How to modify BNC.tersoff to be properly used with pair_style tersoff/zbl?

Hello everyone,

Since LAMMPS has the pair style tersoff/zbl, I want to modify the BNC.tersoff which is already existing in the potential directory to be compatible with the tersoff/zbl pair style. I have read the docs of pair_style tersoff/zbl and viewed the potential files SiC.tersoff, SiC.tersoff.zbl as well as SiC.gw, SiC.gw.zbl. It seems the only modification needed is that Z_i, Z_j, ZBLcut and ZBLexpscale values should be properly added in the BNC.tersoff file. With these values, when the pair_style tersoff/zbl is invoked, LAMMPS and this pair style would do the calculation itself as if it automatically spline the tersoff potential with ZBL potential.

It seems that adding four columns to each entry is not much complicated, but I am not very confident about my understanding. Could you please tell me how should I modfiy BNC.tersoff to be compatible with pair style tersoff/zbl? I would appreciate that if you can give me some guide.

Thanks very much in advance!

With My Best Regards,

Liu

Your understanding is correct. Why doubt yourself before you even start?

Ray

Hi Ray,

Thanks very much for your reply!

I think it would be better for me to plot the energy vs distance both before and after the stiffness of the potential. I have tried pair_write before, but it does not support tersoff potential. Thus, it seems I need to write the equations and plot them by myself. I am still thinking about how to accomplish this. Without the comparison of them in a figure, I am not quite sure whether the modification is right or not.

I found that the r_C and A_F used in SiC.tersoff.zbl and SiC.gw.zbl is 0.95 angstrom and 14 (1/angstrom). I think it may be safe for me to use the same values. But if possible, could you please give me some suggestions about how to decide them in BNC.tersoff? Some better choices may exist. Does the atoms of different type have effect on the value to be chosen? Any help would be much appreciated!

By the way, my previous email has be rejected by the LAMMPS mail list, so I send it again. Sorry for the bothering.