Annabella SelloniAqueous Solutions – Metal Oxide Interfaces from Machine Learning Molecular Dynamics

Aqueous solution-oxide interfaces play a critical role in many environmental, biological, and energy-relevant processes. As the molecular scale properties of these interfaces are still challenging to probe experimentally, many studies rely on computer simulations to complement the experimental observations and obtain an atomistic understanding. In particular, molecular dynamics (MD) simulations based on machine learning potentials (MLPs) capable to reproduce the accuracy of first-principles methods have recently emerged as a promising approach for this purpose. In this talk, I shall discuss recent applications of MLP-MD to understand the structure and properties of aqueous solution-metal oxide interfaces. Focusing on the interfaces of titanium dioxide, a prototypical oxide with a prominent role in energy applications, specific topics will include a characterization of the Electrical Double Layer at the TiO₂-electrolyte interface under different pHs and external electric fields [1, 2], and the effects of organic compounds such as formic acid and methanol on the structure and chemistry of water at the interface [3].