If It’s Not Broke, Break It, and then Break It Again: Understanding [Cp*Rh] Catalysts for Hydrogen Evolution by Investigation of Remarkably Inert Analogues

Abstract

Monomeric half-sandwich rhodium hydride complexes are often proposed as intermediates in catalytic cycles, but relatively few such compounds have been isolated and studied, limiting understanding of their properties. In this thesis, the preparation of a monomeric rhodium(III) hydride complex bearing the pentamethylcyclopentadienyl (Cp*) and bis(diphenylphosphino)benzene (dppb) ligands is reported. The hydride complex is formed rapidly upon addition of weak acid to a reduced precursor complex, Cp*Rh(dppb). Single-crystal X-ray diffraction data for the [Cp*Rh] hydride, which were previously unavailable for this class of compounds, provide evidence of the direct Rh–H interaction. Complementary infrared spectra show the Rh–H stretching frequency at 1986 cm–1. In contrast to results with other [Cp*Rh] complexes bearing diimine ligands, treatment of the isolated hydride with strong acid does not result in hydrogen evolution. Electrochemical studies reveal that the hydride complex can be reduced only at very negative potentials (ca. –2.5 V vs. ferrocenium/ferrocene), resulting in Rh–H bond cleavage and hydrogen generation. Experimentally determined thermochemical parameters for reactions of the [Cp*Rh] hydride and its reduced form provide a rationale for the observed reactivity differences between the dppb and analogous diimine frameworks that can generate H2 with moderately strong acids. These results are discussed in the context of development of design rules for improved catalysts bearing the [Cp*] ligand.

To gain further insight into the electronic properties of the phosphine-based ligands that favor metal hydrides and limit catalysis, a second series of [Cp*Rh] complexes supported by the redox-active bidentate diphosphine ligand bis(diphenylphosphino)ferrocene (dppf) is described, with particular attention paid to the outcomes of proton and electron transfer on this framework. Notably, Cp*Rh(dppf) exhibits a quasireversible RhII/I reduction at –0.96 V vs. Fc+/0 rather than undergoing a net 2e– RhIII/I process as is often observed on the [Cp*Rh] platform. This behavior provides access to a species in the relatively uncommon rhodium(II) oxidation state which has been characterized by electron paramagnetic resonance spectroscopy. Protonation of Cp*Rh(dppf) results in formation of an isolable [Cp*Rh] monohydride that is inert to protonolysis, providing a second example of the stabilizing effect bidentate diphosphine ligands have on Rh–H bonds. The quasireversibility of the dppf-centered FeIII/II couple of the rhodium monohydride [Cp*Rh(dppf)H]+ at +0.41 V vs. the ferrocenium/ferrocene redox couple facilitates a rigorous thermochemical analysis of the system, from which we have determined that oxidation centered at the dppf ligand results in dramatically increased acidity of the Rh–H bond by 23 pKa units.