Hi Su Yun,
Please direct all correspondence to LAMMPS user list, thanks.
I don't believe we have CTIP potential in LAMMPS, so if you really
want to use it you will have to implement it. A quick glance of the
paper indicates CTIP is a Buckingham + Coulomb (Ewald sum) +
Streiz-Mintmire charge transfer model. You can try combining
pair_style buck/coul + ksapce ewald + the new S-M model to make up
CTIP.
Have a look at pair_style comb, which also uses a S-M model. However
the variable charge is solved with extended Lagrangian method in COMB
instead of conjugate gradient minimization method in CTIP.
I would suggest you contact the original authors of the paper and
request for their stand-alone code.
Best,
Ray