Ziegler- Natta Catalysts for Propene Polymerization: An Evolutionary Bridge between Heterogeneous and Molecular Catalysis

Detailed mechanistic understanding of industrially relevant heterogeneous catalysts is rare; as a matter of fact, discovery and even optimization are mostly driven by empiricism. Ziegler-Natta catalysts (ZNCs) for the production of isotactic polypropylene are no exception in this respect. Modern ZNCs are complex multicomponent formulations. Nanostructured MgCl2 is used as support, on which a titanium compound (generally TiCl4) and a Lewis base called “Internal Donor” (ID) are co-adsorbed (where ID is an organic donor such as an aromatic or aliphatic mono- or diester, or a 1,3 diether). Activation of the precatalyst is carried out by an AlR3 species (typically AlEt3), usually mixed with another organic Lewis base, generally an alkoxysilane, called “External Donor” (ED). Among the several models that have been proposed to explain donors (and possibly Al-alkyls) effects on the active Ti sites of ZNCs, the most recent and widely accepted one is the so-called “three-site model”. According to this model, the first coordination sphere of Ti, assumed to be in the Ti(III) oxidation state, is always octahedral and C2-symmetric (like in crystalline TiCl3); steric hindrance in the second coordination sphere, on the other hand, can vary depending on whether or not the first-neighbour surface Mg or Ti atom on either side exhibits a coordination vacancy. Here we provide experimental and computational evidence that TiCl4 and the other components of the formulation (i.e., donor and AlR3 species) are more intimate than has been assumed so far, to the point that these systems resemble immobilized molecular catalysts more than conventional heterogeneous ones. Invoking such (metastable) species can explain some long-standing ‘mysteries’ regarding activation and deactivation pathways in ZN catalysis. As a matter of fact, state-of-the-art DFT-D modeling reveals how the formation of metastable TiCl4-μ-AlR3 surface species allows for low-energy activation and reduction pathways of TiCl4. Similarly, direct coordination of the external donor to the Ti can explain how catalyst deactivation mechanisms, e.g. Cl/OR (R = Me, Et, etc.) metathesis, can occur. Last but not least, consequences on active site models for the stereoselectivity in the chain propagation step will also be highlighted.