[Pw_forum] Question on epsilon.x

[Andrea Ferretti]

Hi Mohsen,

epsilon.x computes the matrix elements of the momentum p,
which is not easy to augment in a ultrasoft-pp fashion (at variance with 
operators which are functions of r, i.e. local).

According to the size of system you want to study, you may want to generate 
your own norm-conserving pseudopotentials for Ba and Ti.

The other possibility is instead to use the PAW formalism, where the 
augmentation of p matrix elements is as easy as for r.
(of course this would require the generation of the PAW dataset, and some 
implementation)

Related to the program epsilon.x:
optical response can be also calculated through the matrix elements
of the operator e^(iqr), which can be augmented easily with USPP. 
Unfortunately, using this solution would require to re-write the program almost 
from scratch (not recommended).
Anyway if you want to follow this possibility,
I would then start from a working norm-conserving implementation of it, as the 
ones in Sax or Yambo

hope it helps
Andrea

BTW: in future posts please remember to add your affilication, according to the 
"politeness policy" of this forum


> Dear users,
> at first, i am so sorry that i asked many question about "epsilon.x".
> But i faced with "USPP are not implemented", Which means i can not use 
> ultrasoft pseudopotentials but i want work with Ba and Ti,
> is there any another Ref. for pseudopotentials or can i solve this problem in 
> some way?
> Thanks for your reply
> Mohsen Modaresi
>

-- 
Andrea Ferretti
Oxford University, Department of Materials
Parks Road, Oxford OX1 3PH, UK
Tel: +44 (0)1865 612796;  Skype: andrea_ferretti
URL: http://quasiamore.mit.edu

Please, if possible, don't send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html