Regarding Lattice parameter

Please reply to the list, thanks.

The lattice constant in EAM potential files are just references – you need to read the original paper that published the potential parameters to find the actual lattice constant from the potential.

Ray

In the mentioned paper in eam potential file, it is 3.615 A.

Try NPT with 0.1 K and see what the lattice constant is.

At O.1K it is coming as 3.615005.

But in actual simulation I first equilibrate the system at 300K then start the dynamics.
Thats why i am calculating the equilibrium constant at 300k.
So in the beginning , how much value of lattice should be given. Should it be a/q to 0.1 K temperature or a/q to 300K.

It is pointless to expect lattice constants obtained by different methods (experiment at 300K, MD simulation at 300K, MD simulation at 0.1K, energy minimization at 0K) to agree to high precision. All that you can expect is that they be “similar.” If you want agreement better than 3.6327 A vs. 3.6149 A, you are going to have to compare the same method, apples to apples. In the context of this mailing list, since LAMMPS does not perform experiments, the only reasonable question is “how do I get LAMMPS to reproduce the value listed in paper X, using interatomic potential Y and at temperature Z”

Aidan

Agreed.

Actually My doubt is if I am equilibrating the system at 300K in actual simulation, what should be the lattice constant value I should mention in the lattice command? Should it be mentioned as 3.6327 (which is equilibrium lattice constant found at 300k) or 3.6149 (which is normally taken).

Hi.

I do not think it should matter if you choose either of the values (3.6327 or **3.614)**if you are equilibrating the system. The system will either expand or contract to adjust to the equilibrium value consistent with the chosen inter-atomic potential

You can resolve this by carrying out two equilibrating simulations, each with one of the above values for the lattice parameter. Compare the final values of pressure, KE, PE, Volume averaged over say 10000 steps after equilibration and satisfy yourself that it does not matter. If it does do write back … Ill be surprised!!

Best Regards

Manoj

Thank you Mr. Manoj for your inputs.