[Pw_forum] ground state structure calculation -reg.

[Suresh A]

I have done vc-relax calculation on rutile to find out ground
    state structure of the system. While looking at my input and output file
    1)SAVA told that "symmetry is broken and suggested to solve cutoff
related
    issues"
    2)EMINE told that if high ecutoff values needed for stress tensor
    convergence than better go for equation of state than vc-relax,
    3) Giuseppe suggested something on exxdiv_treatment and questioned
reason
    for high ecutoff value 200/800.
         Now i have repeated my calculation the input file is included here
    &CONTROL
                           title = 'rutile' ,
                     calculation = 'vc-relax' ,
                    restart_mode = 'from_scratch' ,
                          outdir = '/home/basheer/Desktop/relax2/
' ,
                      pseudo_dir =
    '/home/basheer/Desktop/espresso-5.0.1/pseudo/' ,
                         disk_io = 'high' ,
                       verbosity = 'high' ,
                   etot_conv_thr = 1.d-8 ,
                   forc_conv_thr = 1.d-4 ,
                         tstress = .true. ,
                         tprnfor = .true. ,
     /
     &SYSTEM

- Ignored:
                           ibrav = 6,
                               A = 4.59373 ,
                               B = 4.59373 ,
                               C = 2.95812 ,
                           cosAB = 0 ,
                           cosAC = 0 ,
                           cosBC = 0 ,
                             nat = 6,
                            ntyp = 2,
                         ecutwfc = 50 ,
                         ecutrho = 200 ,
                           nosym = .true. ,
                exxdiv_treatment = 'none' ,
     /
     &ELECTRONS
                        conv_thr = 1.d-12 ,
     /
     &IONS
     /
     &CELL
                   cell_dynamics = 'bfgs' ,
                     cell_factor = 2 ,
     /
    ATOMIC_SPECIES
       Ti   47.88800  Ti.pz-mt_fhi.UPF
        O   15.99900  O.pz-mt_fhi.UPF
    ATOMIC_POSITIONS angstrom
       Ti      0.000000000    0.000000000    0.000000000
       Ti      2.296860000    2.296860000    1.479060000
        O      1.402470000    1.402470000    0.000000000
        O      3.191260000    3.191260000    0.000000000
        O      0.894400000    3.699330000    1.479060000
        O      3.699330000    0.894400000    1.479060000
    K_POINTS automatic
      6 6 6   0 0 0

    for this input the output is
    Begin final coordinates
         new unit-cell volume =    365.14180 a.u.^3 (    54.10843 Ang^3 )

    CELL_PARAMETERS (alat=  8.68089163)
       0.950810397   0.000000544   0.000000000
       0.000000544   0.950810397   0.000000000
       0.000000000   0.000000000   0.617420010
     entering subroutine stress ...

              total   stress  (Ry/bohr**3)                   (kbar)     P=
    0.13
       0.00000085  -0.00000001  -0.00000000          0.12     -0.00
-0.00
      -0.00000001   0.00000085  -0.00000000         -0.00      0.12
-0.00
      -0.00000000  -0.00000000   0.00000104         -0.00     -0.00
 0.15
    entering subroutine stress ...

              total   stress  (Ry/bohr**3)                   (kbar)     P=
    -236.66
      -0.00164796  -0.00000002  -0.00000000       -242.42     -0.00
-0.00
      -0.00000002  -0.00164796  -0.00000000         -0.00   -242.42
-0.00
      -0.00000000  -0.00000000  -0.00153033         -0.00     -0.00
-225.12
                       the output shows lattice parameter is 95% what i have
    given in input and also stress tensor is not converged. When i checked
the
    output structure it shows 2 apical bonds(2 Ti-O BONDS) are lower than
other
    four equitorial bonds(4 Ti-O BONDS) but experimental shows two apical
bond
    should be higher than four equitorial bonds. But when i increase
gradually
    ecutoff value 50 to 100 by step by step the output lattice parameter is
98%
    of what i have given initially and stress tensor also converged as
below.
    and 2 apical bonds(2 Ti-O BONDS) are higher than other four equitorial
    bonds(4 Ti-O BONDS) as experimental value shows.

    Begin final coordinates
         new unit-cell volume =    400.44101 a.u.^3 (    59.33924 Ang^3 )

    CELL_PARAMETERS (alat=  8.68089163)
       0.983405137   0.000000000   0.000000000
       0.000000000   0.983405137   0.000000000
       0.000000000   0.000000000   0.632966319
     entering subroutine stress ...

              total   stress  (Ry/bohr**3)                   (kbar)     P=
    0.01
       0.00000006   0.00000000   0.00000000          0.01      0.00
 0.00
       0.00000000   0.00000006   0.00000000          0.00      0.01
 0.00
       0.00000000   0.00000000   0.00000005          0.00      0.00
 0.01

             total   stress  (Ry/bohr**3)                   (kbar)     P=
    0.18
       0.00000123   0.00000000   0.00000000          0.18      0.00
 0.00
       0.00000000   0.00000123   0.00000000          0.00      0.18
 0.00
       0.00000000   0.00000000   0.00000121          0.00      0.00
 0.18
    in this calculation i remove exx_divtreatment, and occupations=smearing
    cards. is this calculation is correct or i have to go for equation
state?
    and whether symmetry is preserved for ecut=100? Which one i have to
 choose
    ecut =50 or 100 because for 100 only i got 98% lattice parameter than my
    input latticeparameter value. while i doing energy convergence test
energy
    is converged at ecut=50, what about this?
                                      Thank You
A.Suresh,
Project Fellow,
Madurai Kamaraj University,
Madurai.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20131114/a151c797/attachment.html>