Graphitization simulation with van der Waals corrections

From TurboGAP
Jump to navigation Jump to search

WARNING: TUTORIAL UNDER CONSTRUCTION!!!!

In this tutorial we are going to carry out simulations of carbon graphitization similar to those in the simple molecular dynamics tutorial, but we will also add a barostat so that the simulation box size changes to mimick the effect of coupling to an external pressure. We will also look at how to use van der Waals (vdW) corrections on top of a GAP force field. Note that vdW corrections are generally not available for all potentials. Those potentials without "proper" vdW corrections (but again, not all) may include fixed pair-wise long-range interactions that allow you to incorporate dispersion effects in a limited way (but we will not cover those in this tutorial).

It is strongly recommended that you do the simple molecular dynamics tutorial before this one.

Prerequisites for this tutorial

  • A TurboGAP installation

Optional

  • gnuplot (for plotting)
  • An ASE installation
  • VMD (for visualization, ASE can handle visualization but it's slow)

Initial configuration

We will start our simulation from a melted carbon sample with 216 atoms, generated as in the simple molecular dynamics tutorial. For convenience, you can download the file from here (melt.xyz). The approximate density of this sample is 2.15 g/cm3, it temperature is about 9000 K and the pressure is over 200 kbar (although this number will vary depending on the potential used for the calculation). Obviously, this initial configuration is very far away from the graphitic carbon that we want to generate. To graphitize this sample, we will follow the recipe of the previous tutorial, but will use:

- barostating, to simulate the graphitization process at atmospheric conditions (1 bar); - van der Waals corrections, to accurately capture the interlayer interaction in these materials.

The GAP used in the previous tutorial does not incorporate vdW corrections. We will instead use the C60 GAP:[1][2]

wget https://zenodo.org/record/4616343/files/gap_files.tar.gz
tar -xvf gap_files.tar.gz

Reference list

  1. H. Muhli and M.A. Caro. GAP interatomic potential for C60. Zenodo: 10.5281/zenodo.4616343 (2021).
  2. H. Muhli, X. Chen, A.P. Bartók, P. Hernández-León, G. Csányi, T. Ala-Nissila, and M.A. Caro. Machine learning force fields based on local parametrization of dispersion interactions: Application to the phase diagram of C60. Phys. Rev. B 104, 054106 (2021).