Allow Gaussian MM charges#1677
Conversation
* finite size gaussian radii for MM charges in QM/MM * add mm gradient in qmmm * add CP-CI for FCI
Conflicts: pyscf/qmmm/itrf.py
|
I just noticed that in libcint, |
|
@fishjojo Does this PR support MM gradients when EDIT: it appears that point-charge model MM gradients are already in the example: https://github.com/pyscf/pyscf/blob/master/examples/qmmm/30-force_on_mm_particles.py I wonder if that can be made into the interface code. EDIT2: okay the code here sets zeta=1e16 if point-charge model is used, then reuse the same hcore gradient code. Will this be inefficient as compared to the above implementation in the example? |
This commit is mostly based on the PR pyscf#1677 and examples/qmmm/30-force_on_mm_particles.py The aim is to provide point-charge MM gradients before PR 1677 is merged. There will be merge conflicts when PR 1677 gets merged.
|
@zc62 yes, the 3 center integral will be less efficient than the 2 center integral. We can add the point charge version to the main code. |
This PR allows to use Gaussian MM charges in QMMM calculations. The nuclear gradients w.r.t MM atoms are also included. This will be used for interfacing PySCF with LAMMPS.
The code was written by Chenghan @MoleOrbitalHybridAnalyst, and was modified by me.
Future work is to perform QMMM calculations with PBC.