京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第22684号 / 理博第4625号 / 新制||理||1665(附属図書館) / 京都大学大学院理学研究科化学専攻 / (主査)教授 北川 宏, 教授 吉村 一良, 教授 有賀 哲也 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

As an important chemical raw material, alcohols can be used as fuels, solvents and chemical feedstocks to produce a variety of downstream products. With limited fossil fuel resources, alcohol synthesis from syngas reactions can be a potential alternative to the traditional petroleum based alcohol synthesis. Among many catalysts active for syngas to alcohol processes, alkali promoted Mo2C has shown promising performance. More interestingly, the alkali promoter was found to play an important role in shifting the reaction selectivity from hydrocarbons to alcohols. However, limited understanding of the mechanism of this alkali promoter effect is available due to the complexity of syngas reaction mechanism and low content of alkali added to the catalysts. In this thesis, we performed a comprehensive investigation of the alkali promoter effect with density functional theory (DFT) calculations as our primary tool. We first examine various Mo2C surfaces to determine a representative surface structure active to alkali adsorption. On this particular surface, we develop a syngas reaction network including relevant reaction mechanisms proposed in previous literature. With energetics derived from DFT calculations and a BEP relation, we predict the syngas reaction selectivity and find it to be in excellent agreement with experimental results. The dominant reaction mechanism and selectivity determining steps are determined from sensitivity analysis. We also propose a formation mechanism of alkali promoters on Mo2C catalysts that shows consistency between experimental IR and DFT computed vibrational frequencies. Finally, the effect of alkali promoters on the selectivity determining steps for syngas reactions are investigated from DFT calculations and charge analysis. We are able to rationalize the role of alkali promoters in shifting the reaction selectivity from hydrocarbons to alcohols on Mo2C catalysts.

Notre dépendance à l'égard des combustibles fossiles et la diminution des ressources pétrolières nous imposent la recherche de sources renouvelables d'énergie et de produits chimiques. La synthèse Fischer-Tropsch permet de répondre à la demande en carburants propres et renouvelables grâce à l'utilisation de gaz de synthèse issu de la biomasse. L'objectif de ce travail est de contribuer à la compréhension du mécanisme de l'hydrogénation du CO sur des catalyseurs au cobalt et à l'identification du site actif par des études spectroscopiques DRIFTS operando. Ce système permet d'observer les différentes espèces adsorbées à la surface du catalyseur pendant la réaction : CO pontés et linéaires, formiates, carboxylates et hydrocarbures. Nos travaux ont montré qu'une fraction des formiates dits « rapides» peut expliquer la formation du méthanol dans nos conditions de réaction. L'ajout dans le mélange H2+CO d'un élément minéral typique de la biomasse, le chlore sous forme de trichloréthylène, a révélé que, l'activité diminuait. La bande des CO pontés étant la plus impactée et se déplaçant vers les hauts nombres d'onde, l'effet du chlore a été notamment associé à un effet électronique sur le cobalt. L'adsorption du chlore étant réversible, nous avons également étudié l'effet de l'étain. Ce métal n'adsorbe pas le CO dans nos conditions et peut s'adsorber à la surface du cobalt. L'étain empoisonne sélectivement la formation des CO pontés et limite fortement la chimisorption de l'hydrogène. Une relation linéaire entre la vitesse de formation des produits et la proportion de CO pontés est observée, révélant l'importance des CO pontés pour la réaction d'hydrogénation du CO / Our dependence on fossil fuels and the decrease of oil resources warrant the search for renewable energy sources and chemicals. Fischer-Tropsch synthesis enables meeting the requirements for cleaner and renewable fuels through the use of syngas obtained from biomass.The objective of this work was to contribute to the understanding of the mechanism of CO hydrogenation on cobalt-based catalysts and the identification of the active site by operando DRIFT spectroscopy. Different species were adsorbed on the surface of the catalyst under reaction conditions: bridged and linear CO, formates, carboxylates and hydrocarbons. Our resutls shows that so-called “fast formate” can account for the formation of methanol under our reaction conditions. The study of a typical biomass element, chlorine, revealed that the activity decreased under trichloroethylene,. The CO bridged band being the most affected band and shifting to higher wavenumber, the chlorine effect was partly associated with an electronic effect on cobalt. Chlorine adsorption being reversible, tin poisoning was also studied. This metal does not adsorb CO under our conditions. Tin addition to cobalt selectively poisons bridged CO and greatly limits the chemisorption of hydrogen. A linear relationship between the rate of formation of products and the proportion of CO bridged is observed, highlighting the importance of CO bridged