Center for Molecular Modeling - J. Dambre

Modeling Electronic Response Properties with an Explicit-Electron Machine Learning Potential

leen — Mon, 21 Aug 2023 09:49:07 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

Journal of Chemical Theory and Computation

18, 3, 1672-1691

2023

Abstract

Explicit-electron force fields introduce electrons or electron pairs as semiclassical particles in force fields or empirical potentials, which are suitable for molecular dynamics simulations. Even though semiclassical electrons are a drastic simplification compared to a quantum-mechanical electronic wave function, they still retain a relatively detailed electronic model compared to conventional polarizable and reactive force fields. The ability of explicit-electron models to describe chemical reactions and electronic response properties has already been demonstrated, yet the description of short-range interactions for a broad range of chemical systems remains challenging. In this work, we present the electron machine learning potential (eMLP), a new explicit electron force field in which the short-range interactions are modeled with machine learning. The electron pair particles will be located at well-defined positions, derived from localized molecular orbitals or Wannier centers, naturally imposing the correct dielectric and piezoelectric behavior of the system. The eMLP is benchmarked on two newly constructed data sets: eQM7, an extension of the QM7 data set for small molecules, and a data set for the crystalline beta-glycine. It is shown that the eMLP can predict dipole moments, polarizabilities, and IR-spectra of unseen molecules with high precision. Furthermore, a variety of response properties, for example, stiffness or piezoelectric constants, can be accurately reproduced.

Green Open Access

DOI

http://dx.doi.org/10.1021/acs.jctc.1c00978

Private attachment

acs.jctc_.1c00978.pdf

Modeling electronic response properties with an explicit-electron machine learning potential

leen — Thu, 11 May 2023 10:55:43 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

Journal of Chemical Theory and Computation

Volume 18, Issue 3, Pages 1672-1691

2023

Abstract

DOI

https://doi.org/10.1021/acs.jctc.1c00978

Private attachment

acs.jctc_.1c00978.pdf

Modeling Electronic Response Properties with an Explicit-Electron Machine Learning Potential

maarten — Tue, 15 Feb 2022 09:24:18 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

Journal of Chemical Theory and Computation (JCTC)

18 (3), 1672–1691

2022

Abstract

Explicit-electron force fields introduce electrons or electron pairs as semiclassical particles in force fields or empirical potentials, which are suitable for molecular dynamics simulations. Even though semiclassical electrons are a drastic simplification compared to a quantum-mechanical electronic wave function, they still retain a relatively detailed electronic model compared to conventional polarizable and reactive force fields. The ability of explicit-electron models to describe chemical reactions and electronic response properties has already been demonstrated, yet the description of short-range interactions for a broad range of chemical systems remains challenging. In this work, we present the electron machine learning potential (eMLP), a new explicit electron force field in which the short-range interactions are modeled with machine learning. The electron pair particles will be located at well-defined positions, derived from localized molecular orbitals or Wannier centers, naturally imposing the correct dielectric and piezoelectric behavior of the system. The eMLP is benchmarked on two newly constructed data sets: eQM7, an extension of the QM7 data set for small molecules, and a data set for the crystalline β-glycine. It is shown that the eMLP can predict dipole moments, polarizabilities, and IR-spectra of unseen molecules with high precision. Furthermore, a variety of response properties, for example, stiffness or piezoelectric constants, can be accurately reproduced.

DOI

https://doi.org/10.1021/acs.jctc.1c00978

Private attachment

published.pdf

Explicit Electrons in Machine Learning Potentials

toon — Thu, 22 Jan 2026 11:20:36 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

ISBN/ISSN:

Invited talk

Brussels, Belgium

Sunday, 25 May, 2025 to Saturday, 28 June, 2025

Conference reference

ML4SPEC25 Belgian-German WE-Heraeus Seminar:

Incorporating long-range interactions and polarization in machine learning potentials with explicit electrons

leen — Mon, 19 Sep 2022 10:40:12 +0000

M. Cools-Ceuppens

Thu, 15/09/2022

Ghent University AULA, Voldersstraat, Ghent

Supervisors

prof. dr. ir. T. Verstraelen; prof. dr. ir. J. Dambre

Attachment (private)

PhD-thesis.pdf

The eMLP: a novel machine learning potential to model electronic properties with explicit-electrons

maarten — Fri, 07 Jan 2022 12:31:34 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

ISBN/ISSN:

Talk

Conference / event / venue

AutoCheMo International Reactive Force Field Workshop

Ghent, Belgium

Wednesday, 8 December, 2021 to Thursday, 9 December, 2021

Comparing different machine learning force fields: a case study of aluminium

maarten — Mon, 06 May 2019 07:48:30 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

ISBN/ISSN:

Poster

Conference / event / venue

The 34th Winter School in Theoretical Chemistry: Machine Learning

Helsinki, Finland

Monday, 10 December, 2018 to Thursday, 13 December, 2018

Comparing Different Machine Learning Force Fields: A Case Study of Aluminium

maarten — Mon, 06 May 2019 07:45:53 +0000

M. Cools-Ceuppens, J. Dambre, T. Verstraelen

ISBN/ISSN:

Poster

Conference / event / venue

MOFSIM 2019

Ghent, Belgium

Wednesday, 10 April, 2019 to Friday, 12 April, 2019