Energy of spin spiral compared to ferromagnetic state
Posted: Mon Mar 08, 2021 10:31 pm
What is the correct way to compare the energy of a calculation with LSPIRAL=.TRUE. to one with LSPIRAL=.FALSE. and ferromagnetic order?
If I set the q-vector of the spiral to zero (QSPIRAL= 0.0 0.0 0.0) I expect to find the above energy difference to be zero in the case without spin-orbit coupling (SOC) and it should be half the anisotropy energy in the case with SOC. Even after making sure calculations have the same number of k-points, same NBANDS, ENINI and ENCUT, and thus the same number of plane waves, the energy differences are not correct. What are the correct settings to make sure these energy differences turn out correct?
If I set the q-vector of the spiral to zero (QSPIRAL= 0.0 0.0 0.0) I expect to find the above energy difference to be zero in the case without spin-orbit coupling (SOC) and it should be half the anisotropy energy in the case with SOC. Even after making sure calculations have the same number of k-points, same NBANDS, ENINI and ENCUT, and thus the same number of plane waves, the energy differences are not correct. What are the correct settings to make sure these energy differences turn out correct?