[Pw_forum] Issue with Raman Spectra

Dasmahapatra, Atreyi atreyi.dasmahapatra at mavs.uta.edu
Thu Jul 16 21:02:00 CEST 2015


Dear all,


I am a new user in Quantum Espresso.


I am trying to generate the Raman spectra for a polymorph of Indium Oxide (space group 60, pnma, 20 atoms in unit cell). My input file is as follows: InO.ph.pnma.in


phonons of InO at Gamma

 &inputph

  tr2_ph=1.0d-10,

  prefix='InO_pnma_opt',

  epsil=.true.,

  trans=.true.,

  lraman=.true.,

  elop=.true.,

  amass(1)=114.818,

  amass(2)=15.999,

  outdir='./',

  fildyn='InO_pnma_opt.dynG',

  fildrho='InO_pnma_opt.drho',

 /

0.0 0.0 0.0


The problem is the job always crashes with an error as this :


Computing Pc [DH,Drho] |psi>


     Derivative coefficient:  0.001000    Threshold: 1.00E-12


 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     Error in routine cdiaghg (89):

     S matrix not positive definite

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


     stopping ...


Optimization of the structure seems fine. I have tried changing the diagonalization scheme from "david" to "cg" ; however nothing seems to work.


For reference, my optim.in input file is:


&CONTROL

                       title = 'InO' ,

                 calculation = 'vc-relax' ,

                restart_mode = 'from_scratch' ,

                      outdir = './' ,

                      wfcdir = './' ,

                  pseudo_dir = './' ,

                      prefix = 'InO_pnma_opt' ,

                     disk_io = 'default' ,

                   verbosity = 'default' ,

               etot_conv_thr = 1.0E-6 ,

               forc_conv_thr = 1.0D-3 ,

                       nstep = 120 ,

                      iprint = 1,

                     tstress = .true. ,

                     tprnfor = .true. ,

                     wf_collect = .true.

 /

 &SYSTEM

                       ibrav = 0,

                   celldm(1) = 1.8897259886,

                         nat =  20   ,

                        ntyp = 2 ,

                     ecutwfc = 60.0 ,

                       nosym = .false. ,

 /

 &ELECTRONS

            electron_maxstep = 200,

                    conv_thr = 1.0D-8  ,

                 startingpot = 'atomic' ,

                 startingwfc = 'atomic' ,

                 mixing_mode = 'plain' ,

                 mixing_beta = 0.2D0 ,

                 mixing_ndim = 8,

             diagonalization = 'david' ,

 /

 &IONS

 /

 &CELL

   press = 0.0,

 /

CELL_PARAMETERS (alat=  1.88972599)

     5.8035313053775530    0.0000000000000000    0.0000000000000000

     0.0000000000000000    3.2090750352897586    0.0000000000000000

     0.0000000000000000    0.0000000000000000   13.2422623740116698

ATOMIC_SPECIES

In 114.818 In.pw-d-mt_fhi.UPF

O 15.9994 O.pw-mt_fhi.UPF


ATOMIC_POSITIONS (crystal)

In      0.763704465 0.250000000 0.550824250

In      0.736295535 0.750000000 0.050824250

In      0.236295535 0.750000000 0.449175750

In      0.263704465 0.250000000 0.949175750

In      0.191331978 0.250000000 0.188335168

In      0.308668022 0.750000000 0.688335168

In      0.808668022 0.750000000 0.811664832

In      0.691331978 0.250000000 0.311664832

O       0.012337968 0.250000000 0.406560633

O       0.487662032 0.750000000 0.906560633

O       0.987662032 0.750000000 0.593439367

O       0.512337968 0.250000000 0.093439367

O       0.145992425 0.250000000 0.778480451

O       0.354007575 0.750000000 0.278480451

O       0.854007575 0.750000000 0.221519549

O       0.645992425 0.250000000 0.721519549

O       0.380594657 0.250000000 0.566366134

O       0.119405343 0.750000000 0.066366134

O       0.619405343 0.750000000 0.433633866

O       0.880594657 0.250000000 0.933633866


K_POINTS automatic

4 8 2 0 0 0



..


Please help !


- Atreyi Dasmahapatra
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150716/7cbe1c55/attachment.html>


More information about the users mailing list