The ABINIT Users Mailing List ( CLOSED )

[Jorge Iniguez <jiniguez@nist.gov>]

Hi Aldo,

> I have notice that one of the new features in Abinit, it is related
> to noncollinear magnetism. Can somebody comment on this?
> how good is it? do you happen to have a simple example to try ?
> could you include spin-orbit correction?

I do not know which "new" features you are talking about... but I can tell
you that both non-collinear magnetism and spin-orbit haver been around for
quite some time. You will find sample inputs among the tests, starting
from version 2.

I spent some time testing the non-collinear magnetism implementation and
got convinced it works fine (although convergence is sometimes tricky and
very small tolerances are generally required; note that the energy
differences involved in rotating spinors around are usually veeery small).
Regarding the the spin-orbit coupling, it is implemented for a couple of
pseudopotential types, and I do not know how well tested it is. I was
scheduled to test it in detail, but at some point along the way I got
convinced that was not such an interesting thing to do, as what one
eventually wants is to make spin-orbit work within the new PAW
implementation. But I do not know when that is going to happen (at least,
I do not know when I will have time to work on it :)

Best,
Jorge


------------------------------------------------------------------------------
   Jorge Iniguez             NIST Center for Neutron Research
                             100 Bureau Drive, Mail Stop 8562
                             Gaithersburg, MD 20899-8562
                             Phone: 1-301-975-8367 Fax: 1-301-921-9847
                             E-mail address: jiniguez@nist.gov
                             http://www.ncnr.nist.gov/staff/jorge
------------------------------------------------------------------------------