Inferential closed-loop control of particle size and molecular weight distribution in emulsion polymerization of styrene

Abstract

In this work, simultaneous inferential control of particle size distribution (PSD) and molecular weight distribution (MWD) in a semi-batch emulsion polymerization reactor of styrene has been addressed. Using a comprehensive dynamic model for PSD and MWD predictions and performing a sensitivity analysis, it has been revealed that free surfactant and chain transfer agent (CTA) concentrations in the reactor are the most suitable candidates for inferential control of PSD and MWD, respectively. To control concentrations of these species in the reactor, their inlet feed flow rates are used as manipulated variables. It is assumed that the concentration of CTA is measured infrequently and therefore an open-looped observer, based on the reaction calorimetry, has been designed to estimate the CTA concentration. The infrequent measurements of CTA concentration are used to correct its estimation. As the online measuring of the ionic free surfactant concentration is also difficult, solution conductivity which is a good indication of free surfactant concentration is used for control purposes. Simulation results show that the performance of the proposed control scheme is satisfactory even in the presence of model mismatch. POLYM. ENG. SCI., 2010. © 2010 Society of Plastics Engineers