Machine learning surrogate models of many-body dispersion interactions in polymer melts

many-body dispersion; van der Waals interaction; machine learning force field; surrogate modeling; polymer melts; deep neural network; Computer Science - Learning; Physics - Computational Physics

Abstract :

[en] Accurate prediction of many-body dispersion (MBD) interactions is essential for understanding the van der Waals forces that govern the behavior of many complex molecular systems. However, the high computational cost of MBD calculations limits their direct application in large-scale simulations. In this work, we introduce a machine learning surrogate model specifically designed to predict MBD forces in polymer melts, a system that demands accurate MBD description and offers structural advantages for machine learning approaches. Our model is based on a trimmed SchNet architecture that selectively retains the most relevant atomic connections and incorporates trainable radial basis functions for geometric encoding. We validate our surrogate model on datasets from polyethylene, polypropylene, and polyvinyl chloride melts, demonstrating high predictive accuracy and robust generalization across diverse polymer systems. In addition, the model captures key physical features, such as the characteristic decay behavior of MBD interactions, providing valuable insights for optimizing cutoff strategies. Characterized by high computational efficiency, our surrogate model enables practical incorporation of MBD effects into large-scale molecular simulations.

Disciplines :

Engineering, computing & technology: Multidisciplinary, general & others

Author, co-author :

SHEN, Zhaoxiang ; University of Luxembourg

SOSA, Raul Ian ; University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Physics and Materials Science (DPHYMS)

LENGIEWICZ, Jakub ; University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE)

TKATCHENKO, Alexandre ; University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Physics and Materials Science (DPHYMS)

BORDAS, Stéphane ; University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE)

External co-authors :

yes

Language :

English

Title :

Machine learning surrogate models of many-body dispersion interactions in polymer melts

Publication date :

01 April 2026

Journal title :

Machine learning: science and technology

eISSN :

2632-2153

Publisher :

IOP Publishing

Volume :

Issue :

Pages :

025040

Peer reviewed :

Peer Reviewed verified by ORBi

Additional URL :

https://iopscience.iop.org/article/10.1088/2632-2153/ae545a

Funders :

Fonds National de la Recherche Luxembourg

Data Set :

https://doi.org/10.5281/zenodo.15012728

Available on ORBilu :

since 15 April 2026

Statistics

Number of views

7 (1 by Unilu)

Number of downloads

3 (1 by Unilu)

