Skip to main content Accessibility help

Parameterization of empirical forcefields for glassy silica using machine learning

  • Han Liu (a1), Zipeng Fu (a1) (a2), Yipeng Li (a1), Nazreen Farina Ahmad Sabri (a1) and Mathieu Bauchy (a1)...


The development of reliable, yet computationally efficient interatomic forcefields is key to facilitate the modeling of glasses. However, the parameterization of novel forcefields is challenging as the high number of parameters renders traditional optimization methods inefficient or subject to bias. Here, we present a new parameterization method based on machine learning, which combines ab initio molecular dynamics simulations and Bayesian optimization. By taking the example of glassy silica, we show that our method yields a new interatomic forcefield that offers an unprecedented agreement with ab initio simulations. This method offers a new route to efficiently parameterize new interatomic forcefields for disordered solids in a non-biased fashion.


Corresponding author

Address all correspondence to Mathieu Bauchy at


Hide All
1.Huang, P.Y., Kurasch, S., Alden, J.S., Shekhawat, A., Alemi, A.A., McEuen, P.L., Sethna, J.P., Kaiser, U., and Muller, D.A.: Imaging atomic rearrangements in two-dimensional silica glass: watching silica's dance. Science 342, 224227 (2013).
2.Huang, L. and Kieffer, J.: Challenges in modeling mixed ionic-covalent glass formers. In Molecular Dynamics Simulations of Disordered Materials, edited by Massobrio, C., Du, J., Bernasconi, M., Salmon, P.S. (Springer Series in Materials Science; Springer International Publishing: New York, 2015), pp. 87112.
3.Du, J.: Challenges in molecular dynamics simulations of multicomponent oxide glasses. In Molecular Dynamics Simulations of Disordered Materials, edited by Massobrio, C., Du, J., Bernasconi, M. and Salmon, P.S. (Springer Series in Materials Science; Springer International Publishing: New York, 2015), pp. 157180
4.Bauchy, M.: Deciphering the atomic genome of glasses by topological constraint theory and molecular dynamics: a review. Comput. Mater. Sci. 159, 95102 (2019).
5.Yu, Y., Wang, B., Wang, M., Sant, G., and Bauchy, M.: Revisiting silica with ReaxFF: towards improved predictions of glass structure and properties via reactive molecular dynamics. J. Non-Cryst. Solids 443, 148154 (2016).
6.Li, X., Song, W., Yang, K., Krishnan, N.M.A., Wang, B., Smedskjaer, M.M., Mauro, J.C., Sant, G., Balonis, M., and Bauchy, M.: Cooling rate effects in sodium silicate glasses: bridging the gap between molecular dynamics simulations and experiments. J. Chem. Phys. 147, 074501 (2017),
7.Ganster, P., Benoit, M., Delaye, J.-M., and Kob, W.: Structural and vibrational properties of a calcium aluminosilicate glass: classical force-fields vs. first-principles. Mol. Simul. 33, 10931103 (2007).
8.Behler, J.: Perspective: machine learning potentials for atomistic simulations. J. Chem. Phys. 145, 170901 (2016).
9.Carré, A., Ispas, S., Horbach, J., and Kob, W.: Developing empirical potentials from Ab initio simulations: the case of amorphous silica. Comput. Mater. Sci. 124, 323334 (2016),
10.Bartók, A.P., Kermode, J., Bernstein, N., and Csányi, G.: Machine learning a general-purpose interatomic potential for silicon. Phys. Rev. X 8, 132 (2018). .
11.Sundararaman, S., Huang, L., Ispas, S., and Kob, W.: New optimization scheme to obtain interaction potentials for oxide glasses. J. Chem. Phys. 148, 194504 (2018),
12.Wang, M., Anoop Krishnan, N.M., Wang, B., Smedskjaer, M.M., Mauro, J.C., and Bauchy, M.: A new transferable interatomic potential for molecular dynamics simulations of borosilicate glasses. J. Non-Cryst. Solids 498, 294304 (2018),
