xB1 e=0.992

xB2 W=0.999

It may be useful at this point to locate on the ternary diagram several points to get a preliminary feel for the design problem that we are facing. The feed has a composition 0/84/16 mol% B/E/W (benzene/ethanol/water). So the F point is located on the ordinate in region labeled region 1 in Figure 5.17c. One of the desired products is very pure water, which is located at the bottom left corner in region 1. However, the other desired product is very pure ethanol, which is located at the top corner in region 2. This is on the other side of a distillation boundary, so the separation cannot be achieved in a single simple distillation column. The decanter and the second column are added so that the distillation boundary can be crossed.

Now that the complexity of the VLLE is apparent, let us develop a simulation of a flowsheet to produce high-purity ethanol and water. The flowsheet will have two distillation columns and a decanter. There are two recycle streams back to the first column: organic phase from the decanter and distillate from the second column. A simple conceptual flowsheet is given in Figure 5.18.

5.2.2 Process Flowsheet Simulation

The first column does not have a condenser, so the appropriate "stripper" icon is selected from the many possible types under RadFrac, as shown in Figure 5.19. The three streams fed to this column are added with control valves. The stream Feed is specified to be 0.06 kmol/s with a composition of 84 mol% ethanol. The other two streams fed to column C1 are unknown. We must make some reasonable guesses of what the flowrates and the compositions of the organic reflux and the recycle from the top of the second column will be.

One way to estimate these compositions is to recognize that the overhead vapor from the first column will have a composition that is close to that of the ternary azeotrope: 53.06/27.49/19.45 mol% benzene/ethanol/water (B/E/W). We set up a simulation

Ternary Azeotrope Ethanol

Figure 5.19 Selecting a stripper column.

Fnr Hpln nrpQÇ F1

Figure 5.19 Selecting a stripper column.

with a stream with this composition feeding a decanter operating at 313 K. The predicted compositions of the organic and aqueous liquid phases are tabulated as follows:

Composition (mol%)




Organic Aqueous

84.35 7.24

14.14 47.04

1.51 45.72

The composition of the organic reflux should be close to the composition of the organic liquid phase. The composition of the feed to the second column should be close to that of the aqueous liquid phase. Since essentially all the water in the feed comes out the bottom of the second column at a high purity, the amount of water removed from the feed is only (0.06 kmol/s)(0.16) = 0.0096 kmol of water/s (per second). Therefore, as a first estimate, we can use the composition of the aqueous liquid phase for a guess of the recycle composition.

The next issue is guessing the flowrates of the reflux and the recycle. One brute-force way to do this is to guess a recycle flowrate and then find the flowrate of organic reflux to column C1 that is required to keep water from leaving in the bottoms. When this is achieved, the resulting aqueous phase is fed to the second column, and the calculated distillate D2 is compared with the guessed value of recycle (both in flowrate and composition). The composition of the organic phase from the decanter is also compared to the guessed composition. Compositions are adjusted and a new guess of the recycle is made.

É1--Q1 Components ±1 J/J Properties ±1 J/) Flowsheet

^Specifications] Flash Options | Utility |

É1--Q1 Components ±1 J/J Properties ±1 J/) Flowsheet

^Specifications] Flash Options | Utility |

-Flash specifications m-M streams

| O Utilities

j C2

-Valid phases-

-Flash specifications

-Valid phases-


o Hcurves

Ö Dynamic

Figure 5.20 Heat exchanger specifications.

The simulation of the first column is extremely tricky, as we demonstrate below. A very small change in the organic reflux can produce a drastic change in the product compositions. Multiple steady states also occur; the same reflux flowrate can give two different column profiles and product compositions.

The number of stages in column C1 is set at 31. Note that since there is no condenser, the top tray is stage 1. Organic reflux is fed at the top. Recycle is fed at stage 10. Fresh feed is fed at stage 15. Column pressure is set at 2 atm because we need a control valve on the overhead vapor line.

The vapor from column C1 goes through a valve V12 and to a heat exchanger HX. The conditions specified in the HX block on the Input item (Fig. 5.20) are the exit temperature of 313 K and a 0.1 atm pressure drop (entering a negative number for Pressure means a pressure drop).

A decanter is then inserted on the flowsheet by clicking on Separator at the bottom of the window and selecting Decanter, as shown in Figure 5.21a. The operation of the decanter is specified by clicking the Input item under the Decanter block. The pressure is set at 1 atm and adiabatic operation is selected (heat duty is zero as shown in Fig. 5.21b). Under the item Key components to identify 2nd liquid phase, the benzene component is specified by moving benzene into the right Key components window (see Fig. 5.21b).

A very small amount of benzene will be lost in the two product streams, so a small makeup stream of fresh benzene is added to the organic phase from the decanter before it is fed to the first column as reflux.

Finally, a second column C2 is added in the normal way. A 22-stage column is specified with feed on stage 11 and operating at 1 atm. The final flowsheet with all the pumps and valve installed is shown in Figure 5.22.

5.2.3 Converging the Flowsheet

The fresh feed is 0.06 kmol/s with a composition of 84 mol% ethanol and 16 mol% water. Essentially all the ethanol must come out in the bottoms B1 from the first column. So in the setup of this column, a bottoms flowrate is fixed at (0.06)(0.84) = 0.0504 kmol/s. This column has only 1 degree of freedom because it has no condenser or internal reflux. The organic reflux will eventually be adjusted to achieve the desired purity of the ethanol bottoms product (99.92 mol% ethanol). Note that both benzene and water can appear in the bottoms as impurities.

Figure 5.21 (a) Inserting a decanter; (b) specifying conditions in the decanter.

Likewise, essentially all the water must come out in the bottoms B2 of the second column. So in the setup of this column, the bottoms is fixed at (0.06)(0.14) = 0.0096 kmol/s. Initially the reflux ratio is fixed at 2 as the other degree of freedom. This will be adjusted later to achieve the desired purity of the water product (99.9 mol% water).

The first guesses of the compositions of the recycle and reflux are inserted in the Input of these streams. First guesses of reflux and recycle flowrates are made of 0.12 and 0.06 kmol/s, respectively. The simulation is run giving a bottoms composition of 21 mol% benzene and 7 x 10"4 mol% water in the first column. The water is driven overhead, but there is too much benzene in the bottoms because the organic reflux flowrate is too large. Reflux flowrate is reduced from 0.12 to 0.1 to 0.09 kmol/s (as shown in Table 5.2), which reduces the benzene impurity in the bottoms. However, when the reflux flowrate is reduced to 0.08 kmol/s, there is a drastic change in the bottoms

Brevery Flowsheet
Figure 5.22 Full flowsheet.

composition. Now the water is not driven out in the overhead. It comes out in the bottoms because there is not enough benzene to entrain the water overhead.

Now, if the reflux is increased back to 0.09 kmol/s, the column does not converge to the same steady state that it had previously at this flowrate. The flowrate must be increased to about 0.11 kmol/s to reestablish the desired low water content in the bottoms. This multiple steady-state phenomenon is one of the severe complexities that simulations of distillation columns experience when highly nonideal VLLE relationships are involved.

Figure 5.23 compares temperature and composition profiles at two different steady states. The reflux flowrate is 0.082 kmol/s in both cases. The fresh feed and recycle are identical. The bottoms flowrate is the same. However, the bottoms composition is drastically different.

Obviously the steady state indicated by the solid lines in Figure 5.23 is the desired one. The bottoms purity with this steady state is 99.27 mol% ethanol. The impurities are

TABLE 5.2 Effect of Changing Reflux Flowrate


Bottoms Composition

Bottoms Composition

Was this article helpful?

0 0

Post a comment