More statistics

Scopus citations^®

Scopus citations^®
without self-citations

OpenCitations

OpenAlex citations

WoS citations^™

Bibliography

Liu Y Huang Y Duan X 2019 Van der waals integration before and beyond two-dimensional materials Nature 567 323 33 323–33 10.1038/s41586-019-1013-x
Dill K A 1990 Dominant forces in protein folding Biochemistry 29 7133 55 7133–55 10.1021/bi00483a001
Autumn K Sitti M Liang Y A Peattie A M Hansen W R Sponberg S Kenny T W Fearing R Israelachvili J N Full R J 2002 Evidence for van der waals adhesion in gecko setae Proc. Natl Acad. Sci. 99 12252 6 12252–6 10.1073/pnas.192252799
Grimme S Hansen A Brandenburg J G Bannwarth C 2016 Dispersion-corrected mean-field electronic structure methods Chem. Rev. 116 5105 54 5105–54 10.1021/acs.chemrev.5b00533
Tkatchenko A Scheffler M 2009 Accurate molecular van der waals interactions from ground-state electron density and free-atom reference data Phys. Rev. Lett. 102 073005 10.1103/PhysRevLett.102.073005
Ambrosetti A Alfé D DiStasio R A Tkatchenko A 2014 Hard numbers for large molecules: toward exact energetics for supramolecular systems J. Phys. Chem. Lett. 5 849 55 849–55 10.1021/jz402663k
Reilly A M Tkatchenko A 2013 Seamless and accurate modeling of organic molecular materials J. Phys. Chem. Lett. 4 1028 33 1028–33 10.1021/jz400226x
Marom N DiStasio R A Atalla V Levchenko S Reilly A M Chelikowsky J R Leiserowitz L Tkatchenko A 2013 Many-body dispersion interactions in molecular crystal polymorphism Angew. Chem. Int. Edn 52 6629 32 6629–32 10.1002/anie.201301938
Tkatchenko A DiStasio R A Car R Scheffler M 2012 Accurate and efficient method for many-body van der waals interactions Phys. Rev. Lett. 108 236402 10.1103/PhysRevLett.108.236402
Tkatchenko A Ambrosetti A DiStasio R A 2013 Interatomic methods for the dispersion energy derived from the adiabatic connection fluctuation-dissipation theorem J. Chem. Phys. 138 074106 10.1063/1.4789814
Hauseux P Nguyen T T Ambrosetti A Ruiz K S Bordas S P A Tkatchenko A 2020 From quantum to continuum mechanics in the delamination of atomically-thin layers from substrates Nat. Commun. 11 1651 10.1038/s41467-020-15480-w
Hoja J Ko H-Y Neumann M A Car R DiStasio R A Tkatchenko A 2019 Reliable and practical computational description of molecular crystal polymorphs Sci. Adv. 5 eaau3338 10.1126/sciadv.aau3338
Poier P P Lagardére L Piquemal J-P 2022 O(n) stochastic evaluation of many-body van der waals energies in large complex systems J. Chem. Theory Comput. 18 1633 45 1633–45 10.1021/acs.jctc.1c01291
Muhli H Ala-Nissila T Caro M A 2025 Atom-wise formulation of the many-body dispersion problem for linear-scaling van der waals corrections Phys. Rev. B 111 054103 10.1103/PhysRevB.111.054103
Cornell W D Cieplak P Bayly C I Gould I R Merz K M Ferguson D M Spellmeyer D C Fox T Caldwell J W Kollman P A 1995 A second generation force field for the simulation of proteins, nucleic acids and organic molecules J. Am. Chem. Soc. 117 5179 97 5179–97 10.1021/ja00124a002
Christen M et al 2005 The gromos software for biomolecular simulation: Gromos05 J. Comput. Chem. 26 1719 51 1719–51 10.1002/jcc.20303
Jorgensen W L Tirado-Rives J 1988 The opls [optimized potentials for liquid simulations] potential functions for proteins, energy minimizations for crystals of cyclic peptides and crambin J. Am. Chem. Soc. 110 1657 66 1657–66 10.1021/ja00214a001
Vanommeslaeghe K et al 2010 Charmm general force field: a force field for drug-like molecules compatible with the charmm all-atom additive biological force fields J. Comput. Chem. 31 671 90 671–90 10.1002/jcc.21367
Noé F Tkatchenko A Müller K-R Clementi C 2020 Machine learning for molecular simulation Annu. Rev. Phys. Chem. 71 361 90 361–90 10.1146/annurev-physchem-042018-052331
Unke O T Chmiela S Sauceda H E Gastegger M Poltavsky I Schütt K T Tkatchenko A Müller K-R 2021 Machine learning force fields Chem. Rev. 121 10142 86 10142–86 10.1021/acs.chemrev.0c01111
Poltavsky I Tkatchenko A 2021 Machine learning force fields: recent advances and remaining challenges J. Phys. Chem. Lett. 12 6551 64 6551–64 10.1021/acs.jpclett.1c01204
