[Pw_forum] Problem in DOS results

Padmaja Patnaik padmaja_patnaik at yahoo.co.uk
Mon Jul 5 11:54:09 CEST 2010


I am calculating for SiC. Below is the datas you have asked for. 

 &system
    ibrav=1, celldm(1) =16.4824, nat=64, ntyp=2,
    ecutwfc = 32.0,
    occupations='smearing', smearing='methfessel-paxton', degauss=0.02

 /
 &electrons
    mixing_beta = 0.7
 /
ATOMIC_SPECIES
 Si 28.086  Si.pz-vbc.UPF
 C  12.00   C.pz-vbc.UPF
ATOMIC_POSITIONS
 Si    0.000   0.000   0.000
 Si    0.250   0.250   0.000
 Si    0.250  -0.250   0.000
 Si   -0.250   0.250   0.000
 Si   -0.250  -0.250   0.000
 Si    0.250   0.000   0.250
 Si    0.250   0.000  -0.250
 Si   -0.250   0.000   0.250
 Si   -0.250   0.000  -0.250
 Si    0.000   0.250   0.250
 Si    0.000   0.250  -0.250
 Si    0.000  -0.250   0.250
 Si    0.000  -0.250  -0.250
 Si    0.500   0.000   0.000
 Si    0.000   0.500   0.000
 Si    0.000   0.000   0.500
 Si    0.500   0.250   0.250
 Si    0.500   0.250  -0.250
 Si    0.500  -0.250   0.250
 Si    0.500  -0.250  -0.250
 Si    0.250   0.500   0.250
 Si    0.250   0.500  -0.250
 Si   -0.250   0.500   0.250
 Si   -0.250   0.500  -0.250
 Si    0.250   0.250   0.500
 Si    0.250  -0.250   0.500
 Si   -0.250   0.250   0.500
 Si   -0.250  -0.250   0.500
 Si    0.500   0.500   0.000
 Si    0.500   0.000   0.500
 Si    0.000   0.500   0.500
 Si    0.500   0.500   0.500
 C     0.125   0.125   0.125
 C    -0.125  -0.125   0.125
 C     0.375  -0.125   0.125
 C    -0.375   0.125   0.125
 C    -0.125   0.375   0.125
 C     0.125  -0.375   0.125
 C     0.125   0.125  -0.375
 C    -0.125  -0.125  -0.375
 C     0.375   0.375   0.125
 C    -0.375  -0.375   0.125
 C     0.375  -0.125  -0.375
 C    -0.375   0.125  -0.375
 C    -0.125   0.375  -0.375
 C     0.125  -0.375  -0.375
 C     0.375   0.375  -0.375
 C    -0.375  -0.375  -0.375
 C    -0.125   0.125   0.125
 C     0.375   0.125   0.125
 C    -0.125  -0.375   0.125
 C    -0.125   0.125  -0.375
 C     0.375  -0.375   0.125
 C     0.375   0.125  -0.375
 C    -0.125  -0.375  -0.375
 C     0.375  -0.375  -0.375
 C     0.125  -0.125  -0.125
 C    -0.375  -0.125  -0.125
 C     0.125   0.375  -0.125
 C     0.125  -0.125   0.375
 C    -0.375   0.375  -0.125
 C    -0.375  -0.125   0.375
 C     0.125   0.375   0.375
 C    -0.375   0.375   0.375
K_POINTS {automatic}
 2 2 2 0 0 0

I have repeated runs for different smearing, different k_points also.
 Sending you one of the input files.


---------------------------------------------------------------------

Message: 1
Date: Mon, 5 Jul 2010 10:34:54 +0200
From: Gabriele Sclauzero <sclauzer at sissa.it>
Subject: Re: [Pw_forum] Problem in DOS results
To: PWSCF Forum <pw_forum at pwscf.org>
Message-ID: <FC6CD37D-AD91-4AC6-81A9-F89D070980A2 at sissa.it>
Content-Type: text/plain; charset="iso-8859-1"

Dear Padmaja,

   are you sure that your calculation has been converged wrt the k-point sampling? I believe that if you build the supercell correctly and change the k-point sampling accordingly you'll get the same DOS. In principle this should hold even if you didn't converge wrt the size of the k-point mesh, though not completely sure that this is true also for tetrahedra method occupations. 

Please post at least the atomic positions, lattice parameters and the content of the K_POINTS input card, otherwise it will be impossible to help you further...

GS

Il giorno 04/lug/2010, alle ore 21.56, Padmaja Patnaik ha scritto:

> Dear all
> 
> I am studying the properties of a compound semiconductor using super cell calculations. I have generated super cells by expanding the lattice constant 2times, 3 times and 4 times in all the directions. My calculations gives correct results for 2 times and 4 times extended super cells but not for 3 times extended onel. I have made sure that the generated atomic co-ordinates are correct. I have tried fixing all the required parameters like, k points, ecutwfc and nbnd. But still its not working. after doing the scf calculations I do the DOS with occupations=tetrahedra. But the DOS plot for the second case is found to be much different from expected while for all other cases the results match. What can be the probable mistake i am doing? Please suggest.
> 
> Padmaja Patnaik
> Research Scholar
> Dept of Physics
> IIT Bombay
> Mumbai, India
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum


? Gabriele Sclauzero, EPFL SB ITP CSEA
   PH H2 462, Station 3, CH-1015 Lausanne

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20100705/a87dfa32/attachment-0001.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 1753 bytes
Desc: not available
Url : http://www.democritos.it/pipermail/pw_forum/attachments/20100705/a87dfa32/attachment-0001.bin 

------------------------------

Message: 2
Date: Mon, 5 Jul 2010 10:42:34 +0200
From: Gabriele Sclauzero <sclauzer at sissa.it>
Subject: Re: [Pw_forum] Total magnetization and local magnetic moment
To: PWSCF Forum <pw_forum at pwscf.org>
Message-ID: <422DE4FD-F788-4D7E-87B0-9975E478EA02 at sissa.it>
Content-Type: text/plain; charset="iso-8859-1"


So, have you managed to converge the DOS first?

Again, if don't post some specific information about your system, as the input files, part of the output files and a sketch of your procedure, it's almost impossible to localize the problem.

GS


Il giorno 04/lug/2010, alle ore 23.09, Padmaja Patnaik ha scritto:

>  Dear all
> 
> I am getting some surprising results. I am calculating for a semiconductor doped with a transition metal ( using single impurity atom in a super cell). The total magnetization value obtained is more than one and so also the local moment on the impurity atom. But when i performed calculations with relaxation, the total magnetic moment increases but the local moment goes to almost zero value, which is very surprising. I expect the local magnetic moment to be almost of the same value as the total. I have also checked other atoms in the super cell, all are showing zero local moments but still the total magnetic moment is increased after relaxation. Let me mention here I am taking the total magnetic moment from the scf.out file and local moments from pdos.out file.
> 
> 
> Please suggest me how to interpret this result? As per my understanding if a sample is showing ferromagnetic results after adding some impurity then it should be reflected on the local magnetic moment of the impurity atom too. In my results, this is the case before relaxation but not after relaxation, which is quite surprising.
> 
> Regards
> Padmaja Patnaik
> Research Scholar
> Dept of Physics
> IIT Bombay
> Mumbai, India
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum





      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20100705/5f516a5d/attachment.htm 


More information about the Pw_forum mailing list