| Considerable emphasis is given today to reduce the cost of chemical plants by simplifying their design. One way is to reduce the number of units by combining two or more steps into a single vessel. Although there were a few spectacular successes, large amounts of money have been spent on unsuccessful applications. In this thesis, an attempt is made to provide a systematic framework and quantitative criteria to evaluate the chance of success before large expenditures are made. Our method focuses on several criteria. (1) Does the combination of two steps into one unit create a synergism that improves both of them? If so, how strong is the synergism? (2) Are there any detrimental interaction due to this combination? (3) Do the two processes to be integrated share any dominant variables (state variables that strongly impact on process performance)? If so, is there a range of values, for these variables, acceptable for both processes?; For the majority of processes, examination of criteria 2 and 3 shows penalties, arising from the integration, that overweigh the advantages. It is often difficult to modify one of the processes; e.g. while membrane reactors are inherently attractive, it is hard to find membranes effective under reaction conditions.; Although the results have general applications, the thesis focuses on reactive distillation, a subject that achieved wide academic and industrial interest, but has had few large-scale implementations. We show that reactive distillation has large potential for processes involving azeotropes, as it allows implementation of counter-current reaction with miscible reactants in a single column. One of the main problems is that the temperature required for distillation is not suited for the reaction, due to either thermodynamic constraints or lack of catalysts. A main achievement of the thesis is the development of a new design concept, a non-reactive distillation column with several external reactors cascaded into it, able to retain the advantages of reactive distillation without suffering from its limitations. Finally, we show, using isomerization, that combining reaction and distillation, unlike a membrane, does not necessarily shift reaction conversion beyond equilibrium, but this depends on the specific reactive system. |
