[Pw_forum] negative values in the electron-phonon calculation.

[Eyvaz Isaev]

Hi, 

In addition to Stefano's comment I can just add that
your degauss=0.005 seems to be very low. 
Probably there was a mistype, because for Al (see
example 1) degauss=0.05 is used.

Best regards,
Eyvaz.

--- Stefano Baroni <baroni at sissa.it> wrote:

> Matteo: I do not know if there is anything wrong in
> your data and, in  
> the affirmative, what may be wrong.
> As a general remark, negative frequencies (better,
> "imaginary"  
> frequencies: i.e. negative squared frequencies),
> signal the instability  
> of a crystal structure towards a deformation along
> the phonon  
> distortion pattern. In principles, there is nothing
> wrong in this.  
> Whether this is due to a mistake or not, is a matter
> of physics ...
> 
> Hope this helps.
> Stefano Baroni
> 
> On Sep 9, 2004, at 12:47 PM, matteo giantomassi
> wrote:
> 
> > Dear Pwscf users,
> > I want to calculate the electron-phonon
> interaction in CaAlSi using
> > LDA pseudopotentials.
> > The structure has been optimized and all phonon
> frequences are positive
> > but some values of lambda(q) e gamma(q) are
> negative!.
> >
> >
>
***********************************************************************
> 
> > ************************
> > This is the test script i'm using:
> >
>
***********************************************************************
> 
> > ************************
> >
> > #/bin/sh
> > . ../environment_variables
> >
> > # self-consistent calculation
> >
> > cat > caalsi.scf.in << EOF
> >  &control
> >     calculation='scf'
> >     restart_mode='from_scratch',
> >     tstress = .true.
> >     tprnfor = .true.
> >     prefix='caalsi',
> >     pseudo_dir = '$PSEUDO_DIR/',
> >     outdir='$TMP_DIR/'
> >  /
> >  &system
> >     ibrav=  4, celldm(1) =7.93912,
> celldm(3)=1.18356, nat=  3, ntyp= 3,
> >     ecutwfc =20, nbnd=12,
> >     occupations='smearing', degauss=0.006,
> smearing='gaussian',
> >  /
> >  &electrons
> >     conv_thr =  1.0d-8
> >     mixing_beta = 0.7
> >  /
> > ATOMIC_SPECIES
> > Ca  40.078      Ca.pz-n-vbc.UPF
> > Al  26.98154       Al.pz-vbc.UPF
> > Si  28.08550    Si.pz-vbc.UPF
> >
> > ATOMIC_POSITIONS crystal
> >  Ca 0 0 0
> >  Al 0.333333333 0.666666666 0.5
> >  Si 0.666666666 0.333333333 0.5
> > K_POINTS automatic
> >  12 12 12     0 0 0
> > EOF
> > $PW_ROOT/bin/pw.x < caalsi.scf.in > caalsi.scf.out
> >
> > #
> > # These are k-points of a (666) uniform grid in
> the irreducible#  
> > Brillouin Zone
> > qpoints="0.1666667,0.0962250,0.0000000
> >          0.3333333,0.1924501,0.0000000
> >          0.5000000,0.2886751,0.0000000
> >          0.1666667,0.2886751,0.0000000
> >          0.3333333,0.3849002,0.0000000
> >          0.3333333,0.5773503,0.0000000
> >          0.0000000,0.0000000,0.1408181
> >          0.1666667,0.0962250,0.1408181
> >          0.3333333,0.1924501,0.1408181
> >          0.5000000,0.2886751,0.1408181
> >          0.1666667,0.2886751,0.1408181
> >          0.3333333,0.3849002,0.1408181
> >          0.3333333,0.5773503,0.1408181
