The ETBE column described in Chapter 3 (a total of 10 ideal stages) was used as the basis for the dynamic model presented here although the individual models (as described in Sections 7.2.2-7.2.6) can be combined in other ways to simulate any other column configuration.
The steady state model of the reactive distillation process used a modified set of MESH equations. This technique was retained in the dynamic model although several simplifying assumptions were added to offset the additional complexity required in the dynamic case. The following assumptions were introduced:
• chemical equilibrium is attained on all reactive stages;
• DIB formation is negligible;
• all stages are at their bubble point so that the temperature is a function of composition only;
The assumption of chemical equilibrium allows the reaction kinetics to be neglected, simplifying the modelling of the reactive stages. Negligible DIB formation allows the number of components to be reduced from five to four and the elimination of equations describing the DIB reaction. The bubble point assumption effectively permits the enthalpy derivatives to be excluded from the model - a common assumption in both simple and rigorous distillation simulation. Finally, an ideal vapour phase allows fugacity coefficients to be neglected. These assumptions slightly reduce the absolute accuracy of the model but make it a more effective dynamic simulation tool by improving the ratio of real time to simulation time.
The various physical property routines that were used are indicated in Table 7.1 and are identical to those used for steady state simulation. A recently published reaction equilibrium expression, specifically derived for ETBE, was used (Jensen and Datta, 1995) in preference to older information or equations based on MTBE synthesis. Similarly, the latest vapour pressure data was used (Krahenbiihl and Gmehling, 1994).
Table 7.1 - Physical Property Routines and Sources
Was this article helpful?