Skip to Content.
Sympa Menu

forum - Re: [abinit-forum] piezoelectric constants of GaN

forum@abinit.org

Subject: The ABINIT Users Mailing List ( CLOSED )

List archive

Re: [abinit-forum] piezoelectric constants of GaN


Chronological Thread 
  • From: matthieu verstraete <matthieu.jean.verstraete@gmail.com>
  • To: forum@abinit.org
  • Subject: Re: [abinit-forum] piezoelectric constants of GaN
  • Date: Sun, 29 Mar 2009 17:35:10 +0200
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; b=VCw7eXJzAQfajL/KaLHmvvWdGH6XJ1uaMwzDbIqRT8U1zTPmwjCVzlLfuHPHfvqmhv 1DSv87Krp4LHbnTDIekghhh6xP+EYomoXDAGC42v18Y3NqYjHaX+GGAJltjxl/WgIoQl E+mGhV4NBQnCAYUOJJcKnE+RYJg75/9Ir6WLw=

Hello Annie,

you should check the abinit output, to see if the phonon SCF cycles
have terminated correctly, and if the frequencies calculated at that
level are already imaginary. If so you have a real problem: the ASR
correction will be much too large (1.e-2 Ha). Have you turned it on in
the anaddb and/or abinit calculations? You should also send us the
anaddb input and abinit output, as well as platform and version
information.

You get the same negative frequencies for both pseudopotentials? This
is surprising. Normally there are large differences in the ASR
breaking between different pseudopotentials.

That you get the same clamped ion numbers is good but just means that
the strain perturbation was done correctly. You still have a problem
with the phonons. Can you compare those to the litterature as well
(and if you have q/=0)?

