POMs-GibbsNet: A Deep Neural Network for the Prediction of Thermodynamic Parameters of Polyoxometalates.
- 1Institute of Chemical Research of Catalonia (ICIQ), Tarragona, Spain.
- 2Departament de Química Física I Inorgànica, Universitat Rovira i Virgili, Tarragona, Spain.
- 3Hylleraas Centre for Quantum Molecular Sciences, Department of Chemistry, University of Oslo, Oslo, Norway.
Abstract
Polyoxometalates (POMs) are versatile metal-oxo clusters with remarkable structural diversity and applications spanning catalysis, energy storage, and materials science. However, accurately predicting their thermodynamic properties remains computationally demanding due to the large system sizes and the cost of density functional theory (DFT) frequency calculations required to obtain Gibbs free energies. In this work, we present POMs-GibbsNet, a deep neural network designed to predict the thermal free energy correction ( G-E ) of polyoxometalates by combining graph-based molecular representations with DFT-derived descriptors. Molecular structures were encoded using variants of the AABBA autocorrelation graph kernel, which captures atom-atom, bond-bond, and atom-bond correlations together with quantum-derived features including QTAIM descriptors. POMs-GibbsNet was trained on a dataset of 1419 POM structures spanning 19 chemical systems to obtain good results ( R2=0.993 and RMSE = 1.9 kcal · mol⁻¹. To evaluate the performance, a benchmark against 10 classical regressors was performed. Additionally, an uncertainty study was added to the methodology. External validation using POMSimulator reaction networks comprising over 10,819 reactions shows excellent agreement between predicted and DFT reaction energies ( R2>0.99 ), with errors typically within a few kcal mol⁻¹. Error cancellation was also studied during this step. Notably, simplified atom-atom autocorrelation descriptors already provide robust performance while reducing model complexity. These results demonstrate that POMs-GibbsNet enables efficient estimation of thermodynamic corrections in POM systems, providing a powerful framework for accelerating the computational exploration of polyoxometalate speciation and self-assembly across large chemical spaces.
Related
- Covalent Functionalization of Dawson-Type Polyoxometalates to Achieve a Zero to Three-Dimensional Transition for Enhanced Proton Conductivity
- DEEP LEARNING HELMHOLTZ ENERGIES FOR THERMODYNAMIC PROPERTY PREDICTION
- A comparative study on the immobilization of organic-inorganic polyoxometalates hybrids onto gold electrodes: Direct versus indirect, covalent versus electrostatic grafting
- The Disproportionation of Hydrogen Peroxide by Prototype Keggin Polyoxometalates
- M(IV)- and M(V)-Substituted {MW5} Lindqvist-Type Polyoxometalates from Plenary [W6O19]2– or WO3·H2O: A Gateway to Wider Systematic Reactivity Studies
BibTeX
@article{buils2026pomsgibbsnet,
title = {{POMs-GibbsNet: A Deep Neural Network for the Prediction of Thermodynamic Parameters of Polyoxometalates.}},
author = {Jordi Buils and Carles Bo and David Balcells},
journal = {Journal of computational chemistry},
year = {2026},
month = oct,
doi = {10.1002/jcc.70510},
url = {https://europepmc.org/article/MED/42904606},
note = {CC BY},
}
RIS
TY - JOUR
TI - POMs-GibbsNet: A Deep Neural Network for the Prediction of Thermodynamic Parameters of Polyoxometalates.
AU - Jordi Buils
AU - Carles Bo
AU - David Balcells
PY - 2026
DA - 2026/10/01
JO - Journal of computational chemistry
DO - 10.1002/jcc.70510
UR - https://europepmc.org/article/MED/42904606
AB - Polyoxometalates (POMs) are versatile metal-oxo clusters with remarkable structural diversity and applications spanning catalysis, energy storage, and materials science. However, accurately predicting their thermodynamic properties remains computationally demanding due to the large system sizes and the cost of density functional theory (DFT) frequency calculations required to obtain Gibbs free energies. In this work, we present POMs-GibbsNet, a deep neural network designed to predict the thermal free energy correction ( G-E ) of polyoxometalates by combining graph-based molecular representations with DFT-derived descriptors. Molecular structures were encoded using variants of the AABBA autocorrelation graph kernel, which captures atom-atom, bond-bond, and atom-bond correlations together with quantum-derived features including QTAIM descriptors. POMs-GibbsNet was trained on a dataset of 1419 POM structures spanning 19 chemical systems to obtain good results ( R2=0.993 and RMSE = 1.9 kcal · mol⁻¹. To evaluate the performance, a benchmark against 10 classical regressors was performed. Additionally, an uncertainty study was added to the methodology. External validation using POMSimulator reaction networks comprising over 10,819 reactions shows excellent agreement between predicted and DFT reaction energies ( R2>0.99 ), with errors typically within a few kcal mol⁻¹. Error cancellation was also studied during this step. Notably, simplified atom-atom autocorrelation descriptors already provide robust performance while reducing model complexity. These results demonstrate that POMs-GibbsNet enables efficient estimation of thermodynamic corrections in POM systems, providing a powerful framework for accelerating the computational exploration of polyoxometalate speciation and self-assembly across large chemical spaces.
SN - 0192-8651
SN - 1096-987X
ER -