CCL: StateAveraged CASSCF calculations
- From: "Johannes Hachmann"
<jh##chemistry.harvard.edu>
- Subject: CCL: StateAveraged CASSCF calculations
- Date: Thu, 17 Nov 2011 12:06:32 -0500
Sent to CCL by: "Johannes Hachmann" [jh(-)chemistry.harvard.edu]
Hi Saman,
If you try to obtain an excited state in a variational method you often get
root-flipping, i.e., your optimization collapses down to the ground state.
In certain cases you can utilize spin or spatial symmetry constraints to
prevent this but otherwise this is a touch problem.
If you take a ground state result and assume that the higher roots are the
excited states it leads to bad excitation energies because the expansion is
tailored towards one state and may not represent the other states well,
which leads to an unbalanced description.
Therefore, the weights should be chosen according to what you want to get
out of your calc.
There is a way around root-flipping which is described here:
http://jcp.aip.org/resource/1/jcpsa6/v127/i8/p084109_s1
This paper also contains useful references. Helgaker's 'Molecular
Electronic-Structure Theory' book discusses this in some technical detail as
well, if I remember correctly.
Best wishes
Johannes
-----------------------------------------------
Dr. Johannes Hachmann
Postdoctoral Fellow
Aspuru-Guzik Research Group
Harvard University
Department of Chemistry and Chemical Biology
12 Oxford St, Rm M104A
Cambridge, MA 02138
USA
eMail: jh _ chemistry.harvard.edu
-----------------------------------------------
> -----Original Message-----
> From: owner-chemistry+jh==chemistry.harvard.edu _ ccl.net
> [mailto:owner-chemistry+jh==chemistry.harvard.edu _ ccl.net]
On Behalf
> Of Saman Mandegar mandegar_saman::yahoo.com
> Sent: Thursday, November 17, 2011 10:08
> To: Hachmann, Johannes
> Subject: CCL: StateAveraged CASSCF calculations
>
>
> Sent to CCL by: "Saman Mandegar" [mandegar_saman[]yahoo.com]
>
> Dear All,
>
> I would be very happy to have the ideas about state average
> calculations in CASSCF. I am not pretty sure why state average
> calculations are done. What is the reason? Why it is necessary for
> excited state computations? And how we determine the weighting
> factor for each state for example in SA3-CAS(4,3) calculation? Any
> help and pointer to literature is greatly appreciated.
>
> Regards,
> Saman
>
>
>
> -= This is automatically added to each message by the mailing script
> =-
> To recover the email address of the author of the message, please
> change> Conferences:
> http://server.ccl.net/chemistry/announcements/conferences/