[Pw_forum] Forces in full LDA+U scheme are not yet implemented

[Burak Himmetoglu]

Hi Izaak,

You need to set lda_pu_kind = 0 to use U and J in the simplified version of
DFT+U where forces are implemented.

Best regards,

Burak Himmetoglu

Post-Doctoral Associate
Materials Department
University of California at Santa Barbara
CA 93106

On Wed, Feb 27, 2013 at 8:31 AM, Izaak Williamson <izaakw89 at yahoo.com>wrote:

> Hello everyone,
>
> I am using the DFT+U approach to compare my system with experimental
> results. To do so, I want to optimize the effective on-site Coulomb value
> (U) and keep the exchange value (J) fixed while simultaneously relaxing the
> structure using vc-relax. When I try and submit the vc-relax (or even
> relax) calculations, I get the following error:
>
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>      Error in routine force_hub (1):
>       forces in full LDA+U scheme are not yet implemented
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> This doesn't seem right to me. Is it true what the error says, that the
> force calculations are not yet implemented in the "rotationally invariant
> scheme of Liechtenstein et. Al., using Hubbard_U and Hubbard_J"? Perhaps I
> have made a mistake in my input file. The system portion of my input file
> is given:
>
> &SYSTEM
>     ibrav            = 0
>     celldm(1)        = 1.889725989
>     nat              = 73
>     ntyp             = 4
>     ecutwfc          = 60
>     ecutrho          = 600
>     occupations      = 'smearing'
>     degauss          = 0.001
>     nspin            = 2
>     starting_magnetization(1) = 0.5
>     starting_magnetization(2) =-0.5
>     tot_magnetization= 0.0
>     lda_plus_u       = .t.
>     lda_plus_u_kind  = 1
>     Hubbard_U(1)     = 2.6
>     Hubbard_U(2)     = 2.6
>     Hubbard_J0(1)    = 1.2
>     Hubbard_J0(2)    = 1.2
> /
>
> NOTE: Whenever lda_plus_u_kind is set to 1, any relax calculation gives
> the above error. When it is 0, it runs fine.
>
> Is it necessary to include "lda_plus_u_kind = 1" parameter or will the
> code still incorporate U and J if I set it to 0 (simplified version of
> Cococcioni and de Gironcoli using Hubbard_U)?
>
> Thanks in advance.
>
> --
> Izaak Williamson
> Graduate Student
> Department of Materials Science and Engineering
> Boise State University
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>



-- 
Burak Himmetoglu
Research Associate
Department of Materials
University of California at Santa Barbara, CA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20130227/3e9f64e6/attachment.html>