[Pw_forum] How to calculate unoccupied states using cp.x

[Hongjun Xiang]

Dear all,
   I would like to obtain the band gap of a system.
As is well-known, pw.x can give eigenvalues of unoccupied states.
When I tried to use cp.x with the "damp" method to calculate the band 
gap of water molecule by increasing the nbnd value to 8, all eigenvalues 
of unoccupied states are zero.

My input file:
===============================
# Electronic damped dynamics

&CONTROL
   title = ' Water Molecule ',
   calculation = 'cp',
   restart_mode = 'from_scratch',
   ndr = 51,
   ndw = 51,
   nstep  = 100,
   iprint = 10,
   isave  = 100,
   tstress = .FALSE.,
   tprnfor = .TRUE.,
   dt    = 4.0d0,
   etot_conv_thr = 1.d-9,
   ekin_conv_thr = 1.d-4,
   prefix = 'h2o_mol'
   pseudo_dir = './',
   outdir = './tmp/'
/

&SYSTEM
   ibrav = 14,
   celldm(1) = 10.0,
   celldm(2) = 1.0,
   celldm(3) = 1.0,
   celldm(4) = 0.0,
   celldm(5) = 0.0,
   celldm(6) = 0.0,
   nat  = 3,
   ntyp = 2,
   ecutwfc = 70.0,
   nbnd=8
/

&ELECTRONS
   emass = 300.d0,
   emass_cutoff = 2.5d0,
   orthogonalization = 'ortho',
   electron_dynamics = 'damp',
/

&IONS
   ion_dynamics = 'none',
   ion_temperature = 'not_controlled',
/


ATOMIC_SPECIES
  O 16.0d0 O.BLYP.UPF
  H 1.00d0 H.fpmd.UPF

ATOMIC_POSITIONS (bohr)
    O     5.0099    5.0099    5.0000  0 0 0
    H     6.8325    4.7757    4.9999  1 1 1
    H     4.7757    6.8325    4.9998  1 1 1

==================================================

The output file reads:
    ...
    Eigenvalues (eV), kp =   1 , spin =  1
   -24.48  -12.17   -8.43   -6.36    0.00    0.00    0.00    0.00
    ...

=================================================

I also tried the CG method:

&ELECTRONS
   electron_dynamics = 'damp',
   tcg=.true.
/
======================================
The output file in case of nbnd=8 reads:
    ...
    Eigenvalues (eV), kp =   1 , spin =  1
   -24.35  -12.22   -8.40   -6.38    0.00    0.00    0.00    0.00
    ...
======================================
The output file in case of nbnd=4 reads:
    ...
    Eigenvalues (eV), kp =   1 , spin =  1
   -24.48  -12.18   -8.43   -6.36
    ...
=====================================
It seems that when nbnd is larger than the number of the occupied 
states, CG gives even wrongs occupied eigenvalues.

Would someone please tell me if it is possible to obtain correct 
eigenvalues of unoccupied states using cp.x? If yes, how?

P.S., I am using espresso-3.2.

Thank you very much.

Best regards,
Hongjun Xiang

=============================================================
H. J. Xiang
Postdoctoral Research Associate
National Renewable Energy Laboratory
http://www4.ncsu.edu/~hxiang/
=============================================================