The ABINIT Users Mailing List ( CLOSED )

[occupierliu@yahoo.com.cn]

Dear Yukihiro Okuno:
   I optimized the atoms' position as the parameters setted as 
following: 
        optcell 1
        ionmov  3
        ecut    80.0
        ngkpt   6  6  6
and I got the new atomic position as the piezoelectricity calculation's
input:
  acell  7.1922696447E+00   7.1922696447E+00   7.6597671716E+00
  xred  0.0  0.0  0.0
        1/2  1/2  1/2
        1/2  1/2  0.0
        1/2  0.0  1/2
        0.0  1/2  1/2
But the results of the piezoelectricity calculation almost drive me mad!
Using the old input file I posted, I think it same as yours refering to 
the test_v4#67, I got the piezoelectricity tensor is Zero. Some output
are as following:

Effective Charge tensor:

         1         1    4.004116E+00    0.000000E+00   -1.177500E-15
         1         2   -4.507438E-14    4.004116E+00    2.183685E-15
         1         3    8.620324E-15    0.000000E+00    3.757992E+00
         2         1    7.557723E+00    0.000000E+00   -3.647670E-15
         2         2    4.016568E-14    7.557723E+00   -7.803550E-16
         2         3    3.227916E-17    0.000000E+00    6.301300E+00
         3         1   -2.874587E+00    0.000000E+00    5.243542E-15
         3         2    2.975662E-14   -2.874587E+00   -3.616463E-15
         3         3   -2.451651E-14    0.000000E+00   -5.079627E+00
         4         1   -2.434699E+00    0.000000E+00    3.060379E-15
         4         2   -2.696833E-14   -6.223946E+00    7.509596E-15
         4         3   -1.565399E-14    0.000000E+00   -2.482469E+00
         5         1   -6.223147E+00    0.000000E+00   -2.948038E-15
         5         2    2.371206E-14   -2.434699E+00   -1.652028E-15
         5         3   -3.984927E-15    0.000000E+00   -2.481954E+00

Proper piezoelectric constants(clamped ion)(Unit:c/m^2)

      0.00000000      0.00000000      0.00000000
      0.00000000      0.00000000      0.00000000
      0.00000000      0.00000000      0.00000000
      0.00000000      0.00000000      0.00000000
      0.00000000      0.00000000      0.00000000
      0.00000000      0.00000000      0.00000000
      
And There is a worning:
accoustic sum rule violation met:the eigenvalues of accoustic mode
are too large at Gamma point
increase cutoff energy or k-points sampling.
I think this worning caused by calculation in DataSet2 becaused it is
not conversed.

This program costs 15 days on the machine with PIV3.0. 
I have no idear what's wrong with this calculation. Do you have some
suggestions or new results?

Regards
Yong Liu