Theoretical Insights into the Light Alkanes Dehydrogenation and Aldehydes Hydrogenation on Transition Metal and Metal Oxide Surfaces

Because of their extensive use as chemical building blocks, light olefins, such as propylene and ethylene, are among the essential types of compounds in the chemical industry. As a result, the demand for these building blocks has increased steadily over the last few years. For instance, the direct formation of propylene from propane is a well-established commercial process, which, based on energy consumption, is environmentally preferred to the current large-scale sources of propylene from steam cracking and fluid catalytic cracking. Moreover, there is still a big window for catalyst improvement, such as \ce{C-H} bond activation, reducing propane and hydrogen adsorption on surface sites, and minimizing coke formation.

By using density functional theory (DFT) calculations, it is investigated the C--H bond activation of light alkanes (methane, ethane, propane, \textit{n}-butane) on transition metals (TMs), metal oxides (MOs), and single-atom-doped-metal oxides (M$_1$-MOs) surfaces, as well as the hydrogenation of aldehydes on palladium surfaces. This thesis develops simple and highly accurate models for predicting the transition state energy $\left( \Delta E_{TS} \right)$ of the C--H bond activation of light alkanes (C$_1$-C$_4$) on TMs, MOs, and M$_1$-MOs using the final state energy $\left( \Delta E_{FS} \right)$ as a descriptor. ... mehrIn the case of the TM surfaces, the models cover the non-oxidative, O-, and OH-assisted C--H bond activation on closed-packed and stepped surfaces. Here, the variations in $\Delta E_{TS}$ between alkanes were primarily attributed to differences in dispersion contributions determined by the carbon-chain length. In the case of MOs and M$_1$-MOs, the C--H bond activation can be explained based on the Lewis acid-base properties of the surfaces. The linear scaling relationships (LSRs) generated are universal with no functional restrictions and cover a broad structural diversity of MOs and M$_1$-MOs catalyst surfaces, including the effect of superficial oxygen vacancies, several dopants, and more than one phase of the same MO, and different active sites. Thus, it is confirmed that the $\Delta E_{FS}$ is one of the most general descriptors to estimate $\Delta E_{TS}$ on TMs, MOs, and M$_1$-MOs when LSRs are employed. The LSRs generated in this work are expected to pave the way toward the computational development of new and better catalyst materials, guiding future experiments as a first-hand tool in heterogeneous catalysis.

Finally, the fundamental understanding offered by DFT calculations serves as a cornerstone in the complete process of computer-aided catalytic design. This is demonstrated by combining adsorption studies (substrate adsorption and temperature-programmed desorption) and DFT calculations, which are used to elucidate the factors affecting the hydrogenation of various aldehydes on a Pd-based catalyst. The importance of the side chain in determining the reactivity of the carbonyl group is demonstrated. Although the adsorption mode and strength are affected by the substrate structure, these aspects have proven not to be the decisive factors in the conversion of the carbonyl group.