to zinc: COATINGS ^Zn" -0.763 V, EL = - 0.44 V These include glass coatings, porcelain coatings and high-temperature coatings. Any breaks in the coating cause the anodic oxidation of Zn to occur. Zn^Zn2++2e The electrons are consumed by the iron substrate which acts as a cathode. The potential is made more negative by electrons and a cathodic reaction is forced to occur on it. 2H++2e->H2t A fine film of H 2 is formed on the surface. The steel being cathodic does not corrode. Thus, by acting as a sacrificial material, zinc corrodes while the steel substrate is protected. 7.4 S C O P E OF COATINGS Sky is the limit for the coating market, The following are the major target areas: Aerospace, power plants (turbines) and aircraft. Oil and mining industry. Equipment, such as pumps, valves, drilling rigs, slurry pumps, etc. Information storage: discs (magnetic coatings), TV display systems. Desalination plants: brine heaters, heat exchangers, circulation pumps, valves, etc. Automotive industry: gears, valves, pistons, panels, etc. Solar energy: photovoltaic cells and solar cells. Biotechnology and surgical implantation: artificial hearts, valves and joints. Utilities: all household appliances, washing machines, kitchenware and all electrical appliances. Pipelines: oil, gas and utilities pipelines. Transportation: decks, bridges and railcars. 7 .3.4 CATHODIC COATINGS In this type of coating, the metals which are deposited are electropositive to the substrate. For instance, for copper coated steel, copper (£° = +0.337 V) is positive to steel ( F =-0.440 V). The coatings must be pore-free and thick. Electroplated coatings are generally pore-free and discontinuities are not observed. However, if the coating contains a flaw (crater), it acts as the anode with respect to the substrate. Consequently, electrons flow from the crater to the noble coating. At the crater, hydrogen is evolved. 2H++2e^H2t Often an intermediate layer is put in between the substrate and the noble coating, such as the nickel-chromium coatings. 7.5 PAINTING, COATING AND LINING (a) A paint is a pigmented liquid composition containing drying oils alone or drying oils in combination with resins which combine Coatings with oxygen to form a solid protective and adherent film after application as a thin layer. (b) A coatingis any material composed essentially of synthetic resins or inorganic silicate polymers which forms a continuous film over a surface after application and is resistant to corrosive environments. (c) A liningis essentially a film of material applied to the inner surface of a vessel or pipeline designed to hold the liquids or slurries. 3 85 passive surface on the substrate, such as steel. Pigments, such as chromate salts and red lead are the examples of inhibitive pigments, but have health hazards. (2) Impervious Primer This primer used in impervious coating systems makes the coating much more impervious to the passage of CO2, oxygen, air, ions and electrons. Conventional thin film coatings cannot prevent oxygen and water permeation. The primer in an impervious system is used with a thick coal tar enamel to form a highly impervious coating. The pigments are generally metal salts, chromates of zinc, lead and strontium, but Cr and Pb contents have health hazards. 7.6 PAINT COATING SYSTEM The coating system comprises: (a) The primer (b) The intermediate coat (c) The top coat The primer is the most important component of the coating system as the rest of the coating is applied on the primer. In many paint systems, such as those containing a good proportion of natural oils, the pigments maybe inhibitive. However, some pigments, such as red lead in linseed oil (red lead primer) react with the oil to produce soap and protect the steel surface although they do not act inhibitively. The following are the functions of primers: (3) Cathodically Protective Primer One good example is a primer containing zinc. As zinc is anodic to steel, zinc corrodes in preference to steel and protects the steel from corrosion. Experience has shown that a zinc-rich primer can double the life of a chlorinated rubber or an epoxy top coat. The three types of primers described above form the basis of three important coating systems: (a) impervious coating system for equipment requiring immersion, (b) inhibitive system for application in marine (1) It must be strongly bonded to the substrate. atmosphere and (2) It must be resistant to corrosive environ- (c) cathodically protective system for severe corments and suppress corrosion. rosive environments. (3) It must provide good adhesion to the intermediate coat or the top coat. Primers are allowed to stand for a sufficiently long length of time before any coating is applied. Primers can be inhibitive, impervious or cathodic, as described below, and their applications are dependent on the environment encountered by the metallic structure. 7.7 PAINT COATING 7.1) COMPONENTS (TABLE (1) Inhibitive Primers The pigment contained in the primer reacts with absorbed moisture in the coating and form a (a) Vehicle. This is the liquid portion of the paint in which the pigment is dispersed. It is composed of binder and the solvent or thinner or both. (b) Binder. This binds the pigments in the coating in a homogeneous film. The binder also binds the total coating to the substrate. 3 86 Principles of Corrosion Engineering and Corrosion Control The composition of paints Function Protects the surface from corrosion by providing a homogeneous film. It controls the desired film properties. Dissolves the binder. It provides the means of applying the painting. The solid resins are transformed to a liquid state. Assist in improving the quality of the paint. Provides opacity, color and also resistance to corrosion. Used for a wide range of purposes in conjunction with pigments to extend the properties of the binders. Examples: talc, china clay, silica and barytes. such as tung oil, forms a tough solid film in air. (f) Extenders. They are added to improve the application properties of the paint. Table 7.1 summarizes the major components of paints. Paints and coatings can be divided into two categories: convertible type which need a chemical reaction, such as oxidation or polymerization, and the non-convertible type which are formed by evaporation of the solvents. The former category includes alkyds, epoxy, esters, polyesters, urethenes, silicon and other resins. Table 7.1 Component Resin (binder) Solvent Additives Pigments 7.8 COMPOSITION AND F U N C T I O N S OF PAINT COATING C O M P O N E N T S There are several ways of classifying binders (resins). One such classification is given below: (a) Drying oil types, such as alkyd (b) Epoxy, polyurethane and coal tar (two-pack chemical resistance) (c) Vinyl and chlorinated rubber (one-pack chemical resistance) (d) Bituminous coatings (e) Lacquers. Extender (inorganic matter) The binders provide the basis for the generic terminology of paint. The physical and chemical properties of paints are determined by the binders. (c) Pigment. A pigment not only provides a pleasing color but protects the binder from the adverse effect of ultraviolet radiation on the coating. (d) Solvent. The purpose of the solvent is to provide the surface with a coating material in a form in which it can be physically applied on the surface. (e) Additives. These are used to modify the properties of the coating, such as reducing drying time and enhancing the desired properties. One example is a plasticizer which makes the film flexible. Similarly a drier may be added. It is a substance, such a compound of lead, manganese, which when introduced in drying oils reduce their time. A drying oil, 7 .8.1 DRYING OILS These are natural binders and used as they are or with a certain amount of processing. This group contains natural drying oils, such as linseed oil, soya bean oil and coconut oil. They are modified with synthetic resins to obtain useful properties. The composition of major oils used in coatings is given in Table 7.2. Oils are classified as drying, semi-drying and non-drying on the basis of concentration of fatty acids. Fatty acids are called saturated if no double bond is present, unsaturated if one double bond is present and polyunsaturated if more than two double bonds are present. Coatings Table 7.2 Composition of major oils used in surface coatings Saturated acid Tung Linseed Soya bean Castor oil Dehydrated castor oil Tall Coconut 6 10 14 2-4 2-4 3 89-94 Oleic acid 7 20-24 22-28 90-92* 6-8 30-35 6-8 9,12 Linoleic acid 4 14-19 52-55 3-6 48-50 35-40 0-2 9, 12, 15 Linolenic acid 3 48-54 5-9 0 0 2-5 0 3 87 Conjugated acid orv*** 0 0 0 40-42** 10-15*** 0 * Principally ricinoleic acid, not oleic **Conjugated 9,11 linoleic acid ***Conjugated linoleic acid, isomerized linolenic acid and 9, 11, 13 eleostearic acid, proportions dependent on source and degree of refinement The following are typical structures: A drying oil must contain at least 50% polyunsaturated fatty acid. For instance, linolenic acid is a polyunsaturated fatty acid [CH3 • CH2CH = CH (a) Oil (Fatty acids) = CH2CH = CH2CH2CH = CH(CH 2 ) 7 COOH]. Linseed oil contains over 50% of 9,12,15 linolenic CH2—O—R acid, hence it is a good drying oil. Structures of I some fatty acids are shown in Table 7.2. CH —O—R Common to all oils is that they are treated I with various chemicals and heated to manufacture CH2—O—R alkyd resins. Oleoresinous binders are those which are (b) Glycerol manufactured by heating oils and either natural CH2OH or pre-formed synthetic resins together. The ole| oresinous vehicles have largely been replaced by CHOH alkyds and other synthetic resins. | CH2OH (c) Phthalic anhydride 7 .8.2 ALKYD RESINS The alkyd resin is made through reacting an oil with an acid and alcohol. Alkyd resins are the most extensively used synthetic polymers in the coating industry. The principal raw materials are oils, fatty acids, polyhydric alcohols and dibasic acids. Examples of the above are: (1) Oils - castor oil, soya bean oil, etc. (2) Polyhydric alcohol (glycerol). (3) Dibasic acid (phthalic acid). 3 88 Principles of Corrosion Engineering and Corrosion Control The most common binders used are chlorinated rubber, vinyls, silicones and urethanes. They are widely used paints, successfully applied under a wide range of conditions. They also belong to the convertible type of coatings. An oil derived fatty acid is chemically combined into a polymer structure. All natural fats are triglycerides [ 1 mole of glycerine + 2 molecules of fatty acid (saturated)]. Oils are formed from unsaturated fatty acids. All saturated fatty acids contain double bonds. To form an alkyd resin, a drying oil fatty acid, such as oleic acid, or linoleic acid is added to a polyhydric alcohol and a dibasic acid. CH 2 -OH I CH-OH I CH2-OH Glycerol is a traditional alcohol used in alkyd production. Phthalic acid is a typical dibasic acid. 7 .8.3 EPOXY ESTERS [Phthalic Anhydride + Glycerol + Drying Oil + Fatty Acid —> Alkyd] A simplified structure of oil modified alkyd resin is shown below: — FA — G — P A — G — P A — FA G—PA—G—PA PA I PA FA — G PA I I I FA I I G I I P A—FA FA=FattyAcid G =Glycerol PA=Phthallic Anhydride The alkyd has important properties over the original drying oil. Alkyds are often mixed with other generic types of binders to enhance properties. They also belong to the class of convertible type of coatings. A number of epoxy-based coatings are available, such as air drying (epoxy esters), catalyzed (amine or polyamide), epoxy co-polymers (coat tar, phenolic, amine or polyamide), heatcured (epoxy-phenolic) and high-build epoxybased coatings. The group of synthetic resins produce some of the strongest adhesives in current use. One component epoxy, is epoxy ester paints which are cured by oxygen. A special class of epoxies which has come out in recent years is epoxy mastics which are solventless materials. They need less surface preparation and can be cured down to — 10°C. They form very homogeneous coatings. Another class is termed coal-tar epoxy, in which some epoxy is replaced by tar. They are used for tank lining. Epoxy resins are combined chemically with drying oils to form epoxy esters. Epoxy resins are formed by the reaction between diphenylolpropane (bisphenol A) and epichlorohydrin (DPP + ECH). The epoxy group H2C —O —CH2 is found at both the ends of the chain. Amines or polyamides are used as catalysts for epoxy resins to cause cross-linking of molecules. These catalyzed epoxy coatings have a good resistance to alkalies solvents and acids. Both the terminal epoxide group and the secondary hydroxy groups of solid epoxy resins can be reacted with fatty acids to produce the epoxy ester. The epoxide group is terminal in the molecule and the resistance is due to the C-C bond and ether linkage in the polymer. Adhesive properties are because of the polar nature of the polymer. Their overall properties are superior to alkyds. They have improved water and hydrocarbon resistance. They have, however a tendency to 'chalk.' The reaction between DPP and epichlorohydrin in presence of an alkali produces epoxy resin as shown in Section 7.8.3. Coatings 7 .8.4 URETHANE OILS ALKYDS 3 89 AND URETHANE The term polyurethanes denotes a class of polymers containing the group: — NH — C — O II O The urethane oils and urethane alkyds constitute about 50% of the total products of alkyds. They correspond in composition to the conventional air drying alkyds discussed earlier, the only difference being that the dibasic acid (phthalic acid) used in the case of air drying alkyds is replaced by an isocyanates ( R - N = C = O). The most common isocyanate used is toluene diisocyanate (abbreviated as T.D.I.)- The T.D.I, reacts with an active hydrogen atom obtained from hydroxy containing compounds, such as polyethers and vegetable oils. The performance depends upon the characteristics of isocyanate and hydroxy compound used. They do not react with moisture. The following is the sequence of preparation of urethane oils: CH2OR CH2OH CH2OR CH2OR The alkyd molecule retains the fatty acids groups. The oil undergoes oxidation, hence the curing of this alkyd takes place by oxidation of oils. These alkyds have a high resistance to abrasion. The polyurethane resins have a good resistance to water, solvents, organic acid and alkaline. They have a high degree of abrasive resistance. 7 .8.5 SILICONE ALKYDS There are numerous types of paints based on silicone which are combined either with alkyd or acrylic resin. Pure silicone resists temperature up to 600° C and has excellent weathering properties. The copolymer is obtained by reaction between the hydroxyl groups in the alkyd resins with the hydroxyl groups in the silicon intermediates. The silicon intermediate has a low molecular weight. Silicone resin contains a relatively high percentage of reactive hydroxyl (Si - OH) or aloxyl (Si - OR) groups attached to silicon. Thus a copolymer structure is obtained which combines alkyd with silicone. = Si - O H+HOC - ALKYD -> I I I I I | CH2OR I CH2OH | CHOR + CHOH CH2OR CH2OH Drying oil Glycerol CH OH + CH OH Alcoholysis = Si-0-C-ALKYD+H20 The silicone intermediates commercially available for polymer modifications are Intermediate A R R Intermediate B R R Mixed mono- & diglycerides I C HOH R I Si—OH I I — Si—O—Si M eO-Si—O —Si—O- \/ \oN I Si O-Si I R I OMe H O—Si l / 0/ I R, I R CHOH — Urethane oil This combines the properties of the alkyds with the properties of silicones. Silicone alkyds are highly resistant to weather and heat because of silicon contents. Paints made from silicone modified resins have excellent weatherability, with good chemical resistance. Silicone copolymers can withstand most severe weather conditions. They are extensively used for coating of steel coils. 