[QE-users] Wrong position of fermi level in bulk Cr2O3
Yuvam Bhateja
yuvamb16 at gmail.com
Fri Jul 24 09:27:25 CEST 2020
Thank you so much sir for the clarification.
Actually I am interested in performing adsorption on Cr2O3 slab, so before
forming a surface I wanted to check the bulk unit cell, so that I am
assured that I will be choosing the appropriate bulk structure for slab
formation.
And for adsorption part, change in charge density and band gap will be a
prominent factor.
I used Normcon PP with PBE, so I get a very accurate band gap but wrong
position of Fermi level.
Later by using USPP with PW91, Fermi level was at HUMO but the band gap was
underestimated and I even observed some additional states which if I'm not
wrong are called ghost states.
So, I am quite confused which method is efficient for me.
I will highly appreciate if you could advice me on this.
Regards
Yuvam Bhateja
On Fri, 24 Jul 2020, 12:43 pm Pietro Delugas, <pdelugas at sissa.it> wrote:
> I am not sure I have understood what the problem is.
>
>
>
> If all you need is the plot of the density of states and want to set in
> the plot the position of the Fermi level, my suggestion, this being an
> insulator, is to avoid to indicate a value for the Fermi level, just put a
> VBT ( valence band top) in correspondence of the HOMO and a CBM (
> conduction band minimum ) in correspondence of the LUMO.
>
>
>
> Usually if you are more interested in the behavior of the holes ( p-type)
> condition you shift the plot in order that VBT is at 0, or if you are more
> interested on electrons (n-type) in conduction band you shift the plot in
> order to have the CBM at 0.
>
>
>
> If you are interested in both and want to indicate the Fermi Level you
> can only indicate the intrinsic one which is in the middle of the gap
> between VBT and CBM.
>
> HTH
>
> regards
>
> Pietro
>
>
>
>
>
> Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for
> Windows 10
>
>
>
> *From: *Yuvam Bhateja <yuvamb16 at gmail.com>
> *Sent: *Thursday, July 23, 2020 6:01 PM
> *To: *Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
> *Cc: *Pietro Delugas <pdelugas at sissa.it>
> *Subject: *Re: [QE-users] Wrong position of fermi level in bulk Cr2O3
>
>
>
> Thank you for the reply.
>
>
>
> I am not sure how to solve this problem.
>
> I tried using different pseudopotentials with diffrent XC. Even tried
> increasing U value.
>
>
>
> In most of the papers, pure bulk Cr2O3 is used.
>
>
>
> Can you please advise me?
>
>
>
> Thank you in advance.
>
>
>
> Regards
>
> Yuvam
>
>
>
> On Thu, 23 Jul 2020, 6:08 pm Nicola Marzari, <nicola.marzari at epfl.ch>
> wrote:
>
>
>
> Indeed! Although I would feel better about using as Fermi energy the
> zero temperature limit of the chemical potential, that would be
> (homo+lumo)/2? (well, the band edges).
>
> nicola
>
> PS: of course the lumo is wrong even in exact DFT, and the homo is
> correct only in exact DFT, and wrong in approximate DFT.
>
> On 23/07/2020 14:35, Pietro Delugas wrote:
> > Dear Yuvam
> >
> > When one uses smeared occupations in systems with a large gap (
> > significantly larger than the smearing) the final value of the Fermi
> > level returned from the calculation may be any value within the band gap
> > sufficiently far from the band edges (HOMO and LUMO). All the values
> > in this range yield the same values for the occupations and thus of the
> > ground state density. If you wish you can decide to follow the
> > convention of the other papers and refer to the HOMO as the Fermi level.
> >
> > In fact in insulators the thermodynamic Fermi level depends on extrinsic
> > conditions such as doping or defect concentrations and on the
> > temperature. All things which are obviously not taken into account in
> > your calculation.
> >
> > Kind regards
> >
> > Pietro
> >
> > Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for
> > Windows 10
> >
> > *From: *Yuvam Bhateja <mailto:yuvamb16 at gmail.com>
> > *Sent: *Thursday, July 23, 2020 12:44 PM
> > *To: *Quantum ESPRESSO users Forum <mailto:
> users at lists.quantum-espresso.org>
> > *Subject: *[QE-users] Wrong position of fermi level in bulk Cr2O3
> >
> > Hello dear experts,
> >
> > Hope you are doing well.
> >
> > I was trying to calculate the DOS in hexagonal Cr2O3, the DOS is coming
> > very well with a bandgap of 3.09, but the problem is the Fermi level is
> > at the edge of LUMO whereas I've observed that it should be at HUMO edge
> > in few already published papers.
