complex values from the effective charge function

Dear lammps users

I want to make two .eam-files containing the potentials presented in

H. Chamati et al., Surf. Sci. 600 1793 (2006) (Embedded-atom potential for Fe and its application to self-diffusion on Fe (100))

and

Y. Mishin et al., Phys. Rev. B 63, 224106 (2001) (Structural stability and lattice defects in copper: Ab initio, tight-binding, and embedded-atom calculations)

As you know, .eam-files contain the effective charge function Z®, which in lammps is, according to the manual, defined as

r*phi = 27.2 * 0.529 * Zi * Zj. (1)

In the two articles phi have some negative values. Thus, solving Z® from formula (1) will generate some complex values. I can avoid the complex values by using the absolute values of phi but is that really how I shall do it?

Is there any other way to calculate the effective charge?

Sincerely Dan

My recollection is that the Finnis/Sinclair format for EAM
files (eam.fs suffix), allows for a more general definition of phi.

I.e. you input phi directly, unlike for the .eam files where you

input Z.

Steve