3 90 7 .8.6 Principles of Corrosion Engineering and Corrosion Control LACQUERS The vinyl chloride resin coatings are wellknown for their durability. The vinyl copolyThese are solutions of organic film forming mate- mer contains 86% vinyl chloride and 14% vinyl rials in organic solvents from which the solvent acetate. The material containing the vinyl group evaporates after application to a substrate. The when combined with catalysts react to form long solid film is formed after evaporation. The prop- chain polymers. erties of lacquers depend on the type of resin used. The lacquers dry very rapidly. The following are the major types of lacquers: H „C1 H JCI H H v r =r Polymerization / K \ • • • • Cellulose derivatives Polyvinyl chloride Chlorinated rubber Polyurethane elastomers c PVC-PVA Copolymer c c (1) Cellulose Nitrate The structure of cellulose nitrate is shown below: The above is a double bond polymerization process. One vinyl chloride molecule is made to react with another molecule to form the PVC polymers. H OH CH 2 ON0 2 Nitrocellulose nitrate /H H ^ ci cr 0 = C - CH, X H W m The raw material is the wood pulp or cotton linters which are the sources of cellulose. The cellulose is nitrated and heated in water until the desired characteristics re-obtained. It is used as a film former for lacquers. X I Or ^H 1 1 O = C - CH3 The vinyl acetate and vinyl chloride monomers react together to form a polyvinyl polychloride acetate resin. A typical resin is bakelite which contains 14% vinyl acetate and 86% (2) Polyvinyl Chloride vinyl chloride. Vinyl coatings are inert to alkalies, water, oils Vinyl chloride is a gas produced by reacting and greases. Vinyl solutions containing PVC are ethylene or acetylene with hydrochloric acid. used for pipelines, dams and several important The reaction replaces one hydrogen atom in industrial applications. They have an excellent ethylene with a chlorine atom which makes it resistance to water and weather. non-burning. Polymerization produces polyvinyl chloride. Polyvinyl chloride is a linear chain polymer with bulky chlorine side groups. It is (3) Chlorinated Rubber conveniently represented as below: H CI It is derived from natural rubber which is a polymer of isoprene. Tetrachloroisoprene is produced by chlorination of natural rubber. A chlorinated isoprene molecule and its conversion to chlorinated isopropane molecule is shown below. C —C H H Coatings CH3 CH2— C = CH — CH2 Isoprene 3 91 I — CH 2 —C--CH— CH2 I 1 methacrylic acid or copolymers with ethyl, -butyl or methacrylate. Thermoplastic acrylics are prepared by the co-polymerization of a mixture of acrylic and methacrylic monomers. H CH, _[_ Methyl Methacrylate CH3 I ~CH= C —CH = CH3 1 CH —C =CH~- C H CI + C12^ CH3 1 CH— C—CH"- C H I l > • I I 1 1 CI CI CI CI Chlorinated Isoprene Molecule 1 1 l CI 1 l +C H c=o 0-CH3 This provides strong resistance to inorganic acids, water and general chemicals and is heat resistant and possesses good durability. They dry at high speed and show a good versatility for blending with other coatings. Chlorinated rubbers are thermoplastic which means that when they are subjected to high temperature, they become soft. Above 60° C, chlorine may be released. Resistance to hydrocarbons is poor. They have a water white color and show a high resistance to change in color with time. Generally acrylics when co-reacted with other resins, such as epoxy and isocyanide, show a greater durability. They have found a wide use in trains, aeroplanes and buses. 7 .8.7 BITUMINOUS MATERIALS (4) Acrylics The acrylics have been, to a great extent, replaced by chlorinated rubber and vinyl paints for use both in land-based and marine industry. They have a general chemical structure shown below, where R and R' are either hydrogen or alkyl group. R I —C COORl CH9 in Acrylic polymers are mainly used for their excellent strength, chemical and weather resistance. They are made by polymerization of esters of acrylic and methacrylic acid. The main resins are polymers of methyl and esters of acrylic and Asphalt is a component of crude oil, and coal tar is a residue of coal tar distillation. The above two are generally combined with a solvent, such as aliphatic or aromatic hydrocarbons, to form a lacquer. They dry by solvent evaporation. Water emulsions are prepared by suspending minute particles of bitumen in water and emulsifying with fillers, such as manganese and aluminum silicates, coal dust and limestone dust. The bitumen is heated to a temperature 175-245° C for hot metal applications. They provide the best moisture and chemical resistance. They are economical and used for foundries, underground storage tanks, truck chassis, radiators and water proofing. They have a good resistance to non-oxidizing acid, hence, they are attacked by non-polar solvents. When coal is heated to 1095°C in the absence of air, it decomposes into gas and coke. Coal tar is formed and gas is condensed. All coal tar coatings are subjected to degradation by ultraviolet radiation and cracking. 3 92 Principles of Corrosion Engineering and Corrosion Control Asphalt is a hard natural brittle resin. 7 .8.9 ADVANTAGES AND Asphaltic compounds are composed of complex D I S A D V A N T A G E S O F O R G A N I C polymeric aliphatic hydrocarbons. They have a good water and ultraviolet resistance. Coatings C O A T I N G S based on bituminous resins fall into four major Table 7.3 shows the advantages and disadvantages groups: of various organic coatings. (1) Solid asphalt. Solid blocks of asphalts are melted and applied over the substrate. (2) Solvent cutbacks. Bitumen is dissolved by a solvent to produce a non-convertible coating by solvent evaporation. It is called as cutback asphalt. (3) Emulsion. The bitumens are emulsified with water and applied as top coat. (4) Varnishes. The bitumen replaces some or all parts of a resin in oleoresinous paints (oils combined with resin to obtain paints, such as alkyd, phenolics, etc.). The bitumens and tars both have been used extensively in blast tanks. INORGANIC S I L I C A T E S The inorganic silicates form a class of binders which are either dissolved in water or solvent and which react with water to give an inorganic film. They are prepared by fusing a mixture of sand or silicon oxide and typically a mixture of potassium or sodium carbonate. The main use of the silicate binders is in conjunction with a high loading zinc dust as a corrosion control primer. They can be classified in three groups: (a) Post-cured water-based alkali metal silicates (b) Self-cured water-based alkali metal silicates (c) Self-cured solvent-based alkali silicates 7 .8.8 HEAT CONVERSION BINDERS Examples of heat conversion binders include asphalt or coal tar which are melted and applied on the surface. They also include the resins which do not solvate until the heat is applied. They also include high molecular weight thermoplastics or semi-thermoset resins, such as epoxies. Examples: Fusion Bonded Epoxy (F.B.E.). The steel is decontaminated, abrasive blast cleaned and heated by electrical induction heaters to the prescribed temperatures, before uniformly applying epoxy powder. Another example is the use of hot applied coal tar enamels used mainly as pipeline coatings. Organosols and Plastisols. Vinyl coatings of high film thickness can be applied by dispersing polymers of vinyl chloride in a plasticizer (plastisol) or by dispersing in a plasticizer and an organic diluent (organosol). They provide flexiblefilmsof thickness up to 5 mm. Organosols are used mainly for spray applications on furniture and tool handles, etc. Plastisols are used in coil industry and containers. (1) Post-cured Water-based Alkali Metal Silicates They come as a three-package system consisting of zinc dust, silicate binder and the curing solution. The pigment portion is metallic zinc or metallic zinc modified with up to 20% of a coloring pigment. The coating has three components: zinc dust, silicate solution and curing solution. (2) Self-cured Water-based Alkali Metal Silicates Silicates, such as sodium and potassium, are combined with colloidal silicon to accelerate the speed of curing. They are based on high ratio of sodium silicates (Na20-Si02) potassium silicates. Lithium hydroxide colloidal silica and quaternary ammonium silicates are also included, but their higher cost tends to restrict their use. The corrosion resistance of solvent-based coatings is a little inferior to the post-cured coating. Coatings Table 7.3 Advantages and disadvantages of various organic coatings Advantages Good wetting, good flexibility, low cost Good durability, good gloss, low cost Good resistance to moisture, moderate cost Excellent resistance to water, excellent durability Good performance, good resistance to water Excellent chemical resistance, excellent adhesion, good gloss retention Excellent heat resistance Disadvantages 3 93 Type of coating (1) Drying oil (2) Alkyds (3) Epoxy Esters (4) Chlorinated rubber (cured by evaporation) (5) Bitumens (cure by evaporation) (6) Polyurethane Slow drying, poor chemical and solvent resistance Poor chemical resistance Poor gloss retention, tendency to chalk Poor heat resistance Poor resistance to hydrocarbons Relative high cost (7) Silicones Low resistance to chemicals, high cost (3) Self-cured Solvent-based Alkali Silicates These are solvent born alkyl silicates coatings in which zinc is incorporated. The binders are mostly modified alkyl silicates consisting of partially hydrolyzed silicates (generally ethyl-silicate type). The following are the ingredients of a typical ethyl-silicate zinc-rich coating: (a) Pigments: zinc dust (38% wet volume; 75% dry volume) (b) Solvent: ethyl alcohol (C2H3OH) (c) Binder: ethyl-silicate (partially polymerized) Reaction Chemistry Firstly the silicate film is concentrated as the solvent is evaporated. The binder (tetraethyl-silicate) hydrolyzes to form silicic acid which reacts with zinc to form silica-oxygen zinc polymer. Zinc reacts and combines with the matrix. The unreacted zinc surrounds the large molecules. A film of silica matrix will surround zinc polymer. O c/l o 1 1 1 oy 1 o 1 1 o 1 1 1 o— si -- o ~~Si 1 Zn 1 1 o Z1 n 1 o I 0 Fe o Zn 1 1 O O I x/O Si ^ O 1 ^0 Si —O —Si —O —Si — O —Zn —O I 0 Zn I 0 Fe The binder becomes extremely hard and corrosion resistant. The inorganic zinc silicates have an excellent resistance to chemicals, wears and weathering. They also afford cathodic protection to metal substrate. The coatings can be applied at higher humidities and lower temperatures. 3 94 Principles of Corrosion Engineering and Corrosion Control 7 .9 CLASSIFICATION OF BINDERS Binders are classified according to their chemical reactions. The binders belong to one of the following types: (a) Oxygen reactive binders. The resins belonging to this category produce coatings by intermolecular reaction with oxygen. Urethane alkyds, epoxy esters and silicone alkyds are examples. (b) Lacquers. The coatings are converted from a liquid material to a solid film by evaporation, e.g. PVC copolymers, chlorinated rubbers and bituminous material. (c) Heat conversion binders. Materials, such as sulfate or coal tar, are applied in a melted condition. Other examples are organisols and plastisols. Table 7.4 shows composition and application (d) Co-reactive binders. Binders, such as epoxies and polyurethanes, are formed by properties of binders and lacquers. combination of two low molecular weight resins. A solid film is formed after they react upon application on the solid surface. (e) Condensation binders. Binders, such as phenolic resins, release water during polymerization. Phenolic resins are the product of phenol and formaldehyde. (f) Coalescent binders. These include emulsion resins (insoluble) particles in water. These are two-phase systems of two immiscible liquids, where small droplets of one form the dipersoid phase in the other (continuous phase). The particle size is generally 0.1-0.5 |xm. (g) Inorganic binders. These are generally inorganic silicate polymers used with zinc dust to provide high quality coating. The inorganic silicates are prepared by fusing a mixture of sand or silica and a mixture of sodium or potassium carbonate or zinc oxide. Table 7.4 Type Oils Composition and application properties of binders and lacquers Composition Linseed oil, soya bean oil, castor oil, tall oil, coconut oil Oils + natural or synthetic resins Drying oil modified glycerol phthalate (polybasic acid + polyhydric alcohol + oil (oil does not react directly)) Glycerine + Phthalic acid + Fatty acids -> Alkyd Ester (i) Fatty acids of vegetable oils + epoxy resin -> Epoxy ester resin Advantages Economical and easy to apply Economical and easy to apply Good resistance to weather Disadvantages Poor weather and poor corrosion resistance Inadequate corrosion resistance Inadequate corrosion resistance Oleoresinous Alkyds Epoxy High corrosion resistance of 2-pack epoxy coal tar epoxy resistant to marine conditions High cost and poor resistance to weathering (ii) Principal type of epoxy resin is formed by reacting diphenyl propane (bisphenol) with epichlorohydrin (Contd) Coatings 3 95 Table 7.4 {Contd) Type Composition Advantages Disadvantages (iii) Coal tar epoxy are combination of basic epoxy resin with coal tar Polyurethane (i) Toluene di-isocyanatelinseed oil Tdi + polyester pre-polymer (isocyanate group) + hydroxy compound in oils polyesters, or polyhydric alcohols. (ii) Typical example is the reaction between TDI and polypropylene glycol to form an isocyanide terminated pre-polymer Silicone Silicone resin silicone/alkyd resin High resistance to corrosion Adequate tendency for discoloration Vinyl Acrylic Phenolic Chlorinated rubber Vinyl chloride vinyl acetate co-polymer Methyl methacrylate copolymer Phenol formaldehyde resin Chlorinated rubber + a plasticizer (made by reaction of rubber with chlorine) Good weather resistance good chemical resistance Resistance to acids, alkalies and salts Good weather resistance Good chemical resistance Good resistance to mineral oils, acids and alkalies Bituminous materials Asphalts, Coal tars Nitrocellulose Nitrocellulose + alkyd Resistance to water. Good adherence. Adequate resistance to rusting. Economical Dries fast Poor resistance to solvents Poor chemical resistance Tendency for discoloration Upper temperature limit is 80°C. Limited resistance in ultraviolet radiation Upper limit 65°C Grease Grease, oils Economical Limited resistance to solvent Only used for temporary protection 3 96 7.lO Principles of Corrosion Engineering and Corrosion Control PIGMENTS (e) Durability. This is the degree to which the pigments can withstand the effect of conditions without loss in color retention time. (f) Gloss. The pigments should provide a good glossy surface. Gloss is the degree to which a surface simulates a perfect mirror image in its capacity to reflect incident light. Pigment is a substance which is dispersed in the