Many-Body Interactions in Ice

Abstract

Many-body effects in ice are investigated through a systematic analysis of the lattice energies of several proton ordered and disordered phases, which are calculated with different flexible water models, ranging from pairwise additive (q-TIP4P/F) to polarizable (TTM3-F and AMOEBA) and explicit many-body (MB-pol) potential energy functions. Comparisons with available experimental and diffusion Monte Carlo data emphasize the importance of an accurate description of the individual terms of the many-body expansion of the interaction energy between water molecules for the correct prediction of the energy ordering of the ice phases. Further analysis of the MB-pol results, in terms of fundamental energy contributions, demonstrates that the differences in lattice energies between different ice phases are sensitively dependent on the subtle balance between short-range two-body and three-body interactions, many-body induction, and dispersion energy. By correctly reproducing many-body effects at both short range and long range, it is found that MB-pol accurately predicts the energetics of different ice phases, which provides further support for the accuracy of MB-pol in representing the properties of water from the gas to the condensed phase.

Chemical accuracy in modeling halide ion hydration from many-body representations

“…[92] for a recent review). Among existing many-body PEFs for water, MB-pol has been shown to correctly predict the vibration-rotation tunneling spectrum of the water dimer [103], the energetics, quantum equilibria, and infrared spectra of small clusters [104,[106][107][108], the structural, thermodynamic, and dynamical properties of liquid water [105,109], the energetics of different ice phases [110], infrared and Raman spectra of liquid water [111][112][113], the vibrational sum-frequency generation spectrum of the air/water interface [114,115], the infrared and Raman spectra of ice I h [116]. More recently, molecular configurations extracted from classical (MD) and quantum path-integral molecular dynamics (PIMD) simulations with MB-pol have been used to determine the electronic band gap of liquid water, both in the bulk and at the air/water interface, through many-body perturbation theory electronic structure calculations [117].…”

Chemical Accuracy in Modeling Halide Ion Hydration from Many-Body Representations

“…[92] for a recent review). Among existing many-body PEFs for water, MBpol has been shown to correctly predict the vibration-rotation tunneling spectrum of the water dimer [103], the energetics, quantum equilibria, and infrared spectra of small clusters [104,[106][107][108], the structural, thermodynamic, and dynamical properties of liquid water [105,109], the energetics of different ice phases [110], infrared and Raman spectra of liquid water [111][112][113], the vibrational sum-frequency generation spectrum of the air/water interface [114,115], the infrared and Raman spectra of ice I h [116]. More recently, molecular configurations extracted from classical (MD) and quantum pathintegral molecular dynamics (PIMD) simulations with MB-pol have been used to determine the electronic band gap of liquid water, both in the bulk and at the air/water interface, through many-body perturbation theory electronic structure calculations [117].…”

Electron-Hole Theory of the Effect of Quantum Nuclei on the X-Ray Absorption Spectra of Liquid Water

“…NQEs are taken into account via path integral molecular dynamics (PIMD) simulations [43] with the MB-pol manybody potential energy function [44][45][46]. Rigorously built upon the many-body expansion of the interaction energy [42], MB-pol enables the accurate modeling of the properties of water across different phases [47,48], from the dimer [44] and small clusters [49], to liquid water [46,50] and ice [51,52]. We use the self-consistent enhanced static Coulomb-hole and screened exchange (COHSEX) approximation [53] with maximally localized Wannier functions, which greatly enhances the computational efficiency of XAS calculations without compromising the accuracy of the results [54,55].…”