Keith J A Vassilev-Galindo V Cheng B Chmiela S Gastegger M Müller K-R Tkatchenko A 2021 Combining machine learning and computational chemistry for predictive insights into chemical systems Chem. Rev. 121 9816 72 9816–72 10.1021/acs.chemrev.1c00107
Schütt K T Chmiela S Lilienfeld O A V Tkatchenko A Tsuda K Müller K-R 2020 Machine learning meets quantum physics Lecture Notes in Physics Springer
Peivaste I et al 2025 Artificial intelligence in materials science and engineering: current landscape, key challenges and future trajectories Compos. Struct. 372 119419 10.1016/j.compstruct.2025.119419
Chmiela S Sauceda H E Müller K-R Tkatchenko A 2018 Towards exact molecular dynamics simulations with machine-learned force fields Nat. Commun. 9 3887 10.1038/s41467-018-06169-2
Schütt K T Kindermans P-J Sauceda H E Chmiela S Tkatchenko A Müller K-R 2017 Schnet: a continuous-filter convolutional neural network for modeling quantum interactions Proc. 31st Int. Conf. on Neural Information Processing Systems pp 992 1002 pp 992–1002 (available at: https:/dl.acm.org/doi/abs/10.5555/3294771.3294866)
Schütt K T Sauceda H E Kindermans P-J Tkatchenko A Müller K-R 2018 Schnet – a deep learning architecture for molecules and materials J. Chem. Phys. 148 241722 10.1063/1.5019779
Unke O T Meuwly M 2019 Physnet: a neural network for predicting energies, forces, dipole moments and partial charges J. Chem. Theory Comput. 15 3678 93 3678–93 10.1021/acs.jctc.9b00181
Unke O T et al 2024 Biomolecular dynamics with machine-learned quantum-mechanical force fields trained on diverse chemical fragments Sci. Adv. 10 eadn4397 10.1126/sciadv.adn4397
Bartók A P Payne M C Kondor R Csányi G 2010 Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons Phys. Rev. Lett. 104 136403 10.1103/PhysRevLett.104.136403
De S Bartók A P Csányi G Ceriotti M 2016 Comparing molecules and solids across structural and alchemical space Phys. Chem. Chem. Phys. 18 13754 69 13754–69 10.1039/C6CP00415F
Willatt M J Musil F Ceriotti M 2018 Feature optimization for atomistic machine learning yields a data-driven construction of the periodic table of the elements Phys. Chem. Chem. Phys. 20 29661 8 29661–8 10.1039/C8CP05921G
Grisafi A Ceriotti M 2019 Incorporating long-range physics in atomic-scale machine learning J. Chem. Phys. 151 204105 10.1063/1.5128375
Huguenin-Dumittan K K Loche P Haoran N Ceriotti M 2023 Physics-inspired equivariant descriptors of nonbonded interactions J. Phys. Chem. Lett. 14 9612 8 9612–8 10.1021/acs.jpclett.3c02375
Deringer V L Caro M A Csányi G 2020 A general-purpose machine-learning force field for bulk and nanostructured phosphorus Nat. Commun. 11 5461 10.1038/s41467-020-19168-z
Christensen A S Bratholm L A Faber F A von Lilienfeld O A 2020 Fchl revisited: faster and more accurate quantum machine learning J. Chem. Phys. 152 044107 10.1063/1.5126701
Behler J Parrinello M 2007 Generalized neural-network representation of high-dimensional potential-energy surfaces Phys. Rev. Lett. 98 146401 10.1103/PhysRevLett.98.146401
Behler J 2011 Atom-centered symmetry functions for constructing high-dimensional neural network potentials J. Chem. Phys. 134 074106 10.1063/1.3553717
Frank T Unke O Müller K-R 2022 So3krates: equivariant attention for interactions on arbitrary length-scales in molecular systems Advances in Neural Information Processing Systems vol 35 29400 13 29400–13 10.5555/3600270.3602402
Frank J T Unke O T Müller K-R Chmiela S 2024 A euclidean transformer for fast and stable machine learned force fields Nat. Commun. 15 6539 10.1038/s41467-024-50620-6
Batzner S Musaelian A Sun L Geiger M Mailoa J P Kornbluth M Molinari N Smidt T E Kozinsky B 2022 E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials Nat. Commun. 13 2453 10.1038/s41467-022-29939-5
Batatia I Kovacs D P Simm G Ortner C Csányi G 2022 Mace: higher order equivariant message passing neural networks for fast and accurate force fields Advances in Neural Information Processing Systems vol 35 11423 36 11423–36 10.5555/3600270.3601100
Anstine D M Isayev O 2023 Machine learning interatomic potentials and long-range physics J. Phys. Chem. A 127 2417 31 2417–31 10.1021/acs.jpca.2c06778