Matthieu

On Sun, Mar 29, 2009 at 2:15 PM, <annie_beya@yahoo.fr> wrote:
> Dear All,
>
> I am trying to calculate the piezoelectric constants of GaN using the
> response
> function theory.
> When I run anaddb, I find the same  piezo coefficients e_33 and e_31 for the
> clamped ion as those published by Bernardini et al. in PRB 56 R10024 (1997)
> where they used the berry phase approach.
> But the  values for the relaxed ions are not realistic and I have the
> following
> message "piezo9 : WARNING -
>  Unstable eigenvalue detected in force constant matrix at Gamma point
>  The system under calculation is physically unstable."
>
> I have used both hgh (with ecut= 100 Ha) and pspnc pseudopotentials. and
> have
> the same results. The values that are used for acell are the optimized
> values.
> please find below the input file and the anaddb output file.
>
> Thanks for your help.
>
> Annie
>
> Input file
> ndtset 3
> jdtset 1 2 3
> # Set 1 : initial self-consistency
>
> kptopt1   1
> tolvrs1 1.0d-18
> iscf1 7
>
>
> #set 2 the ddk
> rfelfd2 2  # only derivative of the gs wf
> rfdir2 1 1 1
> nqpt2 1
> qpt2 0.0 0.0 0.0 #only the gamma point
> getwfk2 1
> getddk2 2
> kptopt2 3
> iscf2 -3
> tolwfr2 1.0d-20
>
> # Set 3 : response-function strain calculation
>
> rfphon3 1
> rfatpol3 1 2
> rfelfd3 3
> rfdir3 1 1 1
> nqpt3 1
> qpt3 0 0 0
> rfstrs3  3
> getwfk3 1
> getddk3 2
> kptopt3 3
> tolvrs3 1.0d-16
> iscf3 7
>
> #common input data
> paral_rf 1 # activate parallelization over perturbations
> ngroup_rf 4 # number of groups for  parallelization over perturbations
>
>   ecut   100.0
>   ecutsm 0.5
>   iscf   7
>   npulayit 7
>   diemac 12.0
>
>   nline 15
> iprcel 45
>
>   ixc  1
>  nband   18
>  nbdbuf 0
>  occopt 1
>
>  ngkpt   4 4 10
> nshiftk   1
> shiftk    0.0 0.0 0.5
>
>  nstep   100
>  natom   4
>  ntypat   2
>  znucl  31 7
>  typat   1 1 2 2
>   xred   3.3333333333E-01  6.6666666667E-01  0.0
>              6.6666666667E-01  3.3333333333E-01  5.0005858110E-01
>              3.3333333333E-01  6.6666666667E-01  3.7602730263E-01
>              6.6666666667E-01  3.3333333333E-01  8.7602730263E-01
>
>  acell    5.9779130104E+00  5.9779130104E+00  9.7655186937E+00
>
>
>   rprim
>               8.6602540378E-01  5.0000000000E-01  0.0000000000E+00
>             -8.6602540378E-01  5.0000000000E-01  0.0000000000E+00
>              0.0000000000E+00  0.0000000000E+00  1.0000000000E+00
>
>
> #anaddb output
>
> read the DDB information and perform some checks
>
> -begin at tcpu      0.055  and twall      0.055 sec
>
>  Real(R)+Recip(G) space primitive vectors, cartesian coordinates
> (Bohr,Bohr^-1):
>  R(1)=  5.1770245  2.9889565  0.0000000  G(1)=  0.0965806  0.1672825  
> 0.0000000
>  R(2)= -5.1770245  2.9889565  0.0000000  G(2)= -0.0965806  0.1672825  
> 0.0000000
>  R(3)=  0.0000000  0.0000000  9.7655187  G(3)=  0.0000000  0.0000000  
> 0.1024011
>  Unit cell volume ucvol=  3.0222134E+02 bohr^3
>  Angles (23,13,12)=  9.00000000E+01  9.00000000E+01  1.20000000E+02 degrees
>  Now the whole DDB is in central memory
>
> ================================================================================
>
>  Calculation of the internal-strain  tensor
>
> -begin at tcpu      0.059   and twall      0.058sec
>
>  Force-response internal strain tensor(Unit:Hartree/bohr)
>
>  Atom dir   strainxx    strainyy    strainzz    strainyz    strainxz
> strainxy
>  1  x     0.1629756  -0.1629756   0.0000000   0.0000000  -0.1119998
> 0.0000000
>  1  y     0.0000000   0.0000000   0.0000000  -0.1119998   0.0000000
> -0.1629756
>  1  z    -0.1078308  -0.1078308   0.1989130   0.0000000   0.0000000
> 0.0000000
>  2  x    -0.1629756   0.1629757   0.0000000   0.0000000  -0.1119998
> 0.0000000
>  2  y     0.0000000   0.0000000   0.0000000  -0.1119998   0.0000000
> 0.1629756
>  2  z    -0.1078308  -0.1078308   0.1989130   0.0000000   0.0000000
> 0.0000000
>  3  x     0.1876248  -0.1876249   0.0000000   0.0000000   0.1119998
> 0.0000000
>  3  y     0.0000000   0.0000000   0.0000000   0.1119998   0.0000000
> -0.1876248
>  3  z     0.1078308   0.1078308  -0.1989130   0.0000000   0.0000000
> 0.0000000
>  4  x    -0.1876248   0.1876248   0.0000000   0.0000000   0.1119998
> 0.0000000
>  4  y     0.0000000   0.0000000   0.0000000   0.1119998   0.0000000
> 0.1876248
>  4  z     0.1078308   0.1078308  -0.1989130   0.0000000   0.0000000
> 0.0000000
>
>  instr9 : WARNING -
>  Acoustic sum rule violation met : the eigenvalues of accoustic mode
>  are too large at Gamma point.
>  Increase cutoff energy or k-points sampling.
>  The three eigenvalues are:   -6.937594E-02    4.140198E-03   -6.937595E-02
>
>  instr9 : WARNING -
>  Unstable eigenvalue detected in force constant matrix at Gamma point
>  The system under calculation is physically unstable.
>
>  Displacement-response internal strain tensor (Unit:Bohr)
>
>  Atom dir   strainxx    strainyy    strainzz    strainyz    strainxz
> strainxy
>  1  x     1.6258106  -1.6258105   0.0000000   0.0000000  -0.5457645
> 0.0000000
>  1  y     0.0000000   0.0000000   0.0000000  -0.5457645   0.0000000
> -1.6258106
>  1  z    -0.5722507  -0.5722507   1.0556173   0.0000000  -0.0000001
