A hybrid potential "NEP+ILP" for TMDs

[BBBuZHIDAO]

Summary
This PR is to commit a hybrid potential, NEP+ILP. The potential could simulate 2D materials: TMDs ($\mathrm{MoS_2}$, $\mathrm{MoSe_2}$, $\mathrm{WS_2}$, $\mathrm{WSe_2}$ ...) and their heterostructures. Users just need to set a group method to identify the molecular layers and a sublayer group method to identify the sublayer of each molecule, and then the interlayer potential will be set as ILP and intralayer potential will be set as NEP.

Modification

• Add two potential files, ilp_nep_tmd.cu and ilp_nep_tmd.cuh.
• Check this potential in force.cu.

Usage

• Use two group methods to identify the layers and sublayers, and add these group methods to the second line of ILP potential. For example, if you use group method 0 to identify the molecular layers and method 1 for sublayer. Assume you just simulate $\mathrm{MoS_2}$, then the ILP potential file should be:

ilp_nep_tmd 2 Mo S
0 1
5.57945037  9.37766246  2.02722246  144.15177543  97.97857017 89.43759676 2.05903055  5.12205490  491850.31619468 1.0 4.0   16.0
3.62715212  19.97137510 7.58503123  76.10193100 3.31749559  45.72032846 0.94747037  4.41042503  150597.85771629 1.0 4.0     16.0
3.62715212  19.97137510 7.58503123  76.10193100 3.31749559  45.72032846 0.94747037  4.41042503  150597.85771629 1.0 4.0     16.0
3.16140219  8.09326275  1.95314020  4.58676418  118.06546646  58.80941584 0.21536658  4.29960013  148811.24340892 1.0 4.0   16.0

• Add both ILP and NEP potential files to run.in file and make sure the ILP file in the front:

potential ilp.txt nep.txt

• Make sure element list in NEP potential file is the same as ILP potential file, including the ORDER.

Test

• Energy conservation test in bilayer MoS2:
  
• DEBUG flag test:
  

Others
If any questions, please feel free to let me know. THANKS! 😺

[brucefan1983]

remember to test -DUSE_TABLE