TURBOMOLE Users Forum
TURBOMOLE Modules => Molecular Properties, Wavefunction Analysis, and Interfaces to Visualization Tools => Topic started by: woland on December 20, 2013, 12:23:02 PM
-
Hello!
I am a beginner Turbomole user.
I have a set of structures which were optimized (geometry) with other quantum chemistry package than Turbomole.
I wanted to calculate S0 and S1 energies for that molecules (without any geometry optimization) with RI-CC2 method (ricc2 --> method cc2). That step I have done successfully.
Unfortunately, I need also to obtain at RI-CC2 level of theory Mullikan charges for S0 and S1 states.
I have read some topics (http://www.turbo-forum.com/index.php?topic=312.0 (http://www.turbo-forum.com/index.php?topic=312.0) and indicated in this one). I've tried to solve my problem using *.cao files from single point excitation energies calculations and ricc2 -proper or ricc2-fanal option (I've added in control file:
$anadens
cc2-xs1d-ur-1a-001.cao
$pop
)
Calculations were failed. I have 6.4 version of Turbomole.
What should I do to calculate that charges?
My control file for RI-CC2 excited states energies (from TMolex - slightly modified by adding exprop section and pople CAO):
$maxcor 400
$coord file=coord
$scfconv 7
$cbas file=auxbasis
$cdspectrum a.u.
$atoms
n 6 \
cbas = n def2-SV(P) \
basis = n 6-31G*
c 1-5 \
cbas = c def2-SV(P) \
basis = c 6-31G*
h 7-14 \
cbas = h def2-SV(P) \
basis = h 6-31G*
$cosmo_out file=out.cosmo
$title Obl.en.S1-S5
$basis file=basis
$pople CAO
$operating system unix
$denconv 1d-7
$excitations
irrep=a nexc=4
spectrum states=all operators=diplen,qudlen,angmom,dipvel
exprop states=all operators=diplen,qudlen,angmom,dipvel
$spectrum a.u.
$tmole
$ricore 100
$optimize
internal off
redundant off
cartesian on
global off
$symmetry c1
$grad file=gradient
$scfmo none file=mos
$ricc2
cc2
$scfdamp start=0.700 step=0.050 min=0.100
$freeze
implicit core= 6 virt= 0
$subtitle
SP,CC2/RI,charge 1,c1,6-31G*,EX-STATE
$rundimensions
dim(fock,dens)=5797
natoms=14
nshell=52
nbf(CAO)=106
nbf(AO)=106
dim(trafo[SAO<-->AO/CAO])=136
rhfshells=1
$closed shells
a 1-22 ( 2 )
$scfiterlimit 100
$thize 0.10000000E-04
$thime 5
$scfdump
$scfintunit
unit=30 size=0 file=twoint
$scfdiis
$scforbitalshift automatic=.1
$drvopt
cartesian on
basis off
global off
hessian on
dipole on
nuclear polarizability
$forceinit on
diag=default
$forceapprox file=forceapprox
$energy file=energy
$lock off
$end
Calculation sequence:
dscf>dscf.out
ricc2 > ricc2.out
Output files I have stored:
auxbasis
basis
cc2-tm0f-1a-001.cao
cc2-tm0f-1a-002.cao
cc2-tm0f-1a-003.cao
cc2-tm0f-1a-004.cao
cc2-tmf0-1a-001.cao
cc2-tmf0-1a-002.cao
cc2-tmf0-1a-003.cao
cc2-tmf0-1a-004.cao
cc2-xs1d-ur-1a-001.cao
cc2-xs1d-ur-1a-002.cao
cc2-xs1d-ur-1a-003.cao
cc2-xs1d-ur-1a-004.cao
cdspectrum
control
coord
dscf.out
dscf.statistics.parallel
energy
exstates
hp_mpi_appfile
master
metastase
mos
restart.cc
ricc2.out
slave1.output
slave2.output
spectrum
statistics.ricc2
What should I do next to obtain that Mullikan charges?
-
Hi!
You would have to run gradient calculations in the state in which you are interested.
Best,
Arnim
-
Thank you for your quick reply ;-)
In accordance with your suggestion I tried to do it for the ground state.
I have prepared control file with TMolex to Gradient&Energy calculations:
$coord file=coord
$scfconv 7
$cbas file=auxbasis
$atoms
n 6 \
cbas = n def2-SV(P) \
basis = n 6-31G*
c 1-5 \
cbas = c def2-SV(P) \
basis = c 6-31G*
h 7-14 \
cbas = h def2-SV(P) \
basis = h 6-31G*
$cosmo_out file=out.cosmo
$title test
$basis file=basis
$pople CAO
$operating system unix
$denconv 1d-7
$tmole
$ricore 500
$optimize
internal off
redundant off
cartesian on
global off
$symmetry c1
$grad file=gradient
$scfmo none file=mos
$ricc2
cc2 geoopt model=cc2
$scfdamp start=0.700 step=0.050 min=0.100
$freeze
implicit core= 6 virt= 0
$subtitle
SP,CC2/RI,charge 1,c1,6-31G*,
$rundimensions
$rundimensions
dim(fock,dens)=5797
natoms=14
nshell=52
nbf(CAO)=106
nbf(AO)=106
dim(trafo[SAO<-->AO/CAO])=136
rhfshells=1
$closed shells
a 1-22 ( 2 )
$scfiterlimit 30
$thize 0.10000000E-04
$thime 5
$scfdump
$scfintunit
unit=30 size=0 file=twoint
$scfdiis
$scforbitalshift automatic=.1
$drvopt
cartesian on
basis off
global off
hessian on
dipole on
nuclear polarizability
$forceinit on
diag=default
$forceapprox file=forceapprox
$energy file=energy
$lock off
$last step define
$end
Calculation sequence:
dscf>dscf.out
ricc2 > ricc2.out
Output files I have stored:
auxbasis
basis
cc2-gs1d-ur-1a-000.cao
cc2-gsdn-1a-000-total.cao
control
coord
dscf.log
energy
gradient
gsdens
mos
restart.cc
ricc2.log
statistics
statistics.ricc2
And... there are no Mullikan charges in any output file ;/
-
I have just solved the problem by myself.
If Cartesian basis functions are used (mean: $people CAO) instead of spherical basis functions, population analysis is skipped without any information in output file ($pop does not work).