binder to give specific physical and chemical properties. The following are the functions of pigments: (i) Aesthetic appeal: color or special effects. (2) Surface protection: provides resistance to (3) (4) Titanium dioxide. It is manufactured in two modifications: (a) anataseand (b) rutile (Table 7.5). It is a non-inhibitive inert white pigment. It gives good opacity, tinting strength and gloss. (5) It is unaffected by mineral acids and organic acids at high temperature. White lead [2PbC03-Pb(OH)2]. It provides a tough, elastic and durable paint film. It, how(6) ever, reacts with acidic binders. It is a health hazard. Zinc oxide. It is a good absorber for ultraviolet radiation and protects the resin from break7 .10.1 S E L E C T I O N O F down. It has a good fungicidal and mildew resistance. PIGMENTS Lithophone [ZnS 30%/BaSO4 70%]. It has excellent dispensing properties in the resin. It is (a) The pigment used must be compatible with a very strong pigment. It has good opacity, the type of resin used, as the pigment selected good tinting strength, but poor resistance to must not be affected by the solvent, diluent chalking. or any ingredient of the paint. (b) Performance of the paint. A pigment must be compatible with the application of the paint. For instance, titanium dioxide is excellent (B) Black Pigments for industrial plants whereas carbon black Carbon black. Carbon blacks are prepared is not. by burning carbonaceous substances and (c) Opacity or transparency. The selection of pigment depends on the requirement of opacity. If opacity is required, black, brown Table 7.5 Properties of titanium dioxide and green pigments can be used. Transparency can be achieved with yellow and red Property Rutile Anatase iron oxide pigments. (d) Pigment blending capability. Several organic Tinting strength 1000 1200 and inorganic colored pigments can be Density 4.2 kg/dm2 3.9 kg/dm2 blended to produce different blends of colors. Color Slight cream Pure white For instance, a mixture of yellow and blue Resistance to Poor Very good color produces a green color and a red and chalking white combination produces a pink pigment. corrosion and weathering, e.g. iron oxide and chrome oxide. Corrosion inhibition: many pigments, such as chromates, act as inhibitor by helping in forming a barrier film. Film reinforcement: many particles of pigments provide toughness and strength to the paint film, e.g. silicates and oxides. Increased coverage: many pigments increase the volume of the solids present without decreasing the effectiveness of the pigment. Adhesion: many pigments, such as aluminum flakes, increase the adhesion characteristics. 7 .10.2 T Y P E S O F PIGMENTS (A) White Pigments Coatings 397 allowing flue gases to impinge on acid surface. (F) Yellow Pigments They are classified as high, medium and low color blacks depending on the inten- Lead chrornate [pigment yellow 34 - PbCrC>4, PbSC^]. The following are the important sity of the blackness. The fine particle size chromate pigments: (0.010-0.25 |xm) is used for the painting of motor cars. The coarser particle size is used • Primrose chrome (greenish-yellow lead for decorative paints. They disperse in paints sulfochromate) easily. • Lemon chrome (yellow lead sulfochroGraphite. It has a low tinting strength and gives low intensity blacks. mate) Black iron oxide [ferroso-ferric oxide (FeO• Middle chromes (reddish-yellow chromates) F^C^)]. Black iron oxide gives a coarse particle size pigment. It is a cheap and inert pigment. It is resistant to ultraviolet radiation. They have good opacity and tinting strength. Lead chromates with lead sulfate are used to provide rays of color from pale primrose to scarlet. They are economical and have inhibitive properties (C) Brown Pigments largely used in automotive coatings. Iron oxides [Fe2C>3]. These are transparent and have a low tinting strength. Yellow iron oxide [hydrated ferric oxide Calcium plumbate. It is made by heating lime with Fe203-H20]. These have excellent resistance lead oxide in a stream of air. It is resistant to to weathering and chemical attack. corrosion. It provides an excellent adhesion Monoazo yellow (2-nitro-p-toluidine coupled with and toughness. a acetoacetanilide) or (hansa) yellow. They Monoazo-brown and diazo-brown. Both the pighave high tinting strength and a good chemiments have a good resistance to corrosion. cal resistance. They are economical and stable against light. (D) Blue Pigments (G) Orange Pigments Basic lead chromate [xPbCr04, yPbO with some Ultramarine blue [Na7Al6Si6024S2]. It is resistant lead sulfate PbS04]. They are toxic like other to heat. It has a low tinting strength and low chrome pigments. They darken on exposure opacity. It has a poor resistance to acid and a to sunlight. They are resistant to alkalies. good resistance to alkali. Pthalocyanine [copper-phthalocyanine]. These have to flocculate in mixture with other pigments. They have high tinting strength, (H) Red Pigments excellent resistance to chemicals, ultraviolet radiation and weathering. These consist of pigments of mineral origin with the exception of red lead and also contain Fe2C>3 which gives the red color. (E) Green Pigments - Brunswick Greens Red iron oxides [natural and synthetic ferric oxide ( Fe 2 0 3 )] [AI2O2, Si0 2 , 2H 2 0 Fe 2 0 3 and Chrome greens. They are economical and have ^ 0 3 ] . Haematite is the red iron oxide availa high opacity. They are toxic. They may able in earth's crust. The synthetic type has become blue on exposure. They offer a good an excellent resistance to chemicals. It has heat resistance. excellent opacity and light fastness. 3 98 Principles of Corrosion Engineering and Corrosion Control Zinc phosphate [Zn(P04)2 • 4H2O]. It is highly corrosion inhibiting. It is non-toxic and is highly resistant to weathering. It provides an excellent adhesion to the intermediate coating. Diazo-redpigments [pyrazolone]. It has an excellent resistance to solvents and light fastness. (I) Special Effect Pigments Aluminum flakes [leafy and non-leafy types]. These metallic pigments provide an excellent protection to metal substrate subjected to marine environment. The difference between these two types is shown in Table 7.6. Fluorescent pigments [rhodamine]. Fluorescent dyes are made by incorporating fluorescent dyes in resins. They fluoresce by absorbing radiation in the ultraviolet range at a particular frequency and retransmitting the energy at lower frequencies. They are highly expensive and have a poor opacity. They have a poor resistance to weathering. (K) Red Lead [ Pb 3 0 4 ] It is used extensively as an inhibiting primer for steel. It is economical and resistant to atmosphere. It is a health hazard. A summary of important pigments is shown in Table 7.7. 7 .1 1 (J) Corrosion I nhibiting Pigments SOLVENT It is defined as a substance by means of which a solid is brought to a fluid state. Solvents are They prevent corrosion by providing a bar- organic compounds of low molecular weight. rier to the environments. The following is the A solvent dissolves the binder in the paint. After application a solvent is no longer required and example. should be evaporated from the paint film, however, before it evaporates, it must dissolve the Table 7.6 Difference between leafy and non- binder thus giving the paint suitable application leafy aluminum flakes properties. Solvents are two types - hydrocarbons, such as aliphatic and aromatic compounds and chemLeafy Non-leafy ical solvents, such as alcohols, ethers, ketones, (1) The leafy pigments (1) They remain esters, etc. Chemical solvents contain C, H and O flow to the surface of uniformly molecules. the coating film distributed throughout the film (2) They provide a gray (2) They give the 7 .11.1 H Y D R O C A R B O N color without luster appearance of a bright aluminum S OLVENTS surface (3) They are highly (3) They prevent the Normal paraffin, -hexane, boiling point 69°C. resistant to heating surface from the up to 1000°F effect of actinic rays of sun H H H H H H (4) They have excellent I IIIII durability and reduce H— C— C— C— C— C— C—H the transfer rate of moisture H H HH H H Coatings 3 99 Table 7.7 Color Black Summary of important pigments Pigment Iron oxide Carbon black Copper chromite Manganese dioxide Aniline black Lead chromate Cadmium sulfide Yellow iron oxide Nickel azo yellow Formula FeO, F e 2 0 3 C C uCr 2 0 4 M n0 2 Pyrazine ring PbCr0 4 -PbS0 4 CdS Fe 2 0 3 -H 2 0 Nickel complexes of p-chlor aniline coupled with 2-4 dihydroxyquinoline Na7Al6Si6024S2 KFe-Fe(CN)6 Cobalt chromium aluminate Copper phthalocyanine — Chromium sequioxide (Cr 2 0 3 ) (xPbCr0 4 , yPbO) Aluminum Silicate/Ferric Oxide Al 2 0 3 -2Si0 2 -2H 2 0-Fe 2 0 3 Synthetic ferric oxide (Fe0 3 ) Toluidine in red Yellow Blue Green Orange Red Ultramarine blue Prussian blue Cobalt blue Phthalocyanine blue Indanthrone Chromium oxide green Chrome orange Red iron oxides (pigment red 102) Cadmium red Toluidine red H,C N02 Toluidine Red White (A) Opaque (B) Transparent Titanium dioxide Zinc oxide Antimony oxide Aluminum hydrate Gypsum Mica Chalk Barium carbonate T i0 2 (Opaque) ZnO Sb203 Al(OH)3 C aS0 4 -2H 2 0 Hydrous aluminum potassium silicate C aC0 3 B aC0 3 4 00 Principles of Corrosion Engineering and Corrosion Control Hexylene glycol (HG), boiling point 192 to 200°C. OH OH Iso-paraffin, 2-methyl-pentane, boiling point 60.3°C. H H H H H H—C— C—C—C—C — H H H H H I I I II I I CH3 I I CH3 C H 3 — C — C H 2 _ CH H—C—H Cyclo-paraffin (naphthene), boiling point 81°C. cyclohexane, 7 .11.3 P R O P E R T I E S SOLVENTS / H2C H2C c OF \ CH2 CH2 The following are the main properties of solvents: (A) Solvency The more complex the polymer, the stronger is the solvent required to dissolve it. White spirit is suitable for oil and alkyd resins, and toluene is a good solvent for chlorinated rubber as it has higher solvent power. SOLVENTS 1 H2 7 .11.2 C H E M I C A L The following are the major groups: Methyl iso-butyl carbinol (MIBC), boiling point 130 to 133°C. OH CH. CH CH, CH-) CH CH. (B) Viscosity and Solid Content The higher the solvent power the less solvent is needed to achieve a particular viscosity so that solid contents remain high. (C) Ease of Application Diacetone alcohol (keto-alcohol): DAA. Boil- Solvents with a high rate of evaporation have a ing point 166°C with decomposition to acetone, tendency to form a dry mist during spray application. With brush application the evaporation O OH must be slow. CH3 C CH2 C I CH3 I CH3 (D) Drying Physical drying of the paint is slowed down by a slower evaporation rate. Rapid evaporation may lead to several defects. Ethanol, boiling point 78.3°C. CH3—CH2—OH Coatings 4 01 (E) Boiling Points a nd Evaporative Rates The boiling point is related to the rate of evaporation. As a rule, the lower the boiling point, the higher the rate of evaporation. (F) Flash Point Flash point indicates a fire risk. Paints with flash points below 23°C are considered highly flammable. (G) Maximum Acceptable Concentration (MAC) Value It indicates the danger limit which the body can accept without any hazard to the body. The concentration is given in per cubic meter of air. A lower value indicates a higher toxicity than a higher value. Following is a summary of MAC value of important solvents (Table 7.8). Their evaporation, however, added to environmental pollution. Hence, great emphasis has been laid by environmentalists to lower the volatile organic contents (VOC) of the solvents to minimize their hazardous influence on the environment. In recent years, coating formulations requiring lesser solvents or no solvents have been commercially developed. The VOC regulations imposed by environmental protection agency (EPA) have seriously restricted the choice for formulation of coatings. The solvent used must be compatible with the formulation of the coating to obtain a good quality and highly resistant film. Because several resins may be incorporated in the formulation, a proper solvent combination must be selected to improve the compatibility of the resins used in the formulation. Although the solvents evaporate, they may affect the coating by creating porosity, discoloration, poor gloss and poor adhesion. Hence, selection of solvents is of a prime importance. (H) Solvent Retention 7.12 THINNERS (DILUENTS) If any amount of solvent is retained in the coating it would cause poor adhesion and retard the They are added before the application to achieve the right viscosity for the paints. These thinners chemical resistance of the coating. must have the same solvent power as the most volatile element of the particular paint. The rate of evaporation must be compatible with the appli(I) Compatibility cation method. Table 7.9 summarizes the boiling Prior to 1960s the industrial coatings were solvent- points and flash points for important solvents and borne coatings. The resins were primarily dis- diluents. solved in acids diluted with organic solvents. These solvents were subsequently evaporated. 7.13 Table 7.8 Solvent Toluene Ethanol Butyl acetate Xylene MAC value of important solvents Flash point (°C) 6 12 25 27 MAC (ppm) 100 1000 200 100 ADDITIVES These are anti-corrosive pigments enhancers, anti-foams, anti-settling agent, anti-skidding agents, driers and preservatives, which are added in small amount to improve the properties of the paints. Examples of additives are calcite (CaC03), quartz (SiC^) and kaolin clay (Al4(OH)8 S14O10). A comparison of different types of coatings is shown in Table 7.10. 