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

The analysis of first-principles molecular dynamics simulations allows us to accurately predict the physical and chemical properties of fluids. This method can be applied to any atomic analysis in physics and chemistry. To begin, extract each specific set of physical properties using one or more dedicated Python scripts from the package.

Run all the scripts at the command line. They all employ a series of flags, which are as consistent as possible from one script to another. Transform the output of the MD simulation performed in a first-principles code into a UMD file, then transfer the umd files into xyz files to facilitate visualization on various other packages, like VMD or VESTA.

Reverse the UMD file into VASP-type POSCAR files using the umd2poscar. py script, selecting snapshots of the simulations with the predefined frequency. Run the gofrs_umd.

py script to compute the pair distribution function for all the pairs of atomic types A and B.The output is written in one ASCII file tab separated with the extension gofrs.dat. Extract the average interatomic bond distances as the radii of the first coordination spheres. For this, identify the position of the first maximum of the pair distribution functions by plotting the gofrs.

dat file in a spreadsheet application and searching for the maxima and minima for each pair of atoms, then identify the radius of the first coordination sphere as the first minimum of the PDF using the spreadsheet software. Run the speciation script to obtain the connectivity matrix and obtain the coordination polyhedra or the polymerization. Run the speciation_umd.

py script with the flag r0, which samples the connectivity graph at the first level to identify the coordination polyhedra. Run the speciation_umd. py script with the flag r1, which samples the connectivity graph at all depth levels to obtain the polymerization.

Plot the lifetime of each atomic cluster of all the chemical species found in the simulation as found in the papule. dat files. Extract the mean square displacements or MSD of the atoms as a function of time to obtain the self-diffusivity, then compute the MSD using the series of msd_umd.

py scripts and compute the average MSD of each atomic type. Compute the MSD of each atom and of the chemical species. Plot the MSD using a spreadsheet-based software and compute the diffusion coefficients from the slope of the MSD.

Run the vibr_spectrum_umd. py script to compute the atomic velocity velocity autocorrelation or VAC function for each atomic type and perform its fast Fourier transform. Plot the vibrational spectrum from the vibr.

dat file using spreadsheet-like software. Identify the finite value at omega equals zero that corresponds to the diffusive character of the fluid in the various peaks of the spectrum at finite frequency. Run averages.

py to extract the average values and the spread for pressure, temperature, density, and internal energy from the UMD files. Finally, run the full averages. py script to perform the complete statistical analysis including the error of the mean.

Pyrolite is a model multi-component silicate melt that best approximates the composition of the bulk silicate Earth. The UMD package was used to extract several characteristic features of molten pyrolite. The maximum of the silicon-oxygen pair distribution function lies at 1.635 angstrom, which is the best approximation to the bend length.

Using this limit as the silicon-oxygen bond distance, the speciation analysis shows that orthosilicate units that can last for up to a few picoseconds dominate the melt. There is an important part of the melt that shows partial polymerization as reflected by the presence of dimers like disilicate and trimers like Si3Ox units. Their corresponding lifetime is in the order of the picosecond.

Higher order polymers all have considerably shorter lifetime. The different values of the vertical and horizontal steps yield various samplings of the MSD. Even large values of Z and V are enough to define the slopes and thus the diffusion coefficients of the different atoms.

The post-processing time increases dramatically for large values of Z and V.Finally, the atomic velocity autocorrelation functions yield the vibrational spectrum of the melt. Shown here are the contributions of magnesium, silicon, and oxygen atoms, as well as the total value. When attempting this protocol, always check the convergence.

Be sure that the atomic trajectories are long enough such as to capture properly the phenomenon you are interested in. This technique covers post-processing of simulation results. The simulations and their analysis must be done in parallel.