Applications Of Density Functional Theory Calculations In Electrocatalytic Reaction And Electrocatalyst Design

Developing catalysts for oxygen reduction reaction(ORR) with low noble metal contents,high catalytic activity and high stability is the key for large-scale application of proton exchange membrane fuel cell (PEMFC). Rational design of catalysts relies on in-depth understanding of ORR mechanism. Since the "electrode/electrolyte" interfaces are mostly inaccessible to low temperature and ultra-high vacuum conditions, it is difficult to experimentally obtain molecular information of ORR with surface-science techniques. At the same time, the traditional "synthesis-test" catalyst searching mode is a rather cost-ineffective process. However, modern computational chemistry based on density functional theory (DFT) and high-performance computing technology offers a novel opportunity for studying electrocatalysis at the molecular level. In this thesis, we have performed detailed DFT calculations studies on the ORR electrocatalysis, including the reaction mechanism and catalysts design. Considering the increasing importance of the chemical transformation of CO2 in the area of energy and environment, we also performed DFT calculations studies on the chemical and electrochemical reduction processes of CO2. The main contents and results are summarized as follows:1. DFT Study of ORR Mechanism on Pt(111) SurfaceThe DFT geometric optimization, electronic structure and minimum energy paths calculations have been used to investigate the adsorption and dissociation of O2 molecule, and the protonation of the dissociated adsorbates. The results indicated the following:(1) Two molecular chemisorbed intermediates are involved in ORR:the protonated end-on state of OOH* and the unprotonated t-b-t state of OO*. The end-on OOH* is metastable, which can transform to the more stable t-b-t molecular state or dissociates to atomic adsorbates in nearly nonactivated processes. That is, O2 may undergo a sequential protonation and de-protonation process in the earlier stage of reduction; (2) In the presence of the hydrated proton, the chemisorption of O2 molecule is a strong exothermic process, releasing energy larger than the activation energy required for the subsequent dissociation. Therefore, O2 molecule could dissociate without external energy input. In the entire four-electron ORR, the protonation of adsorbed 0 atom to form OH is the slowest step, therefore the rate-determining-step(rds). Such a finding about the rds of ORR can well explain why catalysts that bind atomic oxygen more weakly have better ORR activity, which can not be well accounted for with the earlier view that the rds of ORR is the formation of OOH*.2. Theoretical design of ORR catalysts based on DFT calculationsIn the light of the above reaction mechanism study, we have proposed a multiple-descriptor strategy for rational design of efficient and durable ORR alloy catalysts with low precious metal content. We argued that good alloy catalysts for ORR should simultaneously have negative alloy formation energy, negative surface segregation energy of precious metals and lower oxygen binding strength than pure Pt. By performing detailed DFT calculations on the thermodynamics, surface chemistry and electronic properties of Pt-M and Pd-M alloys (M refers to the non-precious transition metals in periodic table), Pt-based alloys including Pt3V, Pt3Fe, Pt3Co, Pt3Ni, Pt3Cu, Pt3Zn, Pt3Mo, Pt3W and Pd-based alloys including Pd3V, Pd3Fe, Pd3Zn, Pd3Nb, and Pd3Ta were predicted to have improved catalytic activity and durability for ORR, among which some alloys have indeed been reported to have excellent ORR catalytic activity in the literature.The origins for ORR activity enhancement in these alloys have been analyzed in terms of the lattice compression effect and the ligand effect of the alloying transition metals on the electronic property of surface atoms. The results indicated that the lattice compression leads to downshift of d-band center of surface atoms, while the ligand effect could lead to either downshift or upshift of d-band center depending on the lattice and electronic properties of the transition metals. For metals with face-centered cubic lattice and ten d electrons, the ligand effect causes upshift of d-band center of Pt, while the ligand effect of the transition metals with hexagonal lattice structure and ten d electrons would downshift the d-band center of Pt. For transition metals with less than ten d electrons, the ligand effect mostly results in downshift of the d-band center of Pt.3. DFT Calculation Study on the Reduction of CO2 to Hydrocarbons on Cu SurfacesCO2 reduction processes on Cu(111) and Cu(100) single crystal surfaces were studied with DFT calculations on the reaction energy and the minimum energy paths. The results indicated that the possible reaction paths for CO2 reduction on Cu(111) surface are CO2(g)+H*→COOH*→(CO+OH)*, (CO+H)*→CHO*, CHO+H→CH2O*→(CH2+O)*, CH2*+2H*→CH4 or 2CH2*→C2H4, while on Cu(100) surface are CO2(g)→(CO+O)*, (CO+H)*→CHO*, CHO+H→CH2O*→(CH2+O)*,2CH2*→C2H4. On Cu(111) surface, the reaction rate is controlled by steps of CH2O*→(CH2+O)*, CO2(g)+H*→COOH→(CO+OH)* and (CO+ H)*→CHO*, while on Cu(100) surface the rate reaction is controlled by step of (CO +H)*→CHO*.The reaction energies for various steps in the electrochemical reduction of C02 were calculated under different electrode potentials. The results indicated that HCOO and CO are mainly formed when the potential is more positive than-0.50V (vs.RHE). The hydrogenated dissociation of C02 to form CO and the subsequent hydrogenation of CO become increasingly exothermic as the potential goes negative, so that hydrocarbons gradually becomes the favored products in the electrochemical reduction. Under electrochemical conditions, CHO intermediate prefers to dissociate to form CH, rather than to form CH2O intermediate via protonation as does in gas phase reduction. Agreeing with experimental results, our calculations indicated that is preferably formed on Cu(100) surface.