Probing Molecule-Surface Interactions through Thermal Desorption Rates

The accurate description of chemical reaction rates at surfaces is essential for the understanding of heterogeneous catalysis. As industrial catalysts are in general complex, fundamental understanding on how they work requires input from theory, typically using density functional and transition state theory. These methods are the state of-the-art tools to acquire elementary rate constants, which serve as building blocks of kinetic mechanisms. The resulting mechanisms allow us to evaluate the catalyst's activity and selectivity for a particular chemical transformation. Unfortunately, the established theoretical methods for prediction of rate constants lack validation from detailed experiments of which there are few. In the present cumulative thesis, experimental techniques and analysis procedures were developed, allowing to test the established methods for rate constant prediction based on accurate measurements of thermal desorption rates. Kinetic experiments on single crystal surfaces were made possible by the velocity resolved kinetics approach under ultra-high vacuum conditions. I exploited the far-transition-state concept, based on which the molecule's binding energy to and entropy at the surface are directly obtained from thermal desorption rate constants. The binding energies are valuable benchmarks to evaluate the accuracy of the electronic-structure theory. The entropies serve as a critical test to models employed for the description of adsorbate's partition function, essential for the description of thermodynamic and kinetic properties of surface chemistry. In addition, adsorbate entropies serve as a valuable probe for molecule-surface interactions. In this work I present experimental and kinetic modeling studies on four molecule-surface systems H2/Pt(111,332), NO/Pd(111,332), CO/Au(111,332) and NH3/Pt(111,332) which allowed me to make the following conclusions: 1) Thermal desorption rates are a sensitive probe for molecule-surface interactions and can provide information beyond binding energies only. Detailed modeling of adsorbate entropy allows the quantiffcation of diffusion barriers and vibrational relaxation times based on thermal desorption rates. 2) Established models for adsorbate partition functions lack accuracy as they ignore quantum effects and rely on the assumption of decoupled degrees of freedom. Strategies to account for these effects are presented. 3) The desorption rate from surfaces with atomic steps can be dominated by entropic effects, despite the molecule's higher energetic stabilization at steps. These effects will likely be important for surface reactions as well. 4) Elementary parameters obtained from experiments at single crystal surfaces and models developed to explain these experiments have enough transferability to understand aspects of heterogeneous catalysis at industrially relevant conditions and at real catalytic materials. This work provides a perspective that reaction kinetics at surfaces can be a more exact science.2022-12-1