Transferability of atomic energies from alchemical decomposition

Michael J Sahre; Guido Falk von Rudorff; Philipp Marquetand; O Anatole von Lilienfeld

doi:10.1063/5.0187298

Transferability of atomic energies from alchemical decomposition

J Chem Phys. 2024 Feb 7;160(5):054106. doi: 10.1063/5.0187298.

Authors

Michael J Sahre¹, Guido Falk von Rudorff^{2

3}, Philipp Marquetand⁴, O Anatole von Lilienfeld^{1

5

6

7

8

9

10}

Affiliations

¹ Vienna Doctoral School in Chemistry (DoSChem) and Institute of Theoretical Chemistry and Faculty of Physics, University of Vienna, 1090 Vienna, Austria.
² Department of Chemistry, University Kassel, Heinrich-Plett-Str.40, 34132 Kassel, Germany.
³ Center for Interdisciplinary Nanostructure Science and Technology (CINSaT), Heinrich-Plett-Straße 40, 34132 Kassel, Germany.
⁴ Faculty of Chemistry, Institute of Theoretical Chemistry, University of Vienna, Währinger Str. 17, 1090 Vienna, Austria.
⁵ Chemical Physics Theory Group, Department of Chemistry, University of Toronto, St. George Campus, Toronto, M5S 3H6 Ontario, Canada.
⁶ Department of Materials Science and Engineering, University of Toronto, St. George Campus, Toronto, M5S 3E4 Ontario, Canada.
⁷ Vector Institute for Artificial Intelligence, Toronto, M5S 1M1 Ontario, Canada.
⁸ ML Group, Technische Universität Berlin and Institute for the Foundations of Learning and Data, 10587 Berlin, Germany.
⁹ Berlin Institute for the Foundations of Learning and Data, 10587 Berlin, Germany.
¹⁰ Department of Physics, University of Toronto, St. George Campus, Toronto, M5S 1A7 Ontario, Canada.

PMID: 38341696
DOI: 10.1063/5.0187298

Abstract

We study alchemical atomic energy partitioning as a method to estimate atomization energies from atomic contributions, which are defined in physically rigorous and general ways through the use of the uniform electron gas as a joint reference. We analyze quantitatively the relation between atomic energies and their local environment using a dataset of 1325 organic molecules. The atomic energies are transferable across various molecules, enabling the prediction of atomization energies with a mean absolute error of 23 kcal/mol, comparable to simple statistical estimates but potentially more robust given their grounding in the physics-based decomposition scheme. A comparative analysis with other decomposition methods highlights its sensitivity to electrostatic variations, underlining its potential as a representation of the environment as well as in studying processes like diffusion in solids characterized by significant electrostatic shifts.