Developing Viable Catalysts for Ethanol Oxidation Using Electronic Structure Calculations

Farsi, Lida

Etd

Developing Viable Catalysts for Ethanol Oxidation Using Electronic Structure Calculations

Public

Fuel cells enable the conversion of different chemicals directly into electrical energy, and are much more efficient than conventional combustion engines. Direct ethanol fuel cells (DEFCs) use ethanol as a fuel source. However, DEFCs lack an efficient catalyst. In this study, using density functional theory (DFT) we modeled different catalysts, including alloys, to be used for the ethanol oxidation reaction (EOR) in DEFCs. We considered both stability and reactivity of such alloys. We modeled surface segregation energy as an indicator of the alloy's stability under reaction conditions. We modeled bimetallic Pt, Ir, Pd, and Rh alloys and developed a universal model for surface dependent segregation energies in bimetallic alloys. As an indicator of a catalyst's reactivity, we modeled C-C bond breaking in CHxCO (x=1,2,3) as such reactions are the bottleneck of complete oxidation of ethanol. We modeled carbon-carbon bond breaking in transition metals including Pt, Ir, Rh, Au, Ag, alloys including Ir-Rh, Ir-Rh-Sn, metal-metal oxide interfaces including Pt-SnO2, and Pt-Rh-SnO2, and metal oxides including Rh2O3 and recognized Pt-Rh-SnO2 to be the most effective catalyst for the C-C bond breaking. Finally we were interested in the design of bimetallic Ir-based alloys. For this purpose, we combined DFT with statistical physics based methods, specifically cluster expansion, to shed light on the ordering tendency of bimetallic Ir-based alloys. We found that Ir-Pt, Ir-Pd, and Ir-Cu tend to phase separate and do not form an alloy compound, while Ir-Rh, Ir-Ni, and Ir-Cr mix well and form an alloy. We also performed Monte Carlo simulations to investigate temperature effects on non-alloying mixtures. We showed as the temperature rises, non-alloys start to mix well and will form a homogeneous mixture, and as a result such mixtures may still be useful for the EOR. Our results show how atomistic modeling can predict stability and activity of potential alloy catalysts.

Creator