UC Berkeley

Abstract

Electrocatalytic Oxidation of Formate with Rh(III) and Co(III) Electrocatalysts

By

Daniel Louis Kellenberger

Doctor of Philosophy in Chemistry

University of California, Berkeley

Professor John Arnold, Chair

Chapter 1. The hydrogen fuel cell is an environmentally friendly alternative to fossil fuel combustion for the powering of vehicles and other mobile applications. The storage of hydrogen in appreciable densities and the difficulty of its distribution are prohibitive factors for the large scale, societal adoption of current hydrogen fuel cell technologies. A hydrogen fuel cell based on the reversible storage of hydrogen equivalents in liquid, organic substrates would alleviate both of these issues. A hydrogen fuel cell based on the hydrogenated/dehydrogenated pair of formic acid and carbon dioxide is a promising example that would allow for the regeneration of the fuel at an external plant away from the point of release. There exist numerous examples of the dehydrogenation of formic acid to generate dihydrogen. However, the direct electrocatalytic oxidation of formic acid to its constituent protons, electrons, and carbon dioxide byproduct is presented as a more attractive alternative that would allow for the incorporation of a homogeneous electrocatalyst into the fuel cell itself. A generalized mechanism is envisaged that would allow for the direct electrocatalytic oxidation of formic acid by a homogenous, organometallic electrocatalyst.

Chapter 2. The Rh(III)-centered complex [Cp*Rh(bpy)(MeCN)][PF6]2 (Cp* = pentamethylcyclopentadienyl, bpy = 2,2’-bipyridyl) was selected as a possible electrocatalyst for the electrocatalytic oxidation of formate. Analogues for each of the intermediates in the electrocatalytic cycle as presented in Chapter 1 were either isolated, modelled, or directly observed. The Rh(I) complexes Cp*Rh(bpy) and Cp*Rh(phen) (phen = 6,10-phenanthroline) were synthesized and the latter was structurally characterized. Reaction of the Rh(III) complex with formate in acetonitrile resulted in decomposition but coordination of formate was modelled with the isolation of the acetate analogue, [Cp*Rh(bpy)(OAc)][PF6]. The complex [Cp*Rh(6,6’-Me2-2,2’-bipyridyl)(MeCN)][PF6]2 featuring a bulkier chelating ligand was synthesized and monitoring the reaction with formate in acetonitrile by 1H NMR revealed the in situ generation of a Rh(III) hydride, [Cp*Rh(6,6’-Me2-2,2’-bipyridyl)(H)]+. Electrocatalytic oxidation of formate in benzonitrile was achieved as determined by constant potential coulometry experiments conducted at the impressive potential of -900 mV vs. Ag/Ag+.

Chapter 3. A series of Rh(III) electrocatalysts were synthesized of the type [Cp*Rh(chelate)(MeCN)]2+ (1) featuring the chelates (a) 2,2’-bipyridyl, (b) 6,10-phenanthroline, (c) 4,4’-Me2-2,2’-bipyridyl, and (d) 6,6’-Me2-2,2’-bipyridyl to determine the influence of the chelate on the complex’s observed reactivity. Computational chemistry was in agreement with the proposal that the observation of two electrochemical reductions for complexes 1a-c resulted from an equilibrium in solution of 1 with a 16-electron complex, [Cp*Rh(chelate)]2+. The solid state structures of 1a-d determined by single crystal, x-ray diffraction experiments and the gas phase structures calculated with computational chemistry suggested the observed irreversible electrochemical reduction of 1d was due to increased steric effects from its bulkier chelating ligand which would favor dissociation upon reduction. Additionally, a system of calculations were developed to address the competing abilities of the Rh(III) hydrides to act as proton sources or hydride donors with the former being desired for the electrocatalytic oxidation of formate observed in Chapter 2.

Chapter 4. [Cp*Co(bpy)(MeCN)]2+ was synthesized and characterized as a first-row analogue the Rh(III) electrocatalysts studied in Chapter 2 and Chapter 3. A new, simplified synthesis of [Cp*Co(MeCN)3][PF6]2 was achieved by reaction of Cp*Co(CO)2 with two equivalents of AgPF6 in acetonitrile. Reaction of the tris-acetonitrile complex with either 2,2’-bipyridyl or 6,6’-Me2-2,2’-bipyridyl resulted in the generation of the complex [Cp*Co(chelate)(MeCN)][PF6]2. Unlike with the Rh(III) analogues, reaction of the [Cp*Co(bpy)(MeCN)][PF6]2 species with formate generated an isolable Co-formate adduct, [Cp*Co(chelate)(OC(O)H)][PF6]2. The addition of formate to [Cp*Co(bpy)(MeCN)][PF6]2 to generate the Co-formate adduct in situ resulted in the appreciable anodic shift of the oxidation potential of formate by ca. 750 mV.