Comparison of Alternative Control Structures for an Ideal Two-Product Reactive Distillation Column

Abstract

Although steady-state design and open-loop dynamics of reactive distillation columns have been explored in many papers, very few papers have dealt with closed-loop control. Most of these control papers consider reactive distillation columns in which there is only one product, and an excess of one of the reactants is sometimes assumed. This paper explores the closed-loop control of a reactive distillation column in which two products are produced in a single column and stoichiometric amounts of fresh feeds are desired. The reversible reaction is A + B ⇌ C + B. The relative volatilities are favorable for reactive distillation; i.e., the reactants are intermediate boilers between the light product C and the heavy product D. Simple ideal physical properties, kinetics, and vapor−liquid equilibrium are assumed so that the basic control issues of reactive distillation can be explored without being clouded by complexities of a specific chemical system. Six alternative control structures are evaluated via rigorous dynamic simulation. All of the schemes use a composition analyzer in the reactive zone of the column to detect the inventory of one of the reactants so that fresh feed can be manipulated to satisfy the reaction stoichiometry. Therefore, the use of excess reactant is not required. The interaction between design and control is illustrated by the impact of holdup in the reactive zone: increasing holdup (catalyst) improves the dynamic controllability of the process. Single-end temperature control can keep both products at or above specified purity values, even for large disturbances, if reactive-zone holdup is sufficiently large.