> >
> > I used PBE NC SR pseudopotential and also optimized with the same. I
> > used gaussian smearing with a width of 0.001 in VC-relax and tetrahedra
> > in SCF calculation.
> >
> > Here is my input file -
> >
> > &CONTROL
> > calculation = "scf"
> > pseudo_dir = "./"
> > outdir = "./outdir"
> > /
> >
> > &SYSTEM
> > a = 4.6166825e+00
> > c = 1.2457778e+01
> > ecutrho = 4.5000e+02
> > ecutwfc = 4.5000e+01
> > ibrav = 4
> > nat = 30
> > nspin = 2
> > ntyp = 3
> > nbnd = 276
> > starting_magnetization(1) = 0.5
> > starting_magnetization(2) = -0.5
> > starting_magnetization(3) = 0.0
> > tot_magnetization = 0.0
> > lda_plus_u = .true.
> > Hubbard_U(1) = 4.7
> > Hubbard_U(2) = 4.7
> > Hubbard_J0(1) = 1
> > Hubbard_J0(2) = 1
> > /
> >
> > &ELECTRONS
> > conv_thr = 1.00000e-11
> > electron_maxstep = 2000
> > mixing_beta = 7.00000e-01
> > startingpot = "atomic"
> > startingwfc = "atomic+random"
> > /
> >
> >
> > K_POINTS {automatic}
> > 12 12 12 0 0 0
> >
> > ATOMIC_SPECIES
> > Cr1 51.99610 Cr.upf
> > Cr2 51.99610 Cr.upf
> > O 15.99940 O.upf
> >
> > ATOMIC_POSITIONS (angstrom)
> > Cr2 -0.0000027528 2.6654550870 3.9638155751
> > Cr1 2.3083528721 1.3327252495 2.2651020880
> > Cr1 -0.0000027528 2.6654550870 0.1887989665
> > Cr2 0.0000000000 0.0000000000 1.8875092202
> > Cr2 0.0000000000 0.0000000000 8.1164268017
> > Cr1 -0.0000027528 2.6654550870 6.4177165986
> > Cr1 0.0000000000 -0.0000000000 4.3414084067
> > Cr2 2.3083528721 1.3327252495 6.0401185402
> > Cr2 2.3083528721 1.3327252495 12.2690363475
> > Cr1 0.0000000000 0.0000000000 10.5703258606
> > Cr1 2.3083528721 1.3327252495 8.4940197683
> > Cr2 -0.0000027528 2.6654550870 10.1927331413
> > O 1.4445126756 0.0000000398 3.1144588851
> > O 3.0306076646 0.0817435268 1.0381518347
> > O 0.8638388763 1.3327251470 1.0381518347
> > O 3.0306056520 2.5837116627 1.0381518347
> > O -0.7222568310 1.2509851988 3.1144588851
> > O 1.5860942747 2.7471960155 3.1144588851
> > O 3.7528623252 1.3327254382 7.2670694393
> > O 0.7222547656 1.4144695631 5.1907658203
> > O -1.4445122603 2.6654550733 5.1907658203
> > O 0.7222574947 3.9164369542 5.1907658203
> > O 1.5860961469 2.5837115564 7.2670694393
> > O 1.5860937210 0.0817433419 7.2670694393
> > O 1.4445122976 2.6654551364 11.4196834856
> > O -1.5860942579 2.7471959172 9.3433763434
> > O 0.8638382678 3.9981804108 9.3433763434
> > O 0.7222569078 1.2509852624 9.3433763434
> > O -0.7222575680 3.9164369549 11.4196834856
> > O -0.7222547296 1.4144694992 11.4196834856
> >
> > I later tried with pw91 USPP, it fixed the fermi level position (it's
> > now at HUMO edge) but the bad gap is now very underestimated to 2.3 eV.
> >
> > I tried paw but it wasn't working with me.
> >
> > Can someone please help?
> >
> > Regards
> >
> > Yuvam Bhateja
> >
> > IIEST Shibpur
> >
> > India
> >
> >
> > _______________________________________________
> > 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
> >
>
>
> --
> ----------------------------------------------------------------------
> Prof Nicola Marzari, Chair of Theory and Simulation of Materials, EPFL
> Director, National Centre for Competence in Research NCCR MARVEL, EPFL
> http://theossrv1.epfl.ch/Main/Contact http://nccr-marvel.ch/en/project
> _______________________________________________
> 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/20200724/61052a4f/attachment.html>
More information about the users
mailing list