Skip to content

JiaxuanLi-THU/MTP-SiC-Cs

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

25 Commits
 
 
 
 
 
 
 
 

Repository files navigation

MTP-SiC-Cs

This repository contains trained moment tensor potentials (MTPs) developed for the amorphous SiC-Cs system and polycrystalline SiC-Cs system. The potentials enable accurate and efficient atomic simulations to study Cs diffusion in amorphous SiC and polycrystalline SiC with high energy grain boundaries (HEGBs).

Use guide

Software requirements

  1. LAMMPS https://github.com/lammps/lammps.git
  2. MLIP-2 package https://gitlab.com/ashapeev/mlip-2
  3. MLIP-3 package https://gitlab.com/ashapeev/mlip-3

Use of the trained MTP

There are three trained MTPs in the Potential folder: aSiC.mtp, HEGB_L10.almtp, HEGB_L16.almtp

  • aSiC.mtp: level 16 MTP, specialized for amorphous SiC-Cs system
  • HEGB_L10.almtp: level 10 MTP, faster, specialized for polycrystalline SiC-Cs system
  • HEGB_L16.almtp: level 16 MTP, more accurate, specialized for polycrystalline SiC-Cs system
  1. If you want to use aSiC.mtp, please make sure you have already installed the MLIP-2-LAMMPS interface, otherwise see https://gitlab.com/ashapeev/interface-lammps-mlip-2. Add the following content in the LAMMPS input script to define the potential:
    pair_style mlip mlip.ini
    pair_coeff * *
    Detailed useage tutorial of MTP can be found at https://gitlab.com/ashapeev/mlip-2-tutorials
  2. If you want to use HEGB_L10.almtp or HEGB_L16.almtp, please make sure you have already installed the MLIP-3-LAMMPS interface, otherwise see https://gitlab.com/ivannovikov/interface-lammps-mlip-3/. Add the following content in the LAMMPS input script to define the potential:
    pair_style mlip load_from=HEGB_L10.almtp (or HEGB_L16.almtp) pair_coeff * *
  3. Please define the element types in the structural file in the following order: 1-C; 2-Si; 3-Cs

Training set

We also provide the training set in the Trainingset folder.

Computational models

We also provide the models used in this work. Please find them in the Models folder.

Citation

If you utilize the potentials, training set or models in this repository, please cite:

About

MTP for SiC_Cs system

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published