Multiobjective Design of Reactive Distillation with Feasible Regions

Abstract

This work addresses the multiobjective design of complex reactive distillation columns through the use of feasible regions. A cost indicator reflecting energy usage and column size is introduced and used to build the Pareto surface describing the optimal combinations of cost and performance. The study includes the use of superheated and subcooled feed streams and searches for the optimal distribution of feeds and catalyst inside the column. The technique is first illustrated for a base case system with ideal vapor-liquid equilibrium and then for two variants with different volatilities for reactants and products. The best trade-off solutions for each system are identified, the operating conditions analyzed, and a selected number of cases investigated further using rigorous simulation in Aspen Plus. Important insights gained and a few heuristic rules about the distribution of the reaction and feed in the column are presented. Furthermore, a methodology that may potentially contribute to overcome limitations on catalyst volume by increasing the specific reaction turnover is presented.

Performance Indicators in Reactive Distillation Design

“…The number of stages ranges between 29 and 16, the reactive holdup between 3.2 and 8 and the capacity between 90.4 and 41. As discussed earlier (Filipe et al, 2008b) the capacity increases with the number of stages (higher internal flows in larger columns) and, for the same design specification, higher product purities require higher capacities (increased residence time in the reactive trays). Column diameter changes only slightly within each scenario, while between scenarios a significant reduction is observed in C when compared to A and B, which is justified by its lower internal flows due to the feed qualities used.…”

“…To develop this study, six different solutions belonging to the Pareto front relating reactive and capacity (cost indicator) previously generated (Filipe et al, 2008b) were selected. The main assumptions for the optimization methodology developed in GAMS to achieve the Pareto fronts are: steady state operation, constant pressure, vapor-liquid equilibrium at every stage, kinetically controlled reaction occurring in the liquid phase, and negligible heat effects.…”

“…The capacity cost indicator previously defined (Filipe et al, 2006) is based on the size of internal flows, feeds and number of stages, providing an expedient method for the evaluation of a large number of solutions. Equation 1 shows how to calculate the capacity for a column with N stages, where i = 1 refers to the condenser and i = N to the reboiler, V 2 is the vapor condensed at the total condenser, B is the bottom liquid flow and w B and w C are the weights for the boiling and condensing capacity, respectively.…”

“…Design should therefore be preceded by preliminary studies to assess first the potential adequacy of the technology. With this aim some work was developed to create a framework combining feasible regions and optimization techniques to the design and multi-objective optimization of complex reactive distillation columns, for reacting systems with variable degrees of relative volatilities (Filipe et al, 2006;2007;2008b). This led to the consideration of RD columns with distributed feeds, with many of them involving the combination of superheated and subcooled feeds.…”

“…To assess the impact of employing such combined feeds, a cost indicator, designated capacity, was introduced as a novel optimization objective criterion (Filipe et al, 2006). This cost indicator, based on the capacity variables (Jobson et al, 1996), was defined as the weighted sum of liquid and vapor internal flows divided by the total feed to the column.…”