[molpro-user] energy gradient of MR-SO-CISD wavefunction
Zork Zou
zorkzou at yahoo.com.cn
Fri May 13 05:26:41 BST 2011
Hi,
Attached please find the example of geometry optimization with MRCI+SOC. The lowest Omega=3/2 state (2Pi + 2Delta) of the linear HePtF complex was optimized, in which the energies of 2Pi and 2Delta were calculated at the MRCISD+Q level whereas SOC was at the CASSCF level. Since the 2Sigma+ energy was calculated at the CASSCF level (because 2Sigma+ does not contribute to 3/2), the first two spinor states are 3/2 instead of the ground spinor state 1/2.
At some points, the 5/2 state may be a little lower than 3/2. In this case you have to stop the calculation and optimize energy(3) or energy(4) instead. You can also modify the source codes of molpro to identify 3/2 and 5/2 and then swich between energy(1)(2) and energy(3)(4).
The details can be found in Dalton Trans., 39, 2023 (2010)
Best regards,
Wenli
--- On Thu, 5/12/11, Jeong Sik Lim <lim.jeongsik at gmail.com> wrote:
From: Jeong Sik Lim <lim.jeongsik at gmail.com>
Subject: [molpro-user] energy gradient of MR-SO-CISD wavefunction
To: molpro-user at molpro.net
Date: Thursday, May 12, 2011, 12:19 AM
Dear all,
So far, I am doing spin-orbit calculation using ECP. But I am not sure about that energy gradient of MR-SO-CISD wavefunction of specific interested state can be abstracted for geometry optimization? If does, please inform me how to do.
Thank you so much.
-----Inline Attachment Follows-----
_______________________________________________
Molpro-user mailing list
Molpro-user at molpro.net
http://www.molpro.net/mailman/listinfo/molpro-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20110513/fb7899fe/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: heptf-soopt-3.inp
Type: application/octet-stream
Size: 2768 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20110513/fb7899fe/attachment.obj>
More information about the Molpro-user
mailing list