102811batch-distill

102811batch-distill - Batch Dis*lla*on Wankat ch 9 Alembic s*lls for dis*lla*on of brandy Simple batch s*ll Mul*stage

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Batch Dis*lla*on Wankat ch. 9 Alembic s*lls for dis*lla*on of brandy Simple batch s*ll Mul*stage batch dis*lla*on Simple batch dis*lla*on Objec*ves for this lecture •  Describe uses for batch dis*lla*on •  Diagram simple batch dis*lla*on (no column) •  Obtain diﬀeren*al mass balance (Rayleigh equa*on) •  Solve for two special cases: (1) constant K; (2) constant α Recap: •  In batch dis*lla*on, the dis*llate composi*on xD is *me- dependent, requiring a diﬀeren*al mass balance. •  The design equa*on for batch dis*lla*on is called the Rayleigh equa*on: Wfinal ln = F xW ,final ∫x F dxw xD − xw •  Before we integrate, we have to ﬁnd a rela*onship between xD and xW. •  For € simple batch dis*lla*on (no column), this rela*onship is VLE: yD xD K= = xW xW WAIT! K is not constant; K = K(T) € Objec*ves for this lecture •  Evaluate the Rayleigh equa*on for (a) constant rela*ve vola*lity; (b) general case •  Uses of simple batch dis*lla*on: (a)  Solvent switching (b)  Batch steam dis*lla*on • Adding rec*ﬁca*on to batch dis*lla*on Homework this week: 9D4 and 9D6a (due Weds); 9D21 (due Fri) Friday Reading: Ch 9 Recap: •  There is no need for rec*ﬁca*on in batch dis*lla*on for solvent switching, or with batch steam dis*lla*on. •  The Rayleigh equa*on has an analy*cal solu*on for simple batch dis*lla*on if rela*ve vola*lity is assumed constant. •  If α is not constant, but we have (x,y) VLE data, we can integrate numerically, using Simpson’s rule. •  In mul*stage batch dis*lla*on, R, xD, or both, must vary. Objec*ves for this lecture •  Use McCabe- Thiele analysis for mul*stage batch dis*lla*on with constant R •  Determine batch opera*ng *me •  Use McCabe- Thiele analysis for mul*stage batch dis*lla*on with constant xD •  Eﬀect of liquid holdup •  Introduc*on to con*nuous countercurrent extrac*on Mul*stage batch dis*lla*on Constant reﬂux ra*o (variable xD) Constant dis*llate composi*on (variable R) 1 1 •xD 0.9 • 0.8 0.7 • 0.6 y(MeOH) 0.6 y(MeOH) 0.8 • 0.7 • 0.5 0.4 0.5 0.4 0.3 0.3 0.2 dis*lla*on must end when xD,avg = xF 0.1 0 •xD 0.9 0 0.1 0.2 0.3 0.4 0.5 x(MeOH) 0.6 0.7 0.8 0.9 dis*lla*on must end when R = ∞ (L/V = 1) 0.2 0.1 1 0 • 0 0.1 0.2 0.3 0.4 0.5 x(MeOH) 0.6 0.7 0.8 0.9 1 Mul*stage batch dis*lla*on with constant R Given F, xF, xW,ﬁnal, and N, ﬁnd Dtotal, xD,avg 1 4. Use Simpson’s rule to integrate Rayleigh equa*on, then calculate Wﬁnal 5. Solve mass balance for Dtotal and xD,avg y(MeOH) 1. For an arbitrary set of xD values, draw a series of parallel opera*ng lines, each with slope R/(R+1) 2. Step oﬀ N stages on each opera*ng line to ﬁnd its corresponding xW 3. Plot y=1/(xD- xW) vs x=xW; read y- values at x = xF, xW,ﬁnal and (xF+xW,ﬁnal)/2 Assume N = 2 (incl. reboiler) 0.9 0.8 0.7 1 • 2 1 • • 0.6 0.5 2 • 1 • 0.4 0.3 1 • 2• •xD,1 •xD,2 •xD,3 •xD,4 2• 0.2 0.1 xW,4 xW,3 xW,2 0 0 0.1 0.2 0.3 xW,1 0.4 0.5 x(MeOH) 0.6 If xD,avg is speciﬁed instead of xW,ﬁnal: guess xW,ﬁnal, calculate xD,avg, iterate. 