[QE-users] Non-collinear magnetism calculation questions and discrepancy between scf and nscf fermi energy

[Felix Frontini]

 Hi There,

I am a relatively new and unexperienced QE user and am having some difficulty understanding some of the specifics of non-collinear magnetic calculations. I want to perform some simple scf calculations on some non-collinear AFM spin configurations to compare the energy of competing phases in a material I am studying. I need to specify the orientation and magnitude of the moment on 9 sites and while I understand how to specify orientations I am lost as to how to specify each moment. It seems to me that the variable STARTING_MAGNETIZATION is of no use here. Can the variable CONSTRAINED_MAGNETISM be used here?
I also have a secondary question about scf and nscf estimates of the Fermi energy. For the non-magnetic case I have performed a full geometric optimization and also have tested convergence with respect to my k-mesh, degauss value and basis set size yet when I calculate the DOS it gives me a Fermi level that is 1eV lower than for the scf calculation. Is the DOS estimate reliable? Why would these not agree?
Thanks,

Felix Frontini
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200526/a7a7de5f/attachment.html>