[QE-users] Problem with hp.x for f-metals
Abhirup Patra
abhirupp at sas.upenn.edu
Mon Jul 27 17:00:12 CEST 2020
Great! Thanks, Iurii, I will try these suggestions and let you know. By the
way, for a unit cell with 4 atoms, I was able to find the Hubbard U using
the previous input settings I shared with you, the problem merely happens
for a 32 atoms supercell.
The PP is from Opium code, I can try the other PP's as well.
Best,
Abhirup
----------------------------------------------------------------------------------------------
Abhirup Patra
Postdoctoral Research Fellow
Department of Chemistry
University of Pennsylvania
On Mon, Jul 27, 2020 at 4:30 AM Timrov Iurii <iurii.timrov at epfl.ch> wrote:
> Dear Abhirup,
>
>
> > calculation = 'relax'
>
>
> The HP code must be run after a SCF calculation, i.e. calculation = 'scf'.
>
>
> > input_dft = 'PBE'
>
>
> No need to specify this in the input
>
>
> > degauss = 0.05
>
>
> This is too large, try degauss = 0.02
>
>
> > conv_thr = 1.0D-6
>
>
> This is too large, try e.g. conv_thr = 1.0D-10
>
>
> > diago_thr_init = 1e-08
>
>
> I suggest to remove this, and the default will be used
>
>
> > diagonalization = 'cg'
>
>
> I would not use this, try diagonalization = 'david'
>
>
> > ce_srl_gga.upf
>
>
> I do not know this pseudopotential. Is it good? Where did you take it
> from? I suggest to have a look here:
> https://www.materialscloud.org/discover/sssp/table/efficiency
>
>
> Finally, I suggest to try the QE input generator:
> https://www.materialscloud.org/work/tools/qeinputgenerator
>
>
> Greetings,
>
> Iurii
>
>
> --
> Dr. Iurii Timrov
> Postdoctoral Researcher
> STI - IMX - THEOS and NCCR - MARVEL
> Swiss Federal Institute of Technology Lausanne (EPFL)
> CH-1015 Lausanne, Switzerland
> +41 21 69 34 881
> http://people.epfl.ch/265334
> ------------------------------
> *From:* users <users-bounces at lists.quantum-espresso.org> on behalf of
> Abhirup Patra <abhirupp at sas.upenn.edu>
> *Sent:* Saturday, July 25, 2020 7:41:52 AM
> *To:* Quantum ESPRESSO users Forum
> *Subject:* Re: [QE-users] Problem with hp.x for f-metals
>
> Hi Iurii,
>
> Thanks for your prompt reply. Here they are:
> 1.PW
> ------------------------------------------------------
> &CONTROL
> title = 'POSCAR_Ce_alpha_supercell'
> calculation = 'relax'
> restart_mode = 'from_scratch'
> outdir = './'
> pseudo_dir = './'
> prefix = 'POSCAR_Ce_alpha_supercell'
> verbosity = 'high'
> etot_conv_thr = 0.0001
> forc_conv_thr = 0.001
> nstep = 400
> tprnfor = .true.
> /
> &SYSTEM
> ibrav = 0
> nat = 32
> ntyp = 1
> input_dft = 'PBE'
> ecutwfc = 80
> ecutrho = 320
> lda_plus_u = .true.,
> lda_plus_u_kind = 0,
> ! U_projection_type = 'atomic',
> Hubbard_U(1) = 2.00
> occupations = 'smearing',
> smearing = 'gaussian' ,
> degauss = 0.05
> /
> &ELECTRONS
> electron_maxstep = 200
> conv_thr = 1.0D-6
> diago_thr_init = 1e-08
> mixing_mode = 'local-TF'
> mixing_beta = 0.7
> diagonalization = 'cg'
> /
> &IONS
> ion_dynamics = 'bfgs'
> /
>
> ATOMIC_SPECIES
> Ce 140.11600000 ce_srl_gga.upf
> ATOMIC_POSITIONS crystal
> Ce 0.00000000000000 0.00000000000000 0.00000000000000
> Ce 0.00000000000000 0.00000000000000 0.50000000000000
> Ce 0.00000000000000 0.50000000000000 0.00000000000000
> Ce 0.00000000000000 0.50000000000000 0.50000000000000
> Ce 0.50000000000000 0.00000000000000 0.00000000000000
> Ce 0.50000000000000 0.00000000000000 0.50000000000000
> Ce 0.50000000000000 0.50000000000000 0.00000000000000
> Ce 0.50000000000000 0.50000000000000 0.50000000000000
> Ce 0.00000000000000 0.25000000000000 0.25000000000000
> Ce 0.00000000000000 0.25000000000000 0.75000000000000
> Ce 0.00000000000000 0.75000000000000 0.25000000000000
> Ce 0.00000000000000 0.75000000000000 0.75000000000000
