regenerate EAM/fs potential

Hi Dear LAMMPS users and reaserchers,
I am going to regenerate a EAM/fs potential (binary alloy Fe-P developed by Ackland et al, https://doi.org/10.1088/0953-8984/16/27/003) by writing a C program.
I do read the pair_style EAM/fs command, but i have a question. what is the sequence of the tables? i suppose that it is as follows (in EAM/fs format):

F-Fe
phi-Fe
V-Fe

F-P
phi-PP
V-PP

phi-FeP
V-FeP

when i read the output C code, i see that the F-Fe and phi-Fe are correctly identical with what is reported by Ackland et al, but my other ones is not correct. in addition, they reported 9 tables but i expected 8 table (as above). what is wrong?
Regard
S.M.Zamzamian

Everything LAMMPS knows about EAM potential files formats
is on the pair_eam.html doc page. Including the F-S format.

Steve