Charge Calculation during minimisation

LAMMPS LAMMPS Mailing List Mirror
1

Dear All,
I am trying to minimise a system with two groups of atoms using ReaxFF. During the minimisation I do not want to include Group-1 atoms into the charge calculation scheme using this fix:

fix 1 group2atoms qeq/reax 1 0.0 10.0 1.0e-6 reax/c
But during the minimisation I dump the charges of group1 atoms and observe that they have non zero charges.
Is it possible to exclude a group of atoms from charge calculation during the minimisation?

Regards.
dundar.

2

Of course it is possible. Either your executable is out-dated or there is a mistake in your input script.

Ray

3

I checked my input script,

fix 2 polymer qeq/reax 10 0.0 10.0 1.0e-2 reax/c
fix 1 polymer nvt temp 100.0 100.0 100.0

These are fixes for charge calculation and time integration. They both use polymer group. But when i check the charges on other group graphene atoms I see that each carbon atom has around -0.001 charge.
My executable is dated Feb-1 2014.

Regards.

4

Then the only two explanations are:

  1. Something changed since Feb 1, which is unlikely.

  2. The charges you are seeing were already there before fix qeq/reax.

I have just verified that with the most recent version groups without qeq/reax will have charges unchanged.

If you still think this is due to qeq/reax, please post a small input deck.

Ray

5

Then the only two explanations are:

1) Something changed since Feb 1, which is unlikely.

according to git annotate, it looks like group support was committed
to fix qeq/reax on may 2nd 2014,
which would be in keeping with the may 4th 2014 version listed here:
http://lammps.sandia.gov/bug.html

examples:

bfd6a79f (sjplimp 2010-10-26 16:14:31 +0000 34) #include "force.h"
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 35) #include "group.h"
bfd6a79f (sjplimp 2010-10-26 16:14:31 +0000 36) #include "pair.h"

[...]

bfd6a79f (sjplimp 2010-10-26 16:14:31 +0000 519)
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 520) for( ii =
0; ii < nn; ++ii ) {
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 521) i = ilist[ii];
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 522) if
(atom->mask[i] & groupbit) {
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 523)
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 524) /* init
pre-conditioner for H and init solution
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 525)
Hdia_inv[i] = 1. / eta[ atom->type[i] ];
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 526) b_s[i]
    = -chi[ atom->type[i] ];
ddcab126 (sjplimp 2014-05-02 21:38:42 +0000 527) b_t[i]
    = -1.0;

axel.

6

Right! Thanks again, Axel.

Ray