13.Lane, J.M.D.: Cooling rate and stress relaxation in silica melts and glasses via microsecond molecular dynamics. Phys. Rev. E 92, 012320 (2015).
14.van Beest, B.W.H., Kramer, G.J., and van Santen, R.A.: Force fields for silicas and aluminophosphates based on Ab initio calculations. Phys. Rev. Lett. 64, 19551958 (1990),
15.Carré, A., Horbach, J., Ispas, S., and Kob, W.: New fitting scheme to obtain effective potential from Car–Parrinello molecular-dynamics simulations: application to silica. EPL 82, 17001 (2008).
16.Ercolessi, F. and Adams, J.B.: Interatomic potentials from first-principles calculations: the force-matching method. EPL 26, 583 (1994).
17.Shewchuk, J.R.: An Introduction to the Conjugate Gradient Method Without the Agonizing Pain (Carnegie Mellon University, 1994).
18.Gubernatis, J.E. and Lookman, T.: Machine learning in materials design and discovery: examples from the present and suggestions for the future. Phys. Rev. Mater. 2, 115 (2018).
19.Ramprasad, R., Batra, R., Pilania, G., Mannodi-Kanakkithodi, A., and Kim, C.: Machine learning in materials informatics: recent applications and prospects. NPJ Computat. Mater. 3, 54 (2017).
20.Huan, T.D., Batra, R., Chapman, J., Krishnan, S., Chen, L., and Ramprasad, R.: A universal strategy for the creation of machine learning-based atomistic force fields. NPJ Computat. Mater. 3, 37 (2017).
21.Li, Y., Li, H., Pickard, F.C., Narayanan, B., Sen, F.G., Chan, M.K.Y., Sankaranarayanan, S.K.R.S., Brooks, B.R., and Roux, B.: Machine learning force field parameters from Ab initio data. J. Chem. Theory Comput. 13, 44924503 (2017).
22.Hellström, M. and Behler, J.: Neural network potentials in materials modeling. In Handbook of Materials Modeling, edited by Andreoni, W. and Yip, S. (Springer International Publishing: Cham, 2018), pp. 120.
23.Deringer, V.L. and Csányi, G.: Machine learning based interatomic potential for amorphous carbon. Phys. Rev. B 95, 094203 (2017),
24.Car, R. and Parrinello, M.: Unified approach for molecular dynamics and density-functional theory. Phys. Rev. Lett. 55, 24712474 (1985).
25.Bansal, N.P. and Doremus, R.H.: Handbook of Glass Properties (Elsevier: New York, 2013).
26.Fennell, C.J. and Gezelter, J.D.: Is the Ewald summation still necessary? Pairwise alternatives to the accepted standard for long-range electrostatics. J. Chem. Phys. 124, 234104 (2006).
27.Wright, A.C.: The comparison of molecular dynamics simulations with diffraction experiments. J. Non-Cryst. Solids 159, 264268 (1993).
28.Frazier, P.I. and Wang, J.: Bayesian optimization for materials design. In Information Science for Materials Discovery and Design, edited by (Springer Series in Materials Science; Springer, Cham, 2016), pp 4575.
29.Rasmussen, C.E. and Williams, C.K.I.: Gaussian Processes for Machine Learning, 3. print. Adaptive computation and machine learning (MIT Press: Cambridge, MA, 2008).
30.Liu, H., Fu, Z., Li, Y., Sabri, N.F.A., and Bauchy, M.: Balance between accuracy and simplicity in empirical forcefields for glass modeling: insights from machine learning. J. Non-Cryst. Solids (2019).
31.Wang, B., Yu, Y., Lee, Y.J., and Bauchy, M.: Intrinsic nano-ductility of glasses: the critical role of composition. Front. Mater. 2, 11 (2015).
Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

MRS Communications
  • ISSN: 2159-6859
  • EISSN: 2159-6867
  • URL: /core/journals/mrs-communications
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed