Theoretical Study Of Electrocatalytic Oxygen Reduction Reaction On Pt And PtM Alloys

Since William Grove invented the first crude fuel cell in1839, this new energy technology has been increasingly received great attention. A fuel cell is a device that converts the chemical energy from a fuel into electricity through a chemical reaction with oxygen or another oxidizing agent. As the energy conversion in fuel cell is not a combustion process, and thus it is not subject to the Carnot cycle limit. With higher energy conversion efficiency, less environmental pollution and other significant advantages, the fuel cell technology meets the demand for increasing energy efficiency without polluting the environment. At present, the main purposes of fuel cell research are to reduce the cost of electricity output, as well as to increase stability and power density, in order to satisfy the growing energy demand. Proton exchange membrane fuel cell (PEMFC) is an important fuel cell. As the PEMFC has low operating temperature, fast startup, high specific power, simple structure, convenient operation, etc., it is recognized as the preferred energy of electric automobiles and stationary power plants. Along with the development and application of the key materials, such as perfluorinated sulfonic acid proton exchange membrane and carbon supported platinum, the battery performance and lifetime of PEMFC has significantly increased, and reached the standard of the electric cars. Currently there are hundreds of PEMFC-powered cars, submarines, power plants running at home and abroad, which are used as demonstration.Platinum, as an indispensable electrode material, plays a central role in modern fuel cell applications. For example, the most active catalyst of oxygen reduction reaction (ORR) on cathode in PEMFC is Pt alloys (Pt3Ni), but Pt still cannot be replaced in real application systems of PEMFC. However, the high overpotential (>300mV) in oxygen reduction reactions, which occur in all PEMFC on Pt cathodes, limits the energy efficiency of fuel cells significantly. Moreover, because of the agglomeration and dissolution of metal particles, the lifetime of Pt cathodes places another severe constraint on the long-term application. Therefore, study of ORRs on cathode in fuel cells is considered to be the key to reduce the cost of Pt-based catalyst. If we can understand and identify the active center of the reaction on the atomic scale and establish the mechanism of the reaction, it will provide the correct theoretical guidance of selection and design of new catalysts, as well as accelerate the study progress and save a lot of resources.In recent decades, because the first principle method is able to provide and predict some important chemical properties, it is widely applied to simulation of chemical reactions, such as geometry of intermediates, adsorption energy, energy barrier of the elementary reactions and so on. Nowadays, theoretical studies have achieved great success to describe the reaction on solid/gas interfaces. However, due to the complexity of electrochemical environment, the simulation of electrochemical environment and electrocatalytic reaction is still a challenging area, where more complex problems in the system should be considered. Taking ORRs for example, many factors should be taken into account, including the type and coverage of surface species, salvation effect, hydrated proton, the voltage influence on reaction kinetics, etc. So far, few theoretical research study all the important factors within one framework.In this dissertation, we carry out theoretical study of ORRs on Pt and Pt alloys, through a combination of density functional theory (DFT) calculations and the recently developed continuum salvation model based on modified Poisson-Boltzmann electrostatics (CM-MPB). We try to build a theoretical framework, which can help us completely understand the activity and stability of ORRs, and provide the basis for rational design of ORR catalysts. The key problems of our research include:(1) the internal relations between the catalyst component and its activity and stability;(2) the detailed pathway of ORRs, overpotential and the cause of interesting Tafel behavior;(3) the influences of nanoparticle size and morphology on reaction activity.To date, many alloy materials composed of Pt and other elements, namely PtM alloys, have been tested experimentally (e.g. M=Ni, Co, Fe, Ti, V, Sc, Y, Cu), but it remains high challenges to find alternative catalysts better than Pt with both high activity and high stability. This is partly because the ORR catalyst is required to activate O2under corrosive conditions (the oxidative, acidic and high potential state (～1V vs. SHE)). Fundamentally, due to the lack of an atomistic image of the process, no reliable measure for efficiently screening catalysts is available for the rational selection of the alloying element and the identification of the optimal ratio between Pt and M.We outline the key factors affecting the stability and activity of Pt-based ORR catalysts theoretically. By analyzing a group of Pt2M skin alloys, we show that the surface stability at high electrochemical potentials is a sensitive measure for a first screening of materials, owing to the fact that the dissolution of surface atoms in alloys is significantly facilitated with the increase of O coverage. Because of the similarity of the ORR mechanisms between Pt-based alloys and Pt, the free energy barrier to OOH dissociation, an activity index, can serve for the fine tuning of catalyst composition. From more than20different Pt2M alloys, we identify Mo as a good element, which can alloy with Pt to make a stable surface at high electrochemical potentials. We further show that Pt2Mo possesses a lower kinetic barrier for oxygen reduction (OR) compared to Pt, and the barrier towards OOH dissociation is a key kinetic parameter for OR activity. We show that the key difficulty in the design of OR cathodes is in fact not to achieve a high activity, but to maintain a high stability of the metal surface under the corrosive cathode conditions. Our work demonstrate that it is feasible to evaluate and screen electrocatalysts through kinetic