[lammps-users] problems with "pair_style hybird"

Dear all,

    I was trying to simulate the nanoindentation process of NiTi.Now I have some problems with command "pair_style hybird".Hope someone give me some advise,thank you!#atom type 1 Ni#atom type 2 Ti#atom type 3 Cpair_style hybrid meam/c tersoff lj/cut 10 morse 3
pair_coeff * * library.meam Ni Ti NiTi.meam Ni Ti #potential for Ni-Ni Ti-Ti Ni-Ti
pair_coeff * * SiC.tersoff NULL C #potential for C-C
pair_coeff 1 3 1.0094 1.9875 2.56 #potential for Ni-C
pair_coeff 2 3 3.14 3.759 #potential for Ti-C error:Pair coeff for hybrid has invalid style Best regards

all your pair_coeff statements are incorrect since they do not indicate which pair style they “belong” to.
please see the documentation of pair style hybrid for detailed explanations and examples.

axel.

Thank you Axel, but I meet new problems.

I modified the pair_coeff as follows:

pair_style hybrid meam/c meam/c meam/c tersoff morse 2.56 lj/cut 10
pair_coeff 1 1 meam/c library.meam Ni NiTi.meam Ni
pair_coeff 2 2 meam/c library.meam Ti NiTi.meam Ti
pair_coeff 1 2 meam/c library.meam Ni Ti NiTi.meam Ni Ti
pair_coeff 1 3 morse 1.0094 1.9875
pair_coeff 2 3 lj/cut 3.14 3.759
pair_coeff 3 3 tersoff SiC.tersoff NULL C
Then got an error: Incorrect args for pair coefficients (…/pair_hybrid.cpp:403)
Last command: pair_coeff 1 1 meam/c library.meam Ni NiTi.meam Ni
I don’t know where the problem is. Even if the order of the pair_coeff lines is changed, the same error will still be displayed.

There are four potential functions: meam, tersoff, lj and morse. Can’t every pair_coeff of the potential function be wrong, I really don’t know how to correct it.

you have to pay more attention to the documentation and the physics of your system. there are multiple conceptual and syntactical issues with your input.

manybody potentials like meam/c or tersoff have specific requirements to the pair coeff command, both with the atom type range and the mapping of atom type to element.
also, you need to pay much closer attention to the requirements for the pair_coeff command line for meam/c and specifically how the second .meam file is used and what conditions and restrictions are imposed by it.

this is all described in detail in the documentation, but you seem to be making things up as you think it should be and not the way the documentation says it has to be.
also, your Lennard-Jones epsilon looks unusual for metal units. are you sure it is correct?

axel.