Dear lammps users,
My object is is to reproduce the properties of polymer molecular system. Now, all the bond potential parameters have been given. And then, the non-bond potential is set by the pair_style table command, whose corresponding r range is including r range of the bond potential. So, the molecular simulation should get rid of non-bond pair style interaction between the atoms of 1-2, 1-3 and 1-4 bond so as to avoid twice setting potential parameters. (Note that the 1-2, 1-3 and 1-4 stand for the neighbour atoms of bond length, angle and dihedral, respectively) My question is how lammps is responsible for the work of getting rid of that?
Best regards
D. Chai
Dear lammps users,
My object is is to reproduce the properties of polymer molecular system. Now, all the bond potential parameters have been given. And then, the non-bond potential is set by the pair_style table command, whose corresponding r range is including r range of the bond potential. So, the molecular simulation should get rid of non-bond pair style interaction between the atoms of 1-2, 1-3 and 1-4 bond so as to avoid twice setting potential parameters. (Note that the 1-2, 1-3 and 1-4 stand for the neighbour atoms of bond length, angle and dihedral, respectively) My question is how lammps is responsible for the work of getting rid of that?
Do you have any proof that LAMMPS doesn’t take care of it? Have you checked out the documentation for the special_bonds command?
Axel
Dear Axel,
Thanks for your advice, first. True be told, I have read the doc for the special_bonds command, and I have read your comments on the email, “confusion about special bonds and pair potential” ( http://sourceforge.net/p/lammps/mailman/message/34130271/ ). Yes, lammps could set the weighting coefficients to take care of it for lj, coul, angle, dihedral, etc. However, my non bond potential is given by a tabulated style, which, I think, dose not belong to any specific cases in the special_bonds command. My problem is a little different from the common one, mainly because my non-bond table-style potential is a combination of LJ potential and coulomb potential, and could I still use the special_bonds command to solve the problem? Looking forward to your comments.
Best regards
D. Chai
Hi, lammps users,
My question is still open. After read the doc of special_bonds command once more,
my understanding is that the special_bonds command could not exclude the non-bond tabulated pair potential of 1-2, 1-3 and 1-4.
What should I do when considering using lammps to avoid twice counting the potential. Or, is there some misunderstanding in my mind?
Please give me some advice. Thanks a lot.
Regards
D. Chai
Hi, lammps users,
My question is still open. After read the doc of
the question is open *for you*. and that only because you haven't done
what i asked you to do.
special_bonds command once more,
my understanding is that the special_bonds command could not exclude the
non-bond tabulated pair potential of 1-2, 1-3 and 1-4.
nonsense.
What should I do when considering using lammps to avoid twice counting the
potential. Or, is there some misunderstanding in my mind?
like i wrote before, you have to provide proper proof. i.e. set up an
extremely trivial test system and observe the impact of the
special_bonds command on your forces and energies. or point out in the
source code of pair_style table that it is not doing exclusions.
axel
Dear Axel,
I have a trivial test as what you have told. Yes, the lammps have already considered the problem perfectly.
Thanks for your advice. The question of my own is closed. 
cheers
D. Chai