[Pw_forum] HSE pseudopotentials + BNL functional

[Helen]

Hi,

The changes are localized in PW/exx.f90, flib/functionals.f90, 
Modules/functionals.f90. I've implemented the changes in version 3.2. I need 
to check I didn't make changes elsewhere as I implemented this about a year 
ago, and I don't remember all the details. I set on_double_grid to .false. 
Also in Modules/input_parameters.f90 I set x_gamma_extrapolation=.false.

I can create an example input/output file. I also worked on enabling exx.f90 
to give the band-structure for general k-points and not just high symmetry 
k-points. I did this my giving a very small weight to the desired k-point 
and also including in the k-point list all the generated k+q k-points. It is 
a bit of an art though as I put in the k+q k-points by hand and sometimes 
the exx.f90 generates the symmetry points and sometimes I also need to add 
these by hand. Also sometimes the symmetry points need to have zero weight 
and sometimes the very small weight to give corect results.

Dr Helen Eisenberg
 Post-doctoral researcher
 Fritz Haber Research Center for Molecular Dynamics,
 The Hebrew University of Jerusalem


----- Original Message ----- 
From: "Stefano de Gironcoli" <degironc at sissa.it>
To: <pw_forum at pwscf.org>
Sent: Monday, March 23, 2009 5:27 PM
Subject: Re: [Pw_forum] HSE pseudopotentials + BNL functional


> Dear Dr. Eisenberg,
>
>    If you are willing to share your developments it would be very
> much appreciated.
>
>    Are your modifications localized in a few files (like exx.f90 and
> functionals.f90) ? is the functional implemented in the latest version
> of Quantum-Espresso ?
>    If the annswers are yes the simplest option to contribute your
> code would be to send me and/or to Layla Martin-Samos
> (marsamos at democritos.it) the relevant routines so that we can include
> them in the cvs repository and then revert to you to check that
> everything is fine.
>    An input (and an output) using the new feature similar to those in
> the EXXexample directory could also be useful to keep the feature
> working when
> the code is updated.
>
>    best regards and thank you for your contribution,
>
>    stefano
> -
> Stefano de Gironcoli - SISSA and DEMOCRITOS
>
>
> Quoting Helen <helen at fh.huji.ac.il>:
>
>> Dear all,
>> I've been working on a functional somewhat similar to the HSE
>> functional, the BNL functional (R. Baer and D. Neuhauser, PRL
>> 94,043002 (2005)) which also has a explicit-exchange component. I
>> have implemented this functional into Quantum-Espresso and the full
>> details can be seen in our paper currently submitted for publication
>>  and on the archives: arXive:0903.3830v1.
>> Where should I post the Quantum-Espresso code implemented for the
>> BNL functional?
>> Dr Helen Eisenberg
>> Post-doctoral researcher
>> Fritz Haber Research Center for Molecular Dynamics,
>> The Hebrew University of Jerusalem
>>
>>   ----- Original Message -----
>>   From: shangy
>>   To: PWSCF Forum
>>   Sent: Saturday, March 14, 2009 7:49 AM
>>   Subject: [Pw_forum] HSE pseudopotentials
>>
>>
>>   Dear all:
>>        Does anyone has the Si pseudopotentials generated by the
>> screened-exchange hybrid functional of Heyd, Scuseria, and Ernerhof
>> [HSE, J. Chem. Phys. 118, 8207 (2003)]? Can it work well within QE?
>> If anyone has one, would you share it with me? I'm very very thirsty
>>  for it. Thanks in advance.
>>        Best regards.
>>
>>        shangy
>>        Graduate unversity of Chinese academy of sciences
>>
>>
>> ------------------------------------------------------------------------------
>>
>>
>>   _______________________________________________
>>   Pw_forum mailing list
>>   Pw_forum at pwscf.org
>>   http://www.democritos.it/mailman/listinfo/pw_forum
>>
>
>
>
> ----------------------------------------------------------------
>   SISSA Webmail https://webmail.sissa.it/
>   Powered by Horde http://www.horde.org/
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>