Estimation of copolymerization kinetic parameters by maximum likelihood method

Abstract

AbstractA kinetic study of free‐radical solution copolymerization of isobutyl methacrylate and lauryl methacrylate in benzene, initiated with 2,2‐azoisobutyronitrile (AIBN) was performed at different monomer feed compositions, at low conversion level. To avoid the complications of copolymerization kinetics, the compositional‐averaged rate constant method was applied to dead‐end polymerization in constant and variable volume polymerization systems. A conditional probability approach, with “maximum likelihood function” (MLF), was used for estimating the kinetic parameters from experimental data. This method provides point estimates as well as joint probability regions. The validity of this method was approved for the statistically justified and best estimation of the rate parameters in copolymerization systems. Statistical analysis indicated that, although copolymer composition is well described by the terminal unit model (TUM), the rate data are not. Like many other systems recently examined, the data were well represented by the implicit penultimate unit effect (IPUE) model of Fukuda and coworkers. It was found that the coupled rate parameter k̄p/k̄ is greater than the TUM predictions, whereas most of the previous systems studied have coupled rate parameters lower than those predicted by the TUM. An interactive computer simulator for parameter estimation was developed, which is run in two modes of estimation and prediction and is capable of receiving input data from both keyboard and databank. © 2004 American Institute of Chemical Engineers AIChE J, 50: 1260–1272, 2004