{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

nondest.sterility.testing

nondest.sterility.testing - NON-DESTRUCTIVE ON-LINE...

Info icon This preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: NON-DESTRUCTIVE ON-LINE STERILITY TESTING OF LONG-SHELF-LIFE ASEPTICALLY PACKAGED FOOD PRODUCTS BY IMPEDANCE MEASUREMENTS Stoyan N. Nihtianov Electronics Research Laboratory Delft University of Technology Mekelweg 4, 2628 CD Delft The Netherlands Phone: H31 15 278 6174 E-mail: [email protected] Abstract - This paper presents the results obtained in The Electronics Research Laboratory of Delft University of Technology, The Netherlands, with the application of the impedance measurement technique for the needs of non-invasive sterility tests. Two pilot food containers have been chosen: ( 1) Tetra Brik Aseptic (TBA) cartons and (2) Stork Food 8. Dairy Systems (NL) plastic battles. The main characteristics of the TBA cartons are - unstable shape, tolerances in size, packaging laminate comprising paper layer, plastic layers and aluminum foil. The characteristics of the Stark bottles are — relatively stable shape and relatively thick walls. An electric model has been developed for the measured impedance. A relation is derived between the resistive component of the measured impedance and the conductivity of the food in TBA cartons. It has been also demonstrated that with the impedance technique it is possible to acquire information about some important properties of the food container during the sterility testing procedure. The microbiological tests with a number of bacterial species proved the potential of this new non-invasive sterility testing technique. I. INTRODUCTION Non-invasive sterility testing of packaged food products is becoming one of the most dynamic branches in the field of food science and food technology. This new way of testing is expected to improve the safety of each consumer and to save an enormous amount of ready-for-use food and packaging material, which are currently sacrificed by producers for the need of destructive sterility testing. [1][2][3] The automated on-line non—invasive sterility testing can be applied shortly after production, but it Gerard C. M. Mcijer Electronics Research Laboratory Deift University of Technology Mekelweg 4, 2628 CD The Netherlands Phone: ++31 15 278 6174, E-mail: [email protected] also can be carried out in a fast way, immediately before consumption, which could be of interest when a big amount of packaged sterile food has to be used after long period of storage. The non~invasive sterility testing is a much more sophisticated procedure than the current invasive testing. We have to derive information about the sterility of the packaged food without opening the food container. This brings two main difficulties: (1) all the peculiarities of the package should be taken into account, (2) only physical effects that accompany bacterial growth and that can easily be detected from outside the food container can be used for the needs of sterility testing. In addition, the results must be independent of the alterations of the test conditions and the testing method must be applicable in an industrial environment. For 100% testing the speed of the testing procedure has to be compatible with the speed at which the packages are produced. Finally, no changes of the packaging technology, which could worsen the reliability and the productivity and increase the production costs, are acceptable. In [4] is presented a brief overview of some methods for non-invasive sterility testing — ultrasound imaging method, ultrasound Doppler method, calorimetric method and volumetric method. Crucial factors for their minor industrial applicability is their requirement for considerable changes of the packaging technology (the ultrasonic methods) or the long-lasting testing procedure (the calorimetric and the volumetric method). After a couple of years of intensive research work we can conclude that the impedance method has the potential to meet to the greatest extent the requirements for intensive on—line non-invasive sterility This work is supported by the Dutch Foundation for Technical Sciences (STW), Project DEL.4369. O—7803-5432—X/99/$10.00 © 1999 IEEE 243 testing of liquid foods in a number of different types of food containers. Two pilot food containers were chosen: (1) Tetra Brik Aseptic (TBA) cartons and (2) Stork Food & Dairy Systems (NL) plastic bottles. The main characteristics of the TBA cartons are — unstable shape, tolerances in size, multiple-layered packaging material. The characteristics of the Stork bottles are — relatively stable shape and relatively thick plastic walls (Fig.1). The impedance method is based on the fact that TBA cartons Plastic bottles Fig.1. Pilot food containers for non-destructive sterility testing. microbial metabolism normally causes an increase in both conductivity and permitivity of the nutritious media and therefore, as microorganisms grow, the impedance of the media is tending to decrease. We have