[Pw_forum] vc-relax convergence

Giuseppe Mattioli giuseppe.mattioli at ism.cnr.it
Mon Aug 7 13:18:50 CEST 2017


Dear Manu

> Is EXX-ACE implemented in QE 6.1v by default or do I have to compile it
> separately?.

Honestly, I would not give it for granted! Put the ACE flag in the  
make.inc file, it should not harm. :-)

> If I am not using HSE , do I have to use LDA level theory for TiO2 with
> defects?.

Note that there are tons of papers on TiO2 defects. If you do not want  
to use EXX, then you can try with DFT+U(Ti,O). I've used it here

J. Phys. Chem. C 2010, 114, 21694
J. Phys. Chem. C 2014, 118, 29928

with apparently good results. I would suggest these values or  
something around:

     lda_plus_u=.true.,
     Hubbard_U(1)=3.2, <----Ti
     Hubbard_U(2)=2.0, <----O

HTH
Giuseppe

Quoting Manu Hegde <mhegde at uwaterloo.ca>:

> Hi Stefano and Giuseppe,
>
> Thanks for your feedback.
>
> I just wanted to try 48 atom supercell initially and see how it goes (to
> save computation time as well). I will try 108 atom supercell definitely.
>
> Is EXX-ACE implemented in QE 6.1v by default or do I have to compile it
> separately?. I will try to run it (if our busy HPC facility permits!).
>
> If I am not using HSE , do I have to use LDA level theory for TiO2 with
> defects?.
>
> Thanks
> Manu
> (University of Waterloo)
>
> On Sun, Aug 6, 2017 at 6:08 PM, Giuseppe Mattioli <
> giuseppe.mattioli at ism.cnr.it> wrote:
>
>>
>> On the top of Stefano's answer, I would add that in a 48 atom
>> supercell the interaction between the relaxation sphere of an oxygen
>> vacancy and its periodic images along the 100 or 010 directions is too
>> strong and can lead to an improper description of the electronic
>> properties of the defect that can not be considered as an "isolated
>> defect" anymore. I suggest that you use at least a 3x3x1 108-atom
>> supercell. With the new EXX-ACE implementation you may probably afford
>> an HSE calculation, otherwise remember also that the small-polaron
>> character of excess electrons in TiO2 such as those induced by the O
>> vacancy is not at all well described at the GGA level of theory.
>> HTH
>> Giuseppe
>>
>> stefano de gironcoli <degironc at sissa.it> ha scritto:
>>
>> > yes
>> > start from the perfect relaxed geometry (as obtained from vc-relax)
>> > remove an atom and relax the internal structure.
>> > at the end also perform a vc-relaxation (from the fixed-cell relaxed
>> > geometry)
>> > stefano
>> >
>> > On 06/08/2017 17:33, Manu Hegde wrote:
>> >> Hello Stephano,
>> >> Thanks for your reply. Just to double check what you meant before
>> >> starting the calculations, you mean take the equilibrium values
>> >> from the vc-relax calculation (perfect cell,  only co-ordinates)
>> >> and do the 'relax' calculation with defect?.
>> >> Manu
>> >> (University of Waterloo).
>> >>
>> >>
>> >> On Sun, Aug 6, 2017 at 11:15 AM, stefano de gironcoli
>> >> <degironc at sissa.it <mailto:degironc at sissa.it>> wrote:
>> >>
>> >>    does the energy (enthalpy) decreases as the vc-relaxation
>> >>    progresses ?
>> >>    The local structure around the defect may be quite different. It
>> >>    may just take long.
>> >>
>> >>    I would start by performing a fixed cell relaxation with the cell
>> >>    parameters fixed at the undefected equilibrium values. And only
>> >>    after that I would start the vc-relax calculation from the so
>> >>    relaxed configuration to evaluate the vacancy formation volume.
