[lammps-users] Dislocation minimization at zero degree kelvin

A week back, I have posted a problem on minimization at zero degree kelvin. I could sort out how to minimize a dislocation in a Al crystal.

My problem now is: Splitting distance between partials after minimization is very high and equal to 50 Angstroms, when I used energy tolerance to 1.0e-7.

But the elastic solution for Aluminium is around 16 Angstroms. I am able to get this distance only when I increase the tolerance to 1.e-5.

Is minimization so sensetive to energy tolerance? Am I doing any mistake in i/p. Please help. I am attaching the i/p file

# LAMMPS data file for rigid bodies

99855 atoms
1 atom types

-104 104 xlo xhi
-53 49 ylo yhi
0 73 zlo zhi

This seems wierd, because presumbly with high precision tolerance should have better results. I’m guessing maybe this has sth to do with the potential.

One comment: The splitting distance between two partials should depends on the potential you are using, including shear modulus, stacking fault energy, etc. They vary for different potentials.

Best,

AC

Is minimization so sensetive to energy tolerance? Am I doing any

mistake in i/p

Don't know, but the energy landscape in general is very bumpy for
solids and defects. So it is possible to find different local minima.

Steve