The configurations reported in this database were generated using classical molecular dynamics (MD), using an embedded atom model (EAM) potential, as developed by Sheng et al. [1]. Simulations were performed over a range of temperatures and composition (ratio of P vs Ni). The quantum-mechanical total energy of configurations extracted from these MD simulations were then re-computed using density functional theory (DFT) as implemented in the Quantum ESPRESSO package [2].
More details on the generation of this database and on its use for training neural network interatomic potentials via the AENet package [3] can be found in:
M. S. Khan, N. Artith, and O. Andreussi, "Understanding Structure-Composition-Property Relationships of Ni-P Bulk Metallic Glasses", under review (2025)
Md. Sharif Khan, Department of Chemistry and Biochemistry, Boise State University
Oliviero Andreussi, Department of Chemistry and Biochemistry, Boise State University
The training dataset contains 20,000 structures saved in the XCrySDen structure format (XSF).
For each .xsf structure we have:
- The total energy from a DFT calculation (Quantum Espresso), with details included in the NiP-qe.in file,
- The simulation cell vectors,
- The number of atoms in the configuration,
- For each atom in the system, the element, Cartesian coordinates, and forces on that atom.
The authors thank Dr. Sundeep Mukherjee for the initial discussions that started this research project. N.A. acknowledges a start-up grant (Dutch Sector Plan) from Utrecht University. We thank Boise State for providing the startup funds that supported this research, together with the NSF CAREER award #2306929.
[1] H. W. Sheng, E. Ma, M. J. Kramer, "Relating Dynamic Properties to Atomic Structure in Metallic Glasses," JOM, vol. 64, no. 2, pp. 856–865, 2012. https://doi.org/10.1007/s11837-012-0360-y
[2] P. Giannozzi et al., "QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials", J. Phys.: Cond. Mat., 39, 395502 (2009), https://doi.org/10.1088/0953-8984/21/39/395502
[3] J. López-Zorrilla, X. M. Aretxabaleta, I.W. Yeu, I. Etxebarria, H. Manzano, and N. Artrith,"ænet-PyTorch: A GPU-supported implementation for machine learning atomic potentials training", J. Chem. Phys., 158, 164105 (2023), https://doi.org/10.1063/5.0146803.