to point out that not all microorganisms cause big enough changes of the electrical properties of the food, so that we can reliably detect them. The specificity of the impedance method is often compared with that of the pH sterility testing technique. Virtually, always a change of the pH of the food is accompanied with a change of its electrical properties. The usual implementation of the impedance method requires direct contact between the measuring electrodes and the media under test, which is not possible in the case of non-invasive sterility testing. One practical solution is to introduce electrodes in the container, which is done with the TBA cartons, or to use external electrodes, which is the case with Stork bottles. The sterility of the packaged food, even when guaranteed by a 100% sterility-testing, cannot ensure a long shelf life. What additionaliy is needed is a high- quality food container. It must be tight enough to prevent food spoilage caused by interaction with the surrounding non-sterile environment. To ensure this, an inspection of the tightness of the food container is carried out. We shall show here that with the impedance technique it is possible during the sterility testing procedure to acquire also information about some important properties of the food container itself. which directly influence its quality. This is achievable, because with the impedance measurements we always have two output components (resistive and reactive, or module and phase, or quality factor and resonance frequency) and both of them can be informative. In some cases even more than two useful outputs may be available with a multi-frequency impedance measurement of a singie object. ll. STERILITY TESTING OF TBA CARTONS The building laminate of the TBA food container comprises a paper layer, plastic layers and an aluminum foil. The aluminium foil is a very good conductor and prevents externally created electric fields to enter the container. That means that with external electrodes we cannot sense the conductivity of the contained iiquid food. To solve this problem, a small conductive strip has been fixed around the edge of the packaging material. After building the carton, a part of this strip remains inside the container in galvanic contact with the food and the other part goes on the outside wall, where it is easily accessible (see Fig.2). Figure 3 shows a cross-section of a TBA carton with electric field (current) lines and potential zones Fig.2 Inside and outside view of TBA carton with inserted conductive strip. displayed, when voltage is applied by the impedance measurement equipment between the small electrode (the part of the conductive strip, which is in galvanic contact with the food) and the aluminum foil. 244 The electric field extends inward the container rather than along the wail. This is valid for a very wide range of the conductivity sand the relative dielectric constant 5' of the liquid food. The potential drop associated with the current flow mainly occurs in the vicinity of the External insulating material Polarization effect (CW) Impedance measurement eq uipment Small Electrode Fig.3. A cross-section of a TBA food container with electric field (current) lines and potential zones displayed when voltage is applied by the impedance measurement equipment between the email electrode, fixed on the inner surface of the container wall, and the aluminum layer. small electrode. The current through the inner thermoplastic layer has its highest density in the immediate neighborhood of the small electrode. With increase of the distance from the electrode the current density drops rapidiy. These are the reasons why this method appears to be quite insensitive to the variations in shape of the container, especially when they take place far away from the small electrode. At the same time the measured impedance is influenced by the conductivity of a large area of the food. Figure 4 presents the equivalent electric circuit of the measured impedance in Fig.3. The value of the capacitance Cram depends on the dielectric constant 5’ of the food and the measurement setting. Our experiments proved that for various liquid food products its value is only a few pico-farads and its influence on the measured impedance cannot be sensed at frequencies below 1 MHz.