Coulombic energy in pair_style reax/c

LAMMPS LAMMPS Mailing List Mirror
#1

Hi,

I am trying to convert the potential file of reax from “units real” to “units metal”, because I want it to work with “pair_style comb” in a hybrid way. All energies computed by “compute reax” in “units metal” are consistent with those in “units real”, except those about charges including Coulomb energy and charge equilibration energy. For example, non-zero energies calculated in units real are

Step PotEng bond atom valence vdW coulomb charge_equilibration
0 -114958.81 -131085.71 22009.218 8133.2786 10991.359 -8764.4587 -16242.491

and energies with identical read_data file in units metal are

Step PotEng bond atom valence vdW coulomb charge_equilibration
0 -28905.715 -5681.6967 953.95353 352.5214 476.4567 -8764.4587 -16242.491

We can see the values of coulomb and charge are the same but they should be different because of different units.
In the reax potential file [B. Narayanan et al., Modelling Simul. Mater. Sci. Eng. 2012, supplementary information], I did not find any parameters for the Coulomb’s constant. I guess that this constant might be hard-coded in units real in the source code. How can I solve this problem?

Thanks for your help,
Qifang

#2

Yes, Coulomb and QEq energies have hard-coded units:

reaxc_defs.h:#define KCALpMOL_to_EV 23.02 // 23.060549 //KCAL per MOLe --> ElectronVolt

This is where you make the change.

Ray