Unke O T Chmiela S Gastegger M Schütt K T Sauceda H E Müller K-R 2021 Spookynet: learning force fields with electronic degrees of freedom and nonlocal effects Nat. Commun. 12 7273 10.1038/s41467-021-27504-0
Kabylda A Frank J T Dou S S Khabibrakhmanov A Sandonas L M Unke O T Chmiela S Müller K-R Tkatchenko A 2025 Molecular Simulations With a Pretrained Neural Network and universal Pairwise Force Fields 10.26434/chemrxiv-2024-bdfr0-v2
Shen Z Sosa R I Bordas S P A Tkatchenko A Lengiewicz J 2024 Quantum-informed simulations for mechanics of materials: Dftb+mbd framework Int. J. Eng. Sci. 204 104126 10.1016/j.ijengsci.2024.104126
Strobl G 2007 The Physics of Polymers: Concepts for Understanding Their Structures and Behavior Springer 10.1007/978-3-540-68411-4
Francis L F 2016 Chapter 3 - melt processes Materials Processing Academic 105 249 105–249 10.1016/B978-0-12-385132-1.00003-3
Van Krevelen D W Nijenhuis K T 2009 Properties of Polymers: Their Correlation With Chemical Structure; Their Numerical Estimation and Prediction From Additive Group Contributions Elsevier
Bicerano J 2002 Prediction of Polymer Properties CRC Press 10.1201/9780203910115
Hossain D Tschopp M A Ward D K Bouvard J L Wang P Horstemeyer M F 2010 Molecular dynamics simulations of deformation mechanisms of amorphous polyethylene Polymer 51 6071 83 6071–83 10.1016/j.polymer.2010.10.009
Baig C Mavrantzas V G Kröger M 2010 Flow effects on melt structure and entanglement network of linear polymers: results from a nonequilibrium molecular dynamics simulation study of a polyethylene melt in steady shear Macromolecules 43 6886 902 6886–902 10.1021/ma100826u
Yamamoto T 2019 Molecular dynamics simulation of stretch-induced crystallization in polyethylene: emergence of fiber structure and molecular network Macromolecules 52 1695 706 1695–706 10.1021/acs.macromol.8b02569
Romanos N Megariotis G Theodorou D N 2021 Molecular dynamics simulations of stretch-induced crystallization in layered polyethylene Polym. Crystallization 4 e10172 10.1002/pcr2.10172
Kotelyanskii M Theodorou D N 2004 Simulation Methods for Polymers CRC Press 10.1201/9780203021255
Theodorou D N 2007 Hierarchical modelling of polymeric materials Chem. Eng. Sci. 62 5697 714 5697–714 10.1016/j.ces.2007.04.048
Gartner T E Jayaraman A 2019 Modeling and simulations of polymers: a roadmap Macromolecules 52 755 86 755–86 10.1021/acs.macromol.8b01836
Schmid F 2023 Understanding and modeling polymers: the challenge of multiple scales ACS Polym. Au 3 28 58 28–58 10.1021/acspolymersau.2c00049
Martin T B Audus D J 2023 Emerging trends in machine learning: a polymer perspective ACS Polym. Au 3 239 58 239–58 10.1021/acspolymersau.2c00053
Shen Z Sosa R I Lengiewicz J Tkatchenko A Bordas S P A 2025 Github repository: Ml surrogate models of many-body dispersion (available at: https://github.com/Zhaoxiang-Shen/trimmed-SchNet)
Sosa R I Galante M Tkatchenko A 2025 Power of the many-body force: magnitudes and angles of atomic van der waals dispersion forces in extended molecular systems Small Struct. 6 2500226 10.1002/sstr.202500226
Ambrosetti A Ferri N DiStasio R A Tkatchenko A 2016 Wavelike charge density fluctuations and van der waals interactions at the nanoscale Science 351 1171 6 1171–6 10.1126/science.aae0509
Vaswani A Shazeer N Parmar N Uszkoreit J Jones L Gomez A N u. Kaiser L Polosukhin I 2017 Attention is all you need Advances in Neural Information Processing Systems vol 30 6000 10 6000–10 10.5555/3295222.3295349
Jo S Kim T Iyer V G Im W 2008 Charmm-gui: a web-based graphical user interface for charmm J. Comput. Chem. 29 1859 65 1859–65 10.1002/jcc.20945
Choi Y K Park S-J Park S Kim S Kern N R Lee J Im W 2021 Charmm-gui polymer builder for modeling and simulation of synthetic polymers J. Chem. Theory Comput. 17 2431 43 2431–43 10.1021/acs.jctc.1c00169
Reiter G Sommer J 2003 Polymer crystallization: obervations, concepts and interpretations Lecture Notes in Physics Springer
Reiter G Strobl G 2007 Progress in Understanding of Polymer Crystallization Springer
Loshchilov I Hutter F 2019 Decoupled weight decay regularization (arXiv:1711.05101)
Varrette S Bouvry P Cartiaux H Georgatos F 2014 Management of an academic hpc cluster: the ul experience 2014 Int. Conf. on High Performance Computing & Simulation (HPCS) pp 959 67 pp 959–67
