Dear all,
I trying to calculate chemical shifts in carbon 1s ionization energies following the ?SCF approach according to the protocol that was outlined in J. Chem. Theory Comput. 2011, 7, 4104–4114. So, I would like to do a calculation on CH4 from which one core electron has been ejected and there is now one electron left in the 1s orbital of the CH4(+) cation. The protocol calls for simulating this partially filled 1s core with an ECP (modified sbkjc_ecp). I got it working in another program but I am having trouble with NWChem.
I am not sure how to specify that the ECP-modeled core should have a single electron. When I use "nelec 1" in the ECP definition, the program stops with "The number of electrons 1 in the ECP does not accord with a known set of completely filled shells". Does this suggest that I can use ECPs in NWChem only to replace completely filled shells?
I have tried couple of things (charge keyword in the geometry input; set atomscf:z; uhf/rhf) with no luck. There is an example on how to do core-hole calculations with DFT without ECPs but I eventually want to run coupled cluster jobs. Any suggstions are most appreciated.
Example input:
title "Methane Core-Hole State w/ NWChem 6.5"
scratch_dir /work/nwchem
memory total 1000 mb
# These were suggested in CCL long time ago but did not seem to help me :(
set atomscf:z 1.0
set atomscf:tags_z C
# The core hole state of CH4 is singly charged
charge 1
# This comes from "Geometry_examples"
geometry units au
c 0.0000000 0.0000000 0.0000000
h 1.1828637 1.1828637 1.1828637
symmetry group Td
end
basis spherical
H library cc-pvtz
C library cc-pvtz
end
# Trying to model the remaining singe electron by ECP
ecp
C nelec 1
C ul
1 8.5646800 -0.4468550
C S
0 2.8149700 1.9292600
2 8.1129600 14.8819900
end
# if singlet
# then "The number of electrons 1 in the ECP does not accord with a known set of completely filled shells"
# if doublet
# then "scf: no. of closed-shell electrons is not even! 1"
scf
rhf
singlet
end
task scf energy
|