[Pw_forum] odd number of electrons and magnetization

FARAH MARSUSI marsusi at aut.ac.ir
Mon Jan 16 12:38:11 CET 2017


 Dear Giovanni,

Many thanks for your quick response. I have attached the out file but, because of its large size, it was not sent to forum.

By experiment 1Bm is expected, at which this should also lead to opening of a gap at Gamma point (about 0.5 eV). This latter also has not 
been achieved by PBE and PBE0. 
However, if impose total 
magnetization of 1  to the system, every things will go well. 
nqx
 variables larger than one, 
at least by me, were not possible. Perhaps, q vectors go outside the BZ.
 Therefore, I used a dense k-mesh to compensate this problem.
The input file is attached, and here are some important parts of out file.



 EXX: now go back to refine exchange calculation

     total cpu time spent up to now is     3191.8 secs

     Self-consistent Calculation

     iteration #  1     ecut=    80.00 Ry     beta=0.70
     Davidson diagonalization with overlap
     ethr =  2.01E-09,  avg # of iterations =  8.7

     total cpu time spent up to now is    39050.3 secs

     total energy              =    -252.24700782 Ry
     Harris-Foulkes estimate   =    -268.54994708 Ry
     estimated scf accuracy    <       0.00581274 Ry

     total magnetization       =     0.13 Bohr mag/cell
     absolute magnetization    =     0.20 Bohr mag/cell

and for Fock part of calculations:
iteration #  1     ecut=    80.00 Ry     beta=0.70
     Davidson diagonalization with overlap
     ethr =  1.19E-10,  avg # of iterations =  4.6

     total cpu time spent up to now is   250240.0 secs

     total energy              =    -252.26244594 Ry
     Harris-Foulkes estimate   =    -268.59859464 Ry
     estimated scf accuracy    <       0.00000103 Ry

     By experiment 1Bm is expected, on which this will also
 result in opening of a gap in Gamma point. This latter also has not 
been achieved by PBE and PBE0. However, if impose the total 
magnetization to 1, every things go well. 
The end of the attached 
out file is crashed by the sudden lost of power, which is not important 
and enough data have been obtained. 
nqx variables larger than one, 
at least by me, were not possible. Perhaps, q vectors go outside the BZ.
 Therefor, I used a dense k-mesh to compensate this problem.

Here is some important parts of out files:

.
.
.
EXX: setup a grid of 21 q-points centered on each k-point
     (k+q)-points:
   0.0000000   0.0000000   0.0000000         1    1
   0.0000000   0.1443376   0.0000000         2    1
   0.0000000   0.2886751   0.0000000         3    1
   0.0000000   0.4330127   0.0000000         4    1
   0.0000000  -0.5773503   0.0000000         5    1
   0.1250000   0.2165064   0.0000000         6    1
   0.1250000   0.3608439   0.0000000         7    1
   0.1250000   0.5051815   0.0000000         8    1
   0.2500000   0.4330127   0.0000000         9    1
   0.2500000   0.5773503   0.0000000        10    1
  -0.1250000   0.0721688   0.0000000        11    1
  -0.2500000   0.1443376   0.0000000        12    1
  -0.3750000   0.2165064   0.0000000        13    1
   0.5000000  -0.2886751   0.0000000        14    1
  -0.1250000   0.2165064   0.0000000        15    1
  -0.1250000   0.3608439   0.0000000        16    1
  -0.2500000   0.2886751   0.0000000        17    1
  -0.1250000   0.5051815   0.0000000        18    1
  -0.3750000   0.3608439   0.0000000        19    1
  -0.2500000   0.4330127   0.0000000        20    1
  -0.3750000   0.5051815   0.0000000        21    1

PseudoPot. # 1 for  C read from file:
     /home/marsusi/local/qe-6.0/work/graphene/hybrid/C.pbe-hgh.UPF
     MD5 check sum: 0e67d267521f6aebc6a2c9accb749a71
     Pseudo is Norm-conserving, Zval =  4.0
     Generated in analytical, separable form
     Using radial grid of 1073 points,  1 beta functions with: 
                l(1) =   0

     PseudoPot. # 2 for  F read from file:
     /home/marsusi/local/qe-6.0/work/graphene/hybrid/F.pbe-hgh.UPF
     MD5 check sum: 7c8d7aa99b986d59a5bb8f3e2f4f4241
     Pseudo is Norm-conserving, Zval =  7.0
     Generated in analytical, separable form
     Using radial grid of 1105 points,  1 beta functions with: 
                l(1) =   0

     atomic species   valence    mass     pseudopotential
        C              4.00    12.01100      C( 1.00)
        F              7.00    18.99000      F( 1.00)

     Starting magnetic structure 
     atomic species   magnetization
        C            0.000
        F            1.000

      2 Sym. Ops. (no inversion) found


total magnetization       =     0.19 Bohr mag/cell
     absolute magnetization    =     0.23 Bohr mag/cell

     iteration #  2     ecut=    80.00 Ry     beta=0.70
     Davidson diagonalization with overlap
     ethr =  1.31E-09,  avg # of iterations =  3.9

     total cpu time spent up to now is   257285.4 secs

     total energy              =    -252.26244618 Ry
     Harris-Foulkes estimate   =    -268.59859485 Ry
     estimated scf accuracy    <       0.00000174 Ry

     total magnetization       =     0.19 Bohr mag/cell
     absolute magnetization    =     0.23 Bohr mag/cell

     iteration #  3     ecut=    80.00 Ry     beta=0.70
     Davidson diagonalization with overlap
     ethr =  1.31E-09,  avg # of iterations =  3.0