>> >>    The local arrangement around the defect should be already more or
>> >>    less ok (if the supercell is large enough).
>> >>
>> >>    HTH
>> >>    stefano
>> >>
>> >>
>> >>
>> >>    On 06/08/2017 16:40, Manu Hegde wrote:
>> >>>    Hello,
>> >>>
>> >>>    I am trying to optimize anatase TiO2 supercell with 48 atoms. The
>> >>>    supercell converges in 15 BFGS cycles for a perfect cell. But
>> >>>    when I introduce defect (by removing one of the oxygen atom), the
>> >>>    vc-relax is not t all converging. I am using the same parameters
>> >>>    in both the cases. Can anyone give some feedback on this.?.  Here
>> >>>    is my input,
>> >>>
>> >>>
>> >>>    &control
>> >>>     calculation = 'vc-relax' ,
>> >>>    restart_mode = 'from_scratch' ,
>> >>>    outdir = '/work/mhegde/tmp_tiodef/',
>> >>>    pseudo_dir = '/home/mhegde/tio_project/def_tio/',
>> >>>    prefix = 'tioan',
>> >>>     verbosity = 'high',
>> >>>
>> >>>     /
>> >>>     &system
>> >>>     ibrav = 0,
>> >>>     nat = 47,
>> >>>      ntyp = 2,
>> >>>     ecutwfc =40,
>> >>>     ecutrho=400,
>> >>>     occupations='smearing',
>> >>>    degauss=0.003
>> >>>
>> >>>     /
>> >>>     &electrons
>> >>>    conv_thr = 1.D-6,
>> >>>    mixing_beta=0.1
>> >>>    /
>> >>>    &ions
>> >>>    ion_dynamics='bfgs'
>> >>>
>> >>>    /
>> >>>    &cell
>> >>>    cell_dynamics='bfgs'
>> >>>
>> >>>    /
>> >>>    ATOMIC_SPECIES
>> >>>       Ti  47.867000  Ti.pw91-nsp-van.UPF
>> >>>       O   15.99900 O.pw91-van_ak.UPF
>> >>>
>> >>>    ATOMIC_POSITIONS angstrom
>> >>>    Ti 0.00000000       2.84182500 1.20970000
>> >>>    Ti 1.89455000       2.84182500 3.62910000
>> >>>    O  0.00000000       0.94727500 1.99939216
>> >>>    O  0.00000000       2.84182500 4.41879216
>> >>>    O  1.89455000       0.94727500 2.83940784
>> >>>    Ti  -1.89455000       0.94727500  -3.62910000
>> >>>    Ti 0.00000000       0.94727500  -1.20970000
>> >>>    O -1.89455000      -0.94727500  -2.83940784
>> >>>    O -1.89455000       0.94727500  -0.42000784
>> >>>    O  0.00000000      -0.94727500  -1.99939216
>> >>>    O  0.00000000       0.94727500  -4.41879216
>> >>>    Ti 0.00000000      -0.94727500 1.20970000
>> >>>    Ti 1.89455000      -0.94727500 3.62910000
>> >>>    O  0.00000000      -2.84182500 1.99939216
>> >>>    O  0.00000000      -0.94727500 4.41879216
>> >>>    O  1.89455000      -2.84182500 2.83940784
>> >>>    O  1.89455000      -0.94727500 0.42000784
>> >>>    Ti  -1.89455000      -2.84182500  -3.62910000
>> >>>    Ti 0.00000000      -2.84182500  -1.20970000
>> >>>    O -1.89455000       2.84182500  -2.83940784
>> >>>    O -1.89455000      -2.84182500  -0.42000784
>> >>>    O  0.00000000       2.84182500  -1.99939216
>> >>>    O  0.00000000      -2.84182500  -4.41879216
>> >>>    Ti  -3.78910000       2.84182500 1.20970000
>> >>>    Ti  -1.89455000       2.84182500 3.62910000
>> >>>    O -3.78910000       0.94727500 1.99939216
>> >>>    O -3.78910000       2.84182500 4.41879216
>> >>>    O -1.89455000       0.94727500 2.83940784
>> >>>    O -1.89455000       2.84182500 0.42000784
>> >>>    Ti 1.89455000       0.94727500  -3.62910000
>> >>>    Ti  -3.78910000       0.94727500  -1.20970000
>> >>>    O  1.89455000      -0.94727500  -2.83940784
>> >>>    O  1.89455000       0.94727500  -0.42000784
>> >>>    O -3.78910000      -0.94727500  -1.99939216
>> >>>    O -3.78910000       0.94727500  -4.41879216
>> >>>    Ti  -3.78910000      -0.94727500 1.20970000