> Ce 0.50000000000000 0.25000000000000 0.25000000000000
> Ce 0.50000000000000 0.25000000000000 0.75000000000000
> Ce 0.50000000000000 0.75000000000000 0.25000000000000
> Ce 0.50000000000000 0.75000000000000 0.75000000000000
> Ce 0.25000000000000 0.00000000000000 0.25000000000000
> Ce 0.25000000000000 0.00000000000000 0.75000000000000
> Ce 0.25000000000000 0.50000000000000 0.25000000000000
> Ce 0.25000000000000 0.50000000000000 0.75000000000000
> Ce 0.75000000000000 0.00000000000000 0.25000000000000
> Ce 0.75000000000000 0.00000000000000 0.75000000000000
> Ce 0.75000000000000 0.50000000000000 0.25000000000000
> Ce 0.75000000000000 0.50000000000000 0.75000000000000
> Ce 0.25000000000000 0.25000000000000 0.00000000000000
> Ce 0.25000000000000 0.25000000000000 0.50000000000000
> Ce 0.25000000000000 0.75000000000000 0.00000000000000
> Ce 0.25000000000000 0.75000000000000 0.50000000000000
> Ce 0.75000000000000 0.25000000000000 0.00000000000000
> Ce 0.75000000000000 0.25000000000000 0.50000000000000
> Ce 0.75000000000000 0.75000000000000 0.00000000000000
> Ce 0.75000000000000 0.75000000000000 0.50000000000000
>
> K_POINTS automatic
> 2 2 2 0 0 0
>
> CELL_PARAMETERS {bohr}
> 18.23207773008069 0.00000000000000 0.00000000000000
> 0.00000000000000 18.23207773008069 0.00000000000000
> 0.00000000000000 0.00000000000000 18.23207773008069
>
> 2. HP
> ----------------------------------------------------
> &inputhp
> prefix = 'POSCAR_Ce_alpha_supercell',
> outdir = './',
> nq1 = 2, nq2 = 2, nq3 = 2,
> conv_thr_chi = 1.0d-05,
> thresh_init = 1.0d-10
> niter_max = 400
> iverbosity = 2
> perturb_only_atom(1) = .true.
> /
> I hope this will be of help.
>
> Best,
> Abhirup
>
> ----------------------------------------------------------------------------------------------
> Abhirup Patra
> Postdoctoral Research Fellow
> Department of Chemistry
> University of Pennsylvania
>
>
> On Sat, Jul 25, 2020 at 1:25 AM Timrov Iurii <iurii.timrov at epfl.ch> wrote:
>
>> Could you provide the input and output files for PW and HP calculations?
>>
>> Greetings,
>> Iurii
>>
>> --
>> Dr. Iurii Timrov
>> Postdoctoral Researcher
>> STI - IMX - THEOS and NCCR - MARVEL
>> Swiss Federal Institute of Technology Lausanne (EPFL)
>> CH-1015 Lausanne, Switzerland
>> +41 21 69 34 881
>> http://people.epfl.ch/265334
>> ------------------------------
>> *From:* users <users-bounces at lists.quantum-espresso.org> on behalf of
>> Abhirup Patra <abhirupp at sas.upenn.edu>
>> *Sent:* Saturday, July 25, 2020 6:20:09 AM
>> *To:* Quantum ESPRESSO users Forum
>> *Subject:* [QE-users] Problem with hp.x for f-metals
>>
>> Hello,
>>
>> I am trying to compute Hubbard U using hp.x for Cerium, however, it seems
>> that after few iterations the code stopped with the following message:
>>
>> atom # 1 q point # 1 iter # 395
>> Fermi energy shift (Ry) = -0.5530E+01 -0.2375E-07
>>
>> WARNING: The Fermi energy shift is too big!
>> This may happen in two cases:
>> 1. The DOS at the Fermi level is too small:
>> DOS(E_Fermi) = 0.1303E+04
>> This means that most likely the system has a gap,
>> and hence it should NOT be treated as a metal
>> (otherwise numerical instabilities will appear).
>> 2. Numerical instabilities due to too low cutoff
>> for hard pseudopotentials.
>>
>> Stopping...
>>
>> Any help would be appreciated.
>>
>> Best,
>> Abhirup
>>
>>
>> ----------------------------------------------------------------------------------------------
>> Abhirup Patra
>> Postdoctoral Research Fellow
>> Department of Chemistry
>> University of Pennsylvania
>> _______________________________________________
>> 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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200727/c1e6c1b9/attachment.html>
More information about the users
mailing list