[QE-users] Error with pw2gw.x

Shivesh Sivakumar shiveshsivakumar at gmail.com
Fri Jul 17 02:52:36 CEST 2020 

Hello,

Sorry to bring up this topic again, but I realized the finite difference
method (in BerkeleyGW) for calculating optical selection rules require
wavefunction data from both the original k-grid and q-shifted grid. These
calculations are proving to be very computationally expensive. I have the
wavefunction data from the scf calculation (with HSE) for my material but
another calculation with q-shifted grid again requires lot of computational
power. Is there any other way to calculate the dipole matrix element for
HSE calculations on QE?

Best,
Shivesh

On Mon, Jul 13, 2020 at 2:22 PM Shivesh Sivakumar <
shiveshsivakumar at gmail.com> wrote:

> Thank you, Paolo!
>
> Best,
> Shivesh
>
> On Mon, Jul 13, 2020 at 11:57 AM Paolo Giannozzi <p.giannozzi at gmail.com>
> wrote:
>
>> Which routine? pw2gw.x ? input documentation is here:
>> PP/Doc/INPUT_pw2bgw.def
>>
>> Paolo
>>
>> On Mon, Jul 13, 2020 at 8:16 PM Shivesh Sivakumar <
>> shiveshsivakumar at gmail.com> wrote:
>>
>>> Dear Paolo,
>>>
>>> Thank you very much for the answer. I was able to solve that issue. As a
>>> sidenote, what does the routine take as the input? Does a QE wavefunction
>>> generated by pp.x suffice as an input wavefunction?
>>>
>>> Best,
>>> Shivesh
>>>
>>> On Mon, Jul 13, 2020 at 12:03 AM Paolo Giannozzi <p.giannozzi at gmail.com>
>>> wrote:
>>>
>>>> The code that produces data for BerkeleyGW is pw2bgw.x, not pw2gw.x
>>>>
>>>> On Sun, Jul 12, 2020 at 11:29 PM Shivesh Sivakumar <
>>>> shiveshsivakumar at gmail.com> wrote:
>>>>
>>>>> Hello users,
>>>>>
>>>>> I posted this earlier regarding pw2gw.x and wasn't able to find a
>>>>> solution. I would be glad if someone could suggest what I can potentially
>>>>> do.
>>>>>
>>>>> I am trying to calculate the momentum matrix elements for my material
>>>>> with HSE and HSE+U etc, which is not supported in QE. I already calculated
>>>>> the matrix elements with PBE, but I want to see if the optical selection
>>>>> rules change when the hybrid functional is turned on.
>>>>>
>>>>> I set out to do this with the help of BerkeleyGW- They outline a
>>>>> relatively simple way that takes into account only the wavefunctions in the
>>>>> inner product (i.e. the dipole matrix element). This 'finite difference'
>>>>> method is mentioned in the BerkeleyGW manual ( on arxiv, equation 46).
>>>>>
>>>>> I tried porting my wavefunctions from QE to GW using pw2gw.x but I am
>>>>> encountering some errors. Please bear with me while I outline the procedure
>>>>> I used, and the doubts I have.
>>>>>
>>>>> 1) Use pp.x to save the wavefunction in a readable format. Here is my
>>>>> pp.x input. To start with, I just want to calculate the matrix element
>>>>> between the band edges at gamma point. The input below generates the
>>>>> wavefunction at VBM GAMMA.
>>>>>
>>>>> &INPUTPP
>>>>> prefix='CrPs4SCF'
>>>>> outdir='./'
>>>>> filplot='wfn'
>>>>> plot_num=0
>>>>> kpoint=1
>>>>> kband=92
>>>>> /
>>>>>
>>>>>
>>>>> &PLOT
>>>>> output_format=5
>>>>> iflag=3
>>>>> filepp='./plots'
>>>>> fileout='./exportgcf'
>>>>> /
>>>>>
>>>>> This seems to generate the wavefunction just fine. Did not encounter
>>>>> an issue here.
>>>>>
>>>>> 2) To convert this wavefunction to a format that BGW can read, I used
>>>>> pw2gw.x. Here's the input:
>>>>>
>>>>> &input_pw2bgw
>>>>>    prefix = 'CrPs4SCF'
>>>>>    outdir = './'
>>>>>    real_or_complex = 2
>>>>>    wfng_flag = .true.
>>>>>    wfng_file = 'realwfnGW'
>>>>>    wfng_kgrid = .false.
>>>>>    wfng_nk1 = 4
>>>>>    wfng_nk2 = 6
>>>>>    wfng_nk3 = 1
>>>>> /
>>>>>
>>>>> This seems to lead to an error :   Error in routine pw2gw (1):
>>>>>  reading inputpp namelist.
>>>>>
>>>>> Would someone be able to tell me where I might have gone wrong? I am
>>>>> also confused as to what input the pw2gw file reads - Is it the
>>>>> wavefunction directly from my first step i.e. the output 'plots' from my
>>>>> pp.x calculation? or should I copy paste the whole SCF directory that