4 02 Principles of Corrosion Engineering and Corrosion Control Boiling points and flash points of solvents and diluents (thinners) Main types Boiling point or boiling range (°C) 80/100 to 140/160 187-212 160-200 160-200 Table 7.9 Class Flash point (closed cup) (°C) -20 to 27 Aliphatic hydrocarbons (1) Special boiling points spirits, free from aromatics (2) Aromatic-free mineral spirits (odorless white spirit) (3) Mineral spirits (white spirit, contains about 15% aromatics) (4) Aromatic white spirit: contains about 40% aromatics (1) Toluene (2) Xylene (3) High boiling aromatics (solvent napthas), such as Aromasol H andSolvesso 150 (1) Ethanol (2) Isopropanol (isopropyl alcohol. IPA) (3) Isobutanol (isobutyl alcohol) (4) Butanol (butyl alcohol) (1) (2) (3) (1) (2) Ethyl acetate Butyl acetate Isobutyl acetate Methyl ethyl ketone (MEK) Methyl isobutyl ketone (MIBK) 58 39 41 6 27 42 and 66 Aromatic hydrocarbons 111 144 170-200 to 195-215 Alcohols 78 82 108 118 11 127 118 80 117 135 171 12 12 27 35 -4 25 18 -4 15 40 Esters Ketones Glycol derivatives (1) Ethyl glycol (glycol monoethyl ether, cellosolve oxitol) (2) Butyl glycol (glycol monobutyl ether, butyl cellosolve, butyl oxitol) (3) Ethyl glycol acetate (cellosolve acetate, oxitol acetate) 61 156 51 7 .13.1 A D V A N C E S I N C OATINGS, W A T E R B O R N E , S O L V E N T - F R E E A ND S O L V E N T - L E S S C OATINGS terminologies related to the above coatings. A brief description of each type is given below to clear the confusion. (A) Waterborne Coatings By the rapid advancement in coating technol- The waterborne coatings gather popularity in ogy, a considerable confusion has been caused by recent years because of their avoidance of Coatings 4 03 Table 7.10 Coating type (curing mechanism) Summary of selected protective coatings Advantages Disadvantages Drying oils (oxidation) Very good application properties Very good exterior durability Outstanding wetting and penetration qualities Excellent flexibility Good film build per coat Relatively inexpensive Excellent exterior durability Low cost Excellent flexibility Excellent adhesion to most surfaces Easy to apply Very good gloss retention One-package coating - unlimited pot life Hard, durable film Very good chemical resistance Moderate cost Rapid drying Excellent durability, gloss and color retention Very good heat resistance Hard film Rapid drying and recoating Excellent chemical resistance Excellent water resistance Good gloss retention Excellent chemical and solvent resistance Excellent water resistance Very good exterior durability Hard, slick film Excellent adhesion Excellent abrasion resistance Slow drying Soft films - low abrasion resistance Poor water resistance Fair exterior gloss retention Poor chemical and solvent resistance Poor chemical and solvent resistance Fair water resistance Poor heat resistance Alkyd (oxidation) Epoxy ester (oxidation) Poor gloss retention Film yellows and chalks on aging Fair heat resistance Poor solvent resistance High cost Low film build per coat Blasted surface desirable Poor solvent resistance Poor heat resistance Low film build per coat Blasted surface desirable Two-package coating - limited pot life Curing temperature must be above 50°F Poor gloss retention Film chalks and yellows on aging Sandblasted surface desirable Fair acid resistance (Contd) Acrylic (evaporation) Chlorinated rubber (evaporation) Epoxy (polymerization) 4 04 Principles of Corrosion Engineering and Corrosion Control (Contd) Advantages Very good chemical and solvent resistance (fumes and spills) Hard, abrasion-resistant film High film build per coat Excellent adhesion, particularly to aged, intact coatings Easily topcoated after extended periods of of time with a variety of coating types May be applied directly to clean, dry concrete surfaces Disadvantages Two-package coating - limited pot life Curing temperature must be above 50° F Not recommended for immersion service (chemicals, solvents, water) Develops surface chalk on aging Table 7.10 Coating type (curing mechanism) Epoxy emulsion (evaporation, polymerization) Silicone (polymerization heat required) Excellent heat resistance Good water resistance and water repellency Must be heat-cured Very high cost Poor solvent resistance Requires blasted surface Limited chemical resistance a continuous phase and a dispersed phase health-hazardous solvents, environmental friend(discontinuous phase). liness, ease of cleaning and lack of repulsing odor. Waterborne epoxy coatings and waterborne Emulsion. This is a two-phase liquid system in which small droplets of one liquid are immispolyurethane are being used widely in industry. cible and dispersed in a second continuous Latex is a very popular waterborne coating. It liquid phase. is a stable suspension of polymeric particles in water. The water forms a continuous phase. Once Suspension. Dispersion in which the dispersed phase is a solid and the continuous phase is a applied, the water evaporates and polymer parliquid. ticles are forced together overcoming the streak and ionic forces. The particles form a contin- Foam. Dispersion of a gas in a liquid or a solid. uous film by coalescence. The particle size in Aerosol. Dispersion of solid or liquid in gaseous media (air). latex has a very strong influence on the coating properties. Suspension polymerization is a technique for preparing polymer dispersed in water. The monomer is dispersed in water in (B) Solvent-less and Solvent-free small droplets in the media and is maintained Coatings by rigorous stirring. Polymerization occurs in each monomer droplet. Some of the inter-relative Both types are based on low molecular epoxy terms for polymer in liquid media are defined resin. Solvent-free coatings, for instance epoxy below. and polyurethane, do not contain an active solvent unlike the conventional solvent-borne Dispersion. This is a stable two-phase mix- coatings (coatings in which a solvent is used to ture, comprising at least two compounds, dissolve the binder). Solvent-free coatings offer Coatings the following advantages: • • • • • Outstanding impact and wear resistance High impermeability to water Excellent resistance to immersion Excellent abrasion resistance Environmental friendly 4 05 than the organic coatings. Metallic coatings protect the substrate either by acting as a barrier between the environment and the substrate or by corroding preferentially with reference to the substrate. The following are the major methods used for applying metallic coatings: (a) (b) (c) (d) (e) (f) (g) (h) Electroplating or electrodeposition Diffusion Spraying Hot dipping Chemical vapor deposition Physical vapor deposition Welding, cladding, bonding Miscellaneous methods. ELECTROPLATING They have been successfully used in ships and oil tankers. The solvent-less coatings contain a very low solvent content (5.2 volume%). The coatings are based on low molecular epoxy resins. Health regulations now often specify their use, to avoid solvent vapors. 7 .14.1 7 .13.2 S H O P P R I M E R S WASH PRIMERS AND (A) Shop Primers (Non-designated Prefabrication Primers) Steel plates, profiles and steel products are passed through a prefabrication priming plant where steel is coated by a special primer after being cleaned in a centrifugal blast machine. The binders used in the prefabrication primer are alkyd, epoxy, zinc silicate. Iron oxide and zinc are used as pigments. Zinc is used both in organic and inorganic prefabrication primers. Iron oxide is used only in organic binders. (B) Wash Primers They are used as primers in degreased aluminum wherein the acid part would etch the surface of aluminum and increase adhesion. They are also applied as primers on a metallized surface prior (A) Vat Plating to coating. They are widely used in automobile coatings. The electro-deposition process is performed in tanks or vats having capacities up to thousands of liters. The workpiece is fixed in a jig and suspended in the electrolyte. Inert anodes which do 7.14 M E T A L L I C C O A T I N G S not dissolve in the electrolyte are suspended a few centimeters from the workpiece. The DC current The use of metallic coatings is only justified by is supplied by a transformer-rectifier (4-8 V). longer life as they are substantially more expensive Thin coatings obtained are mainly used for Metallic coatings are obtained in a conducting substrate by electro-deposition. The metal which is coated is exposed to a solution container containing a salt of the coating metal in a specially designed tank. The metal which is to be electroplated (workpiece) is made the cathode whereas the anode consists of a rod or sheet of the coating metal. The cathode is connected to the negative terminal of a DC power source and anode to the positive terminal. A specified voltage is given to the system depending on what metal is to be electro-deposited. The anode contains the rod of the metal which is to be electro-deposited, for instance, a copper rod is made the anode if copper is to be electro-deposited. If an anode of inert metal is to be used, a suitable metal salt must be added to the electrolyte. Pure metals, alloys and mixed metals can be electro-deposited (Fig. 7.1). Three methods are used - vat, selective plating and electroless plating. 4 06 Principles of Corrosion Engineering and Corrosion Control Anode rod Figure 7.1 Plating bath decorative purposes. Dense coatings are produced by vat plating. The thickness of the coating is proportional to the current density. (B) Selective Plating In this process, electro-deposition can be made on the desired localized areas without the need for masking and without immersion of components. The anode is mounted in an insulated handle and covered by an absorbent pad soaked in the electrolyte. The work is connected to the negative side of a DC power source and circuit is completed by the contact of absorbent pad with the workpiece. The process makes masking unnecessary. The deposition rate is higher than vat plating and this process enjoys the benefit of portability. The process is illustrated in Fig. 7.2. deposited on the cathode (workpiece). Pure metals, such as copper, nickel, cobalt, gold, silver, etc. can be deposited from their salts by the reduction process. For instance, the following is the reduction chemistry for deposition of nickel: Ni + + +H 2 PO^" + H 2 0 -> Catalyst ->Ni(metal)+2H + +H 2 P0 3 The process is conducted in PTFE-lined stainless steel tank at 90° C. Uniform deposits are obtained. 7 .14.2 DIFFUSION COATINGS In this method, the surface of the metal to be coated is modified by diffusing into it at a high temperature into a metal or an element, which would provide the required resistance when combined with the parent metal. Such coatings are (C) Electroless Plating called 'diffusion coatings' Diffusion coatings can be applied to a range of metals and alloys, such The process involves basically the reduction of metal ions to produce metal atoms which are as nickel, titanium and molybdenum, but the Coatings 4 07 Figure 7.2 Brush or selective plating widest use is on ferrous metals. Examples are zinc diffusion coatings and aluminum diffusion coatings. an electric arc. The resulting molten metal is blown out of the arc by an auxiliary gas stream, as droplets. The process is versatile with low capital investment. In the electrostatic spraying process, the particles released from the spray gun are electrostatically charged and propelled at low velocity by air or revolving spray head. Too much air pressure is to be avoided. This procedure produces a good wrap around without the need to positioning the workpiece. The process is suitable for tubular articles. 7 .14.3 F L A M E S P R A Y I N G The general procedure is to melt the coating material and blow it on to the surface to be coated. The coating material is in the form of very small molten particles or droplets. Four methods based on the form of the coating material are generally used: (1) The coating material is in the form of rod which is melted by an oxyacetylene flame and blown onto the surface to be coated. (2) The coating material is in the form of a powder which is heated by an oxyacetylene frame, atomized and blown onto the surface. A cross section of a typical powder flame gun is shown in Fig. 7.3. (3) The coating is in the form of a wire. The wire is fed into the central orifice of a nozzle. It passes through an oxyacetylene flame and sprayed to the metal surface. (4) The coating material in the form of a wire is heated by passing it through the plasma of 7 .14.4 H O T D I P G A L V A N I Z I N G This is a process by which iron and steel can be treated to prevent rusting. It involves dipping an article into a molten bath of zinc which reacts with the iron and forms a coating. Firstly, heavy greases, paints and lacquers are removed by hot alkali and cold alkali solutions. Surface contamination is removed by grit blasting. Scales and oxides are removed by pickling solution of H2SO4 or HC1. The object to be galvanized is immersed in a molten bath of zinc at 450°C. When immersed 4 08 Principles of Corrosion Engineering and Corrosion Control spiayed malarial Bumlr^ gases Powder and gas \ MiW'umliimliM : • ** 4 yei gases Spray gun Nozzle Figure 7.3 Cross section of a typical powder flame gun in galvanized bath, both iron and steel are immediately wetted by the zinc and react to form an iron zinc layer. At the normal galvanizing temperature, 445-465° C, the initial rate of reaction is very rapid. The main thickness is formed during this period. The normal immersion period is 1-2 min. A hot dipped galvanized coating consists mainly of two parts: an inner layer of zinc and iron in contact with the base metal, and an outer lay of unalloyed zinc. Alloying additions are made to reduce alloy layer formation and improve ductility. When the coating is entirely made of alloy layer, it is called 'gray coating' Firstly, heavy greases, paints and lacquers are removed by hot alkali and cold alkali solutions. Grit blasting is done to remove surface contamination. The oxides are removed by pickling as shown earlier. Various metals, such as aluminum and zinc, are added to the zinc bath to modify the coating. Addition of aluminum produces an even coating and addition of tin improves adhesion. Fluxing is necessary to absorb the impurities on the surface and to ensure that only clean steel comes in contact with molten zinc. In the dry process, the workpiece is first degreased in a hot alkaline solution, and rinsed with hot and cold water, respectively. After rinsing it is pickled in HC1 or H2SO4 to remove the mill scale, oxides and rust. It is rinsed again in water to remove iron salts. After degreasing, the workpiece is immersed in a tank of flux and dried. Drying can be done on hot plates or in ovens at 120°C. The dried workpiece is then immersed in the molten bath of zinc as described above. In the wet galvanizing process, the workpiece is not fluxed and dried prior to immersion in the molten zinc bath. It is directly immersed in a molten bath of zinc containing a blanket of flux on the surface of zinc. The flux cover used in the wet process is composed of zinc ammonium chloride. The workpiece is withdrawn through a flux blanket. The surface of the workpiece is wetted and cleaned as it passes through the flux into the molten bath. The dry process produces a lower quality of dross and a cleaner environment can be maintained in the plant. Less space is required for the wet process and a better surface for galvanizing is produced as the workpiece passes through the flux blanket. A microphotograph of a section of hot dipped galvanized coating in mild steel is shown in Fig. 7.4. Small quantities of 0.005% iron and aluminum reduce the rate of oxidation and improve the brightness. Coatings 7.14.5 MECHANISM OF 4 09 PROTECTION Zinc protects the steel substrate from corrosion by (a) physically protecting the steel from the atmosphere and (b) by sacrificial corrosion of zinc in the environment. Zinc has a potential of —0.763 V and iron a potential —0.44 V. Zinc is clearly anodic to steel. Zinc is oxidized to ZnO in dry air. In the presence of moisture, ZnO reacts further to form Zn(OH) 2 . Figure 7.4 Photomicrograph of a section through a hot dip galvanized coating on mild steel Zn+2H 2 0-^Zn(OH) 2 +H 2 Zn(OH)2 may react further to form a protective film of ZnCC>3 by reacting further with C0 2 . Important Considerations The corrosion rate of Zn is, therefore, retarded. Zinc coating tends to dissolve in chlo• Bath temperature = 445-465°C (830-870°F) ride and sulfur dioxide containing environment. The formation of ZnCC>3 retards the rate of • Period of immersion (generally 1-5 min) • Withdrawal rate (optimum) = 1.5m/min corrosion. When the protective film, however, dissolves (5 ft/min) zinc undergoes oxidation, • Cooling: free circulation of air • Coating thickness (ounces per square foot) Zn -> Zn 2+ + 2e • Coating weight requirements: given in ASTM Z n+2H 2 0->Zn(OH) 2 +H 2 A-123andASTMA-153. 