[QE-users] Problems with symmetry and accuracy thermo_pw calculation

[aleksandr.doma at yahoo.com]

Hello!
Please help with the following. I got the result of calculating the elastic constants of LKBO (orthorhombic, P212121, Space group 19) using thermo_pw (what='scf_elastic_constants', 'elastic algorithm = 'advanced'). The calculation was done, but I found 2 problems:
1) the resulting matrix is ​​asymmetric:    0.1310416661E+04    0.5760227795E+03    0.3377111274E+03    0    0    0    0.5721874370E+03    0.7221687324E+03    0.3757178266E+03    0    0    0    0.3222687958E+03    0.3774439747E+03    0.1122193660E+04    0    0    0    0                                  0                                 0    0.2688418051E+03    0    0      0                                  0               0                0  0.2782351915E+03            0         0    0    0    0   0                                                                0.2810735472E+03

2) the data do not correspond to experimental results.

Some explanations:I noticed that in half SCF calculations, QE identified the symmetry of the crystal as monoclinic (bravais-lattice index = 12 or -12), and in the others - bravais-lattice index = 8 (orthorhombic). At the start, we received the following symmetry information:
     Bravais lattice   8  simple orthorhombic                          Point group number   8 /  38  D_2  (222)
     The coset representatives of the space group with     the standard order of operations are:
     Fractional translations in crystal coordinates
      PGS    Fract. transl. (all)          Fract. transl. (no shift)   SGS
        E   1 -0.0000 -0.0000 -0.0000      0.0000  0.0000  0.0000         E   1       2z   2 -0.5000 -0.0000  0.5000      0.0000  0.0000  0.5000       2z1  65       2x   4  0.5000 -0.5000 -0.0000      0.5000  0.0000  0.0000       2x1  67       2y   3 -0.0000  0.5000 -0.5000      0.0000  0.5000  0.0000       2y1  66
     PGS = Point group symmetry, SGS = space group symmetry      (no shift) means the part of the fractional translation     that cannot be removed by an origin shift
     Fractional translation shift and origin shift in crystal coordinates
      PGS      FT shift                    &Origin shift                 SGS
        E   1 -0.0000 -0.0000 -0.0000      0.0000  0.0000  0.0000         E   1       2z   2 -0.5000  0.0000  0.0000     -0.2500  0.0000  0.0000       2z1  65       2x   4  0.0000 -0.5000  0.0000      0.0000 -0.2500  0.0000       2x1  67       2y   3  0.0000  0.0000 -0.5000      0.0000  0.0000 -0.2500       2y1  66

     The origin shift gives the point of application of the symmetry     Subtract this vector from all atomic coordinates     to have this symmetry centered at the origin

     Space group number   19
     Space group P2_12_12_1   (group number  19).
     The Laue class is D_2h (mmm) 
     In this class the elastic tensor is
     ( c11  c12  c13   .    .    .  )     ( c12  c22  c23   .    .    .  )     ( c13  c23  c33   .    .    .  )     (  .    .    .   c44   .    .  )     (  .    .    .    .   c55   .  )     (  .    .    .    .    .   c66 )
     It requires all six strains.     for a total of 24 scf calculations


In the input file, atomic coordinates were used, the relaxation of which was carried out without considering symmetry (ibrav=0). JTH-LDA data sets were used (ecutwfc = 81 Ry, occupations = 'tetrahedra_opt', conv_thr = 1.D-8).

I would be grateful for any ideas to improve the calculation results.
Thanks,Oleksandr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240819/0384bef4/attachment.html>