Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential.
Journal
Nature chemistry
ISSN: 1755-4349
Titre abrégé: Nat Chem
Pays: England
ID NLM: 101499734
Informations de publication
Date de publication:
07 Mar 2024
07 Mar 2024
Historique:
received:
13
04
2023
accepted:
12
12
2023
medline:
8
3
2024
pubmed:
8
3
2024
entrez:
7
3
2024
Statut:
aheadofprint
Résumé
Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.
Identifiants
pubmed: 38454071
doi: 10.1038/s41557-023-01427-3
pii: 10.1038/s41557-023-01427-3
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Subventions
Organisme : DOE | LDRD | Los Alamos National Laboratory (Los Alamos Lab)
ID : 20210087DR
Organisme : DOE | LDRD | Los Alamos National Laboratory (Los Alamos Lab)
ID : 20210087DR
Organisme : DOE | LDRD | Los Alamos National Laboratory (Los Alamos Lab)
ID : 20210087DR
Organisme : DOE | LDRD | Los Alamos National Laboratory (Los Alamos Lab)
ID : 20210087DR
Organisme : DOE | SC | Basic Energy Sciences (BES)
ID : 89233218CNA000001
Organisme : DOE | SC | Basic Energy Sciences (BES)
ID : 89233218CNA000001
Organisme : DOE | SC | Basic Energy Sciences (BES)
ID : 89233218CNA000001
Organisme : DOE | SC | Basic Energy Sciences (BES)
ID : 89233218CNA000001
Organisme : United States Department of Defense | United States Navy | Office of Naval Research (ONR)
ID : N00014-21-1-2476
Informations de copyright
© 2024. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.
Références
Zuo, Y. et al. Performance and cost assessment of machine learning interatomic potentials. J. Phys. Chem. A 124, 731–745 (2020).
pubmed: 31916773
doi: 10.1021/acs.jpca.9b08723
Behler, J. First principles neural network potentials for reactive simulations of large molecular and condensed systems. Angew. Chem. Int. Ed. 56, 12828–12840 (2017).
doi: 10.1002/anie.201703114
Kulichenko, M. et al. The rise of neural networks for materials and chemical dynamics. J. Phys. Chem. Lett. 12, 6227–6243 (2021).
pubmed: 34196559
doi: 10.1021/acs.jpclett.1c01357
Bartók, A. P. & Csányi, G. Gaussian approximation potentials: a brief tutorial introduction. Int. J. Quantum Chem. 115, 1051–1057 (2015).
doi: 10.1002/qua.24927
Batzner, S. et al. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. 13, 2453 (2022).
pubmed: 35508450
pmcid: 9068614
doi: 10.1038/s41467-022-29939-5
Thölke, P. & Fabritiis, G. D. Equivariant transformers for neural network based molecular potentials. In International Conference on Learning Representations https://openreview.net/forum?id=zNHzqZ9wrRB (2022).
Musaelian, A. et al. Learning local equivariant representations for large-scale atomistic dynamics. Nat. Comm. 14, 579 (2023).
doi: 10.1038/s41467-023-36329-y
Khorshidi, A. & Peterson, A. A. Amp: a modular approach to machine learning in atomistic simulations. Comput. Phys. Commun. 207, 310–324 (2016).
doi: 10.1016/j.cpc.2016.05.010
Yao, K., Herr, J. E., Toth, D. W., Mckintyre, R. & Parkhill, J. The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics. Chem. Sci.9, 2261–2269 (2018).
pubmed: 29719699
pmcid: 5897848
doi: 10.1039/C7SC04934J
Singraber, A., Morawietz, T., Behler, J. & Dellago, C. Parallel multistream training of high-dimensional neural network potentials. J. Chem. Theory Comput. 15, 3075–3092 (2019).
pubmed: 30995035
doi: 10.1021/acs.jctc.8b01092
Kang, P.-L. & Liu, Z.-P. Reaction prediction via atomistic simulation: from quantum mechanics to machine learning. iScience 24, 102013 (2021).
pubmed: 33490920
doi: 10.1016/j.isci.2020.102013
Behler, J. & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Phys. Rev. Lett. 98, 146401 (2007).
pubmed: 17501293
doi: 10.1103/PhysRevLett.98.146401
Haghighatlari, M. et al. NewtonNet: a Newtonian message passing network for deep learning of interatomic potentials and forces. Digital Discov. 1, 333–343 (2022).
doi: 10.1039/D2DD00008C
Batatia, I., Kovacs, D. P., Simm, G., Ortner, C. & Csanyi, G. in Advances in Neural Information Processing Systems (eds Koyejo, S. et al.) 35, 11423–11436 (Curran Associates, 2022).