2A1C13 + 3Fe -> 3FeCl2 + 2A1 S n->Sn 2 ++2e 2H20+02+4e^40H~ (A) Electrogalvanizing In electrogalvanizing, the furnaces, galvanizing pots and cooling tower of hot dipped processes are replaced by a series of electrolytic cells through which steel strip passes. In each of the plating cells, electrical current flows through a zinc solution from anode to cathode (workpiece). Generally soluble anodes made of zinc slabs are used. These anodes are dissolved to provide zinc which is deposited on the cathode (workpiece). Zinc is bonded to steel electrochemically and not chemically as in the hot dipping process. There is no alloy formation in the electrogalvanizing process. Continuous hot dipped galvanizing and electrogalvanizing lines are shown in Fig. 7.5. which releases two electrons. These electrons are accepted by iron which consumes the electrons (Fig. 7.6). Zinc thus becomes the anode and iron (the substrate) becomes the cathode. Hence, while zinc corrodes it protects the steel substrate which consumes the electrons and becomes the cathode. The potential of steel becomes more negative by consuming electrons and a cathodic reaction is formed on its surface. Hence, steel is cathodically protected. Thus, steel is protected by zinc mainly by (a) formation of a partially protective film of Zn on steel surface and (b) cathodically protecting the steel substance from corrosion by sacrificial protection when the zinc protective film breaks down in an aggressive environment. 4 10 Principles of Corrosion Engineering and Corrosion Control (A) CONTINUOUS HOT-DIPPED €Al¥ANiZIN<S LINE COLD ROILED STEEL OR TEMPER ROILED STEEL COATING CONTROL MINIM2E SMNGLE TEMPS* CLEAN HEAT DIP LEva m BMammumztm um TEMPER ROLLED STEEL CLEAN PICKLE ELECTROPLATINS Zn OR.2« ALLOY POST TREAT (C) CONTINUOUS PAINT LINE TEMPER ROLLED STEEL CLEAN mmTREAT OACRO METAL CURE 2INCRO METAL CURE Figure 7.5 Finishing and coatingflowline Water drop Zinc coating {anode) rngs Mmmmmmmm ^^i^^M Figure 7.6 Corrosion mechanism under a water droplet. The Zn coating prevents corrosion 7.14.6 R E S I S T A N C E OF react with CO2 to form Z11CO3 which is GALVANIZED COATINGS TO C O R R O S I O N (a) Environment. Zinc in dry air forms ZnO and in moisture reacts further to form zinc hydroxide (Zn(OH)2). It can further protective. (b) Natural water. In hard water the calcium and magnesium salts act as cathodic inhibitors and insoluble hydroxides of Ca + + and Mg + + are precipitated. The rate of corrosion in soft water is ten times higher than in hard water. Coatings Corrosion rate is reported to be lowest in a pH range of 6 to 12.5. (c) Seawater. The rate of corrosion of glavanized steel is in the range of 5-10mg/dm 2 /d in seawater. (d) Soils. In soils of poor drainage and low resistivity, the life of a galvanized coating becomes shorter compared to that with good drainage and high resistivity soils. (e) Temperature. Reversal of polarity order of iron and zinc in many hot aerated waters at 60° C or above has been observed in many instances. This causes the corrosion of iron to take place in place of zinc. Iron acts as a sacrificial material to zinc in this situation which is very undesirable. Also, water which is high in carbonates and nitrates may bring about the reversal of polarity whereas chloride and sulfate contents generally decreases the tendency of reversal. 4 11 HOT DIPPED Molten baths of aluminum for hot dipping usually contain silicon (7-11%) to retard the growth of brittle aluminum intermetallic layer. The steel strip is heated in the high temperature furnace in an oxidizing atmosphere to remove organic oils followed by heating in a reducing furnace to reduce the oxide layer. The process consists of (a) surface preparation, (b) heat treatment and (c) immersion coating with aluminum. The bath temperature is maintained between 620-710°C. The coating thickness is generally 0.025-0.75 mm. The batch method is shown in Fig. 7.7. 7 .1S.3 CALORIZED COATINGS Articles are heated in a hydrogen atmosphere in contact with a mixture of aluminum oxide and 3% aluminum chloride. After removal from 7.15 ALUMINUM the mixture the articles are heated in the range 620-710°C for 48 h. The coating thickness is genCOATINGS erally 0.025-0.75 mm. The calorized coating must The following methods are available for coating not be deformed. metals with aluminum: (1) (2) (3) (4) (5) (6) (7) Spray aluminizing. Hot dipping. Calorizing. Vacuum deposition. Electroplating. Vapor deposition. Cladding. VACUUM DEPOSITION In vacuum metallizing aluminum is evaporated at high temperature by passing a heavy current through tungsten filaments around which aluminum is wound. The operation chamber is maintained in a high vacuum. The surface is at room temperature or relatively lower temperature. The vaporized aluminum is condensed on Two types of aluminum coatings are generally the surface of the workpiece. Coatings of the order employed: (a) aluminum alloy containing 5-11% of 0.021-0.075 mm are generally applied. Sputter application is similar, however, it is carried out at Si and (b) pure aluminum. a vacuum of 13 N /m 2 at a potential of 1 kV. 7 .1S.1 SPRAYED COATINGS 7 .1S.5 ELECTROPLATED Either pure aluminum or an aluminum alloy is C O A T I N G S applied to the surface to be coated by a pistol fed by either aluminum wire or aluminum powder. Aluminum coatings can be deposited from The coating applied is generally 0.1-0.2 mm thick. organic solution if a sufficient control is 4 12 Principles of Corrosion Engineering and Corrosion Control Unloading Stand Centrifuge Afumtnum coating induction furnace Electrode salt bath furnace Vibrating Inspection Table Hopper (Crane fed) SaltCieaningj bath Overhead. Monorail Rinse Pickle; Rinse! Drying and (Storage Hopper! Figure 7.7 Automatic line for high-production aluminum coating of small parts by bath hot dip method. (By kind permission of ASM, Metals Park, Ohio, USA) maintained. Solutions of aluminum chloride, 7 .is.8 C ORROSION benzoyl chloride, nitrobenzoane and formamide R E S I S T A N C E O F A LUMINUM are used. A current density of 3.2-3.5 kA/m2 is required. Aluminum coating can also be obtained C OATING from aluminum chloride. The bath is operated at 50°C and a current density of 30 A/dm2 is (A) Atmosphere required. Aluminum coatings are more resistant to corrosion in the atmosphere than zinc coatings. The formation of an AI2O3 layer affords protection VAPOR DEPOSITION to the base metals. A film of hydrated aluminum In this process, vapors of a metal bearing com- sulfate is formed which is protective. pound are brought into contact with a heated substrate and a metal compound is deposited on the surface. Steel is exposed to dry aluminum (B) Natural Water chloride in a reducing atmosphere at 1000°C and The air formed oxide/hydroxide film is destroyed aluminum is deposited on the surface. as soon as it is immersed in water, unless the rate of reformation of film from oxygen in water is 2A1C13 + 3Fe -> 3FeCl2+2A1 more than the rate of dissolution of film by CI, NO3 and SO4 - . Soft waters are least aggressive to aluminum. The coatings become passive in a pH 7 .is.7 C LADDING range between 4 and 9 and corrode rapidly in acid and alkaline solution. Bonding of steel to aluminum is obtained by rolling, extrusion or drawing. The bonding of aluminum to steel is obtained by rolling at 540° C. (C) Seawater Aluminum alloy-base metal coated on both sides by pure aluminum is called 'Alclad' The method Aluminum coated steel is subject to pitting of application has a significant effect on the in seawater. The presence of copper in small corrosion resistance of aluminum coating. amounts causes severe microgalvanic corrosion. Coatings Pitting may be initiated by the breakdown of the oxide film at weak points in seawater or brackish water containing a high chloride content. • • 4 13 Passivation film created by immersion in H 2 Cr0 4 Oil film of a lubricant applied electrostatically (0.002 |xm). Tin plating on food cans are extremely thin (0.5 |xm). Additional protection is given by appli7.16 TIN COATINGS cation of lacquers. Can exteriors are painted for decoration and corrosion prevention. Coatings Tin coatings are widely used on non-ferrous met- based on acrylic resins or polyester is used. Inteals, such as copper, lead, cadmium and nickel. rior coatings are based on organosalt, to prevent However, tin coatings on low carbon steel are unacceptably high Sn dissolution with time, for extensively used in the food industry in tin cans. some types of foodstuffs. The base metal for tin-plating is low carbon steel. The term 'tin-plate' is reserved for a low carbon steel strip coated on both sides with a 7 .16.1 M E C H A N I S M O F thin layer of tin. In recent years, the hot dipCORROSION AND CORROSION ping process for tin plating has been replaced by electroplating because of the improved film P R O T E C T I O N O F T I N C A N S properties obtained by the latter process. The metal to be plated is in a strip form (low Tin is cathodic to low carbon steel (£° ,s ++ = carbon steel in this case) of 0.15-0.50 mm thick- 0.136V,££e = - 0.440 V). Hence, steel corrodes in ness. To improve the mechanical properties, the preference to tin. This may be true on the outside strip is a tempered roll. Medium tinning pro- of the tin can but not inside the tin can. cess plant operates on a continuous strip basis, Inside the tin can Sn + + forms complex ions each coil of steel sheet being welded to the next. with the organic liquids food. This causes reversal The steel sheets travel at a speed of 8 m/s. The of polarity and tin becomes active and corrodes steep strip passes through tanks of acid cleaning as Sn + + . The corrosion of the base metal (carbon and alkaline cleaning solutions and pickling, if steel) is thus prevented by the sacrificial action of required. The following is the typical composi- tin coating. The tin ions also inhibit the corrosion tion of a typical electrolyte of an alkaline tinning of base carbon steel by plating on the iron. Tin is both. The bath is operated at 75° C. plated on steel as Sn-Fe alloy which has a more noble potential than steel. Thus, within the food container the corrosion of steel is prevented. Sodium stannate 90 g/1 Sodium acetate 15 g/1 Tin corrodes in oxygenated solutions. Some Sodium hydroxide 8 g/1 oxygen is trapped within the tin can. However, Hydrogen peroxide 0.1 g/1 as soon as the amount of oxygen is consumed by cathodic reduction, the dissolution of tin This may be deposited from alkaline or acid solu- stops. tion. Acid tinning baths contain 70 g/1 of stannous sulfate together with substances, such as phenolAnodic Sn -> Sn 2 + +2 e sulfonic acid gelatins and phenolic compounds. Cathodic 2 H 2 0 + 0 2 + 4 e ^ 4 0 H ~ The electrolyte temperature is 50°C. Four layers are observed on a substrate of tin coated steel. • • • If, however, the amount of oxygen is very large, the magnitude of corrosion of tin would be high and hydrogen would also evolve. The Steel sheet (200-300 |xm) A thin layer of Fe-Sn intermetallic compound amount of oxygen depends upon the volume between the top and bottom contents of the tin on both sides (0.08 mm) can. If hydrogen is liberated, the tin can swells up Free tin layer (0.3 |xm) 4 14 Principles of Corrosion Engineering and Corrosion Control and indicates can failure. A consumer should not 7 . 1 7 . 2 E L E C T R O L E S S N I C K E L buy a bulged can. The presence of sulfur, copper, P L A T I N G phosphorus and silicon have a degrading effect on steel. Several can lacquers are applied to protect In contrast to electrolytic nickel coating, electhe tin can from corrosion failure. The selection troless nickel coating, is deposited without any of lacquer would, however, depend upon the food current as the name indicates. The metal is formed content in the tin can. Thick coatings of tin are by the reduction of nickel ions in solution by a applied to prevent external corrosion. reducing agent. Sodium hypophosphite is used as a reducing agent. The following is the mechanism of reduction. 7.17 NICKEL COATINGS (a) Oxidation of hypophosphite to orthophosphate in the presence of a catalyst. Nickel is one of the most important coating materials because it has the same strength as steel, ( H 2 P0 2 )~ + H 2 0 Catalyst) H + + ( HP0 3 ) = Heat but it is extremely resistant to corrosion. Nickel is most commonly applied electrolytically, but + 2 H (abs) t it can also be applied by chemical and cladding techniques. [H2 is absorbed on the catalyst surface and Several plating baths are used. Three general ( H 2 P0 2 )~ is transformed to (HP0 3 ) = .] purpose baths are commonly used: Watts, Sulfa- (b) Reduction of N i + + by adsorbed hydrogen. mate and Fluoborate. The Watts bath contains the following: N i 2+ + 2 H ^Ni+2H+ absorbed Nickel sulfate, 225-410 g/1 Nickel chloride, 30-60 g/1 Boric acid, 30-45 g/1 Temperature, 46-71°C Current density, 1-10 A/cm2 As the surface of the nickel coating is dull auxiliary brighteners, such as sodium allylsulfonate, diphenyl sulfonate, as well as salts of Zn, Co and Cd are added into the electrolyte. (c) Some hydrogen is used to reduce the amount of (H 2 P0 2 )" to form H 2 0, OH" and P. ( H 2 P 0 2 r +H ( a b s ) -> H 2 OOH" + P (d) Oxidation of hypophosphite to (HPO3 ) = and release of gaseous hydrogen. (H2P02r+H2O^H+HPO^ + H2t Complexing agents are added to prevent the oxidation of reduced nickel and to control the pH. Inhibitors are also added to prevent the decomposition of the solutions in the bath. Typical compositions of alkaline and acid baths are given in Table 7.11. 7 .17.1 C H E M I C A L D E P O S I T I O N OF N I C K E L Steel articles are plated by immersing them in a solution of nickel sulfate or nickel chloride at 70°C and at a pH of 3.5. The workpiece is immersed for five minutes, washed with a solution of sodium carbonate, and heated to 750°C. Metals, such as cobalt, palladium and aluminum are added to catalyze the reaction. A coating thickness of 0.03 mm is produced. The chemically deposited coatings have a higher resistance than electroplated coatings. 7 .17.3 PROPERTIES The corrosion resistance of electroless nickel is superior to that of electrodeposited nickel. The coating may contain 6-12% phosphorus which increases the resistance to corrosion. A uniform coating thickness is obtained by electroless nickel. Coatings Table 7.11 Typical compositions of alkaline and acid baths Alkaline bath Concentration (g/D 45 11 50 100 8.5-10 90-95°C 4 15 Contents Contents Acid bath Concentration (g/1) 21 24 28 2.2 1 4.3-4.6 190-205°C Nickel chloride Sodium hypophosphite Ammonium chloride Sodium citrate pH Temperature Nickel sulfate Sodium hypophosphite Lactic acid Propionic acid Lead pH Temperature The adhesion of electroless nickel coating to the substrate is excellent. The coatings have high strength, limited ductility and a high modulus of elasticity. Nickel coating is not resistant to caustic solutions. 