Thanh-Nam Huynh

Thanh-Nam Huynh

  • 12/2023 - 11/2024: Member of the PhD committee

Applying on-the-fly machine learned force fields (MLFF) for a more accurate assessment of adsorption free energy.

An accurate description of adsorption process is of paramount importance for the understanding of heterogeneous catalytic process. However, the present approaches to calculating adsorption free energies either provide insufficient entropic contributions or are very complex and/or computationally expensive. The harmonic approximation (HA), the method that has been utilized most frequently thus far, is prone to significant deviations. Therefore, this work aims at providing an anharmonic correction to the HA method for a more accurate assessment of adsorption free energy. For this, we focus on the applicability of on-the-fly machine learned force fields (MLFF) to the well-known lambda-path thermodynamic integration. This approach allows us to reduce significantly the needed computational resources while retaining the DFT-level accuracy.