>> >>>    Ti  -1.89455000      -0.94727500 3.62910000
>> >>>    O -3.78910000      -2.84182500 1.99939216
>> >>>    O -3.78910000      -0.94727500 4.41879216
>> >>>    O -1.89455000      -2.84182500 2.83940784
>> >>>    O -1.89455000      -0.94727500 0.42000784
>> >>>    Ti 1.89455000      -2.84182500  -3.62910000
>> >>>    Ti  -3.78910000      -2.84182500  -1.20970000
>> >>>    O  1.89455000       2.84182500  -2.83940784
>> >>>    O  1.89455000      -2.84182500  -0.42000784
>> >>>    O -3.78910000       2.84182500  -1.99939216
>> >>>    O -3.78910000      -2.84182500  -4.41879216
>> >>>
>> >>>    K_POINTS {automatic}
>> >>>
>> >>>     2 2 2   0 0 0
>> >>>
>> >>>    CELL_PARAMETERS angstrom
>> >>>
>> >>>     7.57820000   0.00000000       0.00000000
>> >>>     0.00000000   7.57820000       0.00000000
>> >>>     0.00000000   0.00000000       9.67760000
>> >>>
>> >>>    Thanks,
>> >>>    Manu
>> >>>    (University of Waterloo)
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>    _______________________________________________
>> >>>    Pw_forum mailing list
>> >>>    Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >>>    http://pwscf.org/mailman/listinfo/pw_forum
>> >>>    <http://pwscf.org/mailman/listinfo/pw_forum>
>> >>
>> >>
>> >>
>> >>    _______________________________________________
>> >>    Pw_forum mailing list
>> >>    Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >>    http://pwscf.org/mailman/listinfo/pw_forum
>> >>    <http://pwscf.org/mailman/listinfo/pw_forum>
>> >>
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> Pw_forum mailing list
>> >> Pw_forum at pwscf.org
>> >> http://pwscf.org/mailman/listinfo/pw_forum
>>
>>
>> --
>> ********************************************************
>> - Article premier - Les hommes naissent et demeurent
>> libres et égaux en droits. Les distinctions sociales
>> ne peuvent être fondées que sur l'utilité commune
>> - Article 2 - Le but de toute association politique
>> est la conservation des droits naturels et
>> imprescriptibles de l'homme. Ces droits sont la liberté,
>> la propriété, la sûreté et la résistance à l'oppression.
>> ********************************************************
>>
>>     Giuseppe Mattioli
>>     CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
>>     v. Salaria Km 29,300 - C.P. 10
>>     I 00015 - Monterotondo Stazione (RM)
>>     Tel + 39 06 90672836 - Fax +39 06 90672316
>>     E-mail: <giuseppe.mattioli at ism.cnr.it>
>>     http://www.ism.cnr.it/en/staff/giuseppe-mattioli/
>>     ResearcherID: F-6308-2012
>>
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>


-- 
********************************************************
- Article premier - Les hommes naissent et demeurent
libres et égaux en droits. Les distinctions sociales
ne peuvent être fondées que sur l'utilité commune
- Article 2 - Le but de toute association politique
est la conservation des droits naturels et
imprescriptibles de l'homme. Ces droits sont la liberté,
la propriété, la sûreté et la résistance à l'oppression.
********************************************************

    Giuseppe Mattioli
    CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
    v. Salaria Km 29,300 - C.P. 10
    I 00015 - Monterotondo Stazione (RM)
    Tel + 39 06 90672836 - Fax +39 06 90672316
    E-mail: <giuseppe.mattioli at ism.cnr.it>
    http://www.ism.cnr.it/en/staff/giuseppe-mattioli/
    ResearcherID: F-6308-2012





More information about the users mailing list