[Wien] Some confirmation regarding SO calculation

[Vinayak Mishra]

Dear Prof. Blaha

 Thanks much for your prompt reply but kindly confirm me a little bit more.
I shall be very grateful.
You had asked me whether I have RLO. I do have semi-core p states and
that's why I am using LO but the problem that I am encountering is that
spin-orbit calculation done on the same element by one of my colleague on
crystal structure FCC gives optimum volume which is around 10 a. u.
different than mine. So I was wondering whether it is only large spin -
orbit effect i.e more than 5 Ry (and up to 10 Ry) only or I am missing
something in the specs of the spin-orbit calculation of WINE-2K package
which I described in the previous mail (given below also).
Could you just kindly confirm this.

Thank you very much in advance and kind regards
V. Mishra





Prof Blaha replied:

You cannot expect "unique" answers. All this may depend on the system AND
on the properties
you are interesting.

Adding RLO:  There is a good reason why N (NO) is the "default" for initso.
RLOs are mainly usefull for elements with semicore p-states ( do you have
them ??)
         and for TOTAL ENERGIES.

EMAX in case.in1: This is similar as above: For a bandstructure or DOS
emax=5 will be sufficient in most cases. For E-tot (volume optimization, or
comparison of two structures)  it may not be enough.

Emax in case.inso is only for "printint" (or DOS).

In the end it is always the same:
TEST it yourself for your specific system and propertiy.




Am 29.02.2012 13:23, schrieb Vinayak Mishra:

> Dear WIEN 2K users and Dr Blaha,
>
> I am a regular WIEN2K user. Off late I have started doing spin orbit (SO)
calculation of an element. And I want some small confirmation here.
>

>   For this I have done the normal SR calculation. After this, in order to
include SO effect I invoke the command initso_lapw and generate the
case.inso file by agreeing for
> default settings for ecverything except when it asks to "Add  RLO for
NONE, ALL,  CHOOSE elements ? (N/a/c)" and for this I select "a". This is
how I created inso file. And at the

> same time I observed that in the case.in1 file emax has changed from
default 2.5 to 5. This is the default setting for initso_lapw. Now in order
to run SO calculation I invoke the
> command run_lapw -so -ec 0.00001. And this is how I did the calculation
at one particular value of emax. Now I am not very clear that if I have to
include large spin orbit effect

> (more than 5 Ry, for which maximum recommended value is 10 Ry), should I
change the emax in case.in1 file only or in case.inso file also as I guess
emax in case.inso file is for
> printing purpose only.If not so then what is its purpose.

>
> I am really sorry for bothering but  I just want to confirm this before
going ahead.
>
> Thank you very much in advance for your attention and time.
>
> V. Mishra

>





-- 
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
Vinayak Mishra
Scientific Officer-D
Govt. of India, Dept. of Atomic Energy
Bhabha Atomic Research Centre
Facility for Electromagnetic Systems, C. A. D.

IDA Block 'B', 4th Cross Road, Beside St. Aloysius ITI,
Autonagar, Visakhapatnam - 530 012,  Andhra Pradesh.
Tele (o): +91-891-2892166,  Fax  (o): +91-891-2742460
email : v.mishra.11 at gmail.com
////////////////////////////////////////////////////////////////////////////
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20120302/346290cc/attachment.htm>