calculations.In fact, a complete microscopic mechanistic picture of the ORR is still under debate. Although it is generally regarded that ORR is comprised of three key elementary steps, namely, O2bond breaking, O reduction and OH reduction, a great complexity arises owing to the unique electrochemical conditions and the exact nature of the rate-determining step in ORR is largely elusive. For example, O2dissociation can proceed via either the direct bond breaking (O2adâ†'O+O), or the proton-coupled bond breaking (H++e-+O2adâ†'O+OH). Early experimental studies often assumed that O2dissociation is the rate-determining step as it was observed that the reaction order with respect to molecular oxygen is one. Along this line, a dual-pathway mechanism for O2activation involving the direct and the proton-coupled bond breaking is often suggested in literature, both from experiment and theory. However, with the knowledge from recent surface science studies and theoretical calculations, this assumption is also questioned because O2bond breaking on bare Pt surfaces are found to be facile (can occur at low temperatures, e.g.200K). Instead, other reaction steps were proposed as the rate-determining step, such as the reduction steps involving oxygen atom or hydroxyl groups and the proton transfer to the compact layer. Interestingly, in contrast to the large uncertainty on the overall mechanism, the kinetics of ORR, as represented by the measured Tafel curve, is relatively simple and definitive. According to the observed Tafel curve, i.e. log(j) vs. η(current vs. overpotential) curve for ORR on Pt, two linear regions are identified, one below～0.8V (high η region, where η≥0.43V) with a Tafel slope of～120mV, another above～0.8V (low η region, where η<0.43V) with a slope of-60mV. For ORR at low potentials, the uncertainty lies mainly on whether the proton-coupled O2dissociation or the OH reduction is rate-determining, as both reactions involve only one electron transfer and thus can yield a Tafel slope of～120mV according to the Butler-Volmer relation (j=joe-aηF/RT and the Tafel slope b=2.3RT/aF; with a assumed to be0.5, b=120mV). On the other hand, ORR at the high potentials is more intriguing as a Tafel slope of～60mV implies that more than one electron/elementary-step contributes to the overall kinetics.In this work, the O coverage at the reaction equilibrium, i.e.0.25ML, is identified from ORR kinetics on Pt at the concerned potentials, e.g.～0.8V. A0.25ML O coverage is therefore identified as the low limit for the surface O coverage and hereafter utilized as the initial surface O coverage for investigating the ORR kinetics. Aiming to settle down the reaction mechanism and resolve the key issues on the potential-dependent kinetics, we determine the ORR Tafel kinetics and the polarization curve under electrochemical conditions theoretically. Both the slope and the switch in the curvature are reproduced from theory. Since the Tafel curve perhaps represents the most important kinetic information available from experiment, this theoretical work provides important insight into the proton-coupled electrochemical reaction at the atomic level and demonstrates that the computational electrocatalyst evaluation/screening is now reachable by kinetic information.We demonstrate that the proton-coupled0-0bond breaking (H+e-+O2adâ†'O+OH) is generally the rate-determining step in ORR at concerned potentials. The physical origin of the catalyst poisoning is identified as the competition of OH (from water oxidation) with O2in adsorption at high potentials. The theory here shows that it is possible to markedly reduce the utilization of Pt in active catalysts because only two Pt surface atoms are involved in the key0-0breaking step, while the rest of surface Pt atoms are inactive, terminated by in-situ produced O atoms. Furthermore, it is found that the physical origin for the switch of Tafel slope above～0.85V is the formation of hydroxyl groups on the surface, which competes with the adsorption of molecular O2. The water itself thus acts as the poisoning species at the high potential, when each hydroxyl group on the surface will pin at least two nearby water molecules forming a stable first layer H-bonding network that together with adsorbed O atoms terminates fully the surface. How to avoid the production of hydroxyl at the high potentials is thus a key challenge towards reducing the high overpotential of ORR.In addition, in contrast to single crystal electrodes, industrial catalysts are usually composed of finely dispersed nanoparticle catalyst anchored onto a support. The relationship between structure and activity of the catalyst also rank one of the top concerns in the research of heterogeneous catalysis.On nanoparticle surface, there often exist vertices, edges, and a variety of crystal faces. It is difficult to obtain accurate microscopic electrochemical conditions of Pt nanoparticles experimentally. However, the existing theoretical studies on this issue are generally based on molecular cluster models containing only a small number (no more than35) of atoms, or based on periodic slab/supercell models, which are both different form the nanopartcles. That’s why there are still some key issues for the ORR on the Pt nanoparticles which remain elusive, e.g. weather the ORR activity on Pt nanoparticles located in (on) the{111} or{100} surface, and what is the relationship between Pt nanoparticle size and its activity? In order to study the intrinsic relationship among the particle size, structure and activity of Pt nanoparticles theoretically, we construct the theoretical models of the Pt nanoparticles with different sizes and shapes. By drawing the surface phase diagram, we provide the surface structure of the adsorbed species on the surfaces of Pt nanoparticles under electrochemical environment for the first time. Combined with the basic understanding of ORR on Pt single crystal surface from the former part, we further calculate the surface specific activity (SA) and the mass activity (MA) of Pt nanoparticles. We find that under the working electrochemical potential at0.9V, the truncated octahedral Pt nanoparticles with～2nm D(111)have the highest ORR mass activity.