Application of First-Principles-Based Artificial Neural Network Potentials to Multiscale-Shock Dynamics Simulations on Solid Materials.

Artificial neural-network (ANN) potential trained with first-principles calculations have emerged as a promising approach for molecular dynamics (MD) simulations encompassing large space and time scales while retaining first-principles accuracy. However, the application of ANN-MD has thus far been limited to near-equilibrium processes. Here, we combine first principles-trained ANN-MD with multiscale shock theory (MSST) to successfully describe far-from-equilibrium shock phenomena. Our ANN-MSST-MD approach describes shock-wave propagation in solid with first-principles accuracy, but within 5,000-times shorter computing time. Accordingly, ANN-MD-MSST was able to resolve fine, long-time elastic deformation at low shock speed, which was impossible with first-principles MD due to the high computational cost. This work thus lays a foundation of ANN-MD simulation to study a wide range of far-from-equilibrium processes.

[1]  A. Nakano,et al.  Picosecond amorphization of SiO2 stishovite under tension , 2017, Science Advances.

[2]  A. V. Duin,et al.  ReaxFF: A Reactive Force Field for Hydrocarbons , 2001 .

[3]  G. Ji,et al.  Anisotropic responses and initial decomposition of condensed-phase β-HMX under shock loadings via molecular dynamics simulations in conjunction with multiscale shock technique. , 2014, The journal of physical chemistry. B.

[4]  E. Reed,et al.  Ultrafast transformation of graphite to diamond: an ab initio study of graphite under shock compression. , 2008, The Journal of chemical physics.

[5]  Rajiv K. Kalia,et al.  A crossover in anisotropic nanomechanochemistry of van der Waals crystals , 2015 .

[6]  Burke,et al.  Generalized Gradient Approximation Made Simple. , 1996, Physical review letters.

[7]  Evan J Reed,et al.  A method for tractable dynamical studies of single and double shock compression. , 2003, Physical review letters.

[8]  Nongnuch Artrith,et al.  An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for TiO2 , 2016 .

[9]  Jörg Behler,et al.  Accurate Neural Network Description of Surface Phonons in Reactive Gas–Surface Dynamics: N2 + Ru(0001) , 2017, The journal of physical chemistry letters.

[10]  M. Klein,et al.  Nosé-Hoover chains : the canonical ensemble via continuous dynamics , 1992 .

[11]  Jörg Behler,et al.  Nuclear Quantum Effects in Sodium Hydroxide Solutions from Neural Network Molecular Dynamics Simulations. , 2018, The journal of physical chemistry. B.

[12]  R. Wentzcovitch,et al.  Spin-State Crossover of Iron in Lower-Mantle Minerals: Results of DFT+U Investigations , 2010 .

[13]  Rajiv K. Kalia,et al.  Crystalline anisotropy of shock-induced phenomena: Omni-directional multiscale shock technique , 2016 .

[14]  J. Behler,et al.  Representing molecule-surface interactions with symmetry-adapted neural networks. , 2007, The Journal of chemical physics.

[15]  Kamal Choudhary,et al.  Elastic properties of bulk and low-dimensional materials using Van der Waals density functional. , 2018, Physical review. B.

[16]  Jörg Behler,et al.  Constructing high‐dimensional neural network potentials: A tutorial review , 2015 .

[17]  J. Behler,et al.  Construction of high-dimensional neural network potentials using environment-dependent atom pairs. , 2012, The Journal of chemical physics.

[18]  F. Wakai,et al.  Large increase in fracture resistance of stishovite with crack extension less than one micrometer , 2015, Scientific Reports.

[19]  Shogo Fukushima,et al.  QXMD: An open-source program for nonadiabatic quantum molecular dynamics , 2019, SoftwareX.

[20]  Jerry Wackerle,et al.  Shock‐Wave Compression of Quartz , 1962 .

[21]  Gerbrand Ceder,et al.  Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species , 2017, 1706.06293.

[22]  Blöchl,et al.  Projector augmented-wave method. , 1994, Physical review. B, Condensed matter.

[23]  A. Nakano,et al.  Guidelines for creating artificial neural network empirical interatomic potential from first-principles molecular dynamics data under specific conditions and its application to α-Ag2Se. , 2019, The Journal of chemical physics.

[24]  Raymond Jeanloz,et al.  Equation of state of stishovite and interpretation of SiO2 shock‐compression data , 2003 .

[25]  Rajiv K. Kalia,et al.  Linear-scaling density-functional-theory calculations of electronic structure based on real-space grids: design, analysis, and scalability test of parallel algorithms , 2001 .

[26]  Jörg Behler,et al.  Maximally resolved anharmonic OH vibrational spectrum of the water/ZnO(101¯0) interface from a high-dimensional neural network potential. , 2018, The Journal of chemical physics.

[27]  Michele Parrinello,et al.  Generalized neural-network representation of high-dimensional potential-energy surfaces. , 2007, Physical review letters.

[28]  Priya Vashishta,et al.  A divide-conquer-recombine algorithmic paradigm for large spatiotemporal quantum molecular dynamics simulations. , 2014, The Journal of chemical physics.

[29]  K. Funakoshi,et al.  Curious kinetic behavior in silica polymorphs solves seifertite puzzle in shocked meteorite , 2015, Science Advances.

[30]  F. Wakai,et al.  Strength and toughness of nanocrystalline SiO2 stishovite toughened by fracture-induced amorphization , 2017 .

[31]  J. Behler.  Atom-centered symmetry functions for constructing high-dimensional neural network potentials. , 2011, The Journal of chemical physics.

[32]  E. Reed,et al.  Nanosecond homogeneous nucleation and crystal growth in shock-compressed SiO2. , 2016, Nature materials.

[33]  A. V. van Duin,et al.  Development of the ReaxFF reactive force field for describing transition metal catalyzed reactions, with application to the initial stages of the catalytic formation of carbon nanotubes. , 2005, The journal of physical chemistry. A.

[34]  H. J. Lee,et al.  Ultrafast visualization of crystallization and grain growth in shock-compressed SiO2 , 2015, Nature Communications.