화학공학소재연구정보센터
Industrial & Engineering Chemistry Research, Vol.42, No.15, 3602-3611, 2003
Graph-theoretic method for the algorithmic synthesis of azeotropic-distillation systems
A highly effective algorithmic method is proposed for optimally synthesizing an azeotropic-distillation system from an extensive set of candidate operating units, i.e., functional units. The method has been derived by resorting to the graph-theoretic approach to process-network synthesis based on process graphs (P-graphs); it also resorts to the methodology established in our previous contribution for dividing the residue curve map (RCM) of a material system, i.e., mixture, to be separated into partitioned materials. This allows the entire space of the RCM to be taken into account in composing networks of candidate operating units, thereby preventing the localization of search. Moreover, the RCM is transformed into the flow-rate map, where any material is quantitatively defined by the molar flow rates of its components instead of the concentrations as in the RCM. This renders it possible to eliminate the nonlinearity in the governing equations of the candidate operating units. The efficacy of the method is amply demonstrated through the well-known example of separating ethanol from its aqueous solution with toluene as the entrainer. The method is applicable to other complex processes with phase transition, and/or phase separation, e.g., crystallization, extraction, reactive distillation, and their combinations.