The ABINIT Users Mailing List ( CLOSED )

["Eric Bousquet" <Eric.Bousquet@ulg.ac.be>]

Dear Amini,

In your input file you specify nband=20, however for PbTiO3 with semi-core
electrons the number of occupied band should be equal to 22.
It was shown in the early study of ferroelectric ABO3 compounds that to
describe correctly their properties, these semi-core electrons must be
included in the pseudopotentials.

So, you can check if all of your pseudopotentials contain the valence AND
the semi-core electrons (3s, 3p, 3d and 4s for Ti; 5s, 5p and 6s for Pb;
2s and 2p for O).
If all the semi-core electrons are included, the number of occupied bands
should be equal to 22 for PbTiO3.

I think that your problem comes from these semi-core states.

Regards,

Eric

> Dear matthieu,
>  
> Here is my final results for frequencies of PbTiO3 in cubic phase(with no
> imaginary frequency):
>  
>   Phonon wavevector (reduced coordinates) :  0.00000  0.00000  0.00000
>  Phonon energies in Hartree :
>    1.375306E-05  1.375306E-05  1.611297E-05  4.419942E-04  4.419942E-04
>    4.421844E-04  9.409520E-04  9.409520E-04  9.409555E-04  9.960169E-04
>    9.960169E-04  9.960169E-04  2.658582E-03  2.658628E-03  2.658628E-03
>  Phonon frequencies in cm-1    :
> -  3.018449E+00  3.018449E+00  3.536389E+00  9.700651E+01  9.700651E+01
> -  9.704826E+01  2.065151E+02  2.065151E+02  2.065159E+02  2.186005E+02
> -  2.186005E+02  2.186005E+02  5.834912E+02  5.835014E+02  5.835014E+02
>
>  
> my input file (rf1):
>  
> ndtset  3
> kptopt1   1               
> tolvrs1   1.0d-18      
> iscf1   5            
> #Response Function calculation : d/dk
> rfelfd2   2             
> rfdir2   1 1 1                                
> rfdir 1 1 1
> nqpt2   1
> qpt2   0.0 0.0 0.0   
> getwfk2   -1            
> kptopt2   2         
> iscf2  -3             
> tolwfr2   1.0d-22      
> #Response Function calculation : electric field perturbation and phonons
> rfphon3   1        
> rfatpol3   1 5      
> rfelfd3   3      
>  rfdir3   1 1 1                     
>  nqpt3   1
> qpt3   0.0 0.0 0.0  
> getwfk3   -2           
> getddk3   -1            
> kptopt3   2            
> tolvrs3   1.0d-8
> iscf3   5            
> #######################################################################
> #Common input variables
>
> #Definition of the unit cell
> acell    7.7303653285E+00  7.7303653285E+00  7.7303653285E+00 Bohr    
> rprim  1.0  0.0  0.0   
>        0.0  1.0  0.0   
>        0.0  0.0  1.0   
> #Definition of the atom types
> ntypat 3              
> znucl 82 8 22                             
> #Definition of the atoms
> natom 5               
> typat  1 3 2 2 2      
> xred   0.0    0.0    0.0
>        0.5    0.5    0.5
>        0.5    0.5    0.0
>        0.5    0.0    0.5
>        0.0    0.5    0.5
>
> nband  20               
> #Exchange-correlation functional
> ixc 1             # LDA Teter Pade parametrization
>
> #Definition of the planewave basis set
> ecut    60.0        
> #Definition of the k-point grid
> kptrlatt  6  0  0     
>           0  6  0       
>           0  0  6          
>
> #Definition of the SCF procedure
> nstep 100         
> diemac 6.0     
>  
> Thanks,          
>                       
>
>
> --- On Mon, 12/8/08, matthieu verstraete
> <matthieu.jean.verstraete@gmail.com> wrote:
>
> From: matthieu verstraete <matthieu.jean.verstraete@gmail.com>
> Subject: Re: [abinit-forum] soft mode phonons
> To: forum@abinit.org
> Date: Monday, December 8, 2008, 2:13 PM
>
>
> Hello
>
> Abinit can certainly calculate the soft modes (find publications by Ghosez
> or Vanderbilt, for some examples). If you don't have any, your calculation
> is probably not converged:
>
> 1) if the electronic temperature is too high (providing you are using one)
> 2) if the lattice is strained the frequencies can be pushed positive
> 3) if you don't have enough kpoints or a large enough ecut.
> 4) if your qpoint grid is not fine enough (or does not contain the qvector
> where the phonon goes soft) then you will miss the negative mode,
> depending on the details of the phonon interpolation.
>
> there may be other reasons as well. Are you certain your compound is
> ferroelectric? Are you just reproducing a tutorial or did you adapt it -
>
> Again, and as usual: if you write to forum you should include all system
> details and follow  Nettiquette!!  
> http://www.abinit.org/community/?textetiquette
>
> If you had done so we could have looked at your input and told you which
> was the problem, instead of trying to guess!
>
> Matthieu
>
>
>
> On Sun, Dec 7, 2008 at 6:43 PM, <mozhganamini@yahoo.com> wrote:
>
> Dear all,
>
> I have done RF1 lesson for one of the ABO3 compounds for getting it's
> phonon
> frequencies and I have gotten some real frequencies. But, I know that this
> compound has soft mode in TO1.
> Can Abinit code calculates these soft modes or I did some wrong thing in
> my
> calculations?
>
> Thanks alot in advance,
> mozhgan amini
>
>
>
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Dr. Matthieu Verstraete
>
> European Theoretical Spectroscopy Facility (ETSF) 
> Dpto. Fisica de Materiales, 
> U. del Pais Vasco,            
> Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
> E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390
>
> Mail : matthieu.jean.verstraete@gmail.com
> http://www-users.york.ac.uk/~mjv500
>
>
>
>


Universite de Liege
Institut de Physique, Bat. B5
Allee du 6 aout, 17
B-4000 Sart Tilman
BELGIUM

Phone : ++(32) (0)4-366.37.22
Fax   : ++(32) (0)4-366.36.29

http://www.ulg.ac.be/phythema