Model-Based Design of Single-Site Olefin Polymerization Catalysts
W. Nicholas Delgass1, James M. Caruthers1, Mahdi Abu-Omar2, Kendall T. Thomson1, Venkat Venkatasubramanian1, Gary E. Blau1, Khamphee Phomphrai2, Grigori Medvedev1, Corneliu Stanciu2, Shalini Sharma2, Thomas A. Manz1, Jesmin Haq1, Krista A. Novstrup1, and Balachandra B. Krishnamurthy1. (1) School of Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, IN 47907-2100, (2) Chemistry Department, Purdue University, 560 Oval Drive, West Lafayette, IN 47907
A chemical kinetic model, although an essential ingredient for the understanding of catalytic behavior, does not contain sufficient knowledge to predict improved catalysts. The inherent chemistry of families of catalysts governs relationships between the various rate and equilibrium constants and thus provides the key to the optimal catalyst formulation. The primary objective of Purdue's Discovery Informatics approach to catalyst design is building a quantitative forward model that links descriptors of the catalyst chemistry, through the microkinetic model, to catalyst performance. The forward model evolves as more data and additional chemical insight become available and becomes the repository for knowledge extracted from the experiments and analysis. Progress in implementing the Discovery Informatics approach, and particularly the development of descriptors, will be illustrated with a study of single-site olefin polymerization catalysis; specifically, aryloxide and Cp(Cp*) ligated Ti and Zr catalyst precursors of the form Cp′Ti(OAr)Me2 that are activated for 1-hexene polymerization by B(C6F5)3. This single-site olefin polymerization system has the advantage of a well-defined catalyst site and the opportunity to alter steric and electronic effects by substitution on the aryloxide ring. The kinetics of the batch polymerization were monitored by in situ 1H NMR and product molecular weight distributions were determined by GPC analysis. Rate constants were determined from population balance models for the evolving monomer and molecular weight distribution, and the appropriate nonlinear statistical analysis was used to provide error estimates for the various rate constants. Values for kp, the 1-hexene propagation rate constant, were obtained from the time-dependent 1-hexene concentration data while carefully accounting for effects of the initiation rate and catalyst activation and degradation. To identify descriptors and formulate a forward model, DFT simulations were employed to provide quantitative measures of the steric and energetic consequences of aryloxide substitution. Key elementary steps in the reaction sequence are “docking”of the monomer to form a π-complex and monomer insertion into the growing chain. We postulate that the rate limiting step is access to the Ti site by the 1-hexene. On the basis of literature reports that weakly coordinating anions are known to be associated with higher polymerization rates [1,2], we take the ion pair separation energy as the descriptor related to the energy needed to move the anion away from the cation site to provide access. The second descriptor, associated with the steric limitations on monomer access to the site, is the unrestricted cone angle available for docking. A forward model based on these two descriptors closely correlates kp for 18 Ti-based catalysts and is the first such model of its kind. The details of the kinetic analysis of the batch polymerization and the path forward will also be discussed.
1. Chen, M. C.; Roberts, J. A. S.; Marks, T. J. J Am Chem Soc 2004, 126, 4605-4625. 2. Chen, E. Y. X.; Marks, T. J. Chem Rev 2000, 100, 1391-1434.