(72v) On-Line Optimal Control of Molecular Weight Distribution in Batch and Semi-Batch Free Radical Polymerization Processes

Polymer molar mass distribution is one of the most important polymerization control variables that affect many polymers’ characteristics such as strength and thermal stability. Although there have been full feedback control of reactors on small molecules, polymerization reaction control is less advanced due to lack of understanding of the dynamics of the process, inadequate in situ sensors and nonlinearity in the behavior of the reaction for instance gel affect that influences  kinetic parameters during the reactions.

This contribution presents a complete real time feedback framework to control the conversion kinetics and polymer chain length distribution during the free radical batch polymerization operation. In the first step the process is conducted in batch mode where a detailed polymerization model is applied along with the method of finite molecular weight moments to determine a sequence of reactor temperature set points which lead to the desired molecular weight distribution. The optimization of the model is performed using the gOPT function in gPROMS. In order to have a better controllability on the process the system is also simulated in semi-batch mode where reagents are fed into the reactor during the reaction. The formalism in this case combines kinetics with time dependent processes related to flows of material into and out of the reactor. The objective is to find the monomer (initiator) flow rate to control the average molecular weight and the temperature is adjusted to manipulate the polydispersity. To validate the model and assess the merits of the proposed strategy, free radical polymerization of methyl methacrylate is carried out using 2,2’-azobis(2-methylbutanenitrile) as initiator. Automatic Continuous Online Monitoring of Polymerization reactions (ACOMP) is used to follow the conversion and evolution of the average mass distribution, and multi-detector SEC is applied to cross-check results and measure full distributions of end products.