Numerical Gradients CCSD(T)

[H. -J. Werner]

You have to use either "active" or "inactive" and to express the
geometry by a z-matrix containing the variable you want
to optimize. Please look at the manual!
H.-J. Werner
On Mon, 10 Nov 2003, Luke Thomas wrote:

>I have a question which I hope somebody will be able to help me with.
>I am trying to implement a numerical forces calculation using CCSD(T) on a 
> system with 6 geometric variables.
>I would like to just calculate the gradient in one variable, without 
>calculating the gradient in the other 5.
>I have tried setting the step size for the six variables using
>
>Step, r1=0.01, r2=0.0, r3=0.0 ..... ,
>
>but I am not sure where to putr this command so that MOLPRO sees it.
>
>I have tried
>
>hf
>ccsd(T)
>forces, numerical, step ....., startcmd=hf
>but this does not see the step in r2 as zero.
>
>I have also tried
>
>hf
>ccsd(t)
>forces, numerical, startcmd=hf
>step, ....
>
>But this does not work either.
>
>If I manage to set a set step size of zero for one variable, will it just 
>ignore the variable and not calcualte the gradient, or will it go through 
>the whole gradient calculation without a step size. If so, it there a 
>specific way to use the force command to calculate a gradient in a 
>specified variable.
>
>Many thanks,
>
>Luke Thomas
>University College London
>
>

-- 
Prof. Hans-Joachim Werner
Institute for Theoretical Chemistry
University of Stuttgart
Pfaffenwaldring 55
D-70569 Stuttgart, Germany
Tel.: (0049) 711 / 685 4400
Fax.: (0049) 711 / 685 4442
e-mail: werner at theochem.uni-stuttgart.de