Kinetics of the slurry polymerization of propylene using metallocene catalysts: Experiment and simulation

An experimental reaction polymerization facility was built to study polymerization kinetics using metallocene catalysts and it is described in detail. It has been used to evaluate the performance of methylaluminoxane (MAO)-activated metallocene catalysts in slurry-mode propylene homopolymerization. Experiments using two relatively low activity C2 symmetric catalytic species, rac-ethylenebis(1-indenyl)zirconium dichloride (I) and rac-ethylenebis(4,7-dimethyl-1-indenyl)zirconium dichloride (II) were performed at different process conditions where polymerization kinetics are the rate limiting step. The rates of propylene reaction in a factorial set of experiments at different levels of monomer pressure and reaction temperatures were measured. Mass transfer coefficients for transport of propylene gas into the liquid solvent were calculated at different polymer solid contents in the liquid phase. A chemical mechanism for the homopolymerization of propylene by metallocene catalysts was proposed, and a model formed by a series of population balances was created. Kinetic parameters for polymer chain initiation, deactivation, and propagation were found through non-linear regression methods based on the instantaneous rate of reaction. The method of moments was used to determine estimated number and weight average degrees of polymerization for the polypropylenes produced, based on experimental values from high temperature Gel Permeation Chromatography. Characterization of the final product by 1H Nuclear Magnetic Resonance revealed the nature of the various polymer end-groups. These end-groups assisted in determining the modes of chain termination, and confirmed the feasibility of the proposed chemical mechanism. The dominant end-group in I/MAO based polypropylene was found to be vinylidene, whereas 2- cis-butenyl was for II/MAO. This fact suggested different termination patterns for both systems. Measurement of the initial concentration of the cationic active complexes was identified as an important factor in the determination of kinetic parameters of the model. A method for their estimation is proposed