[Wien] Running optics with hybrid functional

Peter Blaha pblaha at theochem.tuwien.ac.at
Sun Feb 28 09:03:28 CET 2021 

A few comments:

a) For hybrid DFT, NEVER run   x kgen [switches]  yourself (unless you 
really know what you are doing.
The k-mesh must be defined using    runkgenhf_lapw

For the sequence of programs in a scf cycle, you can always examine the 
   :log  file. The exact order depends a bit on various options (eg. 
-redklist), but basically you call lapw2 twice, once after the "lapw1 
step", then comes the "x hf [..] step", and then  lapw2 -hf [..]

Since we also copy/rename some files, it is always recommended to use 
the run_lapw -hf ...  script.

Hybrid-DFT calculations are a bit more tricky than normal DFT and it is 
highly recommended that yu read the corresponding sections in the UG in 
detail. (If you think some info is missing / not understandable, please 
let me know)

Am 27.02.2021 um 22:50 schrieb Gavin Abo:
> The WIEN2k 19.1 (or 19.2) UG [1] on page 55 under "Starting a 
> calculation from another k-mesh" has:
> 
> /Do the calculation with the first k-mesh and ”save” it when it is 
> finished (do not execute clean_lapw since case.vectorhf should be present)./
> 
>  From that, after your scf calculation of step 1 you may need to do a 
> "save_lapw".
> 
> After that, you may have to refer to section "6.5 KGEN (generates k 
> mesh)" starting on page 113.  There is looks like "x kgen -hf" might be 
> needed for creating the reduced k-mesh.  However, perhaps there is a way 
> to use run_kgenhf_lapw [2] described on page 54 of the UG under "Reduced 
> k-mesh for the HF potential" that would be better.  Some experienced 
> with "hf" calculations should have better advice then me on that.
> 
> Under "Flow in run_lapw -hf" on page 52, it looks like "hf" comes after 
> "lapw2" [3].  Since the k-mesh was changed and reduced one to be used, 
> probably -newklist and -redklist, seen under section "7.6  HF 
> (Calculates the hybrid orbitals and eigenvalues)" on page 136, needs to 
> be used simultaneously with hf.
> 
> So, I'm thinking the procedure may need to be something like:
> 
> 1. run_lapw -p -hf -newklist -mode1 -NI
> 2. save_lapw -d scf
> 
> 
> 3. x kgen -hf
> 3.a. x lapw1 -p
> 3.b. x lapw2 -fermi -hf -p
> 3.c. x hf -p -newklist -redklist
> (or 3. run_kgenhf_lapw -redklist)?
> 
> 
> 4. x optic -hf -p
> 5. x joint -hf
> 6. x kram
> 
> 
> [1] http://www.wien2k.at/reg_user/textbooks/usersguide.pdf
> [2] 
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20258.html
> [3] 
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20264.html
> 
> On 2/27/2021 12:39 PM, pboulet wrote:
>> Dear Gavin and Peter,
>>
>> Thank you for your response.
>>
>> Peter: if I understand correctly, I should run x hf -p between my 
>> steps 1 and 2 but with a reduced k-mesh, right?
>>
>> Thank you
>> Best
>> Pascal
>>
>> Pascal Boulet
>> —
>> /Professor in computational materials chemistry - DEPARTMENT OF CHEMISTRY/
>> University of Aix-Marseille - Avenue Escadrille Normandie Niemen - 
>> F-13013 Marseille - FRANCE
>> Tél: +33(0)4 13 55 18 10 - Fax : +33(0)4 13 55 18 50
>> Email : pascal.boulet at univ-amu.fr <mailto:pascal.boulet at univ-amu.fr>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> Le 27 févr. 2021 à 15:21, Peter Blaha <pblaha at theochem.tuwien.ac.at 
>>> <mailto:pblaha at theochem.tuwien.ac.at>> a écrit :
>>>
>>> In the individual steps you are missing the most time consuming step:
>>> x hf -p
>>> Thus you do not have a vectorhf file corresponding to your new k-mesh.
>>>
>>> I'd use a reduced k-mesh (see hybrid calculations in the UG) to save 
>>> time.
>>>
>>> PS: Please note that the momentum matrix elements for hybrid 
>>> calculations are too small due to a missing term and your optics with 
>>> hybrid-DFT will be too small (see a recent paper O.Rubel et al., CPC 
>>> 261 (2021) 107648 and references therein).
>>>
>>>
>>> Am 27.02.2021 um 14:17 schrieb pboulet:
>>>> Dear all,
>>>> I am trying to perform an optic calculation with hybrid (HSE) 
>>>> functional. The problem I face is that I systematically end up with 
>>>> Im(eps)= 0. and Re(eps)=1.0 irrespective of the energy.
>>>> The plasma frequencies w_p_xx and w_p_zz are both Nan.
>>>> I proceed this way:
>>>> 1) Run a SCF calculations with 12x12x12 k-points grid (1728 fbz 
>>>> k-points, 163 ibz k-points):
>>>> run_lapw -p -hf -newklist -mode1 -i 1 -NI
>>>> The run is exceeding the queue limit of 24h so I proceed step by 
>>>> step (1step=13h on 288 cores)
>>>> 2) Run the optic calculations sequence:
>>>> x lapw1 -p
>>>> x lapw2 -fermi -hf -p
>>>> x optic -hf -p
>>>> x joint -hf
>>>> x kram
>>>> Is there anything wrong in the procedure?
>>>> I can provide input/output files if necessary.
>>>> Thank you for your help,
>>>> Pascal
>>>> Pascal Boulet
>>>> —
>>>> /Professor in computational materials chemistry - DEPARTMENT OF 
>>>> CHEMISTRY/
>>>> University of Aix-Marseille - Avenue Escadrille Normandie Niemen - 
>>>> F-13013 Marseille - FRANCE
>>>> Tél: +33(0)4 13 55 18 10 - Fax : +33(0)4 13 55 18 50
>>>> Email :pascal.boulet at univ-amu.fr 
>>>> <mailto:pascal.boulet at univ-amu.fr><mailto:pascal.boulet at univ-amu.fr 
>>>> <mailto:pascal.boulet at univ-amu.fr>>
>>>> _______________________________________________
>>>> Wien mailing list
>>>> Wien at zeus.theochem.tuwien.ac.at <mailto:Wien at zeus.theochem.tuwien.ac.at>
>>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien 
>>>> <http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien>
>>>> SEARCH the MAILING-LIST at: 
>>>> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html 
>>>> <http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html>
>>>
>>> --
>>> --------------------------------------------------------------------------
>>> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>>> Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
>>> Email:blaha at theochem.tuwien.ac.at 
>>> <mailto:blaha at theochem.tuwien.ac.at>   WIEN2k:http://www.wien2k.at 
>>> <http://www.wien2k.at/>
>>> WWW: http://www.imc.tuwien.ac.at <http://www.imc.tuwien.ac.at/>
>>> -------------------------------------------------------------------------
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 

-- 
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at
-------------------------------------------------------------------------

More information about the Wien mailing list