FABIO - AAP 2021
Enabling fast quantum chemical methods for biomass conversion at metal/water interfaces
Résumé: Molecular simulations provide unprecedented, detailed insight in the reactivity at solid/liquid interfaces. These interfaces are of particular importance for biomass conversion, such as alcohol oxidation, hydrogenolysis of polyols and lignin valorization. Simulating the full metal/liquid interface via density functional theory (DFT) is computationally prohibitive. Very recently, we have derived a force-field based strategy to include the solvation effects approximately. However, two main challenges remain: (a) adsorption configurations (including nano-particle shape) might change in explicit aqueous environment compared to gas-phase and (b) the many-body effects of water at the interface are important, but neglected at the force field level. Here, we develop density functional tight-binding (DFTB) for addressing these two issues: DFTB is, in principle, capable of capturing the many-body terms and is roughly three orders of magnitude faster than DFT, so that investigating the flexibility and reorganization at the metal/liquid interface becomes feasible. However, due to the disjoint communities between DFTB developers and modelers of heterogeneous catalysis over metallic surfaces, to date no accurate DFTB parametrization is available for the archetypical Pt/H/C/O systems, which would be needed for investigating the biomass conversion at the Pt interface. By combining the extensive experience with DFT and the solid/liquid interface at LCH with the expertise of DFTB and its parametrization at ILM, we are able to overcome the limitations of DFT (computational expense) and force fields (accuracy, lack of flexibility of surface and adsorbates) to reach an atomistic description of reactions at the metal/liquid interface.
