Re: non-collinear DFT: problem in fixing spin orientation ( No.1 ) |
- Date: 2014/03/17 16:45
- Name: T. Ozaki
- Hi,
It seems that the magnetic moment at each site is almost zero. In this case, the angle of spin does not have any meaning, that is, the angle of 0 degree is the same as 90, 180,...etc.
Also, to fix the spin orietation, the use of "scf.Mixing.Type=rmm-diis" is better than "scf.Mixing.Type=rmm-diisk".
Regards,
TO
|
Re: non-collinear DFT: problem in fixing spin orientation ( No.2 ) |
- Date: 2014/03/19 01:08
- Name: Tamal Goswami <tamal.nbu@gmail.com>
- Respected Sir,
Thanks for your reply. The system contains one transition metal ion which has four unpaired electrons. For other spin orientations the spin moment is correctly predicted but for a particular angle like -90 the spin moment is zero with Rmm-Diis and even with simple scf.mixing.type. Please suggest.
Sincerely,
Tamal
|
Re: non-collinear DFT: problem in fixing spin orientation ( No.3 ) |
- Date: 2014/04/14 15:29
- Name: T. Ozaki
- Hi,
I think the problem has been resolved in OpenMX Ver. 3.7. Could you try to perform the calculation by Ver. 3.7, and report what happens?
Thank you for your cooperation in advance.
Regards,
TO
|
|