[5] The value of the parasitic shunt capacitance CE, (see Fig.4), created between the smalt electrode and the aluminum foil through the thermoplastic material, depends on the surface of the conductive strip. For 245 surface below 100 mm2 the value of CD is comparable with that of 0,000.. If we neglect the influence of Cm, and CE, and we use Cfood EThemopIastic E ‘ material ‘ Cp, RP, Small electrode Fig.4. Equivalent electric circuit of the measured impedance, where: C90; and Rm; represent the polarization effect at the surface of the small electrode, which is in galvanic contact with the liquid; Cpl and RP; represent the impedance of the plastic material between the Al-foil and the liquid; Cfood represents the dielectric properties of the food; Rfoad is the bulk resistance of the liquid; Cp is a shunt capacitance. a simplified serial model for the measured impedance 2;, its resistive component R; equals the sum of Rpm, Rm and RE”. The value of all of them is dependent on the conductivity 0' of the food [4], which gives rise to the sensitivity of this method. The dependence of Rn, on c is less obvious than that of Rm, and Rfm. it can be explained by the spatial redistribution of the current density within the carton when the conductivity of the contained food varies. The capacitive component Cgof the measured impedance is approximately equal to CE”, because CPD, >> Cpl. lll. QUALITY TESTING OF TBA CARTONS The inspection of the tightness of TBA cartons is principally focused on the innermost thermoplastic layer of the multiple-layered packaging laminate. Any possible disintegration of this layer (local leakage) may entail the contents reaching the barrier layer (aluminum foil) or the fibre layer, in which event the other barrier properties of the laminate are lost, even if no actual liquid leakage occurs through the laminate. Locai leakage causes redistribution of the current flow in the packaged food. This effect is very well expressed in the low frequency range, at which a comparatively low-impedance “tunnel" is created between the small electrode and the leakage area. Figure 5 shows the current density in different areas of the carton and the effect of a local leakage on its distribution. To model this effect we add an additional "leakage" branch (RE, Rwy, par) to the equivalent electric circuit in Fig.4 (see Fig.6). For simplicity, the parasitic components C,J and Cfood are not depicted in Fig.6. Their influence on the value of the measured impedance in the low frequency range is negligible. For a great variety of liquid food products, the polarization capacitance Cpa,’ has a much higher value External insulating material Inner thermoplastic layer Aluminium layer Polarization ff 9 act of the leakage area (Cpm') Leakage current (f) path Polarization effect (Opal ) impedance measurement equipment Small Electrode Fig.5. The current density distribution (with different intensity of the grey color) in TBA carton with a low impedance "tunnel" caused by a local leakage. than Op, in the frequency range up to 10 kHz. This is true even for a very small surface of the leakage area, so that the following correlation is valid: [(RfiRpai’)l(Rrood+Rp;)]<(OWE/Cm). It can easily be electrode Fig.6. Equivalent electric circuit of the measured impedance from Fig.3 with additional “leakage” branch, comprising: R; - resistance of the food to the current i’, which flows through the leakage branch; Cm ‘ and Roof- representing the polarization effect at the surface of the local leakage area. shown that, in this case, with decrease in frequency there is an increase in the relative value i’li of the current through the leakage branch (R,, Rpm: Opal). That is why at low frequencies this branch has greater impact on the measured impedance. At the same time the resistance of the leakage branch R, has always higher value than the resistance Rfood. As a result, in the case of local leakage, both the equivalent resistance R; and the equivalent capacitance C; of the measured total impedance between the small electrode and the aluminum foil rise in value. At higher frequencies the influence of the capacitors on the measured impedance drops and with this the influence of the local leakage branch becomes much smaller, because (RyaofiRpor')>>(RfoaJl-Rp,). An additional contribution to this effect is the frequency dependence of the polarization capacitance Cpoi’, which decreases with increase of frequency. In fact, at frequencies above 100 kHz, at which the sterility testing is carried out, the effect of local leakage is already negligible. IV. STERILITY TESTING OF PLASTIC BOTTLES To sense the conductivity of the food in Stork plastic bottles we don’t have to introduce internal electrode, because there are no conductive layers in the laminate. Figure 7 shows a Stork bottle, blow—moulded from high-density polyethyiene, and the electrodes that were used to sense the conductivity changes of the food. The best measurement configuration appeared to be with two large surface electrodes pressing the bottle Seated capsule Electrode 1 Electrode 2 Fig.7. Stork piastic bottle with two external electrodes for non—destructive sterility testing. from outside (see Fig.1 and Fig.7). Figure 8 shows the equivalent electric circuit of the measured impedance between electrodes 1 and 2 (Fig.7). The two electrodes are insulated from the food by a polyethylene layer. Its influence on the measurement result can be expressed by the capacitors C1 and C; and the associated to them loss resistance presented 246 by RC1 and R92. This loss resistance is added to the resistance of the food Rmd and its variation from bottle Cfcoo' Electrode 1 Electrode 2 Fig.8. Equivalent electric circuit of the impedance measured between the two electrodes from F