> 0.0000000
>  2  x    -1.6258106   1.6258106   0.0000000   0.0000000  -0.5457645
> 0.0000000
>  2  y     0.0000000   0.0000000   0.0000000  -0.5457645   0.0000000
> 1.6258106
>  2  z    -0.5722507  -0.5722507   1.0556173   0.0000000  -0.0000001
> 0.0000000
>  3  x    -0.6229205   0.6229205   0.0000000   0.0000000   0.5457645
> 0.0000000
>  3  y     0.0000000   0.0000000   0.0000000   0.5457645   0.0000000
> 0.6229203
>  3  z     0.5722507   0.5722506  -1.0556173   0.0000000   0.0000001
> -0.0000001
>  4  x     0.6229205  -0.6229206   0.0000000   0.0000000   0.5457645
> 0.0000000
>  4  y     0.0000000   0.0000000   0.0000000   0.5457645   0.0000000
> -0.6229202
>  4  z     0.5722507   0.5722508  -1.0556174   0.0000000   0.0000001
> 0.0000001
>
> ================================================================================
>
>  Calculation of the elastic and compliances tensor (Voigt notation)
>
> -begin at tcpu      0.061   and twall      0.061sec
>
>  elast9 : WARNING - :
>  Acoustic sum rule violation met : the eigenvalues of accoustic mode
>  are too large at Gamma point
>  increase cutoff energy or k-points sampling.
>  The three eigenvalues are:   -6.937594E-02    4.140198E-03   -6.937595E-02
>
>  elast9 : WARNING - :
>  Unstable eigenvalue detected in force constant matrix at Gamma point.
>  The system under calculation is physically unstable.
>
>  Elastic Tensor (clamped ion) (unit:10^2GP):
>
>  14.2922293   4.4214181   2.5061853   0.0000019   0.0000070   0.0002773
>   4.4214181  14.2942788   2.5061602   0.0000123  -0.0000035   0.0003425
>   2.5061962   2.5061520  -0.4018496   0.0000163  -0.0000352   0.0000236
>  -0.0000009  -0.0000128  -0.0000309   2.9685846  -0.0000020   0.0000000
>   0.0000000   0.0000000   0.0000000  -0.0000020   2.9685883  -0.0000009
>  -0.0001797  -0.0002196  -0.0000020   0.0000000  -0.0000009   4.9359920
>
>  Elastic Tensor (relaxed ion) (unit:10^2GP):
>  (at fixed electric field boundary condition)
>
>  13.7636139   4.4694689   2.9494283   0.0000019   0.0000070   0.0002773
>   4.4694689  13.7656634   2.9494032   0.0000123  -0.0000035   0.0003425
>   2.9494393   2.9493950  -1.2194895   0.0000163  -0.0000352   0.0000236
>  -0.0000009  -0.0000128  -0.0000310   2.7305637  -0.0000020   0.0000000
>   0.0000000   0.0000000   0.0000001  -0.0000020   2.7305675  -0.0000009
>  -0.0001797  -0.0002196  -0.0000020   0.0000000  -0.0000009   4.6476588
>
>  Compliance Tensor (clamped ion) (unit: 10^-2GP^-1):
>
>   0.0606559  -0.0406439   0.1248093  -0.0000006   0.0000013  -0.0000012
>  -0.0406442   0.0606526   0.1247810  -0.0000009   0.0000016  -0.0000025
>   0.1248103   0.1247802  -0.9318991   0.0000045  -0.0000112  -0.0000112
>   0.0000011   0.0000015  -0.0000091   0.3368609   0.0000002   0.0000000
>   0.0000000   0.0000000   0.0000000   0.0000002   0.3368605   0.0000001
>   0.0000005   0.0000013   0.0000097   0.0000000   0.0000001   0.2025935
>
>  Compliance Tensor (relaxed ion)  (unit: 10^-2GP^-1):
>  (at fixed electric field boundary condition)
>
>   0.0691787  -0.0384071   0.0744243  -0.0000003   0.0000007  -0.0000017
>  -0.0384072   0.0691725   0.0744067  -0.0000007   0.0000011  -0.0000032
>   0.0744249   0.0744063  -0.4600574   0.0000024  -0.0000060  -0.0000076
>   0.0000007   0.0000012  -0.0000048   0.3662247   0.0000003   0.0000000
>   0.0000000   0.0000000   0.0000000   0.0000003   0.3662242   0.0000001
>   0.0000009   0.0000018   0.0000062   0.0000000   0.0000001   0.2151621
>
> ================================================================================
>
>  Calculation of the tensor related to piezoelectric effetc
>  (Elastic indices in Voigt notation)
>
> -begin at tcpu      0.062   and twall      0.062sec
>
>  Proper piezoelectric constants (clamped ion) (unit:c/m^2)
>
>      0.00000000      0.00000000      0.43993237
>      0.00000000      0.00000000      0.43993237
>      0.00000000      0.00000000     -0.89593477
>      0.00000000      0.51206135      0.00000000
>      0.51206135      0.00000000      0.00000000
>      0.00000000      0.00000000      0.00000000
>
>  piezo9 : WARNING -
>  Acoustic sum rule violation met : the eigenvalues of accoustic mode
>  are too large at Gamma point
>  Increase cutoff energy or k-points sampling.
>  The three eigenvalues are:   -6.937594E-02    4.140198E-03   -6.937595E-02
>
>  piezo9 : WARNING -
>  Unstable eigenvalue detected in force constant matrix at Gamma point
>  The system under calculation is physically unstable.
>
>  Proper piezoelectric constants (relaxed ion) (unit:c/m^2)
>
>      0.00000000     -0.00000004     -0.74382882
>      0.00000010     -0.00000002     -0.74382882
>      0.00000008     -0.00000004      1.28772153
>      0.00000000     -0.63730757      0.00000006
>     -0.63730751      0.00000000     -0.00000016
>      0.00000001     -0.00000003      0.00000000
>
> ================================================================================
>



--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
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




Archive powered by MHonArc 2.6.15.

Top of Page