[QE-users] Question regarding mixing of PBE and PBESOL PP

Sun May 5 02:35:10 CEST 2019

Thank you so much for answers and link for PPs!!

________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Holzwarth, Natalie <natalie at wfu.edu>
Sent: Saturday, May 4, 2019 3:00:31 PM
To: Quantum Espresso users Forum
Subject: Re: [QE-users] Question regarding mixing of PBE and PBESOL PP

We are slowly updating our ATOMPAW webpage http://pwpaw.wfu.edu to include LDA and PBESOL PAW datasets that are more reliable when used with Quantum Espresso.    I just added Ca and could add more as needed.       Of course, although similar datasets have been used in the past,   the new ones always need to be tested.

N. A. W. Holzwarth                                       email: natalie at wfu.edu<mailto:natalie at wfu.edu>
Department of Physics                                  web: http://www.wfu.edu/~natalie
Wake Forest University                                 phone: 1-336-758-5510
Winston-Salem, NC 27109 USA                     office: Rm. 300 Olin Physical Lab

On Sat, May 4, 2019 at 12:27 AM Yeon, Jejoon <jyeon at udel.edu<mailto:jyeon at udel.edu>> wrote:

Hello

I ran several DFTs few times before, but I only have limited amount of knowledge and experience in DFTs.

I have Pyrope (Mg3Al2Si3O12) and Grossular (Ca3Al2Si3O12) unit cell, and wish to preform relax and vc-relax. After I get the relaxed atomic structure and cell size, I wish to make an energy curve of Pyrope / Grossular structure w.r.t. compression and expansion of volume, using expanded / compressed unit cells. Purpose of this DFT is to get the database for the parametrization of empirical force field, especially for mechanical properties like stress-strain curve and elastic constants.

Now, I have PBE-PAW(kjpaw) and PBESOL-PAW(kjpaw) PPs for all atom types. I have no problem about Pyrope, I can run it with PBESOL. But Grossular is an issue. For Ca, PBE is only available option from QE PP webpage, there is no PBESOL for Ca yet. Accordingly, if I want to calculate energies of Grossular structure, I have three options as far as I know:

1) input_dft = 'PBE', while mix the PBE (Ca) and PBESOL (Al, Si, O)

2) inout_dft = 'PBESOL', while mix the PBE (Ca) and PBESOL (Al, Si, O)

3) just use PBE PPs for all 4 atom types.

At this point, I'm not sure what would be the good choice for my purpose. I wish to use PBESOL because it is known to have better prediction for solid properties. But I'm not sure if I can mix them with PBE, and if I mix, which input_dft option should I need to choose.

I ran test relaxations for 3 cases, and I found that the final atom coordinates of all 3 cases are very similar, they are mostly the same up to second decimal point numbers. Their final optimized energies are different, but as far as I know I can't compare them, isn't it? I'm not sure on which criteria should I need compare and select the best one from those three options.

Or, if anyone had a similar experience, I welcome any advises or suggestions.

Just in case, following is my example in file.

&CONTROL
  prefix="03_Relax_Grossular_80_800_PBESOL",
  calculation='relax',
  outdir="/scratch/jj/20190501_QEDFT_Grossular_EoS/03_Relax_Grossular_80_800_PBESOL/tmp",
  pseudo_dir="/home/QE_pseudo/",
  restart_mode= 'from_scratch',
  nstep = 1000
/

&SYSTEM
  ibrav=0,
  nat=20,
  ntyp=4,
  ecutwfc=80,
  ecutrho=800,
  occupations='smearing',
  smearing='methfessel-paxton',
  degauss=0.005000,
  input_dft='PBESOL'
/

&ELECTRONS
  diagonalization='cg',
  conv_thr=1d-07,
  mixing_mode='plain',
  mixing_beta=0.100,
/

&IONS
  ion_dynamics = 'bfgs',
/

ATOMIC_SPECIES
  Ca 40.078000 Ca.pbe-spn-kjpaw_psl.1.0.0.UPF
  Al 26.981539 Al.pbesol-n-kjpaw_psl.1.0.0.UPF
  Si 28.085500 Si.pbesol-n-kjpaw_psl.1.0.0.UPF
  O 15.999400 O.pbesol-n-kjpaw_psl.1.0.0.UPF

ATOMIC_POSITIONS {angstrom}
Ca    1.480750    0.000000    2.961500
Ca    2.961500    1.480750    0.000000
Ca    0.000000    2.961500    1.480750
Al    0.000000    0.000000    0.000000
Al    2.961500    2.961500    2.961500
Si    0.000000    2.961500    4.442250
Si    4.442250    0.000000    2.961500
Si    2.961500    4.442250    0.000000
O     0.450148    5.393484    1.791116
O     1.791116    0.450148    5.393484
O     5.393484    1.791116    0.450148
O     3.491016    2.511352    4.752616
O     2.511352    4.752616    3.491016
O     4.752616    3.491016    2.511352
O     5.472852    0.529516    4.131885
O     4.131885    5.472852    0.529516
O     0.529516    4.131885    5.472852
O     2.431984    3.411648    1.170385
O     3.411648    1.170385    2.431984
O     1.170385    2.431984    3.411648

K_POINTS {automatic}
  4 4 4 0 0 0

CELL_PARAMETERS {angstrom}
   5.923000000   0.000000000   0.000000000
   0.000000000   5.923000000   0.000000000
   0.000000000   0.000000000   5.923000000

Thank you

_______________________________________________
Quantum Espresso is supported by MaX (www.max-centre.eu/quantum-espresso<http://www.max-centre.eu/quantum-espresso>)
users mailing list users at lists.quantum-espresso.org<mailto: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/20190505/04644fb4/attachment.html>