Theoretical Study Of Methanol Steam Reforming Catalyzed By PdZn Alloy

With fossil fuels becoming rare everyday, renewable energy attracts more and more attention. As one of the most promising substitutions, hydrogen energy has many merits such as high-efficiency, condensed-density and zero-emission etc. Methanol is an ideal energy carrier which can produce hydrogen in situ for on-board fuel cells. One proposed method is the methanol steam reforming (MSR, CH3OH+H2Oâ†'CO2+H2).Iwasa et al. discovered that Pd/ZnO catalyst displays good stability and exhibits high activity and selectivity towards MSR reaction. Quite interesting, on pure Pd metal, MSR produces exclusively CO. To gain insight into understanding the selectivity difference with Zn modification, it is necessary to make clear the reaction mechanism of MSR on PdZn alloy. Traditional point of view believed that1:1Pd-Zn alloy is the active components. However, recent study proposed that the1:1PdZn alloy might have no activity toward MSR, whereas Pd(111) surface deposited with0.03-0.06ML Zn may be most active. This speculation makes us before MSR reaction mechanism exploration, primary task is to investigate the effect of Zn on the activity of Pd(111) surface and evaluate the activity of Pd(111) surface with low Zn deposition towards MSR.As a participator among the MSR process, a thorough understanding of the behavior of water on PdZn alloy is a prerequisite for obtaining comprehensive knowledge of MSR. Previous theoretical study reported an adsorption energy of water to be only23kJ/mol on the (111) surface of PdZn bulk alloy. Such small adsorption energy implies that water dissociation is difficult on the PdZn(111) or the PdZn(111) is not the active sites for water activation. Recent experiments demonstrate that while water keeps intact on1:1PdZn monolayer alloy, dissociation of water takes place on multilayer of1:1PdZn alloy actively. This experimental conclusion seems to contradict the abovementioned theoretical predictions. We cannot help but asking:is the conclusion that water dissociation occurs on PdZn multilayer correct? If yes, why the multilayer and monolayer PdZn display such striking different behavior toward H2O activation? If not, where can H2O be activated?With these questions in mind, in this dissertation we mainly use density functional theory (DFT) combined with Monte-Carlo (MC) technique to investigate the effect of Zn on the adsorption and reaction of Pd(111), to evaluate the activity on Pd(111) surface with low Zn deposition towards MSR and finally to reveal the microcosmic mechanism of the MSR process.The primary investigation contents are briefly introduced as follows.Firstly we present a systematic theoretical study of the change of surface reactivity induced by incorporation of Zn by examining the CO adsorption on PdZn surface alloy. The results show that Zn atoms in the topmost layer have smaller effect on CO adsorption than the Zn in the second layer, especially for hollow sites. The reason for this phenomenon originates from the strain effect and ligand effect that have different influences on CO adsorption. When Zn atoms are doped on the topmost layer of Pd(111), the strain effect that favors the bonding of CO to the surface alloy offsets almost completely the ligand effect which is unfavorable for CO adsorption. At variance, when Zn atoms are deposited on the subsurface, the strain effect is much weak and cannot standoffs the ligand effect which, in this case, dominates the CO adsorption.To further understand the modification of Zn on the reactivity of Pd(111), we carried out a systematic study of methanol dehydrogenation on a series of Pd-Zn surface alloys using DFT. We find that the variation of binding energy on different substrates can be well rationalized with distribution of Zn:the top-layer Zn enhances the interaction of species like CH3O that binds to the substrate via the oxygen atom. For such species the binding energy increases with the increase of the surface Zn content. For those adsorbates that adsorb on the substrate mainly through the C-Pd interaction, the binding energy decreases with the increase of subsurface Zn concentration. Addition of Zn reduces the activation energy of O-H/C-H bond breaking of CH3OH/CHO whereas it raises the energy barriers of dehydrogenation of CH3O and CH2O. Compared with previous DFT results, we suggest that Zn atoms beyond the third layer have essentially no influence on the adsorption and reaction at the surface.Jeroro et al. observed in the TPD experiment two desorption peaks of CH2O on Pd(111) surface on which low Zn is deposited. One peak centers at210K which is derived from CH2O adsorbed on Pd(111) or normal PdZn alloy surface, and the other at360K. The latter peak corresponds to a binding energy of1.0eV estimated with Redhead formula. To gain the knowledge of surface structure, we performed MC simulation. The results demonstrated that at very low zinc coverage, small surface ensembles of3to5atoms exist preferentially on the surface. Based on these findings, we constructed a series of surface cluster models to mimic the adsorption and reaction of CH2O,"an indicator of MSR mechanism". The results reveal that the360K desorption peak is originated from the formaldehyde adsorbed on the small surface clusters with Pd and/or Zn content. Investigations of CH2O dehydrogenation to CHO and formation of H2COO from CH2O and O show that the supported Zn clusters (very likely as well as clusters dominated with Zn) are most likely the speculated active phases if the Pd(111) surface deposited with0.03to0.06ML Zn is really active for MSR reaction. However, on the surface Zn clusters dehydrogenation of CH3O to CH2O is harder than on1:1alloy surface. Further, the stability of such surface clusters is much lower. Hence, surface Zn clusters, even though they exist in realistic catalysis system, they will not contribute significantly to MSR reactionInvestigations of water adsorption and dissociation in various aggregation forms on multi-and monolayer surface alloy of both flat and stepped surfaces reveal that both the PdZn multi-and mono-layer can activate H2O. On multi-flat PdZn(111) surface, aggregation favors H2O dissociation. In most cases, monolayer surface alloy is more active for water dissociation than the multilayer. On stepped PdZn(221) surfaces, H2O is favorably dissociated on exposed Zn step. Contrary to the point of view of the Austria research groups that the PdZn monolayer cannot activate H2O, our first-principles results clearly demonstrate that as long as the multilayer surfaces are able to dissociate water, water dissociate can take place on the monolayer surface without question. This discrepancy needs further studies from both experimental and theoretical sides. Furthermore, we find that the traditional distinction between covalent and hydrogen bonds is faint in OH-H2O overlayers on PdZn substrate which we have highlighted.Based on the above theoretical studies and the pertinent experiments, we employed the1:1PdZn alloy model to mimic Pd/ZnO catalyst and explored systematically the MSR mechanism. Calculations show that the MSR reaction proceeds from dehydrogenation of CH3OH to CH2O via CH3O, subsequently CH2O facilely reacts with OH or O to yield H2COOH or H2COO species. The yielded products experience stepwise dehydrogenation to produce CO2. The rate-determining step is the dehydrogenation of CH3O. The presence of OH always makes the C-H and O-H bond breaking more exothermic or less endothermic. However, kinetically, the coadsorbed OH reduces the O-H scission barrier, but does not facilitate or even inhibits the C-H rupture. Our results present a vivid microscopic picture of the MSR reaction. They are deemed to shed light on experimental observations, provide detailed thermodynamic and kinetic parameters for further microkinetic simulation of MSR process, and are informative for optimizing the MSR reaction condition and for designing new MSR catalysts of high quality.To recap, this dissertation elucidates the modification of Zn on the chemistry of Pd(111). The Pd(111) surface modified with low Zn concentration is demonstrated to be unimportant for MSR reaction. We show that monolayer surface alloy is more active for water dissociation than the multilayer alloy surface. Finally we furnished a complete energy profile of the MSR process on PdZn(111) surface.