The main objective of this work is to understand the theoretical basis of the working principle of the Hydrogen fuel cell. We seek the physical basis of the Rational Design Technique, the smart way of preselecting materials from the material-pool, implemented in our study anticipating highly promising electrocatalysts for promoting the conversion of chemical energy stored in hydrogen molecules into the electrical energy. It needs the understanding of the relationship among the compositions of the materials under consideration, their electronic structure and catalytic activities. We performed the first principle DFT calculations to achieve the goal. Our work is focused first on the issues in hydrogen oxidation reaction taking place in anode compartment of the cell. Next comes up with the issues with Oxygen Reduction Reaction taking place in cathode compartment. Finally, we focus on mechanisms underlying binding of small molecules on substrates. Platinum perfectly catalyzes hydrogen oxidation reaction on the hydrogen fuel cell anodes. However, it has at least two drawbacks: a) it is too expensive; b) it has a low tolerance to CO poisoning. Pt-Ru bi-functional catalysts are more tolerant to CO, but they are still very expensive. In this work, we performed first-principle studies of stability and reactivity of M/W (110) structures, where M = Pd, Ru, Au monolayers. All three systems are found to be stable: formation energy of MLs is significantly higher than cohesive energy of the M-elements. The calculated binding energies of H, H2, OH, CO, and H2O were used to obtain the reaction free energies. Analysis of the free energies suggests that Au-W bonding does not activate sufficiently Au monolayer, whereas Ru/W (110) is still too reactive for the CO removal. Meanwhile, Pd/W (110) is found to catalyze hydrogen oxidation and at the same time to be highly tolerant to the CO poisoning. The latter finding is explained by the fact that CO binds much weaker to Pd on W (110) than to Pt, while the OH binding is strong enough to ensure CO oxidation. The obtained results are traced to the electronic structure of the systems. Oxygen Reduction Reaction (ORR) is the heart core reaction in fuel cells, Proton Exchange Membrane Fuel cell and DEMFC. However, the reaction is not so obvious and need suitable electrocatalyst. Pt or Pt-based catalysts are found to be the best catalyst so far. But, its cost and shortage make it not feasible economically. Moreover, lower onset potential (maximal electrode potential at which the reaction can proceed) of such catalysts is offering another limitation to fuel cell performance. Research has been conducted in many directions for lowering the cost by replacing the Pt with some other elements of lower cost or reducing the Pt-load in the material; and even more finding the material performing better than Pt. In this paper, we’ve tried to understand the ORR mechanism and look for the material that could be potential option to Pt. Our calculations suggest that for monolayer of Pt on 5 layered slab of Nb or Mo the onset potential is the same as for Pt, while cost of these systems are much lower than that of Pt. Presence of water changes the reaction rate very minimum. Rational design method facilitates the research of selecting the appropriate catalyst and saves time and effort significantly. The result shows that the d-band center model is not accurate to describe the reactivity of the catalyst. For decades, adsorbates’ binding energy (EB) has been used as an indicator of the adsorbate-substrate bond strength (EBF). Thus, although one can compute accurately any E models to gauge bond-strength are developed and applied to rationalize and anticipate EB's because that is a key aspect in the rational search for efficient catalysts. Yet bond-strength alone fails to predict EB trends. Therefore, quantifying and understanding the difference between EB and EBF is essential to catalysts design. Indeed, the adsorbate-substrate bond formation perturbs the substrate’s electronic charge density, which reduces EB by the energy attached to such perturbation:Epert. Here, with the example of carbon monoxide adsorption on metal-doped graphene, we show that Epert may exceed 1 eV and render an unusual situation: although the EB of CO to the Au-doped graphene indicates that binding does not happen, we find evidence of a strong bond between CO and the substrate. Thus, in this case, the large Epert totally disrupt the equivalency between EBF and EB we also propose a method to compute Epert that bypasses dealing with an excited electronic state of the system.
Doctor of Philosophy (Ph.D.)
College of Sciences
Length of Campus-only Access
Doctoral Dissertation (Open Access)
Dhakal, Nagendra, "Computational Approach to Electrocatalysis" (2017). Electronic Theses and Dissertations. 5407.