Hybrid Pair Style with AIREBO Potential

Dear LAMMPS-users group,

I am attempting to use the hybrid pair style with AIREBO and lj/cut/coul/cut potentials. (LAMMPS v. 13Jul11)

Previously, I've successfully modeled graphite with the AIREBO potential. However, when using the hybrid pair style with the aforementioned potentials as follows:

pair_style hybrid airebo 3.0 1 0 lj/cut/coul/cut 10.2
pair_coeff * * /home/chris/src/lammps-13Jul11/potentials/CH.airebo C NULL
pair_coeff 1*2 2 lj/cut/coul/cut 0.00284 3.4

I receive the following error:

ERROR: Pair coeff for hybrid has invalid style.

The log.lammps file indicates the error is occurring at the line containing the pair_coeff for AIREBO (not for lj/cut/coul/cut). As a check, I invoked the hybrid pair style using only AIREBO with Carbon and Hydrogen and still received the error at the same location:

pair_style hybrid airebo 3.0 1 0
pair_coeff * * /home/chris/src/lammps-13Jul11/potentials/CH.airebo C H
ERROR: Pair coeff for hybrid has invalid style

Any input in greatly appreciated.

Thank you very much,

Chris

Chris,
Try this:

pair_coeff * * airebo /home/chris/src/lammps-13Jul11/potentials/CH.airebo C H

Jonathan

Hi everyone,

Jonathan's solution method fixed the problem. I'm assuming that any pair style that calls a file structure under pair_coeff must be respecified when using the hybrid style (ie airebo, eam, etc).

Thanks for the help,

Chris

Quoting Jonathan Lee <[email protected]...>:

All pair coeff commands used with
pair_style hybrid, must add the sub-style argument.

Steve

Thanks, everyone. Makes perfect sense now. Looking back at the hybrid docs page I see the text that specifies this. Since the pair styles can be listed in any order, of course the pair style must be called out in each pair coefficient line. Must have read over that page a dozen times and only now did it click!

Thanks again,
Chris

It's a little odd because at first glance it looks like what you have
proposed should work. I am guessing maybe you need the H? Not sure why,
since I don't think you'd need to assign H if you aren't using H. I would
try the following just as a check:

pair_style hybrid airebo 3.0 1 0 lj/cut/coul/cut 10.2
pair_coeff * * /home/chris/src/lammps-13Jul11/potentials/CH.airebo C H

as was pointed out before, the first issue is the missing "airebo"
after the "* *"

otherwise pair_style hybrid will assume the pair style would be
/home/chris/src/lammps-13Jul11/potentials/CH.airebo, which is
of course inconsistent.

assigning the NULL should be fine.

axel.