TURBOMOLE Users Forum
TURBOMOLE Modules => Aoforce and Numforce => Topic started by: Gess on April 07, 2015, 03:10:16 AM

Dear Users of Turbomole,
I tried to calculate aoforce with B97D/def2SVP and RI approximation. Result was:
CONSTRUCTING first deriv. of <munu>
<muhnu> > RHS
second deriv. of <munu> > Hessian
<muhnu> > Hessian
nucl. rep. > Hessian
integrals are neglected if expon. factor < 3.240440699935191E013
...terminated. cpu: 204.73 wall: 275.66
CONSTRUCTING second deriv. of 2e energy > Hessian
treating Coulomb (and exchange) contribution
integrals will be neglected if total contribution < 3.240440699935191E009
...terminated. cpu: 17254.18 wall: 17254.18
treating exchangecorrelation contribution
MODTRACE: no modules on stack
B97D not implemented for this property
force ended abnormally
~
Does this mean that not all functionals have analytic 2nd derivatives in Turbomole?
Thanks in advance,
Gess.

Hello Gess,
this is correct. The B97D code stems from the group of Stefan Grimme and they implemented the functional up to first derivatives only. You can use NumForce to get the frequencies though.
Regards,
Uwe