Moderators: Global Moderator, Moderator
Hi, everyone,
Now, I am trying to built a two dimensional materials with including the SOC calculation, but I got a magnetic results. However, when I performed a spin polarized calculation, I found that the material is non-magnetic. Any suggestions will be appreciated!
With regards,
Kun Tao
Whether the ground state is magnetic depends on the Hamiltonian. Additionally many XC functionals tend towards symmetry-broken solutions (i.e. magnetic solutions). See for instance the basic tutorial about an isolated oxygen atom.
If you want to impose a nonmagnetic solution, you could probably use symmetry to symmetrize it away, which is like a constraint on the solution. This is then not the true ground state whithin the approximation, but the ground state within the constraint.
It may as well be that the spin-polarized system converged to a metastable nonmagnetic solution (local minimum). You can try to do the spin-polarized calculation with intial magnetic moments (MAGMOM) set to 1.3 x the on-site magnetization (LORBIT=11) that you find in the ncl-SOC calculation. Then compare the total energy of the spin-polarized calculations. Is the nonmagnmetic solution really the ground state?
Best regards,
Marie-Therese