[Pw_forum] negative phonon frequency in GaN bulk

mohnish pandey mohnish.iitk at gmail.com
Sat Apr 10 15:47:05 CEST 2010


Dear PWSCF users,
                                 I am trying to get phonon dispersion
relation of Gallium Nitride. First I have done the vc-relax calculation for
GaN to get equilibrium lattice parameter. Then using that equilibrium
lattice parameter I have done 'scf' calculation after which I have to do
phonon calculation, but the problem is after doing the 'scf' calculation
there is some pressure (greater than the threshold pressure for vc-relax)
calculation. My input, output f for *vc-relax* and input, output for
*'scf'*calculation are
*respectively*

*INPUT FOR VC-RELAX*

&control
    calculation = 'vc-relax',
    restart_mode='restart',
    nstep=200
    outdir='/home/rajpala/Desktop/gan_wur',
    pseudo_dir='/home/rajpala/Desktop/gan_wur'
    prefix='gan',
    tstress = .true.,
    tprnfor = .true.,
  /
  &system
    ibrav=  4, a = 3.3, c = 5.3,nat=  4, ntyp= 2,
    ecutwfc =
30,ecutrho=240,occupations='smearing',degauss=0.01,smearing='gaussian'
  /
  &electrons
   diagonalization='david'
   mixing_mode = 'plain'
   mixing_beta = 0.7
   conv_thr = 1.0d-7
 /
&IONS
  ion_dynamics='bfgs'
  trust_radius_min=1.D-7
/
 &CELL
  cell_dynamics='bfgs'
 /
ATOMIC_SPECIES
  Ga 69.723 Ga.pbe-nsp-van.UPF
  N 14.0067 N.pbe-rrkjus.UPF
 ATOMIC_POSITIONS (crystal)
  Ga 0.00 0.00 0.00
  Ga 1/3 2/3 1/2
  N  0.0 0.0 69/200
  N  1/3 2/3 169/200
 K_POINTS (automatic)
 6 6 4 0 0 0

*SEGMENT OF OUTPUT* *OF VC-RELAX*

!    total energy              =    -395.68385837 Ry
     Harris-Foulkes estimate   =    -395.68385837 Ry
     estimated scf accuracy    <          6.2E-10 Ry

     The total energy is the sum of the following terms:

     one-electron contribution =    -190.48989791 Ry
     hartree contribution      =     134.51657806 Ry
     xc contribution           =    -108.33929602 Ry
     ewald contribution        =    -231.37124250 Ry
     smearing contrib. (-TS)   =      -0.00000000 Ry

     convergence has been achieved in   3 iterations

     Forces acting on atoms (Ry/au):

     atom   1 type  1   force =    -0.00000000    0.00000000    0.00003463
     atom   2 type  1   force =     0.00000000   -0.00000000   -0.00003656
     atom   3 type  2   force =    -0.00000000   -0.00000000    0.00000483
     atom   4 type  2   force =     0.00000000    0.00000000   -0.00000291

     Total force =     0.000051     Total SCF correction =     0.000052
     SCF correction compared to forces is too large, reduce conv_thr


     entering subroutine stress ...

          total   stress  (Ry/bohr**3)                   (kbar)     P=
-0.20
  -0.00000138  -0.00000000  -0.00000000         -0.20     -0.00     -0.00
  -0.00000000  -0.00000138  -0.00000000         -0.00     -0.20     -0.00
  -0.00000000  -0.00000000  -0.00000124         -0.00     -0.00     -0.18


     bfgs converged in  24 scf cycles and  13 bfgs steps
     (criteria: energy < 0.10E-03, force < 0.10E-02, cell < 0.50E+00)

     End of BFGS Geometry Optimization

     Final enthalpy =    -395.6838583715 Ry
     new unit-cell volume =    315.86297 a.u.^3 (    46.80606 Ang^3 )

CELL_PARAMETERS (alat)
   0.973879690   0.000000000   0.000000000
  -0.486939845   0.843404552  -0.000000000
  -0.000000000  -0.000000000   1.585692727

ATOMIC_POSITIONS (crystal)
Ga      -0.000000000  -0.000000000  -0.015912479
Ga       0.333333333   0.666666667   0.484088029
N        0.000000000   0.000000000   0.360941274
N        0.333333333   0.666666667   0.860883176

*INPUT FOR SCF

*&control
    calculation = 'scf',
    restart_mode='restart',
    outdir='/home/rajpala/Desktop/gan',
    pseudo_dir='/home/rajpala/Desktop/gan'
    prefix='gan',
    tstress = .true.,
    tprnfor = .true.,
  /
  &system
    ibrav=  4, a = 3.213802977, c = 5.232785999,nat=  4, ntyp= 2,
    ecutwfc =
30,ecutrho=240,occupations='smearing',degauss=0.01,smearing='gaussian'
  /
  &electrons
   diagonalization='david'
   mixing_mode = 'plain'
   mixing_beta = 0.7
   conv_thr = 1.0d-7
 /
ATOMIC_SPECIES
  Ga 69.723 Ga.pbe-nsp-van.UPF
  N 14.0067 N.pbe-rrkjus.UPF
 ATOMIC_POSITIONS (crystal)
Ga      -0.000000000  -0.000000000  -0.015912479
Ga       0.333333333   0.666666667   0.484088029
N        0.000000000   0.000000000   0.360941274
N        0.333333333   0.666666667   0.860883176
 K_POINTS (automatic)
 6 6 4 0 0 0*
*
*OUTPUT FOR SCF*

!    total energy              =    -395.68329166 Ry
     Harris-Foulkes estimate   =    -395.68329169 Ry
     estimated scf accuracy    <       0.00000005 Ry

     The total energy is the sum of the following terms:

     one-electron contribution =    -190.48963216 Ry
     hartree contribution      =     134.51703338 Ry
     xc contribution           =    -108.33945035 Ry
     ewald contribution        =    -231.37124253 Ry
     smearing contrib. (-TS)   =      -0.00000000 Ry

     convergence has been achieved in   8 iterations

     Forces acting on atoms (Ry/au):

     atom   1 type  1   force =     0.00000000    0.00000000   -0.00006952
     atom   2 type  1   force =     0.00000000    0.00000000   -0.00008099
     atom   3 type  2   force =     0.00000000    0.00000000   -0.00003004
     atom   4 type  2   force =     0.00000000    0.00000000    0.00018054

     Total force =     0.000212     Total SCF correction =     0.000060


     entering subroutine stress ...

          total   stress  (Ry/bohr**3)                   (kbar)     P=
-1.64
  -0.00001229   0.00000000   0.00000000         -1.81      0.00      0.00
   0.00000000  -0.00001229   0.00000000          0.00     -1.81      0.00
   0.00000000   0.00000000  -0.00000878          0.00      0.00     -1.29

*As you can see there is perssure is different in final vc-relax and scf
calculation. After that I am doing phonon calculation and I am getting some
frequencies to be negative.
Can anybody please tell me where I am going wrong.*

Thanks in advance,
Sincerely,
mohnish


-- 
Mohnish Pandey
Y6927262,4th Year dual degree student,
Department of Chemical Engineering,
IIT KANPUR
09235721300
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20100410/b907bf02/attachment.html>


More information about the users mailing list