Dear Peng,
I don't believe this is related to the spin-orbit calculation since if you are using an ECP there is
not a restriction to segmented basis sets since the 2-electron SO integrals are not required. Perhaps
you are doing a state-averaged MCSCF optimization, which might give the error message below?
-Kirk
On Nov 9, 2004, at 9:28 AM, Peng Zou wrote:
Dear Molpro Users,
� I've attached my original question and Dr. Knowles's response. The topic is discussed in another thread (see the link given by Dr. Knowles). But it is not straightforward to us how this is done if the basis set is not in the basisset library and the format of the input. We are sincerely looking forward to any suggestions from those who have met similar problems.
� Thanks.
Best regards,
Peng
Peng Zou Combustion Research Facility, Sandia National Laboratories MS 9055, Livermore, CA 94550 Phone: 925-294-3527 Fax: 925-294-2276
-------- Original Message --------
Subject: RE: basis set input
Date: Tue, 9 Nov 2004 09:58:06 -0000
From: Peter Knowles <[EMAIL PROTECTED]>
Reply-To: <[EMAIL PROTECTED]>
To: 'Peng Zou' <[EMAIL PROTECTED]>
Please post questions like this to the molpro-user mailing list, so that the
right person has a chance to reply, and that everyone else can benefit from
the answer. Your criticism is reasonable, and in fact we are working to
circumvent the need to restrict to segmented contraction basis sets in a
future version of the code.
http://www.molpro.net/molpro-user/archive/all/msg00476.html offers some, but
not complete, help.
Best regards, Peter Knowles
> -----Original Message-----
> From: Peng Zou [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, 09 November, 2004 2:58 AM
> To: [EMAIL PROTECTED]
> Subject: basis set input
>
> Dear Sir,
> We are trying to do some spin-orbit caculation on compounds with
> halogen atoms and planning to use the small-core ECP. An error is
> encounted GENERAL BASIS CONTRACTION NOT IMPLEMENTED IN CADPAC GRADIENTS,
> which requires the basis set being input in its primitive form.
> It is not straight forward to us how this is done if the basis set is
> not in the basis library. It will be of great help if you can show an
> example of the input format, for example for the cc-pVdz-PP basis set.
> Thank you in advance.
> Best regards,
> Peng
