Hi
Dreiding force field employs special charmm like hydrogen bonding potential which is 12-10 form of LJ-like expression multiplied by forth power of cos (theta), theta is a angle between donor, hydrogen and acceptor. It seems like i will have to generate new code for that and integrate that into lammps. It seems like a lot of work, so before beginning for that, i just thought i will ask the community whether anybody has employed this potential before with lammps. This seems like a very useful potential, so i was bit surprised that lammps don’t have it as of now.
Thanks
Chetan
I have some Dreiding code sent to me a couple years back from
someone in the Goddard group. It never made it into the main
branch and is probably a bit out of date now. I'll send it to
you in a separate email.
Steve
Hi Steve
First of all, i apologize for extreme delay in the reply. I earlier thought i would reply you after i get done with everything.
So I have made some changes, removed some errors (i will send the list in the end) and now the code gets compiled with lammps new version successfully and also runs without any errors. But I have few questions that i have been working on for past few days before i make tests:
I think it would be great if i could also add a compute style to calculated hydrogen bond energy of the system and make this hydrogen bond energy add to total potential energy which can be printed later on. I have been working on this front, getting myself familiar with lammps organization. Although i am beginning to feel comfortable with inner code structure of lammps, it is still sort of cloudy. If you could provide tips to me for doing above, that will be great.
Second, I have not submitted trial jobs yet since i would like to have hydrogen bond energy getting printed. Additionally, i am thinking of running job on water molecules 1) with hydrogen bonding potential 2) without hydrogen bonding potential, and see if difference in the radial distribution functions of oxygen atoms makes sense as a test to check the correctness of the code.
If you want me to run additional tests, let me know.
Thanks very much
Chetan
Chetan - sounds like great progress.
I think it would be great if i could also add a compute style to calculated
hydrogen bond energy of the system and make this hydrogen bond energy add to
total potential energy which can be printed later on. I have been working on
this front, getting myself familiar with lammps organization. Although i am
beginning to feel comfortable with inner code structure of lammps, it is
still sort of cloudy. If you could provide tips to me for doing above, that
will be great.
Send me (separately, not to the mail list) the details on how this is
computed, via a compute vs as part of the pair style itself.
Then I'll think about how best to make it part of energy output.
This can be done by a fix, for example.
Steve