Molecular Kinetics of Catalytic Olefin Polymerization: an Integrated Quenched-Flow and High-Temperature Cryoprobe NMR Approach

This work represents a contribution to a better mechanistic understanding of catalytic olefin polymerizations by means of a combination of Quenched-Flow (QF) kinetic studies and NMR polymer analyses. To the best of our knowledge, it is the first time that the two approaches have been integrated, and applied to molecular (homogeneous) as well as heterogeneous catalyst systems. In a first part (Chapter 2), we validated our QF setup and methods in ethene polymerization mediated by a bis(phenoxyimine)Ti(IV) catalyst known for a long-lasting controlled kinetics up to relatively high temperatures. From variable-temperature QF measurements of kp (in nice agreement with the previous literature) we determined the values of enthalpy and entropy of activation; along with a fraction of active Ti close to 100%.As a matter of fact, a similar study in the presence of a well-known ansa-zirconocene, namely rac-Me2Si(2-Me-4-Ph-1-Ind)2ZrCl2, ended up with non-trivial results, indicating a strong dependence of chain propagation kinetics on the nature of the co-catalyst (namely, methylaluminoxane (MAO) with or without the addition of butylhydroxytoluene (BHT) as an AlMe3 trap). In particular, we found that entropy of activation is an extremely important variable in this catalysis, and that catalysts features a rather high enthalpy of activation. Overall, Chapter 2 demonstrates that the QF approach is robust and reliable, but also that one cannot assume tout-court a trivial Cossee-Arlman mechanism for catalytic olefin polymerization, even in case of supposedly simple single-center molecular catalysts. Not surprisingly, heterogeneous Ziegler-Natta catalysts (ZNCs) turned out to be even more complex than homogeneous ones. The ill-defined nature of the active species and their (very) low concentration are not the only complicating factors; as a matter of fact, issues related with the accumulation of 'dormant' sites are also important in the polymerization of propene (and other substituted olefins). In Chapter 3, we benchmarked our system and protocols by investigating propene polymerization in the presence of a 3rd-generation ZNC with ethyl benzoate (EB) as the Internal Donors (ID). The 'heart' of the present work was the kinetic investigation of a 4th-generation ZNC with a diisobutyl phthalate (DIBP) ID, reported and discussed in Chapter 4. We confirmed previous literature indications that pre-contacting the precatalyst with TIBA is necessary here to observe activity under QF conditions; while we cannot propose a univocal explanation for this fact, we suggest that this bidentate ID hinders most Ti centers in the precatalyst, and needs to be removed (in this work by reacting with TIBA) for catalytic activity to develop. This is a first major difference with respect to the catalyst system of Chapter 3 (ID = EB). Another was the major impact of regioirregular 2,1 insertions on the kinetics of propene polymerization, clearly pointed out by 13C NMR analyses of the produced polymers. Indeed, we demonstrated conclusively that the most stereoselective catalytic species in this system rapidly develop into 'dormant' sites due to the accumulation of alpha-Me-branched Ti-Polymeryls. Less stereoselective catalytic species, in turn, seem to feature a less dormant character, and keep on propagating despite the slowing down effect of the regiodefects. As a result, a peculiar time dependence of polymer stereoregularity was observed, the degree of isotacticity declining in the first few seconds of polymerization and ultimately attaining a steady-state level. This finding is key to understand the dramatic impact of H2 added to the system as a chain transfer agent; in fact, the boost of productivity and the strong enhancement of stereoselectivity in the presence of H2 can now be explained in terms of its waking-up action on the most stereoselective catalytic species.