[QE-users] Issue with ONCV pseudos and projwfc.x (lforcet=.true.)

[Giuseppe Mattioli]

Dear Andrea and Guido
I would add further warning: test the pseudos, because I’ve obtained  
different total energies for the same  species and input by building  
the oncvpp code with different compilers, and even ghost states when  
compiling 3.3.1 with gfortran 4.9 some time ago. Lorenzo Paulatto  
rebuilt the entire sg15 library of ONCVPPs with the pseudo wave  
functions, and I’m happily using his files. Maybe you can ask him,  
because I cannot share them at the moment.
HTH
Giuseppe

Guido Menichetti <menichetti.guido at gmail.com> ha scritto:

> Hi Andrea,
>
> Thank you very much for your help and the quick answer.
>
> I will try to re-build the pseudo following your recipe.
>
> Thank you again.
> Have a nice day
>
> G.
>
> Il giorno gio 3 gen 2019 alle ore 12:52 Andrea Ferretti <
> andrea.ferretti at unimore.it> ha scritto:
>
>>
>>
>> Hi Guido,
>>
>> I think the issue can be related to the lack of atomic orbitals in the
>> oncv upf files.
>> From the link you report, I see  number_of_wfc="0"  for all of the
>> pseudo's.
>>
>> (These are not the beta projectors, there by construction, but auxiliary
>> atomic orbitals needed for atomic projections, DFT+U, etc).
>>
>> If you really want to use these pseudos, a possible solution that worked
>> with us in the past is to simply regenerate the pseudo with a newer
>> version of the oncv code.
>>
>> For instance, the following procedure worked with me:
>>
>> * extract the oncv input file from the given UPF files without atomic wfc
>> * download and compile a recent version of the oncv code (v3.2.3 worked
>>    for me)
>> * adapt the PP input file to this version of oncv (if I remeber correctly
>>    a few input flags changed). Select upf output fmt
>> * re-generate the pseudo (a different and larger radial grid will
>>    generated). Atomic wfcs should be there.
>> * possibly check that the two pseudos give almost identical results
>>
>> take care
>> Andrea
>>
>> >
>> > I have an issue with ONCV pseudopotentials (taken from
>> http://www.quantum-simulation.org/potentials/sg15_oncv/) and projwfc.x.
>> >
>> > I used qe-6.3, q-e-develop (updated 29/12/2018) versions to test this
>> issue.
>> >
>> > I used the example in PP/examples/ForceTheorem_example as simple test.
>> > Following the example, I ran pw.x < sr.in > sr.out_sg15 with the
>> scalar-relativistic pseudopotential.
>> > Then I ran pw.x < par.in > par.out_sg15 with the full relativistic
>> pseudo.
>> > At the end I ran projwfc.x < proj_par.in > proj_par.out but it resturns:
>> >
>> > ** On entry to ZGEMM parameter number 13 had an illegal value
>> >
>> > The input and output files together with the pseudos are shared in the
>> following link.
>> >
>> >
>> > I hope you can help me to solve this issue.
>> > Thank you for your work,
>> > Kind Regards
>> >
>> > Guido
>> >
>> > --
>> > ***************************************
>> > Guido Menichetti
>> > Post-Doc researcher in Solid State Physics
>> > Istituto Italiano di Tecnologia
>> > Graphene Labs: Theory and technology of 2D materials
>> > Address: Via Morego, 30, 16163 Genova
>> > Email:   guido.menichetti at iit.it
>> >              guido.menichetti at df.unipi.it
>> >              menichetti.guido at gmail.com
>> > ****************************************
>> >
>> >
>>
>> --
>> Andrea Ferretti, PhD
>> S3 Center, Istituto Nanoscienze, CNR
>> via Campi 213/A, 41125, Modena, Italy
>> Tel: +39 059 2055322;  Skype: andrea_ferretti
>> URL: http://www.nano.cnr.it
>> _______________________________________________
>> users mailing list
>> users at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
>
> --
> ***************************************
> Guido Menichetti
> Post-Doc researcher in Solid State Physics
> Istituto Italiano di Tecnologia
> Graphene Labs: Theory and technology of 2D materials
> Address: Via Morego, 30, 16163 Genova
> Email:   guido.menichetti at iit.it
>              guido.menichetti at df.unipi.it
>              menichetti.guido at gmail.com
> ****************************************



GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>