[Wien] NMR switch for metals

[Peter Blaha]

Dear wien2k users,

The calculation of Knight shifts in metals is quite difficult, as it has 
not only the orbital contribution, but also mainly a spin contribution. 
In addition, the convergence with respect to k-mesh of the different 
contributions is very difficult.
We (in particular Robert Laskowski) is working on this. We have not yet 
come up with a final answer, but I'm confident that we will have in near 
future a working version, which can treat NMR in metals with larger, but 
still reasonable effort.

What is in the distributed code and suggested in previous mails (-metal 
and the suggestion with TEMP) is according to our latest developments 
already a bit outdated and we think we have a better version, but this 
is still under development.

In any case, as a hint for checking: it turned out that calculations of 
the orbital susceptibility (case.xim) is the most difficult term and if 
this gives a diverging result, the computed shift is also very wrong.

So in conclusion: In the present version of the nmr code, be careful 
with NMR in metals ! There is no publication out yet on this topic. Also 
note, that Mauri et al never ever published a second paper with NMR on 
metals. We get different (but already now better) results then in this 
paper, but it is too early for final conclusions.

Peter Blaha


On 09/25/2014 04:13 PM, Mausolf, Bernhard wrote:
> Dear all,
> We are trying to calculate Al NMR shifts for an intermetallic phase
> containing many different Al-positions (wien2k-version 14.1). We want to
> compare the calculated NMR-shifts with our experimental data. At first
> we did calculations without the "-metal"-option of the NMR-modul
> (x_nmr_lapw) and tried to converge the results in the
> case.outputnmr_integ file with respect to the k-mesh and rkmax-value.
> The results look quite reasonable.
>
> At the moment we are using the same struct-file as input to recalculate
> the NMR shifts by using the "-metal"-option referring to this
> mailing-list thread:
>
> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2014-March/020807.html
>
> We changed TETRA to TEMP in the in2-input files but did not start to
> play with the "-kbT"-option yet.
>
> The questions are: If convergence can be achieved with both methods,
> should the results be exactly the same with and without "-metal"-option?
> Are there any differents in the expected accuracy of the mentioned methods?
> Which method should be preferred for metallic cases? And finally, are
> there any publications in addition to PRB 2012(1), 2012(2) and 2014 from
> Mr. Laskowski, that deal with calculation of NMR parameters (especially
> for metals)?
>
> Thank you in advance!
> B. Mausolf
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/staff/tc_group_e.php
--------------------------------------------------------------------------