Schoenholz S S Cubuk E D 2020 Jax m.d. a framework for differentiable physics Advances in Neural Information Processing Systems vol 33 992 1002 992–1002 10.5555/3495724.3496683
Schoenholz S S Cubuk E D 2021 Jax, m.d. a framework for differentiable physics J. Stat. Mech. 124016 10.1088/1742-5468/ac3ae9
Ambrosetti A Reilly A M DiStasio R A Tkatchenko A 2014 Long-range correlation energy calculated from coupled atomic response functions J. Chem. Phys. 140 18A508 10.1063/1.4865104
Kim M Kim W J Gould T Lee E K Lebégue S Kim H 2020 Kim, umbd: a materials-ready dispersion correction that uniformly treats metallic, ionic and van der waals bonding J. Am. Chem. Soc. 142 2346 54 2346–54 10.1021/jacs.9b11589
Hermann J Tkatchenko A 2020 Density functional model for van der waals interactions: unifying many-body atomic approaches with nonlocal functionals Phys. Rev. Lett. 124 146401 10.1103/PhysRevLett.124.146401
Hauseux P Ambrosetti A Bordas S P A Tkatchenko A 2022 Colossal enhancement of atomic force response in van der waals materials arising from many-body electronic correlations Phys. Rev. Lett. 128 106101 10.1103/PhysRevLett.128.106101
Shen Z Sosa R I Lengiewicz J Tkatchenko A Bordas S P A 2025 Machine learning surrogate modeling of many-body dispersion interactions in polymer melts Zenodo 10.5281/zenodo.15012728
Heek J Levskaya A Oliver A Ritter M Rondepierre B Steiner A van Zee M 2024 Flax: a neural network library and ecosystem for JAX (available at: http://github.com/google/flax)
Bradbury J Frostig R Hawkins P Johnson M J Leary C Maclaurin D Necula G Paszke A VanderPlas J Wanderman-Milne S Zhang Q 2018 JAX: composable transformations of Python+NumPy programs (available at: http://github.com/jax-ml/jax)
Chen B Siepmann J I 1999 Transferable potentials for phase equilibria. 3. explicit-hydrogen description of normal alkanes J. Phys. Chem. B 103 5370 9 5370–9 10.1021/jp990822m
Shen Z 2025 Efficient quantum-informed computational frameworks for mechanics of materials: many-body dispersion and machine learning surrogate modeling PhD Thesis University of Luxembourg (available at: https://hdl.handle.net/10993/66537)
Deshpande S Lengiewicz J Bordas S P A 2022 Probabilistic deep learning for real-time large deformation simulations Comput. Methods Appl. Mech. Eng. 398 115307 10.1016/j.cma.2022.115307
Deshpande S Bordas S P A Lengiewicz J 2024 MAgNET: a graph U-Net architecture for mesh-based simulations Eng. Appl. Artif. Intell. 133 108055 10.1016/j.engappai.2024.108055
Deshpande S Sosa R I Bordas S P A Lengiewicz J 2023 Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics Front. Mater. 10 1128954 10.3389/fmats.2023.1128954
Deshpande S Rappel H Hobbs M Bordas S P A Lengiewicz J 2025 Gaussian process regression + deep neural network autoencoder for probabilistic surrogate modeling in nonlinear mechanics of solids Comput. Methods Appl. Mech. Eng. 437 117790 10.1016/j.cma.2025.117790
Papavasileiou P Koronaki E D Pozzetti G Kathrein M Czettl C Boudouvis A G Bordas S P A 2023 equation-based and data-driven modeling strategies for industrial coating processes Comput. Ind. 149 103938 10.1016/j.compind.2023.103938
Vijayaraghavan S Wu L Noels L Bordas S P A Natarajan S Beex L A A 2023 A data-driven reduced-order surrogate model for entire elastoplastic simulations applied to representative volume elements Sci. Rep. 13 12781 10.1038/s41598-023-38104-x
Chen L Berke P Z Massart T J Bordas S P A Beex L A A 2022 An adaptive multiscale quasicontinuum approach for mechanical simulations of elastoplastic periodic lattices Mech. Res. Commun. 126 104019 10.1016/j.mechrescom.2022.104019
Dombroski J P Taylor S W Gill P M W 1996 Kwik: Coulomb energies in o(n) work J. Phys. Chem. 100 6272 6 6272–6 10.1021/jp952841b
Mayer A 2007 Formulation in terms of normalized propagators of a charge-dipole model enabling the calculation of the polarization properties of fullerenes and carbon nanotubes Phys. Rev. B 75 045407 10.1103/PhysRevB.75.045407
Hirshfeld F 1977 Bonded-atom fragments for describing molecular charge densities Theor. Chim. Acta 44 129 38 129–38 10.1007/BF00549096