Understanding the Effect of M(III) Choice in Heterodinuclear Polymerization Catalysts.

Abstract

The ring-opening copolymerization (ROCOP) of epoxides with CO2 or anhydrides is a promising strategy to produce sustainable polycarbonates and polyesters. Currently, most catalysts are reliant on scarce and expensive cobalt as the active center, while more abundant aluminum and iron catalysts often suffer from lower activities. Here, two novel heterodinuclear catalysts, featuring abundant Al(III), Fe(III), and K(I) active centers, are synthesized, and their performance in the polymerization of four different monomer combinations is compared to that of their Co(III) analogue. The novel Al(III)K(I) catalyst exhibits outstanding activities in the cyclohexane oxide (CHO)/CO2 ROCOP, and at 1 bar CO2 pressure it is the fastest aluminum-based catalyst reported to date. The M(III) site electronics for all three catalysts, Al(III)K(I), Fe(III)K(I), and Co(III)K(I), are measured using IR and NMR spectroscopy, cyclic voltammetry, and single crystal X-ray diffraction. A correlation between M(III) electron density and catalytic activity is revealed and, based on the established structure-activity relationship, recommendations for the future catalyst design of abundant Al(III)- and Fe(III)-based catalysts are made. The catalytic performances of both Al(III)K(I) and Fe(III)K(I) are further contextualized against the relative elemental abundance and cost. On the balance of performance, abundance, and cost, the Al(III)K(I) complex is the better catalyst for the carbon dioxide/epoxide ROCOP, while Fe(III)K(I) is preferable for anhydride/epoxide ROCOP.