Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Predicting molecular dipole moments by combining atomic partial charges and atomic dipoles

Journal Article · · Journal of Chemical Physics
DOI:https://doi.org/10.1063/5.0009106· OSTI ID:1681219
 [1];  [1];  [2];  [2];  [1]
  1. Ecole Polytechnique Federale Lausanne (Switzlerland). Lab. of Computational Science and Modeling, IMX
  2. Cornell Univ., Ithaca, NY (United States). Dept. of Chemistry and Chemical Biology
+ Show Author Affiliations

The molecular dipole moment (μ) is a central quantity in chemistry. It is essential in predicting infrared and sum-frequency generation spectra as well as induction and long-range electrostatic interactions. Furthermore, it can be extracted directly—via the ground state electron density—from high-level quantum mechanical calculations, making it an ideal target for machine learning (ML). Here, we choose to represent this quantity with a physically inspired ML model that captures two distinct physical effects: local atomic polarization is captured within the symmetry-adapted Gaussian process regression framework which assigns a (vector) dipole moment to each atom, while the movement of charge across the entire molecule is captured by assigning a partial (scalar) charge to each atom. The resulting “MuML” models are fitted together to reproduce molecular μ computed using high-level coupled-cluster theory and density functional theory (DFT) on the QM7b dataset, achieving more accurate results due to the physics-based combination of these complementary terms. The combined model shows excellent transferability when applied to a showcase dataset of larger and more complex molecules, approaching the accuracy of DFT at a small fraction of the computational cost. We also demonstrate that the uncertainty in the predictions can be estimated reliably using a calibrated committee model. The ultimate performance of the models—and the optimal weighting of their combination—depends, however, on the details of the system at hand, with the scalar model being clearly superior when describing large molecules whose dipole is almost entirely generated by charge separation. These observations point to the importance of simultaneously accounting for the local and non-local effects that contribute to μ; furthermore, they define a challenging task to benchmark future models, particularly those aimed at the description of condensed phases.

Research Organization:
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States). National Energy Research Scientific Computing Center (NERSC)
Sponsoring Organization:
USDOE Office of Science (SC); European Research Council (ERC)
Grant/Contract Number:
AC02-05CH11231
OSTI ID:
1681219
Journal Information:
Journal of Chemical Physics, Journal Name: Journal of Chemical Physics Journal Issue: 2 Vol. 153; ISSN 0021-9606
Publisher:
American Institute of Physics (AIP)Copyright Statement
Country of Publication:
United States
Language:
English

References (68)

