Modeling the effect of electric potential at Ag(111)/water interface through neural network molecular dynamics

We explore the influence of electric potential at the Ag(111)/H2O electrochemical interface using state-of-the-art neural network (NN) potential molecular dynamics and enhanced sampling methods. With a hybrid explicit-implicit solvation model and constant potential algorithm in VASP, grand canonical DFT calculations were employed to provide the energies and forces for training the NN potential at a constant potential (CP) of -1.5 V vs. SHE. Our simulations show that electric potential significantly impacts the dynamics of interfacial water and CO2 adsorption, compared to the potential of zero charge (PZC). This highlights the critical role of electric potential in capturing realistic interfacial behavior under operational conditions.