[QE-users] LIBXC versus internal QE functionals

[Abdesalem Houari]

Dear All,
I'm sorry, I have written my previous email in rush !After checking well the exchange-correlation document (Modules/funct.f90), my first issue seems resolved. 
I found that "pbe"   = "sla+pw+pbx+pbc" (= XC-001I-004I-003I-004I-000I-000I).So the true PBE needs, not only "pbx" exchange + "pbc" correlation, but also "sla" exchange + "pw" correlation. 

Thanks.
A. Houari
 

    On Sunday, 26 May 2024 at 08:54:24 pm GMT+1, Abdesalem Houari via users <users at lists.quantum-espresso.org> wrote:  
 
 Dear All,
I wanted to do a calculation test using PBE-functional from LIBXC and compare it to the internal QE one. As a test, I opted for body centered cubic (bcc) iron.
I'm a bit puzzled by the differences in the results.
- First, for PBE from LIBXC, following the QE user-guide on how to use LIBXC, I put the following input: input_dft='XC-000I-000I-101L-130L-000I-000I'. I obtained a total energy value of -39.08694619 Ry.

- Then, for QE internal PBE functional, after having a look to Modules/funct.f90 file, I deduced the PBE should be forced from the input as follow: input_dft='XC-000I-000I-003I-004I-000I-000I'. Here I was quite surprised  obtaining a different value of the total energy  -31.22123378 Ry... It seems like it is not the correct PBE ?

- When I just use NC-PP (Fe.pbe-mt_fhi.UPF) without forcing the input_dft, I get a similar result to LIBXC-PBE ... The same result is obtained also when I forced the input_dft, but including LDA-exchange (Slater)+correlation (PW) i.e. input_dft='XC-001I-004I-003I-004I-000I-000I'

I'm really confused ! I would be grateful if some could give a bit of explanation, on how the input_dft works exactly ? especially with respect to the Module/funct.f90 file which gives all the internal functionals.  


I would also ask for another issue. It is explained in the PW manuel that input_dft overrides the PP. Normally, if I'm not wrong, whatever the kind of the PP given in the input file (PBE or other; norm-conserving or ultra-soft or PAW potential), the code will discard it and does not look at it at all when input_dft is forced.Here also I'm confused, I forced the input_dft as: input_dft='XC-001I-004I-003I-004I-000I-000I', with two different PP (Fe.pbe-mt_fhi.UPF and Fe.pbe-spn-rrkjus_psl.0.2.1.UPF); and the total energies are completely different !  

Thanks a lot in advance

Sincerely yours.

========================================
Dr. Abdesalem HOUARI-------------------------------------------------------------------------------------------
Department of physics, Theoretical Physics Laboratory
University of Bejaia-06000. Algeria.
E-mail: abdeslam.houari at univ-bejaia.dz & habdslam at yahoo.fr
https://sites.google.com/site/houariabdeslam/homepage
========================================
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users at lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240526/49001718/attachment.html>