>>>>> includes the .xml file and charge density data?
>>>>>
>>>>> Best,
>>>>> Shivesh Sivakumar
>>>>> University of Washington-Seattle
>>>>>
>>>>> On Fri, Jul 10, 2020 at 6:37 PM Shivesh Sivakumar <
>>>>> shiveshsivakumar at gmail.com> wrote:
>>>>>
>>>>>> Hello all,
>>>>>>
>>>>>> I am trying to calculate the momentum matrix elements for my material
>>>>>> with HSE and HSE+U etc, which is not supported in QE. I already calculated
>>>>>> the matrix elements with PBE, but I want to see if the optical selection
>>>>>> rules change when the hybrid functional is turned on.
>>>>>>
>>>>>> I set out to do this with the help of BerkeleyGW- They outline a
>>>>>> relatively simple way that takes into account only the wavefunctions in the
>>>>>> inner product (i.e. the dipole matrix element). This 'finite difference'
>>>>>> method is mentioned in the BerkeleyGW manual (
>>>>>> https://arxiv.org/pdf/1111.4429.pdf, equation 46).
>>>>>>
>>>>>> I tried porting my wavefunctions from QE to GW using pw2gw.x but I am
>>>>>> encountering some errors. Please bear with me while I outline the procedure
>>>>>> I used, and the doubts I have.
>>>>>>
>>>>>> 1) Use pp.x to save the wavefunction in a readable format. Here is my
>>>>>> pp.x input. To start with, I just want to calculate the matrix element
>>>>>> between the band edges at gamma point. The input below generates the
>>>>>> wavefunction at VBM GAMMA.
>>>>>>
>>>>>> &INPUTPP
>>>>>> prefix='CrPs4SCF'
>>>>>> outdir='./'
>>>>>> filplot='wfn'
>>>>>> plot_num=0
>>>>>> kpoint=1
>>>>>> kband=92
>>>>>> /
>>>>>>
>>>>>>
>>>>>> &PLOT
>>>>>> output_format=5
>>>>>> iflag=3
>>>>>> filepp='./plots'
>>>>>> fileout='./exportgcf'
>>>>>> /
>>>>>>
>>>>>> This seems to generate the wavefunction just fine. Did not encounter
>>>>>> an issue here.
>>>>>>
>>>>>> 2) To convert this wavefunction to a format that BGW can read, I used
>>>>>> pw2gw.x. Here's the input:
>>>>>>
>>>>>> &input_pw2bgw
>>>>>>    prefix = 'CrPs4SCF'
>>>>>>    outdir = './'
>>>>>>    real_or_complex = 2
>>>>>>    wfng_flag = .true.
>>>>>>    wfng_file = 'realwfnGW'
>>>>>>    wfng_kgrid = .false.
>>>>>>    wfng_nk1 = 4
>>>>>>    wfng_nk2 = 6
>>>>>>    wfng_nk3 = 1
>>>>>> /
>>>>>>
>>>>>> This seems to lead to an error :   Error in routine pw2gw (1):
>>>>>>  reading inputpp namelist.
>>>>>>
>>>>>> Would someone be able to tell me where I might have gone wrong? I am
>>>>>> also confused as to what input the pw2gw file reads - Is it the
>>>>>> wavefunction directly from my first step i.e. the output 'plots' from my
>>>>>> pp.x calculation? or should I copy paste the whole SCF directory that
>>>>>> includes the .xml file and charge density data?
>>>>>>
>>>>>> Best,
>>>>>> Shivesh Sivakumar
>>>>>> University of Washington-Seattle
>>>>>>
>>>>> _______________________________________________
>>>>> Quantum ESPRESSO is supported by MaX (
>>>>> www.max-centre.eu/quantum-espresso)
>>>>> users mailing list users at lists.quantum-espresso.org
>>>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>>>
>>>>
>>>>
>>>> --
>>>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>>>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>>>> Phone +39-0432-558216, fax +39-0432-558222
>>>>
>>>> _______________________________________________
>>>> Quantum ESPRESSO is supported by MaX (
>>>> www.max-centre.eu/quantum-espresso)
>>>> users mailing list users at lists.quantum-espresso.org
>>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>>
>>> _______________________________________________
>>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso
>>> )
>>> users mailing list users at lists.quantum-espresso.org
>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>
>>
>>
>> --
>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>> Phone +39-0432-558216, fax +39-0432-558222
>>
>> _______________________________________________
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
>> 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/20200716/94ef9afd/attachment.html>

More information about the users mailing list