ABINIT Discussion Forums

Dear Abinit users

The new implementation of DFPT for non-linear response (usepead = 0) works for collinear spin polarized system.
but in the test file, all the test of this implementation use the variable spinmagtarget=0.
I think for that case, the system must be Anti-ferromagnetic, right?

--------

The real problem is, when I try to do the non-linear response for the FM system. I cannot set the spinmagtarget to any other value but zero like the error i got:

--- !ERROR
src_file: m_chkinp.F90
src_line: 2888
mpi_rank: 0
message: |
spinmagntarget was input as 2.00
For a response function run, spinmagntarget is required to be 0.0d0 or the default value.
Action: modify value spinmagntarget or nsppol in input file.


From the tutorial, I know that one can decide the total spin-polarization by using a metallic value for occopt(e.g. 4 7),
But seems the input variable optdriver=5 is contradiction with the occopt not equal to 1 since i got this error:

chkint_prt: ERROR -
Context: the value of the variable usepead is 0.
Context: the value of the variable optdriver is 5.
The value of the input variable occopt is 4, while it must be
equal to 1
Action: you should change one of the input variables occopt,
usepead or optdriver.

I have confirmed that my system is insulating by doing the band structure calculation.
In this situation, I cannot do the non-linear response calculation neither using spinmagtarget nor metallic value occopt.
Is there any other way to determine the spin-polarization in the non-linear response?


Cheers,
Andy

Dear Andy,
We have to look if the spinmagntarget is well included into the non-linear calculation and we'll come back to you.
Best wishes,
Eric

Dear Eric,
Thanks for the reply.
I am looking forward it!
I wonder if there is any other way to set the spin without using spinmagntarget?

Cheers,
Andy

andyamygto wrote: ↑

Wed Apr 08, 2020 3:53 am

I wonder if there is any other way to set the spin without using spinmagntarget?

If not metallic occupation treatment there is no other way...

Ok. I think i will just wait for the good news
Hope that everything is alright.

Cheers
Andy

Dear Andy,
Bad news, the spinmagntarget correction is not included in the non-linear response, meaning FM case cannot be done... Lets see what Lucas is saying about it.
Best wishes,
Eric

Dear Eric,

I see. it's unfortunate to know that FM case cannot be done...
Hope that spinmagntarget correction will be include in the future. (or the metallic occupation)
I'll wait the comment form Lucas.

Regards,
Andy

Dear Andy,

I'm sorry to confirm that there is a problem with the code for a ferromagnetic system. Thank you for bringing it up.

This issue is taken into account now, and I will let you know when the correction will be done.

Regards,
Lucas

Dear Lucas,

Thanks for the info. I'll wait for the good news.
You guys are doing amazing job.

Best wishes,
Andy

Dear all,

Another question I wonder,
Is that means it is impossible to calculate the non-linear properites within DFPT method?
or I can just use occopt = 1 for metal? I think it makes wrong since i can not define the fermi level well.

If there are some advise I will be thankful.

Regards,
Andy