Simulations For The Hydrogen Permeation Processes And Relevant Material Properties In Ceramic Proton Conductors

Ceramic proton conductor materials are intensively studied for their promising applications, such as hydrogen separation, methane reforming, water gas shift and solid oxide fuel cells. New materials are discovered successively, and applied researches are going on gradually. The progresses on the experiments impel the proposal of models and mechanisms. While the development of simulation could give deeper insights into the new materials and relevant applications, and promotes the experiments in return. Therefore, experiment and simulation supplement each other. In this article, simulations based on ceramic proton conductor materials are performed, in combination with reported experimental results. The following topics are discussed, namely, models for metal-ceramic composite hydrogen separation membranes, crystal structure and proton migration pathways in new proton conductor La2Ce2O7, and a concise model for proton conductor electrolyte fuel cells.In chapter1, the background of the whole article is presented, which is divided into two parts:First one is the introduction of proton conductor, focusing on ceramic proton conductor materials and their applications. Second one is a brief summary of material modeling in various levels. Some different kinds of simulation methods are introduced briefly, as well as relevant softwares and tools.In chapter2, models for hydrogen separation membranes are discussed in detail, including cases of both symmetric and asymmetric structures, taking Ni-BZCY composite as example. Expressions and equivalent circuits for each case are obtained separately. Hydrogen permeation rates under a wide range of experimental conditions could be predicted by this model. Performance bottleneck is also determined, which indicates that, the rate limiting step varies from bulk diffusion to surface exchange as the dense membrane thickness decreases. Besides, the proportion of hydrogen flux from Ni phase is at least one magnitude lower than that from ceramic proton conductor phase. For those samples with thin dense membranes, the contribution from Ni is noticeable. As to the asymmetric membranes, the model results demonstrate that, concentration polarization within support is neglectable; while interfacial polarization is much more severe, especially when the dense layer is fairly thin. The increment of permeation rate is quite unproportionate to the reduction of dense layer thickness for membranes thinner than several ten micrometers. On the contrary, methods that enhance the interfacial reactions are more effective, such as grain refining and surface modification. The model may provide some suggestions to the fabrication of hydrogen separation membrane.In chapter3, crystal structure and proton migrations within La2Ce2O7are discussed with the employment of density functional theory. So far, incompatible conclusions on the stable structure are made in previous works, and discussions about proton transportations are scarce. Hereby, judgment on the stable structure is decided, a possible mechanism that resulting in disordered sublattice is proposed. Proton transfer pathways in bulk material are searched; and supercells for surface slabs are built to analyze the effect of Sm doping on interfacial processes. First, By comparing between pyrochlore ordered and fluorite disordered configurations, it turns out that: The pyrochlore distorted structure is not the lowest in energy. The formation of O48f-O8a type anion Frenkel defects tends to stabilize La2Ce2O7and is the induction of disordering. Whereas O48f-O8b type anion Frenkel defects are unfavorable. Since the rise in energy due to cation anti-site is significantly compromised after the formation of O48f-O8a type anion Frenkel defects, the cation anti-sites are also likely to be present in bulk La2Ce2O7at room temperature. However, they are not as dominant as anion ones. This is consistent with previously reported neutron diffraction pattern. Second, stable proton sites and proton reaction pathways in bulk La2Ce2O7are calculated under both ordered pyrochlore and stable disordered fluorite configurations. The results show that the distance between two nearest oxygen atoms is shortened with the presence of an interstitial proton. The calculated energy barriers of proton transfers within bulk material are less than0.6eV, which are lower than experimental formal activation energies. It is likely that the partial structural disorder disfavors the activation energies of proton migrations because of the existence of higher energy barriers. Third, interfacial processes such as hydrogen adsorption and dissociation are discussed, by applying supercells containing Ni, La2Ce2O7slab and vacuum. The relevant energy barriers are above0.7eV, which could explain the reason why the measured formal activation energy is higher than the calculated proton transfer barrier within bulk. Fourth, Sm dopant is added to the above mentioned supercells. It is discovered that the energy barrier of superficial hydrogen transfer is reduced, which facilitates the interfacial steps. The result is supported by the observed phenomena.In chapter4, a concise model for anode support solid oxide fuel cell with proton conductor electrolyte is built. To make use of experimental data, the simulated samples take BaZr0.1Y0.2Ce0.703-δ as proton conductor electrolyte, and Sm0.5Sr0.5CoO3-δ-Sm0.2Ce0.8O2-δ composite as cathode. According to the results, most overpotential derives from electrolyte ohmic resistance and cathode reaction polarization. Besides, cells with proton conductor and oxygen ion conductor electrolyte are compared in the aspect of electrode concentration polarizations.Chapter5is the summary of this article.