Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 58: Solid-Liquid Interfaces: Reactions and Electrochemistry II
O 58.6: Vortrag
Mittwoch, 19. März 2025, 11:45–12:00, H4
Predicting Charge Effects at Electrified Solid/Liquid Interfaces — •Nicolas Bergmann1, Nicéphore Bonnet2, Nicola Marzari2, Karsten Reuter1, and Nicolas G. Hörmann1 — 1Fritz-Haber-Institut der MPG, Berlin — 2Laboratory of Theory and Simulation of Materials, EPFL, Lausanne
Computational modeling of electrified solid-liquid interfaces must account for capacitive contributions at potentials beyond the point of zero charge (PZC) [1]. These contributions can approximately be obtained from first-principles calculations at constant charge, employing a second-order Taylor expansion that involves the surface’s free energy, the work function, and the interfacial capacitance at the PZC [2]. Machine learning (ML) surrogate models have already successfully been employed to predict the PZC energies at significantly reduced computational costs. Here, we extend this by presenting a ML model for work functions and apply this to two standard electrode modeling setups. By including the derivatives of the atomic forces with charge, we stabilize our model and predict the atomic effective charge. As an outlook, we show how this methodology could be utilized to run molecular dynamics simulations at charged conditions and electrochemical barrier calculations.
[1] N. Bergmann, N.G. Hörmann, and K. Reuter, J. Chem. Theory Comput. 19, 8815 (2023).
[2] N.G. Hörmann and K. Reuter, J. Chem. Theory Comput. 17, 1782 (2021).
Keywords: Machine Learning; Electrochemistry; DFT; Computational Chemistry; Modeling