> >          0.0000000,0.0000000,0.2816362
> >          0.1666667,0.0962250,0.2816362
> >          0.3333333,0.1924501,0.2816362
> >          0.5000000,0.2886751,0.2816362
> >          0.1666667,0.2886751,0.2816362
> >          0.3333333,0.3849002,0.2816362
> >          0.3333333,0.5773503,0.2816362
> >          0.0000000,0.0000000,0.4224543
> >          0.1666667,0.0962250,0.4224543
> >          0.3333333,0.1924501,0.4224543
> >          0.5000000,0.2886751,0.4224543
> >          0.1666667,0.2886751,0.4224543
> >          0.3333333,0.3849002,0.4224543
> >          0.3333333,0.5773503,0.4224543"
> > for qpoint in $qpoints ; do
> > #
> > # non self-consistent calculation
> > #
> > qx=`echo $qpoint | cut -d, -f1`
> > qy=`echo $qpoint | cut -d, -f2`
> > qz=`echo $qpoint | cut -d, -f3`
> > cat > caalsi.nscf.in << EOF
> >  &control
> >     calculation='phonon'
> >     restart_mode='from_scratch',
> >     prefix='caalsi',
> >     pseudo_dir = '$PSEUDO_DIR/',
> >     outdir='$TMP_DIR/'
> >  /
> >  &system
> >     ibrav=  4, celldm(1) =7.93912, celldm(3)=
> 1.18356, nat=  3, ntyp=  
> > 3,
> >     ecutwfc =20, nbnd=12, occupations='smearing',
> >     degauss=0.006, smearing='gaussian',
> >  /
> >  &electrons
> >     conv_thr =  1.0d-8
> >  /
> >  &phonon
> >     xqq(1) = $qx, xqq(2) = $qy, xqq(3) = $qz
> > /
> > ATOMIC_SPECIES
> >  Ca  40.078    Ca.pz-n-vbc.UPF
> >  Al  26.98154  Al.pz-vbc.UPF
> >  Si  28.08550  Si.pz-vbc.UPF
> > ATOMIC_POSITIONS crystal
> >  Ca 0 0 0
> >  Al 0.333333333 0.666666666 0.5
> >  Si 0.666666666 0.333333333 0.5
> > K_POINTS automatic
> >     6 6 6    0 0 0
> > EOF
> >
> > $PW_ROOT/bin/pw.x < caalsi.nscf.in >>
> caalsi.nscf.out
> > cat > caalsi.ph.in << EOF
> > phonons of caalsi at $qpoint
> >  &inputph
> >   tr2_ph=1.0d-10,
> >   prefix='caalsi',
> >   amass(1)=40.078,
> >   amass(2)=26.98154,
> >   amass(3)=28.08550,
> >   outdir='$TMP_DIR/',
> >   fildyn='dyn.$qpoint'
> >   fildvscf='caalsidv'
> >  /
> > $qpoint
> > EOF
> > $PW_ROOT/bin/ph.x < caalsi.ph.in >> caalsi.ph.out
> >
> > cat > caalsi.nscf2.in << EOF
> >  &control
> >     calculation='phonon'
> >     restart_mode='from_scratch',
> >     prefix='caalsi',
> >     pseudo_dir = '$PSEUDO_DIR/',
> >     outdir='$TMP_DIR/'
> >  /
> >  &system
> >     ibrav=  4, celldm(1) =7.93912, celldm(3)=
> 1.18356, nat=  3, ntyp=  
> > 3,
> >     ecutwfc =20, nbnd=12, occupations='smearing',
> >     degauss=0.006, smearing='gaussian',
> >  /
> >  &electrons
> >     conv_thr =  1.0d-8
> >  /
> >  &phonon
> >     xqq(1) = $qx, xqq(2) = $qy, xqq(3) = $qz
> > /
> > ATOMIC_SPECIES
> >  Ca  40.078    Ca.pz-n-vbc.UPF
> >  Al  26.98154  Al.pz-vbc.UPF
> >  Si  28.08550  Si.pz-vbc.UPF
> > ATOMIC_POSITIONS crystal
> >  Ca 0 0 0
> >  Al 0.333333333 0.666666666 0.5
> >  Si 0.666666666 0.333333333 0.5
> > K_POINTS automatic
> >     12 12 12   0 0 0
> > EOF
> > $PW_ROOT/bin/pw.x < caalsi.nscf2.in >>
> caalsi.nscf2.out
> 
=== message truncated ===



		
__________________________________
Do you Yahoo!?
Take Yahoo! Mail with you! Get it on your mobile phone.
http://mobile.yahoo.com/maildemo