So the prediction of the EQ stage model isn't accurate in the case of multi-component separation and chemical reaction with vapor-liquid-solid three-phase. It is the rate of mass and heat transfer, and not the equilibrium that limits the separation.

The difference of calculated values between the EQ and NEQ stage models can also be brought out in the composition profiles along the column, which are illustrated in Figs. 28 and 29. Here the studied system is somewhat special and the boiling-point difference of the key light (benzene) and heavy (1-dodecene) components is very high, up to 133.2 °C. This means that these two components are easy to be separated in very few equilibrium stages. So there is an abrupt change of benzene composition in the feed tray for the EQ stage model, as shown in Fig. 28. In the reactive and stripping sections of the SCD column, the benzene composition is almost zero, which leads to a higher temperature and fewer product of z'-ph (i = 2, 3, 4, 5, 6) than actual operation. However, the change of benzene composition in the feed tray in Fig. 29 for the NEQ stage model is somewhat smooth. In many places of the reactive and stripping sections, the benzene composition is away from zero in the NEQ stage model, which leads to a lower temperature and more products of z'-ph (i = 2, 3, 4, 5, 6) than in the EQ stage model. Therefore, the calculated values by the NEQ stage model are more in agreement with the experimental values.

Because the deviation between the experimental and calculated values is large for the EQ stage model, it is deduced that tray efficiency under the operation condition listed in Table 7 is low. That is to say, the assumption of EQ stage is far away from the actual. But for the NEQ stage model there is a little fluctuation of temperatures in the vicinity of the feed tray, which may be due to the great influence of the feed mixture on the mass and heat transfer rates and the limitation of accuracy of thermodynamic equations used in the NEQ stage model.

A comparison of the product composition at the bottom between the experimental and calculated values is also made in Table 8 where x is the mole fraction in the liquid phase. Table 8 shows that the concentrations of both benzene and 1 -dodecene at the bottom are very low. The fact of no benzene and 1-dodecene at the bottom proves that the SCD column is effective for this alkylation reaction. Furthermore, it can be seen from Table 8 that the calculated values by the NEQ stage model are more approximate to the experimental values than by the HQ stage model. So it farther proves that the NEQ stage model is reliable and can be used for the design and optimization of the SCD column.

The NEQ stage model will be preferred for the simulation of a tray reactive distillation column to the EQ stage model. However, as mentioned above, a close agreement between the predictions of EQ and NEQ stage models can be found if the tray efficiency is accurately predicted for the EQ stage model. But in the case that tray efficiency is difficult to obtain, especially for the vapor-liquid-solid three-phase system, the NEQ stage model can play a role.

In the experiment, the conversion of 1-dodcccne 100% and selectivity of 1-dodecene 100% are obtained. Moreover, the weight concentration of 2-ph in the i-ph is up to 35%, which is higher than 25% reported in the fixed-bed reactor [86]. So the SCD column is positive to the selectivity of 2-ph, which is favorable in industry.

Fig. 25. Comparison of calculated and experimental values at 101.3 kPa; the tray is numbered from the bottom to the top; • —Experimental values; ♦ —Calculated values by the NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 25. Comparison of calculated and experimental values at 101.3 kPa; the tray is numbered from the bottom to the top; • —Experimental values; ♦ —Calculated values by the NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 26. Comparison of calculate) and experimental values at 135.7 kPa; the tray is numbered from the bottom to the top; • —Experimental values; ♦ —Calculated values by the "NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 26. Comparison of calculate) and experimental values at 135.7 kPa; the tray is numbered from the bottom to the top; • —Experimental values; ♦ —Calculated values by the "NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 27. Comparison of calculated and experimental values at 162.1 kPa; the tray is numbered from the bottom to the top; V — Experimental values; ♦ — Calculated values by the NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 27. Comparison of calculated and experimental values at 162.1 kPa; the tray is numbered from the bottom to the top; V — Experimental values; ♦ — Calculated values by the NEQ stage model; A — Calculated values by the EQ stage model.

Fig. 28. Composition profile of the SCD column for the EQ stage model; the tray is numbered from the bottom to the top; ♦ —benzene; ■ —1-dodecene; 1 -dodecane; it —/-ph.

Fig. 28. Composition profile of the SCD column for the EQ stage model; the tray is numbered from the bottom to the top; ♦ —benzene; ■ —1-dodecene; 1 -dodecane; it —/-ph.

Fig. 29, Composition profile of the SCD column for the NFQ stage model; the tray is numbered from the bottom to the top; ♦ —benzene; ■ —1-dodecene; 9—1-dodecane; tV —/-ph.

Fig. 29, Composition profile of the SCD column for the NFQ stage model; the tray is numbered from the bottom to the top; ♦ —benzene; ■ —1-dodecene; 9—1-dodecane; tV —/-ph.

•eaclive pump around itRftïj wilhdriwal:

L K l-i,_feed to si Jc reactor

Was this article helpful?

## Post a comment