Skip to content

Documentation and discussion regarding the study of proton conduction in graphanol using deep learning potentials.

Notifications You must be signed in to change notification settings

JKarlJohnson/protonConduction_graphanol

 
 

Repository files navigation

Proton Conduction Graphanol

Documentation and discussion regarding the study of proton conduction in graphanol using deep learning potentials.

Primary Files

  1. cec.py: Script to compute the center of excess charge (CEC) to locate a proton in graphanol.
  2. lmc.py: Script to perform lattice Monte Carlo (LMC) proton transfer simulation for graphanol. The code uses graph models constructed with Networkx.
  3. goh.pb: DeePMD potential (DP) for graphanol.
  4. goh.poscar: Sample POSCAR file of 24C graphanol.

Miscellaneous Files

  1. z-axis-check.py: Script that takes in XDATCAR file from a simulation and checks if all the atoms are within the accepted upper and lower z-planes. This is convenient for quickly checking the stability 2-D surfaces with dangling bonds that are prone to breaking. As for my case, I am interested in studying the proton conduction in graphanol, a lot of simulations are prone to bond breaking, which get hard to detect from mean-squared displacement calculations.
  2. dump2xdatcar-sort.py: Script to convert LAMMPS dump file to XDATCAR. It also sorts the atoms if LAMMPS jumbles the atom index order.
  3. breakXDATCAR.py: Script that takes in a XDATCAR file and then breakes it into smaller XDATCAR files of equal time spans. This is useful if you were to iteratively read individual images of a large XDATCAR file using ase.io. I will be computing something called the hopping rate using these broken XDATCAR files.
  4. hoppingRATE.py:

About

Documentation and discussion regarding the study of proton conduction in graphanol using deep learning potentials.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.9%
  • HTML 0.1%