     Magnetic moment per site:
     atom:    1    charge:    1.6566    magn:    0.0057    constr:    0.0000
     atom:    2    charge:    1.6538    magn:    0.0061    constr:    0.0000
     atom:    3    charge:    1.6487    magn:    0.0100    constr:    0.0000
     atom:    4    charge:    1.6581    magn:    0.0004    constr:    0.0000
     atom:    5    charge:    1.6565    magn:    0.0005    constr:    0.0000
     atom:    6    charge:    1.6496    magn:   -0.0047    constr:    0.0000
     atom:    7    charge:    1.6487    magn:    0.0100    constr:    0.0000
     atom:    8    charge:    1.6523    magn:    0.0091    constr:    0.0000
     atom:    9    charge:    1.6471    magn:    0.0078    constr:    0.0000
     atom:   10    charge:    1.6564    magn:    0.0008    constr:    0.0000
     atom:   11    charge:    1.5643    magn:   -0.0014    constr:    0.0000
     atom:   12    charge:    1.6565    magn:    0.0005    constr:    0.0000
     atom:   13    charge:    1.6538    magn:    0.0061    constr:    0.0000
     atom:   14    charge:    1.6516    magn:    0.0066    constr:    0.0000
     atom:   15    charge:    1.6523    magn:    0.0091    constr:    0.0000
     atom:   16    charge:    1.6603    magn:   -0.0035    constr:    0.0000
     atom:   17    charge:    1.6564    magn:    0.0008    constr:    0.0000
     atom:   18    charge:    1.6581    magn:    0.0004    constr:    0.0000
     atom:   19    charge:    5.2516    magn:    0.0089    constr:    0.0000

          k = 0.0000 0.0000 0.0000 ( 77531 PWs)   bands (ev):

   -31.0291 -25.1365 -22.4376 -22.4359 -22.4292 -22.3251 -22.3225 -22.1038
   -17.7312 -17.6872 -17.6362 -17.6326 -15.2699 -15.2695 -14.6553 -14.4429
   -14.4422 -14.3799 -14.3366 -14.3349 -12.0876 -10.9781 -10.6455 -10.6420
   -10.3364 -10.2964 -10.2963 -10.2244  -8.8573  -8.8543  -8.4620  -8.3201
    -8.3196  -7.8682  -7.8673  -7.1793  -7.1791  -6.2411  -1.8047  -1.7177
    -1.7168  -0.3094   1.2814

  total cpu time spent up to now is   263237.4 secs

     End of self-consistent calculation


Best regards,
F. Marsusi,

Assistant Professor,
Department of Physics,
Amirkabir University of Technology.​
On Mon, 01/16/2017 01:41 PM, Giovanni Cantele <giovanni.cantele at spin.cnr.it> wrote:
> Maybe it is worth providing: i) output file, to have a look to the results ; ii) full input file, to understand whether there is some input-related issue; iii) what are the experimental results you expect to reproduce with QE.
> Based on the only information included in your message, it can be difficult to provide an answer. Are you sure about the choice of the nqx variables? What is the value of the magnetisation at PBE level? I would firstinvestigate the latter, before moving to the PBE0 calculation.
> Giovanni
> 
> > On 16 Jan 2017, at 10:33, FARAH MARSUSI <marsusi at aut.ac.ir> wrote:
> Dear QE users,
> 
> The total magnetization of the system consist of one fluorine adatom on the graphene sheet in a 3 by 3 supercell obtained by both PBE and PBE0  is nearly zero. No fix magnetization imposed to system, in spite of the odd number of electrons, and let it to be determined during the self-consistent cycles. 
> This result is inconsistent with experiments. 
> 
>       ibrav = 4,
>                    celldm(1) = 13.999489019,
>                    celldm(3) = 2.7,
>                          nat = 19,
>                         ntyp = 2,
>                      ecutwfc = 80 ,
>                         nbnd = 43,
>                    input_dft = 'PBE0' ,
>                  occupations = 'smearing' ,
>                      degauss = 0.02 ,
>                     smearing = 'gaussian' ,
>                        nspin = 2 ,
>    starting_magnetization(2) = 1.0,
>             exxdiv_treatment = 'gygi-baldereschi' ,
>        x_gamma_extrapolation = .false. ,
>                         nqx1 = 1 ,
>                         nqx2 = 1 ,
>                         nqx3 = 1 ,
> 
> 
> 
> Best regards,
> F. Marsusi,
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum

-- 

Giovanni Cantele, PhD
CNR-SPIN
c/o Dipartimento di Fisica
Universita' di Napoli "Federico II"
Complesso Universitario M. S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy
e-mail: giovanni.cantele at spin.cnr.it
Phone: +39 081 676910
Skype contact: giocan74

ResearcherID: http://www.researcherid.com/rid/A-1951-2009
Web page: http://people.na.infn.it/~cantele




-- 
This email was Anti Virus checked by Astaro Security Gateway. http://www.sophos.com


        _______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170116/e73e6b81/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: grap.in
Type: application/octet-stream
Size: 2497 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170116/e73e6b81/attachment.obj>


More information about the users mailing list