[Pw_forum] Bug in ph.x sometimes occur in hexagonal phases

[Uri Argaman]

Dear QE users and developers
When I do phonon calculations in hexagonal phases (titanium and zirconium)
in a 4X4X4 q-points grid, QE calculate 12 dynamical matrices. But, in a
specific geometries QE calculate 13 dynamical matrices, two of them are the
same. dyn0 in that case is:
   4   4   4
  13
   0.000000000000000E+00   0.000000000000000E+00   0.000000000000000E+00
   0.000000000000000E+00   0.000000000000000E+00   0.156342786935722E+00
   0.000000000000000E+00   0.000000000000000E+00  -0.312685573871444E+00
   0.000000000000000E+00   0.288675134594792E+00   0.000000000000000E+00
   0.000000000000000E+00   0.288675134594792E+00   0.156342786935722E+00
   0.000000000000000E+00   0.288675134594792E+00  -0.312685573871444E+00
   0.000000000000000E+00   0.288675134594792E+00  -0.156342786935722E+00
   0.000000000000000E+00  -0.577350269189585E+00   0.000000000000000E+00
   0.000000000000000E+00  -0.577350269189585E+00   0.156342786935722E+00
   0.000000000000000E+00  -0.577350269189585E+00  -0.312685573871444E+00
   0.250000000000000E+00   0.433012701892189E+00   0.000000000000000E+00
   0.250000000000000E+00   0.433012701892189E+00   0.156342786935722E+00
   0.250000000000000E+00   0.433012701892189E+00  -0.312685573871444E+00

You can see that the 5'th and the 7'th dynamical matrix are the same. I
want to emphasize that this bug is only at a certain lattice parameter. For
example, this is the input files:
This is the SCF input:
 &control
    calculation='scf'
    restart_mode='from_scratch',
    prefix='zr',
    pseudo_dir = '/home/argaman/espresso_intel/pseudo/',
    outdir='/home/argaman/tmp/tmp5110/',
    tstress = .true. ,
    tprnfor = .true. ,
    verbosity='high',
    wf_collect=.true.
 /
 &system
    ibrav=  4, celldm(1) =5.98584, celldm(3) =1.59905042567 , nat= 2, ntyp=
1,
    ecutwfc =45,ecutrho=225,
    occupations='smearing', smearing="methfessel-paxton", degauss=0.01,
 /
 &electrons
    conv_thr =  1.0d-9,
    electron_maxstep=1000
    mixing_beta = 0.7d0,
 /

ATOMIC_SPECIES
   Zr  91.224 zr_pbe_v1.uspp.F.UPF
ATOMIC_POSITIONS (crystal)
   Zr      0.666666666    0.333333333    0.250000000    0  0  0
   Zr      0.333333333    0.666666666    0.750000000
K_POINTS {automatic}
 18 18 18 0 0 0

This is the phonon input:
phonon  for Zr
 &inputph
  tr2_ph=1.0d-15,
  prefix='zr',
  fildvscf='aldv',
  amass(1)=91.224,
  outdir='/home/argaman/tmp/tmp5110/',
  fildyn='zr.dyn',
  trans=.true.,
  ldisp=.true.,
  nq1=4, nq2=4, nq3=4
 /

I got this bug on versions:
5.0.3, 5.1 and 5.2.0
Thank you
Uri Argaman
Ben-Gurion University
Israel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150826/5ea1d688/attachment.html>