0.7 0.8 0.9 1 Opera*ng *me at constant R (D) tbatch = toperating + tdown depends on vapor ﬂow rate (V), which depends on boilup rate € shut down, cleaning and recharging s*ll pot, restart • if the boilup rate is constant, then V is constant, so D is constant toperating = Dtotal F − Wfinal F − Wfinal = = D V −L V (1 − L / V ) 1 1− L / V = R +1 € toperating = € D= € V R +1 € Dmax = Vmax R +1 (R + 1) F − Wfinal V ( ) • V = Vmax when vapor velocity u = uﬂood • uﬂood depends on column diameter • typically operate at D = 0.75 Dmax Calcula*ng column diameter We want to use the smallest diameter that will not ﬂood. uflood ⎛ྎ σ ⎞ྏ0.2 ρ − ρ L V = Csb,flood ⎜ྎ ⎟ྏ ρV ⎝ྎ 20 ⎠ྏ where σ is surface tension, ρL and ρV and liquid and vapor densi*es, respec*vely. € Csb,ﬂood is the capacity factor, depends on ﬂow parameter FP and tray spacing; obtain from graphical correla*on (e.g., Figure 10- 16 in Wankat 3rd Ed) diameter (feet ) = ( 4V MWv ) πηρVuop (3600) where η is the frac*on of the column cross- sec*onal area available for vapor ﬂow (i.e., column cross- sec*onal area minus downcomer area). € Mul*stage batch dis*lla*on with constant xD Given F, xF, xD (maybe) xw,ﬁnal and N, ﬁnd Rini*al, Rmin, xW,min 1. Draw trial op. lines and step oﬀ N stages to end at xF 3.Find xW,min using (L/V) = 1. Verify xW,ﬁnal > xW,min N = ∞, R = Rmin 0.9 0.8 • 0.7 • /V) min (L •xD • 0.6 y(MeOH) This is trial- and- error except for N = 2 and N = ∞ (Rmin) 1 0.5 0.4 0.3 4. Solve mass balance for Wﬁnal and Dtotal. 0.2 0.1 0 N = 2 (incl. reboiler) xW.min xF 0 0.1 0.2 0.3 0.4 0.5 x(MeOH) 0.6 0.7 0.8 0.9 1 Opera*ng *me with constant xD mass balance: W =F xD − xF xD − xW dW dD − = = V − L = V (1 − L / V ) dt dt dW xD − xF dxw =F dt € x − x 2 dt D w ( ) toperating = F € xD − xF € V xF ∫x W ,final dxW ( (1 − L / V ) xD − xw ) 2 1. Draw a series of arbitrary opera*ng lines, each with a diﬀerent slope L/V 2. Step oﬀ N stages on each opera*ng line to ﬁnd its corresponding xW € 3. Plot y=1/[(1- L/V)(xD- xW)2] vs x=xW; read y- values at x = xF, xW,ﬁnal and (xF+xW,ﬁnal)/2 4. Use Simpson’s rule to evaluate integral, then calculate topera*ng Eﬀect of liquid holdup on the column • usually we can assume vapor holdup is negligible • liquid holdup causes xw to be lower than it would be in the absence of holdup • causes the degree of separa*on to decrease To assess the eﬀect on dis*lla*on • measure the amount of holdup at total reﬂux • perform computa*onal simula*on ...
View Full Document

This note was uploaded on 12/29/2011 for the course CHE 128 taught by Professor Scott,s during the Fall '08 term at UCSB.

Ask a homework question - tutors are online