ig.7. to bottie deteriorates the sensitivity. The larger the surface of the electrodes, the larger the value of the capacitors and the smaller the loss resistance, associated with them. The larger the distance between the electrodes, the larger the resistive component of the measured impedance for a given conductivity of the food, which ensures higher sensitivity. V. EXPERIMENTAL RESULTS Experimental results with TBA cartons The impedance measurements were carried out with HP 4194A impedance Analyzer. Pilot 1—litre TBA cartons (see Fig.1) were used with saline in them. The conductivity of the saline was varied within the range from 0.05 S/m to 2.5 S/m. To assess the sensitivity of the described sterility testing method, we used a reference method to measure conductivity of fluids, described in [6]. The relation between the conductivity 0' and the measured resistance R, when applying this method, is: R=1l4nar=0.282f(a.\/S), where r is the radius and S = 47:12 is the surface of a spherical electrode. To define the relation RAG), which best fits our experimental results, we used polynomial curve fitting. With the least—squares method a second-order relation was generated that repeats the experimental results with maximum tolerance of i1 %, according to the equation (1) — A — B + C J30 J30 2 In equation (1) A, B and C are parameters for a given frequency. The parameter C represents the loss resistance of the thermoplastic layer Rm (see Fig.4), when the conductivity of the food a has an infinite value. The surface S of the small electrode was 40 mmz. The parameters A, B and C change their values with frequency, which makes the sensitivity of the method frequency dependent. Table 1 presents the values of the parameters A, B and C for a few frequencies, sensitivity. From Table 1 it becomes clear that with the presented method we have higher sensitivity to the conductivity changes of the food (expressed by A), when compared to the reference method. The price we pay for this is a small non-linearity introduced by the parameter B and a very small offset (parameter C). In the range of the measured conductivity the maximum non-linearity was less than 2%. at which we obtained the highest TABLE 1 —-—-- ——m—— We tested the sensitivity of the method to variations of the carton's shape by simulating large deviations from its normal shape with UHT-mllk inside. When these variations were at distance larger than 30 mm from the small electrode, the measurement results altered with less than 1%. We also tested how a local internal leakage through the inner thermoplastic layer of 1 liter TBA carton, filied once with tap water and then with UHT—milk, changes the values of RE and 20;. The surface of the small electrode was 70 mmz. Two types of leakage were created a pinhole with a surface of less than 12 mm2 and another leakage area with a surface 70 mm. The two types of leakage were created once 5 mm and then 90 mm far from the small electrode. Figure 9 shows the frequency dependence of RE, C; and (p; = arctg(—1/(R,; .CS)) of the measured impedance 2; when the carton was filled with water. From the experimental results we can conclude that at low frequency the measured capacitance Cg is largely dependent on the surface of the leakage area and only slightly dependent on the distance to the small electrode. The measured resistance is both dependent on the surface of the leakage area and on the distance to the small electrode. From the combined information about R2 and C; we may have an idea about both the surface of the leakage area and how far away it is from the email electrode. The phase shift (p; of the measured impedance combines both the sensitivity of R; and C; to leakage, and that is why it can be considered the best parameter to detect leakage within very wide range of its size and location. To use the measured values of R): and C; as criteria for the existence of leakage in the package, we need information about the conductivity of the food. This information can be obtained during the sterility testing 247 of the food, which takes place at higher frequency. at which the effect of local leakage is negligible. 100000 100 0.1 0.2 0.5 'i 2 5 10 20 50 100 200 500 Frequency, kHz 10000 1000 W 100 10 1 0.1 0.2 0.5 1 o 50 100 200 500 Frequency 1E'z)(kl?l b) Frequency, kHz 0 , 400.1 5 0.2 0.5 ‘l 2 10 20 ‘_..,......-i! .9“ -1 00 (P2 I deg Fig.9. Dependence of R; (a), C; (b) and (p; (c) on the presence of local leakage in a TBA carton versus frequency. Five different situations have been created. no leakage (1); a pinhole leakage at distance 5 2mm (2) and 90 mm (3) from the small electrode; a 70 mm2 leakage at distance 5 mm (4) and 90 mm (5) from the small electrode. Figure 10 shows the relative sensitivity of (p; to local leakage and the relative sensitivity of R40) versus frequency. it can be seen that at frequencies higher 248 than 100 kHz the presence of local leakage will not affect the sterility testing res...
View Full Document

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern