We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cond-mat.mtrl-sci

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Condensed Matter > Materials Science

Title: A deep machine learning potential for atomistic simulation of Fe-Si-O systems under Earth's outer core conditions

Abstract: Using artificial neural-network machine learning (ANN-ML) to generate interatomic potentials has been demonstrated to be a promising approach to address the long-standing challenge of accuracy versus efficiency in molecular dynamics (MD) simulations. Here, taking the Fe-Si-O system as a prototype, we show that accurate and transferable ANN-ML potentials can be developed for reliable MD simulations of materials at high-pressure and high-temperature conditions of the Earth's outer core. The ANN-ML potential for Fe-Si-O system is trained by fitting to the energies and forces of related binaries and ternary liquid structures at high pressures and temperatures obtained by first-principles calculations based on density functional theory (DFT). We show that the generated ANN-ML potential describes well the structure and dynamics of liquid phases of this complex system. The efficient ANN-ML potential with DFT accuracy provides a promising scheme for accurate atomistic simulations of structures and dynamics of complex Fe-Si-O system in the Earth's outer core.
Subjects: Materials Science (cond-mat.mtrl-sci)
Cite as: arXiv:2201.11835 [cond-mat.mtrl-sci]
  (or arXiv:2201.11835v2 [cond-mat.mtrl-sci] for this version)

Submission history

From: Chao Zhang [view email]
[v1] Thu, 27 Jan 2022 22:41:44 GMT (1260kb)
[v2] Thu, 19 May 2022 06:11:53 GMT (1260kb)

Link back to: arXiv, form interface, contact.