Modeling and optimization of a seeded suspension polymerization process

A mathematical model is developed and validated for a seeded suspension polymerization process that manufactures polystyrene beads starting from pretreated polyolefin seeds. For the manufacture of this specialty product, we focus on prediction of the dynamics of particle growth, polymerization rate and particle average molecular weight. Particular emphasis is placed on acquisition of reliable parameters which must be estimated from available analytical measurements of the product. A hybrid approach is proposed for parameter subset selection and estimation. Excellent agreement is obtained with experimental results for different product grades. The monomer feeding rate, which is a key controlling factor of bead properties, is determined based on the proposed model and an optimum semi-batch feeding policy is determined. This improved policy is also implemented and validated by pilot plant experiments.