RSPt Site - The best RSPt in the world!

The best RSPt in the world!

Topics

Home

General News (14/0)

GeekLog (12/0)

User Functions

Don't have an account yet? Sign up as a New User
Lost your password?

Events

There are no upcoming events

Older Stories

Friday25-Apr

RSPt workshop in Sandia (0)

Tuesday01-Apr

Welcome to RSPt Site
Thursday, September 09 2010 @ 10:18 PM CEST

Home »

Use »

symcof generation

symcof generation

sumanta

Registered: 09/14/09
Posts: 6

Tuesday, March 09 2010 @ 05:10 PM CET

Dear RSPt users ,
I am calculating Orbital moment anisotropy for fct Ni . The symcof files it generates for 001 and 110 spin axes have same euler angles resulting -ve orbital moment in 110 direction . Could you please give me any suggestion???

best regards
sumanta

jxw

Registered: 11/09/06
Posts: 13

Thursday, March 11 2010 @ 08:37 PM CET

Spin and local coordinate (which the euler angles specify) axes are only the same when the spin axis is the highest symmetry axis. I gave a routine (symt_base.c) to Olle Eriksson that chooses the spin axis as the local-coordinate z-axis for the first site of a type. This would replace the corresponding routine in the sym directory. This will work in your case, at least.

jxw

Torbjorn

Registered: 11/03/06
Posts: 75

Thursday, April 29 2010 @ 08:19 AM CEST

OK, now I ran into the same problem.

I guess a more satisfactory solution would be to solve it within the band structure program itself by calculating the orbital moment from eigenvectors and structure functions that have been rotated to coordinates where the spin axis is in the z-direction. That means having a set of density coefficients that are only used for the orbital moment, which seems like a waste of space, but it is fairly easy to do.

Alternatively we include the spin-flip terms in the density coefficients and then rotate them directly when calculating the orbital moment. That would take a (small) step towards a non-collinear code but has the drawback that any bug in the implementation will influence all calculations and not just the orbital moment...

Any thoughts John? In particular, were the things I said above correct?

Torbj�rn

Admin

Registered: 10/12/06
Posts: 130

Thursday, April 29 2010 @ 08:36 AM CEST

If I may hijack this thread a little:

I was helping Sumanta a few days ago when I noticed that symt (at least from rspt916) calculated the wrong number of valence electrons. The number from sum(A)-sum(Ncore) was 24 but symt put 60 in the dta files. I have not looked into the problem in detail, at this stage I am more curious whether anyone else has seen this behavior?

/Per

olleerik

Registered: 11/14/06
Posts: 8

Thursday, April 29 2010 @ 09:49 AM CEST

I think the easiest way is to use a symcof generator that calculates Euler angles so that the quantization axis of the spherical harmonics coincides with the spin-quantization axis. Then RSPt can be used as is, and orbital moments should come out OK. Another alternative is to calculate expectation values not only of l_z but of l_+ and l_-, from which l_x and l_y can be obtaned since l_+=l_x+i(l_y) and l_-=l_x -i(l_y). I think the first option is easiest, and is what the symcof-code John sent out does, right ? Sumanta, did I not give it to you ?