Quantum mechanical studies of interactions in model catalytic surfaces

Bond making and breaking are important to heterogenous catalysis. The strength and coordination of the chemisorption bond between the surface and the adsorbate are central to surface chemical processes. Yet state-of-the-art density functional theory (DFT) methods using computationally advantageous non-hybrid exchange-correlation functionals have been shown to break down at a qualitative level in predicting these aspects of chemisorption. The class of systems for which this shortcoming is highlighted is the seemingly unexceptional chemisorption of CO on transition metals. These shortcomings inhibit use of DFT as a predictive modeling tool. In this thesis, a simple post-DFT treatment is developed and tested on a representative variety of transition metal surfaces. The success of this simple extrapolation procedure is measured against experimental data and is shown to reproduce observed site preferences. Being able to mirror experimental details of chemisorption is not the only utility of the extrapolation procedure. This thesis also shows how extrapolated chemisorption energies can be used to make insights about CO chemisorption trends for different metals, coverages, and overlayer patterns. The amount of error associated with uncorrected chemisorption energies is highly dependent on adsorption site geometry. Also, the preferred adsorption site(s) on a particular surface evolve with adsorbate coverage. Therefore, the study of adsorbate-adsorbate interactions presented in this thesis would not have been possible without development of the extrapolation method. The development of the correction method required consistent calculations of a variety of chemisorption systems. The approach of harvesting representative databases of chemisorption energies is also applied to how surface structure affects CO chemisorption. By studying different metals, surface terminations, and strain states, it is shown that decomposing the transition metal d-band into symmetry-relevant components results in a better correlation between electronic structure and chemisorption trends. In addition to understanding how surface modification affects chemisorption bonding, this thesis also employs simple kinetic models to predict how the strain state in different metals affects optimal catalytic conditions. These simple predictions may aid in deepening current understanding of catalyst performance as a function of temperature and other reaction conditions. Bonding between the surface and adsorbates is only half of the challenge in understanding modern catalysts which employ metal dispersed on a support. Interactions between the metal and the support can strongly influence the electronic structure of the active sites on a catalytic surface. In this thesis, Ag clusters on an alumina support are chosen to demonstrate how distinct combinations of noble metal and oxide support can lead to novel electronic states in the supported cluster. Collectively, the works presented in this thesis bring forward a set of accomplishments highlighting the comprehension DFT modeling can provide. The insights obtained accredit these modeling methods further for future studies aimed at both improving fundamental understanding and guiding novel catalyst design