7 .17.4 CORROSION NICKEL RESISTANCE OF COATINGS household appliances and tools. They are widely used in pump bodies, heat exchangers, plates, evaporator tubes, alkaline battery cases and food handling equipment. Nickel tarnishes rapidly in a corrosive environment. Nickel coatings are very useful as long as a decorative appearance is not required. However, if the nickel coatings are to maintain the decorative appearance, they are given a bright chromium overlay. Nickel offers a high degree of resistance to corrosion and oxidation in seawater. Nickel coatings are resistant to dry gases, such as carbon dioxide, hydrogen, ammonia. They are also resistant to carbon tetrachloride, oil, soaps and petrol. Nickel coating increases fatigue strength. Nickel coatings also minimize corrosion fatigue. Nickel coatings are not resistant to nitric acid and environments containing chloride. The life of nickel coatings may be further increased by a thin overlay of microcracked chromium because corrosion would not penetrate the nickel directly but spread laterally. Nickel is generally plated as a part of multilayer coating system. The outermost layer is generally 120 mV more active than the substrate. Hence, if any corrosion were to occur, it would be confined to the top layer as it would be anodic to the substrate. The coating system comprising the dull nickel, bright nickel and chromium layer is called the composite nickel coating. 7 .17.6 C H R O M I U M C O A T I N G S (E° = - 0 . 7 4 V ) As mentioned earlier, a thin overlay of chromium is applied on nickel coating to prevent the corrosion of nickel coatings. Chromium is generally applied in the form of electrodeposit. As chromium coating is resistant to corrosion and has a bright luster, it is used as a decorating coating. Chromium is more active than iron (E° = —0.76 V) and has a tendency to become passive. The electrolyte for chromium plating has the following nominal composition: • • • • • • Chromic acid, 300 g/1 Sulfuric acid, 2.6 g/1 Sodium fluorides, 0.4 g/1 Sodium fluosilicate, 0.2 g/1 Temperature, 90° C Current density, 30 A/dm2 7 .17.5 APPLICATIONS Nickel coatings find wide applications in automotive industry, such as bumpers and exhaust trims, The chromium deposits are quite porous, hence, corrodant can find easy entry. If the layer is more 4 16 Principles of Corrosion Engineering and Corrosion Control than 0.003 mm thick, microcracks appear in the Alternate (if cyanide is to be avoided): layer. In modern practice, a microcracked layer (more than 0.003 mm) is laid over a microcrack • Hydrated cadmium sulfate, 50 g free layer, which is laid on a nickel coating. • Sulfuric acid, 50 g • Gelatine, 10 g • Sulfonated naphthalene, 5 g 7 .17.7 C O R R O S I O N • Current density, 1.8 A/dm2 RESISTANCE Microcracked chromium coatings are resistant to 7 .18.1 C O R R O S I O N corrosion in the atmosphere. The coatings are not RESISTANCE suitable for use in strongly acidic environments. The chromium coatings may not be impressive on In an industrial atmosphere, a 25 |xm thick coattheir own merit but their contribution in increasing may last for one year but in a marine enviing the life of nickel, copper and other coatings is ronment, the life is significantly increased. In the significant. marine environment, the chloride and insoluble carbonates are produced which are not washed out from the surface. They provide good protec7 .17.8 A P P L I C A T I O N S tion to steel in stagnant condition, soft waters and Chromium coatings are mainly used in automo- acid, or alkaline conditions. But cadmium is a tive applications. The other important application health hazard. is in food industry. Tin coating of food cans have been replaced in several instances by chromium coatings. The steel is coated with chromium (0.008-0.01 |xm) thin and an organic top coat 7.19 CONVERSION is laid over the chromium coating. This practice increases the resistance of the outside of the C O A T I N G S containers to corrosion and increased adhesion Conversion coatings refer to the types of coating to steel. which on application convert the substrate into a compound with desirable properties. The surface so prepared provides a high degree of adhesion and corrosion resistance. Some important 7.18 C A D M I U M COATING conversion coatings are described below. Cadmium is anodic to iron (££d = — 0.403 V and E£e = — 0.440V). Cadmium can resist a humid atmosphere better than zinc, but it has 7 .19.1 P H O S P H A T E C O A T I N G S a far less protecting power than zinc. The most common method of applying cadmium coating Phosphate conversion coatings are well-known is by electroplating. Cadmium plating is done and widely used for applications on steel, zinc in cyanide baths containing a mixture of cad- and aluminum. mium oxide and sodium cyanide to produce Na2Cd(CN)4. The following is the formation: • • • • • • Cadmium oxide, 24 g Sodium cyanide, 75 g Sodium hydroxide, 15 g Triethoanolamine, 20 ml Nickel oxide, 0.2 g Current density, 5 A/dm2 7 .19.2 T Y P E S O F COATINGS PHOSPHATE Phosphate coatings are of three types: (a) zinc phosphate, (b) iron phosphates and (c) manganese chromium phosphates. Coatings 4 17 7 .19.3 P R O C E S S O F PHOSPHATING (2) Formation of an insoluble tertiary zinc phosphate from soluble primary zinc phosphate. 3Zn(H 2 P04)2->Zn3(P04)2+4H 3 P04 Phosphate coatings are generally applied by immersion or spraying (Fig. 7.8). The bath contains generally zinc phosphate in phosphoric acid (3) Formation of sludge (ferrous phosphate) by addition of NaN02and an oxidizing agent, such as nitrate. The metallic workpiece passes through the following stages. H 3 P04+Fe3(P0 4 )2 + 3HN02 -> 3FeP04 \ A spray zinc phosphate process for steel and zinc is shown in Fig. 7.9. The following is the process + 3 NO+3H 2 0 cycle: Sludge is the precipitation suspended in the (1) Alkaline cleaning with an alkali (3-8 g/1) at bath. It settles down to the bottom. 82°C. (2) Hot rinsing with water at ITC. (3) Cold rinsing with water at 25°C. (4) Acid pickling by H 2 S0 4 (15 wt%) at 60°C. 7 .19.5 I R O N P H O S P H A T E (5) Zinc phosphate bath at 25°C. C OATINGS (6) Neutralization by NaNC>2. (7) Lubrication with soap at 66°C. The basic composition is modified by addition of chemicals, like fluoride. They do not offer as 7 .19.4 M E C H A N I S M O F Z I N C good resistance as zinc phosphate coatings. The following are the reactions: P HOSPHATING (1) The metal reacts with phosphoric acid producing iron phosphate. 3 Fe+2H 3 P04^Fe3(P0 4 )2 + 3H2 t (a) 2Fe + 3 NaH 2 P0 4 -> 2FeHP0 4 + N a 3 P0 4 + ^"2 T ( b ) 4FeHP0 4 + 0 2 -> 4 F e P 0 4 | + 2 H 2 0 Exhaust Exhaust K Chrornate Rinse Water Rinse Phosphate Coat Water Rinse ~~X Clean J Top View Drying Oven i_ Load and Unload Area -X "1 Figure 7.8 Continuous conveyorized spray line for phosphating. (By permission of ASM, ASM Metals Handbook, Vol. 5, p. 446. By kind permission of ASM, Metals Park, Ohio, USA) 418 Principles of Corrosion Engineering and Corrosion Control Alkaline cfa&rter 1 minute &5PC Vaster Rinse i minute ambient. Utanatei Alkaline Geaner i minute &?€ Water Hlmz I minute ambient D emised Water mmkm mm Rinse 0 3 minute ambient I Water Rinse 1 minute ambient 2 mlnytss 55«C a •«*» Orbing and Painting 7 Figure 7.9 7.19.6 Spray zinc phosphating for steel CHROMIUM PHOSPHATE 7.19.8 CHROMATE COATINGS C ONVERSION COATINGS T he coating bath consists of a n acidic mixture of p hosphate, chromate a nd fluoride. Fluoride acts as a n a ccelerator. It is a pplied t o a luminum. T he m echanism is given below: C hromate coatings a re generally used o n a luminum as well as on zinc and certain other metals. Chromium coating is f ormed as a c onsequence of chemical attack o n a m etal when it is b rought in contact with a n a queous solution of c hlomic acid, chromium salts, hydrofluoric acid, salts, (a) A1+3HF-^A1F 3 + 3 H+ (b) C r 0 3 + 3 H + + H 3 P 0 4 - + C r P 0 4 + 3 H 2 0 phosphoric acid o r m ineral acids. Chromate ions are known for t heir inhibition properties. If a The coating is c haracterized by the formation c hromate coating is to be d eposited, t he p asofAl 2 03-2Cr04-8H 2 0. sivity o n the m etal surface i n s olution must b e b roken down so t hat t he c oating is d eposited on t he m etal surface. This is d one b y a dding ions like chlorides and sulfates which break down 7.19.7 A D V A N T A G E S O F the passivity. When t he passive film is b roken down by chloride ions, hydrogen is released which P HOSPHATING reduces t he c hromate i on to Q ^C^CrC^xIH^O, which is s paringly soluble. It is d eposited o n the P hosphate coatings have a g ood resistance t o c orrosion. They are porous and so make an attrac- m etal surface. T he c hromate treatment process tive base for a pplication of p aints a nd o rganic for aluminum is s hown i n Fig. 7.10. H owever, coatings. T he r ate of c orrosion is s ubstantially chromium is a h ealth hazard. Corrosion resistance. Chromate coatings a re decreased when paints a re a pplied o n p hosphate coatings. They are also used as p rotection against m ostly applied o n a luminum components, parcorrosion i n c ombination with sealing film of oil ticularly o n aircraft components a nd on a luor grease instead of paint. Phosphate coatings find minum objects which are to be coated with paints. application i n m ilitary equipment, cars, refriger- They a re beneficial i n m arine environments. ators, w ashing machines, toys a nd steel strips for Chromium coatings provide an non-porous bond for all paints. p ainting. C oatings 419 3 mfmsies ?S € iS Water Rinse i minute ambient Etching Oe»ier I minute 60PC Wst&r ftfnse 1 mlmite ambient Water Rinse I mfmite ambient Chromate conversion toa&t? a mlmjtes 3CFC water ffcnse l mtttute ambient Desrmsttef S minutes amdier* Aclduiatad final Figure 7.10 Immersion chromate treatment for aluminum. (By permission of oil and colour chemists association, Australia, Paints and their application, Vol. 2, Tafe Educational Books, Kensington, NSW, Australia, 2003) 7.20 ANODIZED COATINGS A nodizing is the formation of a relatively thick coating and a coherent oxide film on the surface of a metal by making the metal the anode in an electrolytic cell. The most common electrolytes used are chromic acid, sulfuric acid, phosphoric acid and mixtures of t heirs. T he following are the typical compositions of the bath for anodizing of aluminum. (a) Sulfuric Acid Bath Sulfuric acid, 10 wt% Water, 90 wt% Temperature, 18°C Duration, 15-30 min Current density, 1.2 A/dm 2 (b) Alumilite Bath Sulfuric acid, 15 wt% Water, 85 wt% Current density, 1.3 A/dm 2 A n umber of trade names have been associated with anodized coatings. Often dyes are applied on anodized coatings before the sealing process. The drying materials are generally hydroxide of copper, cobalt and aluminum. Bright anodizing is a process in which aluminum is polished followed by the protection of the shining aluminum surface with the deposition of a thin film of o xide. Bright anodizing is carried out on aluminum alloys containing some magnesium. No impurity of any kind should be present in the alloy. For bright anodizing a mixture of phosphoric acid and nitric acid containing some acetic acid is used. Reaction: ' A l ^ A l 3 + + 3 e (Anodic) " H2^2H++2e(Cathodic) 7.20.1 CORROSION R E S I S T A N C E OF ANODIZED ALUMINUM A luminum anodized coatings are resistant to corrosion between pH 4 and 8.5. The coatings are 4 20 Principles of Corrosion Engineering and Corrosion Control to 850°C for 5 min. This process is repeated until the required thickness coating is obtained. In the dry process, the object to be coated is heated to fusion temperature of the frit and the powdered frit is spread on the workpiece. The firing temperature is 700° C for iron and steel coating. Steels coated with porcelain enamels can be used up to 1000°C. suitable for outdoor applications. An optimum resistance to corrosion is obtained if the coating is in the thickness range of 18-30 |xm. They show a good resistance to deicing salts. Anodized aluminum is subject to corrosion by magnesium hydroxide, calcium hydroxide, mortar, plasters and cement. Aluminum anodized coatings find a wide use in food packaging and processing industry as well as packaging of pharmaceutical products because of the high resistance of these coatings to the food and pharmaceutical products. A large amount of anodized aluminum is used in food cans. Within a pH range 4-9, anodized coatings resist almost all inorganic chemicals, however, they are subjected to pitting in aerated chloride solutions. Anodized coatings are resistant to halogenated organic compounds, however, in the presence of water these chemicals hydrolyze to produce mineral acids which may destroy the oxide film on the surface. The corrosion resistance of anodized coatings can be further enhanced by sealing the pores of the coating and incorporating inhibitors, such as dichromate in the sealing solution. Several formulations are available for sealings. Chromium compounds should not be used for food and pharmaceutical products. 7 .21.2 C O R R O S I O N R ESISTANCE Porcelain enamels have excellent resistance to most acids and salts but are attacked by alkalies. Resistance to atmosphere is excellent. Recommended coatings for industrial plants are given in Table 7.12. 7.22 FAILURE OF PAINTS AND COATINGS A mechanistic understanding of failures of paints and coatings is important for engineers. The failures can be divided in four categories: (a) (b) (c) (d) Formulation-related failures Adhesion-related failures Application-related failures Design-related failures 7.21 GLASS FLAKE COATINGS An account of important coating failures is Flakes of chemically resistant glass 3-4 microns given below. Only selected failure categories are in thickness are dispersed in alkyl resins. TiC>2 is described here. used as pigment. A glass flake layer of 5 microns is obtained. The use of glass flakes increases the 7 .22.1 F O R M U L A T I O N R E L A T E D impermeability and erosion resistance. F AILURES 7 .21.1 PORCELAIN ENAMELS In the wet process, powdered frit (ground glass) is suspended in water together with clay, and the mix which contains 15% solids. The mix is formed (A) Chalking by melting mixtures of B2O3, Si02, AI2O3, Z r02, Na2CC>3 and PbO. For acid resistance, the pro- (a) Definition. It is the failure of a coating by the portion of Si02 in increased. The mix is sprayed formation of a powdery layer on a coating onto the metal surface to be coated. The worksurface. As the powder is white, the failure is piece is heated to dry the coating followed by firing referred to as chalking (Fig. 7.11). In this category, failures of coating due to improper formulation are described. o VH |-§o T3 PH o nve • I f» H J -H 3o u om ooo *H s O OM PH <» u_ fa u io o -i oo PH fr 3 > s ocm3 " §3 S 2 3 o pq PH PH J-i p m m 12 3 o g m .y PH O o PH '3 » -t «i-H O P LI N PH um r-s ctf D bD 5-H 'N C o o c o a 3 CO CO CO « ^ 8 < o VH 3c co fPH C Q s 3 o m .y PH C Q <N £ o8 R to ^ ow PH (N O < u o Ctf PH c o o a> V -i cro fr G '3 LO g u CO o u oli lor o .3 <D J-H u .2 Xi x2 o 's o •O i U <N 6 o PH w o Tf oH P W o ^ o a3 P-i a o (J in c^ u bD (3D .s o u u .C m O *N o fl O (N PH }-i O c 'S c o '3 1° PH mic 'B '3 '3 (^ o c est< 3 o o Vl § O PH T^ W (N C^ CO -t—> l /> a 6 o u a > P* l -H CO 8 ^ <D 8 ^ ^ 5 D +-» o fl C 8 OJ ^-H J CO ? ^o "S VH m fN 8 C Q S T3 C OJ ^ to <u CO <U w VH gHo| Z .H ^ £3 r j O ed (N ^ fi < D aJ aJ +J 2u <U o T3 . 2 S ' 0 PH O C T3 O L> <U O <U o CO PH O ON lach lectr ical quip OD b .3 c 3 r< 0> < u 0) "d .G C3 (j a t: 11 S'g .§1R CO C/5 ^H T3 c w OH O PH O o m G G O u c/5 os OH w 03 <Ti t o <-> *2 So o G (N 0) w *o S2 I O LTi O •TJH OH & C^ CO (U 03 O N G o •- G LO ^ u G '3^2 2> O OH | CO G w H G oJ o3 CO G *-l CO 03 <L> c o Q* (U GO « 3 o w 53 O O *G G U o nd o #G *N u C/3 .2 H^ IH '3 S "a> G H—» | s§ '£ S .y "d G G o 2 'o 3 ^^ OH t>JD o u J=J o OH^-H -^ on a.