SchNet – A deep learning architecture for molecules and materials journal June 2018
Quantum Mechanical Static Dipole Polarizabilities in the QM7b and AlphaML Showcase Databases dataset January 2019
Quantum mechanical dipole moments in the QM7b, 21k molecules of QM9, and MuML showcase datasets dataset January 2020
Models and source data for MuML dipole fitting dataset January 2020
Representation of the molecular electrostatic potential by a net atomic charge model journal October 1981
An approach to computing electrostatic charges for molecules journal April 1984
Determining atom-centered monopoles from molecular electrostatic potentials. The need for high sampling density in formamide conformational analysis journal April 1990
Comparison of atomic charges derived via different procedures journal December 1993
Gaussian approximation potentials: A brief tutorial introduction journal April 2015
Bonded-atom fragments for describing molecular charge densities journal January 1977
Distributed multipole analysis, or how to describe a molecular charge distribution journal October 1981
A beginner's guide to the modern theory of polarization journal November 2012
Performance and basis set dependence of density functional theory dipole and quadrupole moments journal July 2000
A theoretical analysis of the sum frequency generation spectrum of the water surface journal August 2000
Minimal Basis Iterative Stockholder: Atoms in Molecules for Force-Field Development journal July 2016
How Accurate Is Density Functional Theory at Predicting Dipole Moments? An Assessment Using a New Database of 200 Benchmark Values journal March 2018
Fast and Accurate Uncertainty Estimation in Chemical Machine Learning journal November 2018
Polarizable QM/MM Approach with Fluctuating Charges and Fluctuating Dipoles: The QM/FQFμ Model journal March 2019
Equation of State of Fluid Methane from First Principles with Machine Learning Potentials journal February 2019
PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges journal April 2019
Discovering a Transferable Charge Assignment Model Using Machine Learning journal July 2018
Distributed Multipole Analysis:  Stability for Large Basis Sets journal September 2005
Improved Atoms-in-Molecule Charge Partitioning Functional for Simultaneously Reproducing the Electrostatic Potential and Chemical States in Periodic and Nonperiodic Materials journal July 2012
Hirshfeld-E Partitioning: AIM Charges with an Improved Trade-off between Robustness and Accurate Electrostatics journal April 2013
Beyond Point Charges: Dynamic Polarization from Neural Net Predicted Multipole Moments journal July 2008
Dynamically Polarizable Water Potential Based on Multipole Moments Trained by Machine Learning journal May 2009
Comparison of Density Functional and MP2 Calculations on the Water Monomer and Dimer journal October 1994
Determination of partial atomic charges from ab initio molecular electrostatic potentials. Application to formamide, methanol, and formic acid journal March 1978
Benchmarking Quantum Chemical Methods for the Calculation of Molecular Dipole Moments and Polarizabilities journal May 2014
Chemical shifts in molecular solids by machine learning journal October 2018
Quantum mechanical static dipole polarizabilities in the QM7b and AlphaML showcase databases journal August 2019
Quantum chemistry structures and properties of 134 kilo molecules journal August 2014
Redefining the atom: atomic charge densities produced by an iterative stockholder approach journal January 2008
Machine learning molecular dynamics for the simulation of infrared spectra journal January 2017
Feature optimization for atomistic machine learning yields a data-driven construction of the periodic table of the elements journal January 2018
Electronic Population Analysis on LCAO–MO Molecular Wave Functions. I journal October 1955
On the Non‐Orthogonality Problem Connected with the Use of Atomic Wave Functions in the Theory of Molecules and Crystals journal March 1950
Critical analysis and extension of the Hirshfeld atoms in molecules journal April 2007
Coupled cluster response functions journal September 1990
Density‐functional thermochemistry. III. The role of exact exchange journal April 1993
Gaussian basis sets for use in correlated molecular calculations. IV. Calculation of static electrical response properties journal February 1994
FOHI-D: An iterative Hirshfeld procedure including atomic dipoles journal April 2014
Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning journal June 2018
Alchemical and structural distribution based representation for universal quantum machine learning journal June 2018
Automatic selection of atomic fingerprints and reference configurations for machine-learning potentials journal June 2018
Operators in quantum machine learning: Response properties in chemical space journal February 2019
Incorporating long-range physics in atomic-scale machine learning journal November 2019
Use of the complete basis set limit for computing highly accurate ab initio dipole moments journal January 2020
Inexpensive modeling of quantum dynamics using path integral generalized Langevin equation thermostats journal March 2020
How van der Waals interactions determine the unique properties of water journal July 2016
Accurate molecular polarizabilities with coupled cluster theory and machine learning journal February 2019
Permutationally invariant potential energy surfaces in high dimensionality journal October 2009
Visualization and analysis of atomistic simulation data with OVITO–the Open Visualization Tool journal December 2009
Machine learning of molecular electronic properties in chemical compound space journal September 2013
Using Gaussian process regression to simulate the vibrational Raman spectra of molecular crystals journal October 2019
The Theory of Intermolecular Forces book January 2013
Formulation in terms of normalized propagators of a charge-dipole model enabling the calculation of the polarization properties of fullerenes and carbon nanotubes journal January 2007
Machine-learning approach for one- and two-body corrections to density functional theory: Applications to molecular and condensed water journal August 2013
Interatomic potentials for ionic systems with density functional accuracy based on charge densities obtained by a neural network journal July 2015
Accurate interatomic force fields via machine learning with covariant kernels journal June 2017
Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons journal April 2010
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning journal January 2012
Symmetry-Adapted Machine Learning for Tensorial Properties of Atomistic Systems journal January 2018
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Machine Learning a General-Purpose Interatomic Potential for Silicon journal December 2018
Machine learning unifies the modeling of materials and molecules journal December 2017
Polarization in Kohn-Sham density-functional theory journal June 2018
Nonlinear Light Scattering and Spectroscopy of Particles and Droplets in Liquids journal May 2012

Similar Records

Linear Combination of Atomic Dipoles to Calculate the Bond and Molecular Dipole Moments of Molecules and Molecular Liquids
Journal Article · Tue Dec 21 23:00:00 EST 2021 · Journal of Physical Chemistry Letters · OSTI ID:1963035
Determining Partial Atomic Charges for Liquid Water: Assessing Electronic Structure and Charge Models
Journal Article · Tue Jan 05 23:00:00 EST 2021 · Journal of Chemical Theory and Computation · OSTI ID:1773837
Quantum mechanical static dipole polarizabilities in the QM7b and AlphaML showcase databases
Journal Article · Mon Aug 19 00:00:00 EDT 2019 · Scientific Data · OSTI ID:1577560

Related Subjects

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY
Gaussian processes
coupled-cluster methods
density functional theory
electrostatics
infrared spectroscopy
machine learning
molecular dipole moments
polarization
polypeptide
sum-frequency generation