Polymer reactors and molecular weight distribution. IV. Free‐radical polymerization in a steady‐state stirred‐tank reactor train

This paper reports the simulation of steady‐state styrene polymerization in a single, continuous, stirred‐tank reactor (CSTR) at high viscosity, a train of three CSTR's with and without recycle, and a single, imperfectly mixed CSTR. Monomer conversion and polymer molecular weight distribution (MWD), measured experimentally by gel permeation chromatography, were used as criteria of simulation. Earlier results for the single CSTR were used to modify the catalyst efficiency f and termination rate constant kt, to account for the effect of viscosity, μ. The regression equations obtained were log μ = −1.06 −0.522 log (S + 1) −8.34 log (1 − PF) + 0.565 logr̄w, log (f/fi) = 0.022 log (1 + μ) −0.0285 [log (1 + μ)]2, and log (kt/kti) = 0.060 log (1 + μ) −0.0934 [log (1 + μ]2, where μ is reactor viscosity in centipoise, S is solvent concentration in g‐mole/1., PF is polymer weight fraction, r̄W is weight‐average chain length, and fi and kti are values at zero conversion. The theoretical models for the single CSTR and train of CSTR's simulated the behavior of the real, bench‐scale CSTR's over a range of experimental conditions. The model for the CSTR system was applied to a theoretical mixing model made up of perfectly mixed CSTR's.