^ O 2 uS bO.S ' G ^H N G G O fc* G CO CD »H OH j> 'o CO +-> G G c3 o JH T> (N -*-1 03 53 T3 G T 3 rt G N <> L K a '3 | c3 J «H <> L G str G .& '3 co s u £ •£ OX)©Gs G M O O > O 03 ^ C v> u . .3 r\. 3 <D 8 <-i OHX M ^ ^ - ' : _G; « Ctf <* OH^H ^ G WGS^PHG^W I* J_» ° (N iicone acrylic ^E o3 CO micron G o o c/3 ^u < »H G O Th o <A S-i G .o y G <o u G '-1 on G " s ^H .y '6 co G CO o L 2 O CO 2 O ——G G ^ '3 . . cn p 2^ - H ^G PH 2 © .y Go g G '3 G u LO t^ 03 u bD G u< O 'N o <-0 ^-> Q o 5-t < u -5O — r-! 03 U G O LO i G O ^ '3 ^ SP.S ON G J G CO PH PH •° '§ 2> 03 CO ! 03 • M ^ H|CS o c3 CO C3 CO +-> C/3 CO PQ S 3 a» PQ 3 .3 JP ^2 G ^ ZG G O 03 u o V3 CO <> L VH G O o < * > G ^ 6 2© Go LO CO G o u u 1S 3* u PH § • S r& 3O C CO 's 1 V5 LO #u o c/i '3 ^ 03 U <D ^ G #u G i G .y NP O O ^ T3 o bD p M -G ON G o *3 8^ 2> .5 -p O 3 O 1 CO u <G CN CN CN c3 CO CN +-» rd CO </5 s u o O (N S-i 3 5> *-l m 0> <L> 3 u u CO > o 1 > o< > o CN 2 0 0 ^u ^< w ^ •v. s < u "O fl) •• VD "G CO o O LO r/5 03 o o 22 S G —y •u s g2 o o 6 G QJ Ot!9 i ^S ^ <L> G VH o , 03 s ^ £* *-< - x- co 7 1—1 I 2 XG 2" I I« g CO <L> c/5 PH PH G <u ^ G T 5 T3 OGa G 03 E 4 24 Principles of Corrosion Engineering and Corrosion Control (b) Mechanism. In erosion, it is the high winds or sand particles which hit the coating and disintegrate it. Once the coating is disintegrated and the substrate is exposed, corrosion proceeds rapidly on the substrate, particularly in a humid environment. (c) Prevention. Same as for the chalking. (C) Checking, Alligatoring and Cracking Although different in appearance, all the three failures mentioned above are related to the deterioration of the coating with time. Definitions (a) Checking. Small check-board patterns are formed on the surface of a coating as it ages. The coating becomes harder and the brittleness increases as it ages (Fig. 7.12). Figure 7.11 Chalking (Courtesy: Sigma Coatings) (b) Mechanism. The binder of the coating disintegrates due to severe sunlight and leaves the pigments which were held by it. The binder continues to disintegrate until corrosion starts to develop on the substrate and the coating surface is worn out. A pigment may interact with ultraviolet radiation and breakdown. A change in the color of a pigment indicates chalking. Previous coatings may not have satisfied the porosity of the substrate. (c) Prevention (1) Select an appropriate combination of binder and pigment on the basis of experience. (2) Select pigments with a least tendency for chalking. (3) Remove loose chalk deposits and apply a new coating system. (B) Erosion (a) Definition. It is the failure of a coating due to erosion by high winds and sand particles, mostly encountered in deserts. Figure 7.12 Checking (Courtesy: Sigma Coatings) Coatings (b) Alligatoring. It is a film rupture caused by the application of a hard dry brittle film over a more softer and extensible film. The failed surface of the coating resembles the hide of an alligator, hence the name alligatoring (Fig. 7.13). It is also called crazing. (c) Cracking. The surface of a coating cracks due to weathering and aging are similar to alligatoring, however, in this case the cracks reach the substrate and cause a more serious defect than either checking or alligatoring (Fig. 7.14). Cracking may also be caused by a fast curing rate. Mechanism The basic reason in the above failures is the introduction of stresses in the coating. In the case of checking, stresses are caused by shrinkage of the coating due to weathering, whereas in the case of alligatoring, the stress are caused by the shrinkage of the surface of the coating at a higher rate than the rate at which the body of the coating shrinks. In the case of cracking, stresses are set by continuous weather of the coating. The end result is the rupture of the coating by the stress. Prevention Alligatoring can be prevented by selecting proper formulation and allowing specified drying time between coats. The ingredients selected must 4 25 Figure 7.14 Crakling (Courtesy: Sigma Coatings) show a minimum difference in the rates of expansion and contraction so that stresses created by the above processes are minimized. Cracking can be minimized by selecting a coating which is resistant to weathering and oxidation. Remove all dust and contaminations and prime, fill and then reapply full paint system. The coating selected must also be flexible. Similar prevention is required for checking. The pigments selected for the latter must be inert and reinforcing. (D) Wrinkling Figure 7.13 Alligatoring (Courtesy: Sigma Coatings) Definition. It is the appearance of wrinkles on the surface of the coating (Fig. 7.15). 4 26 Principles of Corrosion Engineering and Corrosion Control Poor application. Porosity can result from improper application. Film thickness. If the film thickness is very high, stresses are introduced by shrinkage and the coating peels off. Examples (A) Blistering (a) Definition. This is the failure of a coating by the development of large or small round hemispherical pimples from the surface either dry or filled with a liquid (Fig. 7.16). (b) Mechanism. The blister may develop by one of the following mechanisms: (1) Absorption of moisture by the soluble pigments. When the soluble pigments absorb moisture, the solution becomes concentrated and water is pulled into all the areas where the pigment is dissolved, by the process of osmosis. Osmosis is the transfer of the liquid through the coating from a lesser concentration to a higher concentration. Blisters filled with liquid are, therefore, formed. If some soluble salts are present between the inter-coats or in the substrate, absorption by osmosis would occur and blisters would be formed. Figure 7.15 Wrinkling (Courtesy: Sigma Coatings) Mechanism. During curing, the surface is cured more rapidly than the body of the coating. Because of the difference in the rate of expansion of the surface of the coating and body of the coating, serious stresses are created in the coating. These stresses cause wrinkles. Such types of failures occur generally on oil-based coatings. Prevention. Use of slow drying solvents and mixtures of silicates to control the drying rate. Thin coating should be applied to avoid excessive thickness. 7 .22.2 ADHESION-RELATED FAILURES The failures described under this heading are due to the lack of adhesion of the coatings to the metal. The following may be the reasons for lack of adhesion. • Poor surface preparation. A good surface preparation is a primary prerequisite for coating. Any dirt, grease, rust or scale, if present, must be removed. Before the application of new coating, the old coatings must be removed completely. This mill scale must be completely removed from the surface. If it is not removed, the oxygen and moisture may penetrate under the scale and cause the surface to corrode. Once the surface corrodes, the substrate and the coating is destroyed. Loss of adhesion may not occur immediately. However, it is dependent on how severe the environment is. Figure 7.16 Blistering (Courtesy: Sigma Coatings) Coatings 4 27 sufficient degree to cause lifting of the (2) Inadequate solvent release. If the coatcoating. ing is a rapid drying type, some of the solvent may not be released and is trapped in the coating. When the temperature of the (c) Prevention. Blistering is a serious type of failure. The only practical remedy is to remove coated surface is raised, the solvent may the coating and repair it, if blistering is localcreate a significant vapor pressure and ized. If blistering occurs all over the surface, cause the formation of blisters. A simicompletely change the coating which is more lar condition may arise when a top coat expensive but provide a better adhesion. is applied over a porous undercoat. Water soluble pigments in humid environ(3) Poor coating adhesion. In the areas of ments must not be avoided. Ensure that poor adhesion, adhesion of coating to prior to application of coating the substrate substrate or adhesion of inter-coats, the is completely dry and free from any conliquids and gases maybe trapped and this taminations. Sometimes blistering may be can lead to the development of blisters. harmless, e.g. blistering of sub-sea coal tar Salt contamination on a surface may epoxy paints on oil rigs can be caused by also cause blistering. They may occur in cathodic protection but can be a harmless extremely wet environment. effect. (4) Improper film thickness. The thickness of the coating must be strictly in accordance with the specifications for a particular coating. (B) Peeling (5) Proper choice of primer. As primer is the foundation of a coating, it is impor- (a) Definition. Peeling is the loss of adhesion tant to select a proper primer according resulting in detachment and curling out of to the intended application of the coating. the paint film from the substrate or between For instance, if the coating is to be applied inter-coats. It may occur by a number of in a marine environment, an inhibitive different causes given below (Fig. 7.17). primer must be selected. For immersion A combination of one or more of the service, an impermeable primer must be following factors may lead to peeling: selected. (6) Weather. High humidity, fluctuating (1) If the tensile strength of a coating exceeds temperatures and wind speeds can lead the bond strength between coating and substrate. to failure of coatings by either creating (2) If the solvent used for one coating softens conditions inappropriate for curing or by the other coating. introducing contamination during coat(3) If corrosion proceeds under the coat. ing application. Inhibitive primers can be (4) Painting of a surface when it is wet with usefully applied to overcome the effect of rain or dew. moisture adsorption. (5) Inadequate removal of the old coat (7) Cathodic protection. If a structure is before the application of a new coat. cathodically protected, hydrogen gas may be formed in sufficient volume. It may (6) Ingress of moisture to the substrate. also be formed by stray currents. Hydrogen vapor pressure pushes the coatings up (b) Mechanism. One or more than one of the and blisters are formed. The size of the factors given above may cause a loss in adheblisters depends on the degree of adhesion of the coating. The paint film is curled sion of the coating, the salt content in from the surface. the blisters and the internal pressure of (c) Prevention. Remove the old coat completely either the liquids or gas inside the blisbefore applying the new coat. Never coat the ters. Generally, blistering leads to rust surface under wet conditions. Do not allow formation. The blister may expand to a moisture to penetrate. Minimize or eliminate 4 28 Principles of Corrosion Engineering and Corrosion Control Figure 7.18 Flaking (Courtesy: Sigma Coatings) (D) Inter-coat Delamination Figure 7.17 Peeling (a) Definition. As the name suggests, this is the loss of adhesion between two coats (Fig. 7.19). (b) Causes of delamination (1) Not using a compatible coating with the old coating during repairs. (2) Lack of removal of contamination from the substrate before applying a new coat. (3) Application of very thick coatings. (4) An increase in the rate of the curing by sunlight, such as in case of coal tar epoxy coatings. the conditions from (1) to (6) above, which cause peeling. (C) Flaking (a) Definition. Flaking is the pulling away of the coating from the substrate because of the hard and brittle nature of the coating (Fig. 7.18). (b) Mechanism. It is similar to the mechanism of (c) Mechanism. One or more of the above factors may come into play and cause delampeeling. The coating which flakes is generally ination. hard and brittle. (c) Prevention. Avoid contamination between (d) Prevention. If the delamination is localized, the coating may be repaired by sanding off coatings and do not apply very thick coatings. and recoating with a compatible paint. HowNever paint a wet surface and use coatings which are compatible. ever, when the delamination spreads over a Coatings 4 29 Figure 7.20 ings) Holidays/Misses (Courtesy: Sigma Coat- Holidays can also be caused by dust particles being incorporated during painting. Remedy. None of the areas must remain uncoated. Figure 7.19 Inter-coat Sigma Coatings) delamination (Courtesy: (B) Pinholes These are minute holes which are formed during the application of a paint, which exposes the underlying substrate. The flow of applied very large area, the best remedy is to reapply the coating all over again, after removing the paint is not continuous due either to poor mixing of two pack materials or insufficient agitation old coating. leading to insufficient wetting of substrate. Remedy. Apply coating evenly and pay proper attention to the surface. Apply coating by brush 7 .22.3 F A I L U R E D U E T O evenly, and allow to dry. Apply another coat of IMPROPER APPLICATIONS paint. (A) Holiday (C) Spattering This is a particular area in which coating has been left uncoated while the rest of the surface has been This is the defect caused when only droplets of the coated (Fig. 7.20). 'Holidays' includes pinholes. paint reach the substrate and a gap of bare metal 4 30 Principles of Corrosion Engineering and Corrosion Control exists between the droplets. The bare metal serves as the 'holiday area" Remedy. Apply paint by spray gun adjusted properly. 7.23 SELECTION OF COATINGS Selection of an appropriate coating system is a prerequisite for the durability of coating. What are the major factors that would influence the choice? There may be several factors, some of them are more important than others. Below is a list of such factors which deserve consideration: (1) (2) (3) (4) (5) (6) Cost and life to the first maintenance. Compatibility with the environment. Resistance to atmospheric pollution. Resistance to extreme climates. Ease of maintenance. Safety problem. (D) Cratering It is the formation of small bowl shaped depressions in the paint or varnish film. It is caused by the falling of impurities in the wet paint. Remedy. It can be removed by addition of antifoamers and bubble releasing agents (for solventbased paints). 7 .22.4 F A I L U R E R E L A T E D T O WELDING AND DESIGN Weld areas can cause failure of coatings by several causes. Some of the causes are given below: (a) Interference of the weld flux with the coating creates a lack of proper adhesion. (b) Splatters (droplets) of welds create cavities which expose the bare metal surface. (c) Underfilm rusting may proceed by creation of cavities. Recommended coatings for industrial plants are given in Table 7.9. 