Multi-Scale Simulation of Surface Segregation and Oxygen Reduction Reaction on Platinum Alloy Surface: Density Functional Theory, Monte Carlo Simulation, and Kinetic AnalysisDuan, Zhiyao (2013) Multi-Scale Simulation of Surface Segregation and Oxygen Reduction Reaction on Platinum Alloy Surface: Density Functional Theory, Monte Carlo Simulation, and Kinetic Analysis. Doctoral Dissertation, University of Pittsburgh. (Unpublished)
AbstractProton-exchange membrane fuel cell (PEMFC) is an electrochemical device for directly converting chemical energy stored in hydrogen fuel to electricity at low temperature. In order to improve the efficiency and reduce the cost of PEMFC, numerous research efforts have been devoted to developing cheaper yet more efficient electrocatalysts by alloying Pt with 3d transition metals. A number of Pt-based alloys were identified to have better activity for catalyzing ORR than pure Pt catalysts. Although Pt-based catalysts have better ORR activity, the underlying reasons for the improvement are unclear and controversial. Atomistic simulation can provide molecular-level information of physical and chemical processes in materials. In my research works, the state-of-art simulation methods were employed to elucidate the surface structures of Pt-based alloys, reaction mechanism of ORR on Pt and Pt-based surfaces, and degradation of Pt and Pt-based nanoparticle catalysts. Based on the Firstly, surface segregation phenomena in Pt3Ti, Pt-Pd, and Pt3Fe alloys were investigated with density functional theory (DFT) and Monte Carlo (MC) simulations. Through the computational study, the driving forces and mechanisms of surface segregation were clarified, and the surface composition profiles were quantitatively predicted. For example, the DFT simulation suggested that off-stoichiometric effect accounted for the experimentally observed Pt segregation to the outermost layer of the Pt3Ti (111). Our MC simulations predicted in a Pt3Ti (111) sample with a Pt concentration slightly above 75 at. %, Pt atoms would segregate to the surface to form a pure Pt outermost layer, while the ordered Pt3Ti crystal structure would be maintained in the second layer and below. The knowledge of Secondly, first-principles DFT calculations was employed to elucidate the reaction mechanism of ORR on Pt and Pt/M (111) and (100) surfaces (M = Ni, Co, Fe). The binding strengths of chemical intermediates involved in ORR are less strongly on Pt/M surfaces compared to the pure Pt couterparts due to the modified electronic structure of the Pt overlayer by the subsurface transition metals. ORR mechanism is also shifted on modified Pt overlayers. It was found that ORR proceeds through OOH dissociation mechanism on Pt (111) surface, while on Pt/M (111) surfaces ORR proceeds through HOOH dissociation mechanism. The significance of the changed ORR mechanism is that ORR activity measured by the barrier of rate-determining step is greatly enhanced on Pt/M (111) surfaces. For example, on Pt/Ni (111) surface, O2 hydrogenation is the rate-determining step with a barrier of 0.15 eV compared to O hydrogenation with 0.79 eV on Pt (111) surface. We In the third part, kinetic Monte Carlo (KMC) algorithm is implemented to study the kinetics of ORR based on the mechanistic information obtained in the second study. The information of the elementary reactions involved in ORR such as the adsorption sites of the reactants and products, activation energies, etc. is input into the KMC code. The KMC simulation can simulate the dynamics of ORR and output the current density (joules/cm2/s) generated from the reactions. Then, the simulated current density which is a measure of ORR activity can be directly compared to experimental measurement. In the study, kinetics of ORR on Pt (111) and Pt (100) surfaces were simulated. The simulated current density of ORR on Pt (111) and Pt (100) at electrode potential 0.8 V is in the same magnitude Expect for the activity issue, Pt nanoparticle catalyst also faces degradation problem due to the highly oxidizing environment in the cathode of PEMFC. In the final part, the degradation of Pt nanoparticle catalyst through Pt dissolution is studied employing grand-canonical Monte Carlo (GCMC) simulation. Pt dissolution process was found to be initialized through the dissolution of under-coordinated Pt atoms sitting on the corners and edges of the nanoparticle. After the initial dissolution of Pt atoms on corners and edges, more under-coordinated Pt atoms are generated and the dissolution process is accelerating. The smaller Pt nanoparticle is more vulnerable to the Pt dissolution process than the larger nanoparticle. A Pt nanoparticle with about 5 nm diameter is stable in the environment. It was also found that Au atoms segregated to the under-coordinated sites would stabilize the nanoparticle Share
Details
MetricsMonthly Views for the past 3 yearsPlum AnalyticsActions (login required)
|
![[feed]](/images/feed-icon-32x32.png){kind=icon}