CCL: Reaction of an anion with a closed shell molecule with CASSCF

Dear Muhammed,
 Unfortunately, I can't shed any light on the technical aspects of the
 CASSCF calculation, but *qualitatively*, what you describe looks very much
 like the "curve crossing model" of chemical reactivity that has been
 used
 by Shaik and Pross and others. If you are looking for qualitative insights,
 I would search "curve crossing model". Good luck with your
 calculations.
 Alan
 On Thu, Jul 2, 2020 at 10:01 AM Muhammed Buyuktemiz mbtemiz3]~[gmail.com <
 owner-chemistry###ccl.net> wrote:
 >
 > Sent to CCL by: "Muhammed  Buyuktemiz" [mbtemiz3^gmail.com]
 > Dear CCLers,
 >
 > I am trying to investigate the electron transfer processes for a reaction
 > between an anion and a closed shell molecule. As a specific example, you
 > may consider the reaction between hydroxide and acyl chloride.
 >
 > Initially, I am placing two species at 10 A apart from each other and
 > presume that the two are infinitely separated, i.e the ion and the acyl
 > chloride don't `feel` each other. Later on I am carrying relaxed geometry
 > optimizations by keeping inter-molecular distance r constant while r is
 > decreased in subsequent calculations.
 > However, I am having difficulty in understanding/interpreting the
 > wavefunction for the infinitely separated species. At 10 A, CASSCF
 > optimized wavefunction* shows two major configurations each with 0.60
 > coefficients. The first configuration shows what I would normally expect,
 > an OH(-) ion and a closed shell Acyl Chloride electronic structure. The
 > problem I have is related to the second configuration; This configuration
 > shows that an electron transfer takes place at 10 A and the electron is
 > located on one of the antibonding orbitals of the acyl chloride, yielding
 > OH(0) + Acyl Chloride(-) type of electronic structure.
 >
 > I assume this is not chemically correct (since the two are infinitely
 > separated) and doesn't allow me to investigate the perturbations induced
 > when OH- approaches to Acyl Chloride**. But most importantly, since Acyl
 > Chloride now has an extra electron located on the antibonding orbital, C-
 > Cl bond breaks up in the geometry optimization calculations.***
 >
 > My initial thought on the problem was that this might be an unbalanced
 > active space issue but various combinations also led to the same outcome.
 > I now realize this may not surprising due to the variational nature of
 > the calculation; OH(0) + Acyl Chloride(-) configuration is as
 > important/stable as the OH(-) + Acyl Chloride configuration. However this
 > sentiment alone does not fix the issue and I am stuck.
 >
 > Is there any way to work around this issue that will allow me to
 > investigate what `happens` to orbitals on the OH- approach?
 >
 > Any suggestions are greatly appreciated. Thank you
 >
 >
 > *: rest of the coefficients are smaller than 0.1.
 > **: this problem also persists for smaller/larger intermolecular
 > distances and choosing an arbitrary 10 A is not a problem.
 > ***: OH(-) -- Acyl Chloride distance is kept at 10 A.>
 >
 >
 --
 Alan Shusterman
 Chemistry Department
 Reed College
 3203 SE Woodstock Blvd
 Portland, OR 97202-8199
 503-517-7699
 http://blogs.reed.edu/alan/
 "Patience, persistence, and a sense of humor." Dave Barrett
 (1956-2017,
 Reed College '77)