7 .23.1 COST It is the most important factor for the selection of coating. The cost is related to the importance of the structure to be coated and life of the coating system. The life of the system would, in turn, depend on several factors, such as surface preparation, quality of the coating, proper application, proper curing, maintenance and monitoring. The cost factor is important but not a deciding factor Prevention in areas of civil and military strategic importance. The first cause (flux interference) can be removed Frequency of maintenance is an important factor. by adopting a proper cleaning process. Welds are the potential holidays in underground pipelines. Weld areas are generally the first regions to show 7 .23.2 C O M P A T I B I L I T Y W I T H coating failures in a non-aqueous environment. T HE E NVIRONMENT Environment consideration is critical to the proper selection of coating system. For instance, for structures close to seawater, an inhibitive primer of proven quality needs to be considered and preferred over a vinyl primer. The coating components should be compatible with the seawater environment. For a desert environment, the system needs to withstand high wind and fine dust particles to avoid erosion and chalking of the coating. (A) Edge Failure The failure of coating on the edge is a design related failure because coating is thinner at the edge compared to the rest of the surface. An edge provides a break in the coating. The coat fails by the development of rust in the coating which penetrates to the substrate. Remedy. Apply an extra coat on the edges. Coatings 4 31 7 .23.3 R E S I S T A N C E TO ATMOSPHERIC POLLUTION The pollution problem may be caused by a large emission of C and CO2 from automobiles, in certain cases SO2 from refineries. In agricultural areas, ammonia may contaminate the atmosphere. These factors deserve consideration before the selection of coating system. to offer an excellent corrosion protection. Any defects, such as cracks and leakages caused by corrosion, may lead to explosion and fires and result in an economical and technical catastrophe. Hence, safety related issues deserve serious considerations. QUESTIONS F U N D A M E N T A L S OF COATING 7 .23.4 R E S I S T A N C E EXTREME CLIMATES TO Multiple Choice Questions 1. Mark the statement which best defines 'paint.' a) A product in a liquid form when applied to a surface forms a dry film having protective, decorative or specific technical properties b) A layer of dried paint film resulting from the application of a paint c) The formation of a tenacious, homogeneous and adherent film on a substrate d) A barrier which prevents the substrate from the environment. 2. In an inhibitive type of coating a) the pigment contained in the primer reacts with the moisture contained in the air and passivates the steel surface b) the primer forms a barrier between the steel surface and the environment c) the moisture does not penetrate to the inhibitive primer d) the water absorbed by the coating does not play any role in the inhibition 3. A zinc-rich primer protects the steel surface by Lack of consideration of climatic factors may lead to premature failure of the coating. For instance, consider three towers, one located in Malaysia (warm and humid with frequent rains), one located in Alaska (with extremely cold climate) and the last one located on the Eastern Coast of Saudi Arabia (with very hot and humid climate with suspended salts and sand particles in the air). In the first situation (warm and humid with frequent rains), the coating system needs to be highly resistant to humidity and wet conditions, and coatings based on epoxy esters binders may offer an ideal choice as these resins are highly resistant to moisture and wet conditions. In the second case, any modest coating system would help, the important point would be to ensure that the coating would not shrink and would maintain its strength in extreme temperatures below freezing. In the third instance, the coating would need to be resistant to humidity, dust particles, heat and corrosive environment. Coatings, such as epoxy esters would be suitable. 7 .23.5 E A S E OF MAINTENANCE A system which requires least maintenance is a better choice even at a high initial capital cost and compromise is to be made with the original life of the system. 7 .23.6 S A F E T Y P R O B L E M S Safety is related to the integrity of a coated structure. The coating system, therefore, needs a) forming a strong bond with the steel surface through alloy formation between zinc and steel b) acting as an impervious barrier between the steel surface and the environment c) cathodic action of zinc-rich primer with the steel surface d) forming a physical bond with the steel surface 4 32 Principles of Corrosion Engineering and Corrosion Control 4. An impervious coating protects the steel sur- 10. Which of the following is not affected by the face by choice of solvents? a) acting as an inert barrier against air, oxygen and carbon dioxide b) forming a chemical bond with the metal surface c) acting as a cathode to the steel surface d) passivating the steel surface 5. Which one of the following is not a function of the primer? a) b) c) d) Binding to metal surface Adhesion of topcoats Chemical resistance Providing a seal for the coating system a) b) c) d) Viscosity Drying speed Gloss Hiding power Conceptual Questions 1. What is the difference between 'paint and 'coating' 2. From a purely functional point of view, state the important components of a coating system? 3. State two materials from each category used for castings: a) metals b) inorganic materials c) organic materials 6. The primary function of a binder is to a) blend the pigments together into a homogeneous film b) provide a pleasant appearance to the coating c) improve the flexibility of coating d) control the thickness of the film 4. State the chronology of organic coatings starting from the raw materials to solid film formation. 5. State three characteristics of PVC coatings. 6. What improvements in the properties of 7. Chlorinated rubber is an example of paints are expected by addition of copolymer binders? a) lacquers 7. What is the main ingredient which determines b) oxygen-reactive binders the properties of the oils? c) co-reactive binders 8. What determines the drying properties of the d) condensation binder oils? 9. State three factors which influence the film 8. A pigment may reinforce a film by properties. a) forming a barrier to ultraviolet radiation 10. Differentiate between active solvents and diluents. b) increasing the thermal coefficient of expansion c) dispersing of pigment particles d) transferring heat to the localized heat M E T A L C O A T I N G S site Multiple Choice Questions 9. In general, solvents 1. Incalorizing a) convert binders into workable fluids b) distribute the pigment in the binder evenly c) are universal d) are not compatible with binders a) the steel strip is heated in a high temperature furnace in an oxidizing atmosphere b) either pure aluminum or an aluminum alloy is applied to the surface Coatings c) articles are heated in a hydrogen atmosphere in contact with a mixture of aluminum oxide and 30% aluminum chloride and after removal from the mixtures, the articles are heated in the range of620-710°C d) articles in an oxidizing atmosphere are heated in a mixture of 30% aluminum chloride and 3% aluminum oxide 2. The process represented by the equation, 2A1C13 + 3Fe -> 3FeCl2+2A1, is called a) b) c) d) vapor deposition process vacuum deposition process electroplating hot dipping d) moisture enters the tin cans 433 7. Corrosion of nickel coating in industrial atmosphere is minimized by a) bright nickel coating which acts as sacrificial coating b) applying a coating thickness of 0.8-1.5 mil c) electrodepositing a very thin layer (0.01-0.03 mil) of chromium coating on top of nickel coating d) applying a coating thickness of 1-10 mil (0.025-0.15 mm) 8. Chromium coatings are most extensively used in a) b) c) d) food industry ship industry aerospace industry steel industry 3. Aluminum coatings are a) more resistance to atmosphere than zinc coatings b) not subjected to pitting in seawater c) not satisfactory for application in soils d) not suitable for application in a marine environment 4. Zinc coatings are generally applied by a) b) c) d) hot dipping hot spraying airless spraying electroplating 9. Zinc coatings protect the steel surface mainly by a) acting as anode to the substrate b) forming an intermetallic compound with steel c) acting as a noble coating inert barrier to environment 10. The main objective of phosphate coating is to a) provide an excellent adherence of paint to steel b) protect steel from corrosion c) remove any unevenness from the surface d) provide an inert barrier for the atmosphere 5. Tin corrodes on the inside of the tin cans because a) tin becomes noble by reacting with the organic food liquids b) tin becomes cathodic to iron c) organic substance at the cathode oxidize d) food is contaminated with bacteria 6. Tin can corrode if a) oxygen is trapped within the tin can b) tin is reduced because of high overvoltage of Sn + + ions c) temperature inside food cans exceeds 50°C Conceptual Questions 1. Which types of coatings is signified by the following coating processes? a) Calorizing b) Chromizing c) Aluminizing State one method for chromizing. 4 34 Principles of Corrosion Engineering and Corrosion Control 2. What is the main difference between the dry 3. Cracking and wet galvanizing process? a) is a surface problem 3. What is the effect of the following factors on b) is less serious than either checking or galvanizing? alligatoring a) Bath temperature c) occurs as the coating ages with time b) Immersion time d) is adhesion-related c) Cooling after galvanizing d) Fluxing 4. Cratering 4. State three advantages and three disadvantages of hot dip aluminum coating. 5. What is the mechanism of corrosion of tin inside the food cans? a) is also called spatter coat b) is used by development of pin holes c) is due to the falling of dust particles in the paint when it is dry d) is also called crawling 5. Retained solvents create blistering because How and Why Questions a) the solvents make the coating solver and 1. Why calorizing, siliconizing and chromizing flexible are considered diffusion coatings? In which b) the vapor pressure of solvent is reduced industry aluminized coatings are used? c) combining of the moisture vapor with the 2. State why galvanized pipes are no longer prosolvent or dissolving into the solvent tective if water heated above 85°C is passed d) rapid evaporation of the solvent through the pipes. 3. How a microcracked chromium coating pro6. Peeling is tects a nickel coating from corrosion? 4. Why phosphating is used as a base for organic a) a design-related failure paints? b) an application failure 5. What are the advantages of using chromium c) occurs because the coating becomes very coatings over tin coatings in food cans? hard and brittle as it ages d) takes place if the topcoat is not compatible with the undercoat FAILURES OF COATINGS 7. Chalking may be controlled by Multiple Choice Questions 1. 'Chalking is a) b) c) d) an application-related failure formulation-related failure design related to failure adhesion-related failure a) using white pigments, such as titanium dioxide b) using blue pigments c) selection of a proper solvent d) controlling reduction in the thickness of the coating 8. Mud cracking may be prevented by 2. Checking a) b) c) d) is surface phenomenon penetrates through the coating is visible to the eyes is an adhesion-related failure a) b) c) d) using a proper pigment to vehicle ratio using blue pigments using highly filled water-base coating apply coatings during wet conditions that are moderate C oatings [20] [21] 435 S U G G E S T E D LITERATURE FOR R E A D I N G [22] ASM, Metals Handbook: Surface Cleaning, Finishing, and Coating. 10th ed., 5, ASM, 2000, [23] Ohio: Metals Park, USA. Surface Coatings, 1 and 2, Oil and Color Chemists Association, Tafe Educational Books, 1974, NSN: Kensington, Australia. American Society of Metals, Metals Handbook: Corrosion, 10th ed., 13, ASM, 2000, Ohio: Metals KEYWORDS Park, USA. Galvanizing Guide, 3rd ed., Zinc Development Association, London, 1971, Australia: Fundamentals of Coating Melbourne. Zinc, Its Corrosion Resistance, Australian Zinc Acrylic resin A synthetic resin formed from either Development Association, 1975. the polymerization or co-polymerization of acrylic Munger, C.G. (1999). Corrosion Prevention by monomer. Protective Coatings, 2nd revised ed. Vincent, L.D. Alkyd resin A synthetic resin made by condensation ed. NACE, Texas: Houston, USA. reaction (release of water) between a polyhydric alcohol Stoye, D. and Freitag, W. eds (1998). (glycerol, etc.) and dibasic acid (or phthalic anhydride). Paints, Coatings, and Solvents, 2nd revised ed. Binder The non-volatile portion of the vehicle of a paint. After drying, it binds the pigment particles New York: Wiley-VCH. Gainger, S. and Blunt, J. eds (1998). Engineer- together with the paint film as a whole. ing Coating Design and Application, 2nd ed. Body It is used to indicate the consistency of the paint. Brush mark Lines of unevenness left after the paint has Cambridge: Abington Publishing, England. Fedrizzi, L. and Bonora, P.L. (1997). Organic been applied by a brush. and inorganic coatings for corrosion prevention, Coverage The rate at which the paint spreads. The rate Paper from Eurocorr 96, Institute of Material, of spreading is expressed in square meters per liter. Cross-linking Establishment of chemical links between London, England. Lamboune, R. (1987). Paint and Surface the molecular chains to form a three-dimensional netCoatings: Theory and Practice. England: Ellis work of polymers. Cross-linking toughens the coating. Diluent A liquid which is used in conjunction with a Horwood. Galvanizers Association, General Galvanizing solvent to improve the film properties. Epoxide resin A resin formed by interaction between Practice, London, 2000, England. Zinc Development Association, Galvanizing epichlorohydrin and bisphenol. Filler A compound for rilling fine cracks and indentaGuide, London, 1988. tion. Satas, D. and Tracton, A.A. (2001). CoatGloss The property by which light is scattered specuings Technology Handbook. New York: Marcel larly. Dekker. Lacquer A fast drying clear or pigmented coating Bierwagen, C. P. (1987). The science of dura- which dries slowly by evaporation of the solvent, e.g. bility of organic coating. Progress Org. Coat, 15, chlorinated rubber. 179-185. Leafing A particular orientation of flaky pigments to Simpson, T.C. (1993). Electrochemical methods form a continuous sheet at the surface of the film. to monitor corrosion degradation of metallic Oil length The percentage of oil in the binder. coatings. Proc. 12th Int. Corrosion Congress, 1, Paint A product in a liquid form which, when applied Texas: Houston. to the surface, forms a continuous film having decoraAmerican Society of Metals, Surface Engineering, tive, protective or other specified properties. ASM Handbook, 6, 2000, Ohio: Metals Park, Phenolic resins A class of synthetic resin produced USA. by the condensation of a phenol with an aliphatic Vincent, L.D. (2004). The Protective Coaters aldehyde, such as formaldehyde. Pigment The insoluble solid particles dispersed in a Handbook. NACE, Texas: Houston, USA. Grainger, S. and Blunt, J. (1998). Engineering paint which gives the desired color and the properties Coatings: Design and Applications, 2nd ed. required, such as color, opacity and durability. Primer The first coat of a painting system which helps NACE. Marshall, A. (1996). Corrosion Inspectors Hand- to bind the coats applied subsequently to the substrate. Primer inhibits corrosion. Red lead is an example. book, 3rd ed. NACE, Texas: Houston, USA. Lesoto, S. et al. (1978). Paint/Coating Dictionary, Fed. Of Soc. For Coating Technology, Philadelphia, Penn. NACE, Coating Collections on CD Rom, Network Version, Texas: Houston, USA. Schweitzer, P.A. (1996). Corrosion Engineering Handbook, New York: Marcel Dekker, USA. NACE, Coa