Mechanisms and microkinetic modeling of CO₂ conversions on multi-functional catalysts

Abstract

Our heavy dependence on fossil fuels allows a large amount of CO₂ to be directly emitted into the atmosphere. There is strong evidence that the rise of atmospheric CO₂ level causes a cascade of severe environmental issues such as ocean acidification, sea level rises, long drought, and intense heat waves. This thesis explores relevant catalysis technologies that will transform CO₂ into a building block species for chemicals production. Catalytic CO₂ utilization faces major limitations because of the chemical stability of this molecule. Multiple technologies such as dry reforming of methane (DRM) and CO₂ hydrogenation have demonstrated their technological and economical potentials to overcome CO₂ conversion’s limitations. Yet, the catalysis science and technology for CO₂ utilization is far from mature. Effective and affordable catalysts suitable for industrial-scale applications are not readily available. Fundamentally, the catalytic reactivities of simple mono-functional catalysts are limited by the so-called Sabatier principle. Moreover, the best performing catalysts often rely on expensive noble metals. Catalyst discovery and design have shifted focus toward composite, bifunctional materials manufactured from earth-abundant elements. Breakthroughs have already been made in ammonia synthesis, CO oxidation, water-gas-shift reaction, and hydrogen production reactions. DRM converts CO₂ and CH₄ (both are potent greenhouse gases) into syngas, a versatile industrial mixture. Currently, DRM catalysts are challenged by inadequate reactivity and short lifetime. In this thesis, systematic investigations were carried out to understand the mechanistic origin of DRM on the dual-site models representative of real-life bifunctional catalysts. An unconventional material, Co₃Mo₃N (a ternary nitride), was the focus in this study. Earlier experimental studies indicated that Co₃Mo₃N is active and durable, but the source of its reactivity and stability remain unclear. The adsorptions, desorption, and surface reactions of DRM intermediates on the Co₃Mo₃N (111) facet were investigated using the quantum mechanical density functional theory (DFT) method. The site preferences and DRM pathways on Co₃Mo₃N are revealed for the first time regarding this catalytic material. My work yielded clear evidence that Co₃Mo₃N promotes CH₄ activation and the oxidations of surface carbonaceous species at its Co site and Co-Mo₂N boundary site, respectively. DFT calculations further showed that, due to the presence of two distinct sites, the OH and CHO intermediates that appear during DRM do not obey the linear scaling relationships, resulting in the oxidation reactions occurring at higher than usual rates. The analyses based on DFT calculations are then corroborated by the mean-field microkinetic modeling (MKM) designed especially for dual-site catalytic systems. My work concluded that bifunctional catalysts containing sites with O affinities are desirable for DRM. The MKM results further clarify that cross-site diffusions of DRM intermediates, i.e., C, O, OH, CO, and CH, play the most prominent role in mitigating coke formation. In addition, solid and liquid Ga containing well-dispersed Ni, Pd, and Ru atoms were modeled for DRM. It was found that Ru trimer embedded in Ga solid solutions yields one of the best H₂ production rates. Then, the topological cluster classification (TCC) analyses on ab initio molecular dynamics (AIMD) simulations suggested that the transition metal solutes (e.g., Ni, Pd) dissolved in liquid Ga prefer the liquid-gas interface at low or moderate temperatures. This thesis also considered indium oxide (In₂O₃) catalysts supported on Zr, Ce, and Pr oxides for methanol production via CO₂ hydrogenation. Experimentally, the highest CO₂ conversion and CH₃OH selectivity were observed on ZrO₂-supported In₂O₃ (Zr-In₂O₃). DFT calculations revealed that a unique bent configuration of CO₂ adsorption at the oxygen vacancy site (O[subscript v]) in Zr-In₂O₃ stabilizes the formate (HCOO) intermediate. Subsequent hydrogenations of HCOO to CH₂O, and then CH₃O are also thermodynamically more favorable over Zr-In₂O₃ than on other oxide catalysts. DFT modeling also showed that the product selectivity depends on the relative activation energies between hydrogenation (for CH₃OH formation) and the C–O bond cleavage (for CO formation) of HCOO. This thesis demonstrated the predictive power of DFT in elucidating the complex surface chemistries on bifunctional catalytic materials. Based on the case studies, DFT, coupled with the microkinetic modeling and molecular dynamics simulation techniques, produced highly valuable knowledge that can be elusive for other research tools. More importantly, the theoretical knowledge will allow researchers to continue the pursuit of more efficient and stable catalysts for CO₂ utilizations so that we will be better equipped to solve some of the most urgent societal issues.