S 335 Lu

Figure 16.5 Effects of stripping and rectifying tray numbers.

(330-355 103 $/year). Therefore, column cost, which decreases with decreasing numbers of separation trays, is the most dominant factor among the three capital cost factors. Thus, the optimum capital cost shifts to smaller numbers of stripping and rectifying trays, but the optimum energy cost shifts to larger numbers of stripping and rectifying trays. The result is a minimum in the TAC curve at a certain number of stripping and rectifying trays. For a390 = 0.95 with constant reactor numbers Next = 3 and constant reactor holdups VR = 80kmol, the optimum number of stripping and rectifying trays are NS = 17 and NR = 7, respectively.

It is interesting to compare these design results with the approximate heuristic design found in some initial work by Kaymak and Luyben.1 The first two columns in Table 16.1 give a direct comparison. The improved design has fewer trays and larger reactors. The TAC is 13% lower.

Other a3go Cases. Figure 16.6 provides the optimum design results for the column/side reactors process for a range of temperature-dependent relative volatilities (a390 = 0.95 - 2). These results show that there is little change in the optimum number of reactors. As the reference a390 decreases, the optimum number of reactors also slightly decreases from four to three. The inlet and outlet temperatures of the reactors decrease with the decrease of the reference a390 value.

B. Kaymak and W. L. Luyben, Design of distillation columns with external side reactors, Ind. Eng. Chem. Res. 43, 8049 (2004).

However, a lower temperature is unfavorable for reaction because the reaction rates get smaller. The result is a rapid increase in the required reactor holdups. A lower reference a390 value results in more difficult separation, as expected. Thus, there is an increase in the required vapor boilups and number of stripping trays. Because the operating pressure at the top is kept constant for all a390 cases, there is no significant change in the number of rectifying trays. Note that the optimum number of stripping trays is larger than the optimum number of rectifying trays. This is caused by the higher temperatures in the lower part of the column, which means lower relative volatilities.

Results for these four different a390 cases are given in Table 16.1. The values of the five design optimization variables are the number of reactors, the holdup of each reactor, the number of stripping trays, the number of rectifying trays, and the number of trays between each liquid trap-out tray. Table 16.1 and the upper graph in Figure 16.7 show that reducing the reference relative volatility increases both capital and energy costs, which results in an increase of TAC.

The lower graph in Figure 16.7 compares the economic optimum steady-state design of the column/side reactor process with those of the reactive distillation column and the multiunit conventional process. The reactive distillation column is the most economical alternative for the a390 = 2, where there is no reaction/separation temperature mismatch. The column/ side reactor process becomes more attractive as the mismatch of reaction / separation temperatures becomes more severe. The distillation column with a side reactor is economically superior for reference relative volatilities that are smaller than 1.5 for this case study.

0(390

Figure 16.7 Comparison of alternative processes.

Figure 16.7 Comparison of alternative processes.

«390

Figure 16.8 Effect of feed tray locations.

Figure 16.8 Effect of feed tray locations.

TABLE 16.2 Results for Reactive Distillation Columns with Optimized Feed Tray Location

TABLE 16.2 Results for Reactive Distillation Columns with Optimized Feed Tray Location

Was this article helpful?

0 0

Post a comment