Systematic Generation of the Optimal and Alternative Flowsheets for Azeotropic-Distillation Systems

A systematic and rigorous method for synthesizing azeotropic-distillation systems, which is of utmost practical importance, is yet to be fully established. The available methods are based mainly on heuristics and graphical procedures. Our experience indicates that even in synthesizing a simple separation network, the structure of the optimal solution may be counterintuitive. For synthesizing a complex network structure necessary for an azeotropicdistillation system, therefore, the probability is very high that the solution generated would be far from the optimal one unless the method is systematic and rigorous. The proposed method is capable of algorithmically synthesizing optimal, near optimal, and other feasible structures for an azeotropic-distillation system from a set of candidate operating units. Initially, the residue curve map of the system is transformed to a unique multidimensional representation to facilitate the systematic partitioning of the feasible regions into lumped materials bounded by the thermodynamic boundaries and pinches. This renders it possible to derive analytical expressions of the resultant materials in terms of the coordinates of these boundaries and to automatically identify with dispatch the candidate operating units, such as separators, mixers, and decanters, for possible inclusion in the system. The processgraph (P-graph) representation of these operating units serves as a basis for the synthesis procedure including combinatorial algorithms. The method is equally applicable to various other complex processes with phase transition and/or phase separation with any number of components. Crystallization, extraction, reactive distillation, and their combinations are examples of such processes. A case study in which ethanol is separated from its aqueous solution with toluene as the entrainer demonstrates amply the efficacy of the method.