system not charge neutral?

LAMMPS LAMMPS Mailing List Mirror
1

Hi

Lammps output is telling me that my system is not charge neutral with net charge of 0.1572. Size of my system is 16635 atoms. I am using ewald summation which requires total charge to be 0.0. Could anyone tell me if 0.1572 charge is okay or I have to reduce it down further? If yes, what is the tolerance?

Thanks
Chetan

2

Hi

Lammps output is telling me that my system is not charge neutral with net
charge of 0.1572. Size of my system is 16635 atoms. I am using ewald
summation which requires total charge to be 0.0. Could anyone tell me if
0.1572 charge is okay or I have to reduce it down further? If yes, what is
the tolerance?

this kind of total charge usually results from truncation/rounding errors
when assembling the data file. so the total error is likely to be scattered
throughout your system. rather than trying to adjust this be changing
individual entries, i would have a look at the code/tool that you have used
to generate the data file and make it output more digits of our put or
check for rounding/truncation errors in any conversion factors used.

whether that remaining charge will have a significant impact on your
simulation or not, is hard to predict. it is an avoidable inconsistency
and thus i personally would always want to eliminate it.

axel

3

Maybe Paul wants to comment on the amount
of excess charge that messes up Ewald.

Steve

4

You can use Ewald in LAMMPS with non-neutral systems, but LAMMPS will give a warning (as you've seen) and make a correction to the energy, seen here in the second term:

    energy -= g_ewald*qsqsum/1.772453851 +
      0.5*PI*qsum*qsum / (g_ewald*g_ewald*volume);

I'd recommend you remove the excess charge from your system. If not, the simulation should be able to run anyway. If the excess charge is small (which seems to be the case for you), it probably won't make a lot of difference.

Paul

5

Thanks, Paul, Axel and Steve.

@ Axel: such total charge is arising mainly out of my need to assign required charges (-1 or 0 ) to a polymer unit (excluding end atoms, which I want to treat as dummy atoms with zero charge.) Since materials studio needs you to include end atoms as well for assigning charge, i need to assign total charge so that charge on polymer unit is as required. Then i assign zero charges on end dummy atoms.

Thanks