Chigaev, M. et al. Lightweight and effective tensor sensitivity for atomistic neural networks. J. Chem. Phys. 158, 184108 (2023).
pubmed: 37158328
doi: 10.1063/5.0142127
Unke, O. T. et al. SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects. Nat. Commun. 12, 7273 (2021).
pubmed: 34907176
pmcid: 8671403
doi: 10.1038/s41467-021-27504-0
Smith, J. S., Nebgen, B., Lubbers, N., Isayev, O. & Roitberg, A. E. Less is more: sampling chemical space with active learning. J. Chem. Phys. 148, 241733 (2018).
pubmed: 29960353
doi: 10.1063/1.5023802
Devereux, C. et al. Extending the applicability of the ANI deep learning molecular potential to sulfur and halogens. J. Chem. Theory Comput. 16, 4192–4202 (2020).
pubmed: 32543858
doi: 10.1021/acs.jctc.0c00121
Young, T. A., Johnston-Wood, T., Zhang, H. & Duarte, F. Reaction dynamics of Diels–Alder reactions from machine learned potentials. Phys. Chem. Chem. Phys. 24, 20820–20827 (2022).
pubmed: 36004770
doi: 10.1039/D2CP02978B
Jiang, B., Li, J. & Guo, H. High-fidelity potential energy surfaces for gas-phase and gas-surface scattering processes from machine learning. J. Phys. Chem. Lett. 11, 5120–5131 (2020).
pubmed: 32517472
doi: 10.1021/acs.jpclett.0c00989
Kolb, B., Zhao, B., Li, J., Jiang, B. & Guo, H. Permutation invariant potential energy surfaces for polyatomic reactions using atomistic neural networks. J. Chem. Phys. 144, 224103 (2016).
pubmed: 27305992
doi: 10.1063/1.4953560
Cooper, A. M., Hallmen, P. P. & Kästner, J. Potential energy surface interpolation with neural networks for instanton rate calculations. J. Chem. Phys. 148, 094106 (2018).
doi: 10.1063/1.5015950
Li, J., Song, K. & Behler, J. A critical comparison of neural network potentials for molecular reaction dynamics with exact permutation symmetry. Phys. Chem. Chem. Phys. 21, 9672–9682 (2019).
pubmed: 30672927
doi: 10.1039/C8CP06919K
Zeng, J., Cao, L., Xu, M., Zhu, T. & Zhang, J. Z. Complex reaction processes in combustion unraveled by neural network-based molecular dynamics simulation. Nat. Commun. 11, 5713 (2020).
pubmed: 33177517
pmcid: 7658983
doi: 10.1038/s41467-020-19497-z
Chen, R., Shao, K., Fu, B. & Zhang, D. H. Fitting potential energy surfaces with fundamental invariant neural network. II. Generating fundamental invariants for molecular systems with up to ten atoms. J. Chem. Phys. 152, 204307 (2020).
pubmed: 32486688
doi: 10.1063/5.0010104
Takamoto, S. et al. Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements. Nat. Commun. 13, 2991 (2022).
pubmed: 35637178
pmcid: 9151783
doi: 10.1038/s41467-022-30687-9
Takamoto, S., Izumi, S. & Li, J. TeaNet: universal neural network interatomic potential inspired by iterative electronic relaxations. Comput. Mater. Sci. 207, 111280 (2022).
doi: 10.1016/j.commatsci.2022.111280
Chen, C. & Ong, S. P. A universal graph deep learning interatomic potential for the periodic table. Nat. Comput. Sci. 2, 718–728 (2022).
pubmed: 38177366
doi: 10.1038/s43588-022-00349-3
Behler, J. Constructing high-dimensional neural network potentials: a tutorial review. Int. J. Quantum Chem. 115, 1032–1050 (2015).
doi: 10.1002/qua.24890
Ren, P. et al. A survey of deep active learning. ACM Comput. Surv. 54, 1–40 (2021).
