CCL:G: Avoided crossing

 Sent to CCL by: Nikita Matsunaga [Nikita.Matsunaga()liu.edu]
 You can calculate minimum energy crossing point between two states of different
 spins, such as singlet and triplet,  in GAMESS. (In $CONTRL RUNTYP=MEX would do
 the trick.)  The states of the same symmetry and spin can mix, and causes
 derivative coupling to be non-zero near or at either conical intersection or
 avoided crossing.  There is no mixing between singlet and triplet states in the
 non-relativistic quantum chemistry, hence the two states have to cross. The
 states can mix when spin-orbit coupling is introduced.
 Nikita
 On Apr 20, 2012, at 8:47 AM, Abrash, Sam sabrash{}richmond.edu wrote:
 >
 > Sent to CCL by: "Abrash, Sam" [sabrash**richmond.edu]
 > As far as I understand it, avoided crossings are always between states of
 the same symmetry, so there wouldn't be an avoided crossing between a singlet
 and a triplet state.
 >
 > Sam
 >
 > Samuel A. Abrash
 > Department of Chemistry
 > University of Richmond
 > Richmond, VA 23173
 > Phone:  804-289-8248
 > Fax:  804-287-1897
 > E-mail:  sabrash|*|richmond.edu
 > Web-page:  http://www.richmond.edu/~sabrash
 > "In 1893 Charles Hinton left Japan to become a mathematics instructor
 at Princeton University, where he invented a baseball-pitching machine that used
 gunpowder to propel the balls, like a cannon.  After several accidents, the
 device was abandoned and Hinton lost his job ..." Terry Pratchett, Ian
 Steward and Jack Cohen, The Science of Diskworld III
 >
 > -----Original Message-----
 >> From: owner-chemistry+sabrash==richmond.edu|*|ccl.net [mailto:owner-chemistry+sabrash==richmond.edu|*|ccl.net] On
 Behalf Of Walid M. I. Hassan walid_m76.:.yahoo.com
 > Sent: Thursday, April 19, 2012 7:37 PM
 > To: Abrash, Sam
 > Subject: CCL:G: Avoided crossing
 >
 >
 > Sent to CCL by: "Walid M. I. Hassan" [walid_m76|-|yahoo.com]
 > Dear Scientists
 >
 > long time since I send email via this list,
 > my question is very simple does anyone know a software to determine the
 > avoided crossing or conical intersection between singlet and triplet
 states,
 > i mean states of different multiplicity
 >
 > In Gaussian 09 manual, in Opt keyword ,under conical option it is written,
 > "Search for a conical intersection or avoided crossing using the
 state-
 > averaged CASSCF method. Avoided is a synonym for Conical. Note that
 > CASSCF=SlaterDet is needed in order to locate a conical intersection
 between
 > a singlet state and a triplet state."
 > I tried this and find it is impossible since in input file you determine a
 > singlet or triplet multiplicity but not both
 >
 > and SlaterDet option just reverse the spin but do not change multiplicity
 > see this part of output to see the configurations
 >
 > when I use singlet state with No SLATER DETERMINANT
 >> I get the following configuration
 >>       Configuration      1 Symmetry 1 111000
 >>       Configuration      2 Symmetry 1 11ab00
 >>       Configuration      3 Symmetry 1 110100
 >>       Configuration      4 Symmetry 1 1a1b00
 >>
 >>
 >> which become as follows with SLATER DETERMINANT
 >>       Configuration      1 Symmetry 1 111000
 >>       Configuration      2 Symmetry 1 11ba00
 >>       Configuration      3 Symmetry 1 11ab00
 >>       Configuration      4 Symmetry 1 1b1a00
 >>       Configuration      5 Symmetry 1 11b0a0
 >>
 >>
 >> for triplet state in both cases i get the following configuration
 >> Configuration      1 Symmetry 1 11aa00
 >>       Configuration      2 Symmetry 1 11a0a0
 >>       Configuration      3 Symmetry 1 1a1a00
 >>       Configuration      4 Symmetry 1 110aa0
 >
 > as you can see SlaterDet option reverse ab to ba but not to aa
 > I want to get conical intersection or avoided crossing between
 > Configuration      1 Symmetry 1 111000 of singlet and
 > Configuration      1 Symmetry 1 11aa00 of triplet state
 >
 > I sent this question to Gaussian team and their reply was not convincing to
 > me, I will put it here for your reference,
 > I think they just go around the problem to claim that there is no problem
 in
 > the manual
 > i think gaussian program can not do it
 > if i am wrong i will appreciate if you guide me
 > If i am write so what program can find this conical intersection for me
 > sorry for long email but i think it won't show my point if it is shorter
 >
 >   Dr. Ibrahim,
 >
 >    Thank you for giving us a chance to comment.
 >
 >    When you look at those configuration lists you are not actually
 > looking a lists of identical things.  The singlet without Slater
 Determinant
 > is a list of configurations, fixed linear combinations of determinants
 > giving singlets.  So there will be no triplet terms in this expansion
 > and the higher roots of the CASSCF equations will all be singlets as
 > well.
 >
 >    The second case lists determinants directly and generates singlets
 > and m(l)=0 triplets.  So configurations 2 and 3 are a pair of determinants
 > which can contribute to both singlets and triplets with the sign difference
 > carried on the CI expansion coefficients, not on the determinants
 > directly.
 >
 >      The third case correspond to a ground state triplet, m(l)=+/-1
 > and no singlet contributions are possible.  Again any excited states will
 > be triplets so this reference is not suitable for the crossing.
 >
 >      So the program can compute the crossing you desire by using a
 > singlet reference with SlaterDet included in the option list.  You
 > should still solve for the first few roots to confirm that you have
 > the second root corresponding to the triplet you expected.http://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp-:-//www.ccl.net/chemistry/sub_unsub.shtmlhttp-:-//www.ccl.net/spammers.txt>;
 >
 仁城苔炳 松永
 Nikita.Matsunaga]_[liu.edu