Data-driven many-body models enable a quantitative description of chloride hydration from clusters to bulk

Caruso, Alessandro; Paesani, Francesco

doi:10.1063/5.0059445

Cited by 36 publications

(71 citation statements)

References 122 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The n -body energies, ε n B , are defined recursively for 1 < n ≤ N by the expression It has been shown that the MBE converges quickly for nonmetallic systems, for example, molecular systems with large band gaps and/or localized electron densities. Water, our system of interest, has a band gap of ∼9 eV, and the sum of the two-body (ε 2B ) and three-body (ε 3B ) energy contributions corresponds to ∼90–95% of the total interaction energy. − , Given its relatively large band gap and the fast convergence of the MBE, several many-body potentials derived from high-level quantum-mechanical data have been reported in the literature for water − and more recently for various aqueous systems. − …”

Section: Theorymentioning

confidence: 99%

Assessing the Interplay between Functional-Driven and Density-Driven Errors in DFT Models of Water

Palos

Lambros

Swee

et al. 2022

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

We investigate the interplay between functional-driven and density-driven errors in different density functional approximations within density functional theory (DFT) and the implications of these errors for simulations of water with DFT-based data-driven potentials. Specifically, we quantify density-driven errors in two widely used dispersion-corrected functionals derived within the generalized gradient approximation (GGA), namely BLYP-D3 and revPBE-D3, and two modern meta-GGA functionals, namely strongly constrained and appropriately normed (SCAN) and B97M-rV. The effects of functional-driven and density-driven errors on the interaction energies are first assessed for the water clusters of the BEGDB dataset. Further insights into the nature of functional-driven errors are gained from applying the absolutely localized molecular orbital energy decomposition analysis (ALMO-EDA) to the interaction energies, which demonstrates that functional-driven errors are strongly correlated with the nature of the interactions. We discuss cases where density-corrected DFT (DC-DFT) models display higher accuracy than the original DFT models and cases where reducing the density-driven errors leads to larger deviations from the reference energies due to the presence of large functional-driven errors. Finally, molecular dynamics simulations are performed with data-driven many-body potentials derived from DFT and DC-DFT data to determine the effect that minimizing density-driven errors has on the description of liquid water. Besides rationalizing the performance of widely used DFT models of water, we believe that our findings unveil fundamental relations between the shortcomings of some common DFT approximations and the requirements for accurate descriptions of molecular interactions, which will aid the development of a consistent, DFT-based framework for the development of data-driven and machine-learned potentials for simulations of condensed-phase systems.

show abstract

Section: Theorymentioning

confidence: 99%

Assessing the Interplay between Functional-Driven and Density-Driven Errors in DFT Models of Water

Palos

Lambros

Swee

et al. 2022

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…[21][22][23] Sophisticated many-body inter atomic potentials extensively tted to high level quantum chemistry calculations on small clusters is a promising pathway currently being explored to overcome the challenge of parameter determination of classical methods. [34][35][36] 3.3 Ion-solvent clusters (ISCs)…”

Section: Continuum Solvent Models (Csm)mentioning

confidence: 99%

Towards predictive design of electrolyte solutions by accelerating ab initio simulation with neural networks

2022

View full text Add to dashboard Cite

show abstract

“…[21][22][23] Sophisticated many-body inter atomic potentials extensively fitted to high level quantum chemistry calculations on small clusters is a promising pathway currently being explored to overcome the challenge of parameter determination of classical methods. [34][35][36]…”

Section: Classical Molecular Dynamics (Cmd)mentioning

confidence: 99%

Towards predictive design of electrolyte solutions by accelerating ab initio simulation with neural networks.

Zhang

Pagotto

Duignan

2022

Preprint

View full text Add to dashboard Cite

Electrolyte solutions play a vital role in a vast range of important materials chemistry applications. For example, they are a crucial component in batteries, fuel cells, supercapacitors, electrolysis and carbon dioxide conversion/capture. Unfortunately, the determination of even their most basic properties from first principles remains an unsolved problem. As a result, the discovery and optimisation of electrolyte solutions for these applications largely relies on chemical intuition, experimental trial and error or empirical models. The challenge is that the dynamic nature of liquid electrolyte solutions require long simulation times to generate trajectories that sufficiently sample the configuration space; the long range electrostatic interactions require large system sizes; while the short range quantum mechanical (QM) interactions require an accurate level of theory. Fortunately, recent advances in the field of deep learning, specifically neural network potentials (NNPs), can enable significant accelerations in sampling the configuration space of electrolyte solutions. Here, we outline the implications of these recent advances for the field of materials chemistry and identify outstanding challenges and potential solutions.

show abstract

Data-driven many-body models enable a quantitative description of chloride hydration from clusters to bulk

Cited by 36 publications

References 122 publications

Assessing the Interplay between Functional-Driven and Density-Driven Errors in DFT Models of Water

Assessing the Interplay between Functional-Driven and Density-Driven Errors in DFT Models of Water

Towards predictive design of electrolyte solutions by accelerating ab initio simulation with neural networks

Towards predictive design of electrolyte solutions by accelerating ab initio simulation with neural networks.

Contact Info

Product

Resources

About