Sivaraman, G. et al. Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide. NPJ Comput. Mater. 6, 104 (2020).
doi: 10.1038/s41524-020-00367-7
Smith, J. S. et al. Automated discovery of a robust interatomic potential for aluminum. Nat. Commun. 12, 1257 (2021).
pubmed: 33623036
pmcid: 7902823
doi: 10.1038/s41467-021-21376-0
Yoo, P. et al. Neural network reactive force field for C, H, N, and O systems. NPJ Comput. Mater. 7, 9 (2021).
doi: 10.1038/s41524-020-00484-3
Zaverkin, V., Holzmüller, D., Steinwart, I. & Kästner, J. Exploring chemical and conformational spaces by batch mode deep active learning. Digit. Discov. 1, 605–620 (2022).
doi: 10.1039/D2DD00034B
Young, T. A., Johnston-Wood, T., Deringer, V. L. & Duarte, F. A transferable active-learning strategy for reactive molecular force fields. Chem. Sci. 12, 10944–10955 (2021).
pubmed: 34476072
pmcid: 8372546
doi: 10.1039/D1SC01825F
Ang, S. J., Wang, W., Schwalbe-Koda, D., Axelrod, S. & Gómez-Bombarelli, R. Active learning accelerates ab initio molecular dynamics on reactive energy surfaces. Chem 7, 738–751 (2021).
doi: 10.1016/j.chempr.2020.12.009
Guan, X. et al. A benchmark dataset for hydrogen combustion. Sci. Data 9, 215 (2022).
pubmed: 35581204
pmcid: 9114378
doi: 10.1038/s41597-022-01330-5
Warshel, A. & Weiss, R. M. An empirical valence bond approach for comparing reactions in solutions and in enzymes. J. Am. Chem. Soc. 102, 6218–6226 (1980).
doi: 10.1021/ja00540a008
Baskes, M. Determination of modified embedded atom method parameters for nickel. Mater. Chem. Phys. 50, 152–158 (1997).
doi: 10.1016/S0254-0584(97)80252-0
van Duin, A. C. T., Dasgupta, S., Lorant, F. & Goddard, W. A. ReaxFF: a reactive force field for hydrocarbons. J. Phys. Chem. A 105, 9396–9409 (2001).
doi: 10.1021/jp004368u
Brenner, D. W. et al. A second-generation reactive empirical bond order (REBO) potential energy expression for hydrocarbons. J. Phys. Condens. Matter 14, 783–802 (2002).
doi: 10.1088/0953-8984/14/4/312
Senftle, T. P. et al. The ReaxFF reactive force-field: development, applications and future directions. NPJ Comput. Mater. 2, 15011 (2016).
doi: 10.1038/npjcompumats.2015.11
Schreiner, M., Bhowmik, A., Vegge, T., Busk, J. & Winther, O. Transition1x—a dataset for building generalizable reactive machine learning potentials. Sci. Data 9, 779 (2022).
pubmed: 36566281
pmcid: 9789978
doi: 10.1038/s41597-022-01870-w
Wang, L.-P. et al. Discovering chemistry with an ab initio nanoreactor. Nat. Chem. 6, 1044–1048 (2014).
pubmed: 25411881
pmcid: 4239668
doi: 10.1038/nchem.2099
Wang, L.-P. in Computational Approaches for Chemistry Under Extreme Conditions 127–159 (Springer, 2019).
