Mathematical Model of the Continuous Heterogeneous Process for the Production of High-Impact Polystyrene

C. LUCIANI; D. ESTENOZ; G. MEIRA

Buenos Aires (Argentina)

Congreso; XXII Interamerican Chemical Engineering Congress; 2006

High-Impact Polystyrene (HIPS) is a heterogeneous material produced via polymerization of styrene in the presence of an elastomer such as polybutadiene (PB) or styrene-butadiene diblock copolymers. Typically, HIPS is composed by about 80% of free polystyrene (PS), 2% of residual PB, and 18% of graft copolymer, exhibiting core-shell or salami morphologies (rubber particles dispersed in a free PS matrix, with a single or numerous PS occlusions, respectively). Although the industrial process for the production of HIPS has been known for over 50 years, not only the morphology development but also the relationships between the morphological and molecular structure and final properties of HIPS remain uncertain. This is a consequence of the complex physico-chemical phenomena that take place during the process (i.e.: phase separation and phase inversion). Therefore, most theoretical developments have been focused on homogeneous models which consider the reaction mixture as a single phase. Recently, a pair of heterogeneous mathematical models has appeared in the literature (Casís et al., 2005; Luciani et al., 2005). Casís et al. (2005) predicted both composition and molecular structure in two phases, while Luciani et al. (2005) also estimated the morphology evolution along a batch bulk polymerization. However, neither of them is capable of describing the continuous process which is the most common method for producing HIPS [i.e.: series of continuous stirred tank reactors (CSTRs)]. In this work, a first attempt to model the continuous heterogeneous process for the production of HIPS is presented. The volume of the first CSTR is divided in a set of segregated elements which are treated as pseudo-batch reactors (with a particular residence time). The heterogeneous approach can be applied to each of these pseudo-batch reactors. Thus, the CSTR exhibits a discreet residence time distribution. A similar procedure is applied along the reactor train, subjecting each pseudo-batch reactor to a subsequent subdivision in segregated elements. The global composition, molecular and morphological properties of the final HIPS are estimated as an average of final set of segregated (or pseudo-batch reactor) elements. Simulated results are adjusted and validated with measurements of conversion, grafting efficiency, free PS average molecular weights, and average diameter of particles from samples taken in an industrial plant. The simulated results adequately reproduce the measurements. The model presents an interesting tool to describe the complex phenomena carried out during the industrial polymerization of HIPS, and to design alternative operation policies with the aims of optimizing and controlling the industrial process.