error in dscf with ghost atom (with basis)

[marand]

Dear users!

I am trying to compute the vertical spectrum of bithiophene using the additional rydberg orbitals centred on the centre of charge of the molecule, which in his case is point 0. 0. 0. (C2h symmetry). Therefore I placed there the ghost atom q for which I added the diffuse rydberg orbitals to the basis set (in the file basis).

However, dscf program crashes with the message:

          ------------------
          density functional
          ------------------
Dummy atoms with grid-point found.
DFT quadrature threshold dgrenze tightened by factor 1/10.

Becke-Half-and-Half-LYP hybrid functional: BH-LYP
exchange:    1/2 (LDA + Becke (B88)) + 1/2 HF
correlation: Lee-Yang-Parr (LYP)
Dummy atoms with grid-point found.
DFT quadrature threshold dgrenze tightened by factor 1/10.

Dummy atoms with grid-point found.
DFT quadrature threshold dgrenze tightened by factor 1/10.

Dummy atoms with grid-point found.
DFT quadrature threshold dgrenze tightened by factor 1/10.

spherical integration : Lebedev's spherical grid
spherical gridsize    :           5
   i.e. gridpoints    :         590
radial integration    : Chebyshev 2nd kind (scaling 3)
DIFFUSE GRIDS switched on:
   radial gridsize plus 4 -->  12, rho scaled by 4.0000
integration cells     :           9
partition function    : becke
partition sharpness   :           3

found dummy atom(s) with basis: included in dft
no grid for q

MODTRACE: no modules on stack

fatal error in grinp.
dscf ended abnormally
dscf ended abnormally
program stopped.

I have tried to find the reason in the manual files, but unfortunately without success.
I would greatly appreciate some help, cos I am quite at a loss.
Yours!
Marcin Andrzejak
P.S.: the valence basis set is aug-cc-pVDZ

[Arnim]

Hi,

q is a dummy atom not a ghost atom. You would have to select a defined element in order to have a defined DFT grid, e.g. he, and then set the charge to zero.

Cheers,

Arnim

[marand]

THX! It helped
Marcin