Deringer, V. L. & Csányi, G. Machine learning based interatomic potential for amorphous carbon. Phys. Rev. B 95, 094203 (2017).
doi: 10.1103/PhysRevB.95.094203
Powles, R. C., Marks, N. A. & Lau, D. W. M. Self-assembly of sp
doi: 10.1103/PhysRevB.79.075430
Tomas, C. D., Suarez-Martinez, I. & Marks, N. A. Graphitization of amorphous carbons: a comparative study of interatomic potentials. Carbon 109, 681–693 (2016).
doi: 10.1016/j.carbon.2016.08.024
Lei, T. et al. Mechanism of graphene formation via detonation synthesis: a DFTB nanoreactor approach. J. Chem. Theory Comput. 15, 3654–3665 (2019).
pubmed: 31117479
doi: 10.1021/acs.jctc.9b00158
Chen, Z., Sun, W. & Zhao, L. Combustion mechanisms and kinetics of fuel additives: a ReaxFF molecular simulation. Energy Fuels 32, 11852–11863 (2018).
doi: 10.1021/acs.energyfuels.8b02035
Miller, S. L. & Urey, H. C. Organic compound synthesis on the primitive earth. Science 130, 245–251 (1959).
pubmed: 13668555
doi: 10.1126/science.130.3370.245
Saitta, A. M. & Saija, F. Miller experiments in atomistic computer simulations. Proc. Natl Acad. Sci. USA 111, 13768–13773 (2014).
pubmed: 25201948
pmcid: 4183268
doi: 10.1073/pnas.1402894111
Los, J. H., Ghiringhelli, L. M., Meijer, E. J. & Fasolino, A. Improved long-range reactive bond-order potential for carbon I construction. Phys. Rev. B 72, 214102 (2005).
doi: 10.1103/PhysRevB.72.214102
Srinivasan, S. G., Van Duin, A. C. & Ganesh, P. Development of a ReaxFF potential for carbon condensed phases and its application to the thermal fragmentation of a large fullerene. J. Phys. Chem. A 119, 571–580 (2015).
pubmed: 25562718
doi: 10.1021/jp510274e
Wang, J. et al. A deep learning interatomic potential developed for atomistic simulation of carbon materials. Carbon 186, 1–8 (2022).
doi: 10.1016/j.carbon.2021.09.062
Sorensen, C., Nepal, A. & Singh, G. P. Process for high-yield production of graphene via detonation of carbon-containing material. US patent 9,440,857 (2016).
Cooper, S. P., Mathieu, O., Schoegl, I. & Petersen, E. L. High-pressure ignition delay time measurements of a four-component gasoline surrogate and its high-level blends with ethanol and methyl acetate. Fuel 275, 118016 (2020).
doi: 10.1016/j.fuel.2020.118016
Brickel, S., Das, A. K., Unke, O. T., Turan, H. T. & Meuwly, M. Reactive molecular dynamics for the [Cl–CH3–Br]
doi: 10.1088/2516-1075/ab1edb
Li, J., Chen, J., Zhang, D. H. & Guo, H. Quantum and quasi-classical dynamics of the OH + CO → H + CO
pubmed: 25669543
doi: 10.1063/1.4863138
Smith, J. S. et al. Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning. Nat. Commun. 10, 2903 (2019).
pubmed: 31263102
pmcid: 6602931
doi: 10.1038/s41467-019-10827-4
Allen, A. E. A. et al. Learning together: towards foundational models for machine learning interatomic potentials with meta-learning. Preprint at arXiv https://doi.org/10.48550/arXiv.2307.04012 (2023).
Eckhoff, M. & Reiher, M. Lifelong machine learning potentials. J. Chem. Theory Comput. 19, 3509–3525 (2023).
pubmed: 37288932
pmcid: 10308836
doi: 10.1021/acs.jctc.3c00279
Rezajooei, N., Thien Phuc, T. N., Johnson, E. & Rowley, C. A neural network potential with rigorous treatment of long-range dispersion. Digit. Discov. 2, 718–727 (2023).
doi: 10.1039/D2DD00150K
Ko, T. W., Finkler, J. A., Goedecker, S. & Behler, J. A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer. Nat. Commun. 12, 398 (2021).
pubmed: 33452239
pmcid: 7811002
doi: 10.1038/s41467-020-20427-2
Bommasani, R. et al. On the opportunities and risks of foundation models. Preprint at arXiv https://doi.org/10.48550/arXiv.2108.07258 (2022).
Smith, J. S. et al. The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules. Sci. Data 7, 134 (2020).
pubmed: 32358545
pmcid: 7195467
doi: 10.1038/s41597-020-0473-z
Smith, J. S., Isayev, O. & Roitberg, A. E. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost. Chem. Sci. J. 8, 3192–3203 (2017).
