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

Abstract

Despite the key role that ionic solutions play in several natural and industrial processes, a unified, molecular-level understanding of how ions affect the structure and dynamics of water across different phases remains elusive. In this context, computer simulations can provide new insights that are difficult, if not impossible, to obtain by other means. However, the predictive power of a computer simulation directly depends on the level of 'realism' that is used to represent the underlying molecular interactions. Here, we report a systematic analysis of many-body effects in halide-water clusters and demonstrate that the recently developed MB-nrg full-dimensional manybody potential energy functions achieve high accuracy by quantitatively reproducing the individual terms of the manybody expansion of the interaction energy, thus opening the door to realistic computer simulations of ionic solutions.

Highly Accurate Many-Body Potentials for Simulations of N2O5 in Water: Benchmarks, Development, and Validation

“…These potential energy functions, termed MB-nrg (for "many-body energy"), employ the same underlying functional forms for the intermolecular many-body interactions as MB-pol. The first MB-nrg models have been devised to describe the interactions of water with monoatomic ions [57][58][59] and, more recently, to describe CO2/water and CH4/water mixtures. 60,61 Similarly to MB-pol, the MBnrg potential energy functions have been successfully used in studies across phases from clusters [62][63][64][65][66][67] to bulk mixtures.…”

Temperature and Nuclear Quantum Effects on the Stretching Modes of the Water Hexamer

“…Finally, MB PESs for halides in water have recently become available. 36 Perhaps simple methods as introduced herein could be useful in understanding nuclear quantum effects on aqueous ion solvation.…”

MB-Fit: Software Infrastructure for Data-Driven Many-Body Potential Energy Functions

“…[124][125][126] Because it is currently not possible to perform CM5 and XDM calculations at the DF-FNO-CCSD(T) level of theory, in order to guarantee the same representations of V pol , V 2B disp , and V elec for the two sets of TTM-nrg and MB-nrg PEFs, the CM5 and XDM calculations for all MB-PEFs were carried out with the ωB97M-V functional since it was shown to consistently provide the closest agreement with CCSD(T) data for various molecular interactions. 57,93,96,[138][139][140][141] Both the CM5 and XDM calculations were carried out with Q-Chem v5.1 121 using the aug-cc-pVTZ 137 basis set. All 1B and 2B energies were calculated using Psi4 122 at the corresponding QM level of theory, including counterpoise correction for the BSSE.…”

“…2 and 5 assess the ability of the TTM-nrg and MB-nrg PEFs to reproduce 1B and 2B energies that were the target of the training process, one of the greatest challenges for ML PEFs is to preserve the same accuracy for many-body energies and/or molecular systems that are not included in the training sets. Combining explicit data-driven representations of short-range low-order interactions with implicit (mean-field-like) many-body representations of high-order and long-range interactions, it has been shown that the MB-pol PEF [58][59][60][61]142 for water as well as the MB-nrg PEFs for ions in water 92,95,107,141 and various molecular fluids 97,98 are able to correctly reproduce each individual term of the MBE of Eq. 1.…”