Chemistry of complex materials: Designing a high-dimensional NN potential gives maximally resolved vibrational spectra

By | 2019-02-13

Authors:  Vanessa Quaranta, Matti Hellström, Jörg Behler, Jolla Kullgren, Pavlin D. Mitev, and Kersti Hermansson

Unravelling the atomistic details of solid/liquid interfaces, e.g., by means of vibrational spectroscopy, is of vital importance in numerous applications, from electrochemistry to heterogeneous catalysis. A reactive high-dimensional neural network potential based on density functional theory calculations was developed and then used to sample the interfacial structures by means of Molecular dynamics simulations. In the second step, one-dimensional potential energy curves have been generated for a very large number of configurations to solve the nuclear Schrödinger equation. Finally, we have been able to identify substantial correlations between the stretching frequencies and hydrogen bond lengths for all species.

J. Chem. Phys. 148, 241720 (2019);