Analyzing Melts and Fluids from Ab Initio Molecular Dynamics Simulations with the UMD Package.

Area of Science:

• Computational chemistry
• Materials science
• Geophysics

Background:

• Ab initio molecular-dynamics simulations generate complex data.
• Analyzing fluid dynamics, especially in natural systems like melts and supercritical fluids, requires specialized tools.
• Existing analysis methods can be cumbersome and lack standardization.

Purpose of the Study:

• To develop an open-source Python package for analyzing ab initio molecular-dynamics simulations.
• To provide a standardized format (UMD files) for storing simulation data.
• To enable the computation of key structural, transport, and thermodynamic properties.

Main Methods:

• Development of a Python-based open-source package with libraries for file formats and crystallography.
• Implementation of scripts for command-line execution.
• Proposal and utilization of a Universal Molecular Dynamics (UMD) file format for atomic trajectories and thermodynamic data.

Main Results:

• The UMD package computes pair-distribution functions, bond lengths, interatomic connectivity, and chemical speciation.
• Analysis includes chemical species lifetime, mean-square displacements, diffusion coefficients, vibrational spectra, and viscosity.
• The package facilitates statistical analysis of simulation results.

Conclusions:

• The developed Python package offers a comprehensive and accessible solution for analyzing molecular dynamics simulations of fluids.
• It simplifies the study of natural systems like silicate melts and supercritical fluids.
• The open-source nature and standardized UMD format promote wider adoption and data sharing in the scientific community.