Zubatyuk, R., Smith, J. S., Leszczynski, J. & Isayev, O. Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network. Sci. Adv. 5, eaav6490 (2019).
pubmed: 31448325
pmcid: 6688864
doi: 10.1126/sciadv.aav6490
Smith, J. S., Lubbers, N., Thompson, A. P. & Barros, K. Simple and efficient algorithms for training machine learning potentials to force data. Preprint at arXiv https://doi.org/10.48550/arXiv.2006.05475 (2020).
Seung, H. S., Opper, M. & Sompolinsky, H. Query by Committee. In Proc. Association for Computing Machinery (1992). https://doi.org/10.1145/130385.130417
Kühne, T. D. et al. CP2K: An electronic structure and molecular dynamics software package—Quickstep: efficient and accurate electronic structure calculations. J. Chem. Phys. 152, 194103 (2020).
pubmed: 33687235
doi: 10.1063/5.0007045
Kohn, W. & Sham, L. J. Self-consistent equations including exchange and correlation effects. Phys. Rev. 140, A1133–A1138 (1965).
doi: 10.1103/PhysRev.140.A1133
Becke, A. D. Density-functional exchange-energy approximation with correct asymptotic behavior. Phys. Rev. A 38, 3098–3100 (1988).
doi: 10.1103/PhysRevA.38.3098
Lee, C., Yang, W. & Parr, R. G. Development of the Colle–Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 37, 785–789 (1988).
doi: 10.1103/PhysRevB.37.785
VandeVondele, J. & Hutter, J. Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases. J. Chem. Phys. 127, 114105 (2007).
pubmed: 17887826
doi: 10.1063/1.2770708
Goedecker, S., Teter, M. & Hutter, J. Separable dual-space Gaussian pseudopotentials. Phys. Rev. B 54, 1703–1710 (1996).
doi: 10.1103/PhysRevB.54.1703
Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).
pubmed: 20423165
doi: 10.1063/1.3382344
Jadrich, R. B., Ticknor, C. & Leiding, J. A. First principles reactive simulation for equation of state prediction. J. Chem. Phys. 154, 244307 (2021).
pubmed: 34241343
doi: 10.1063/5.0050676
Fetisov, E. O. et al. First-principles Monte Carlo simulations of reaction equilibria in compressed vapors. ACS Cent. Sci. 2, 409–415 (2016).
pubmed: 27413785
pmcid: 4919768
doi: 10.1021/acscentsci.6b00095
van der Maaten, L. & Hinton, G. Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008).
Kim, S. et al. PubChem 2023 update. Nucleic Acids Res. 51, D1373–D1380 (2022).
pmcid: 9825602
doi: 10.1093/nar/gkac956
Hagberg, A. A., Schult, D. A. & Swart, P. J. Exploring network structure, dynamics, and function using NetworkX. In Proc. 7th Python in Science Conference (eds Varoquaux, G., Vaught, T. & Millman, J.) 11–15 (2008).
Larsen, A. H. et al. The atomic simulation environment—a Python library for working with atoms. J. Phys. Condens. Matter 29, 273002 (2017).
doi: 10.1088/1361-648X/aa680e
Thompson, A. P. et al. LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. Comp. Phys. Comm. 271, 108171 (2022).
doi: 10.1016/j.cpc.2021.108171
Musaelian, A. et al. Learning local equivariant representations for large-scale atomistic dynamics. Nat. Commun. 14, 579 (2023).
pubmed: 36737620
pmcid: 9898554
doi: 10.1038/s41467-023-36329-y
Stukowski, A. Visualization and analysis of atomistic simulation data with OVITO–the open visualization tool. Model. Simul. Mat. Sci. Eng. 18, 015012 (2009).
doi: 10.1088/0965-0393/18/1/015012
Martínez, L., Andrade, R., Birgin, E. G. & Martínez, J. M. PACKMOL: a package for building initial configurations for molecular dynamics simulations. J. Comput. Chem. 30, 2157–2164 (2009).
pubmed: 19229944
doi: 10.1002/jcc.21224
Liu, D. C. & Nocedal, J. On the limited memory BFGS method for large scale optimization. Math. Program. 45, 503–528 (1989).
doi: 10.1007/BF01589116