From owner-chemistry@ccl.net Mon Sep 9 09:47:00 2019 From: "Xiaoge Su xs14_-_nyu.edu" To: CCL Subject: CCL: CASSCF spin density Message-Id: <-53839-190909035652-26194-8P6nObCXv/bTxqoOoS+H7g a server.ccl.net> X-Original-From: "Xiaoge Su" Date: Mon, 9 Sep 2019 03:56:49 -0400 Sent to CCL by: "Xiaoge Su" [xs14^^nyu.edu] Dear CCL users, I am working on a molecule that is considered to be a singlet biradical, and we applied a CASSCF calculation on the molecule. The result has shown that the molecule is multi-referenced, and the next step we would like to do is to render the alpha-beta spin-density, this would be a strong evidence showing the separation of alpha-beta electrons. With the attached Molpro input file, we were unable to generate the spin density of our molecule. All density values in the cube file were zero or numbers close to zero. Base on publications, it seems that Molpro is able to output the alpha- beta spin density difference, we read the manual but we were not able to find a way of doing so. Could anyone please give us some help? Is there any other softwares that could be used for this calculation? The molpro input file is attached. Best, Xiaoge ***,test gprint,basis,orbitals,civector; symmetry,nosymmetry; geometry={ >>GEOMETRY } basis=6-31G* hf; {multi; frozen,6; closed,19; occ,25; rotate,25.1,28.1,0; orbprint,10; dm,1; locorb,2150.2; print,density; } {pop;density,2150.2,type=spin;individual} {cube,spindensity; density,2150.2,type=spin; } From owner-chemistry@ccl.net Mon Sep 9 11:21:00 2019 From: "Mariusz Radon mariusz.radon_+_gmail.com" To: CCL Subject: CCL: CASSCF spin density Message-Id: <-53840-190909112008-32254-V2Go/Y7RXdjsiLM6SqiuaA=server.ccl.net> X-Original-From: Mariusz Radon Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=utf-8 Date: Mon, 9 Sep 2019 17:19:59 +0200 Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Sent to CCL by: Mariusz Radon [mariusz.radon*_*gmail.com] > On 09 Sep 2019, at 09:56, Xiaoge Su xs14_-_nyu.edu wrote: > > > Sent to CCL by: "Xiaoge Su" [xs14^^nyu.edu] > Dear CCL users, > > I am working on a molecule that is considered to be a singlet biradical, > and we applied a CASSCF calculation on the molecule. The result has shown > that the molecule is multi-referenced, and the next step we would like to > do is to render the alpha-beta spin-density, this would be a strong > evidence showing the separation of alpha-beta electrons. > > With the attached Molpro input file, we were unable to generate the spin > density of our molecule. All density values in the cube file were zero or > numbers close to zero. Dear Xiaoge: That is the correct and exact result: for any singlet, including an open-shell one, the spin density should, indeed, be zero everywhere in space! Hence, you cannot obtain any other spin density from CASSCF calculations because this theory correctly takes into account the spin symmetry of the wave function. Please, bear in mind that the spin polarization that is obtained from broken-symmetry DFT is an artifact of using a single-determinantal Kohn-Sham wave function (the BS wave function of an “open-shell singlet" is not an eigenfunction of S^2, in fact, it is does not really describe a singlet state but rather a mixture of the singlet with a triplet). The spin polarization you expected to find may be very diagnostic in case you interpret the results of DFT calculations, but the exact wave function and a CASSCF one does not have any such property for the singlet state. Might be a bit counterintuitive, but it’s true. If you would like to look at the biradical character of your CASSCF wave function, look down to natural orbitals an their occupation numbers as well as into the CAS-CI coefficient of the wave function - in case of biradical you should see the mixing between two (or more) different singlet configuration directly. Best wishes, Mariusz Radon -- Mariusz Radon, Ph.D. Department of Inorganic Chemistry Faculty of Chemistry Jagiellonian University Gronostajowa 2, 30-387 Krakow, Poland room: C1-06 phone: 48-12-686-24-89 e-mail: mradon-,-chemia.uj.edu.pl; mariusz.radon-,-uj.edu.pl web: http://www.chemia.uj.edu.pl/~mradon ORCID: https://orcid.org/0000-0002-1901-8521 > > Base on publications, it seems that Molpro is able to output the alpha- > beta spin density difference, we read the manual but we were not able to > find a way of doing so. Could anyone please give us some help? Is there > any other softwares that could be used for this calculation? > > The molpro input file is attached. > > Best, > Xiaoge > > > > > ***,test > > gprint,basis,orbitals,civector; > > symmetry,nosymmetry; > > geometry={ >>> GEOMETRY > } > > basis=6-31G* > > hf; > > {multi; > frozen,6; > closed,19; > occ,25; > rotate,25.1,28.1,0; > orbprint,10; > dm,1; > locorb,2150.2; > print,density; > } > > {pop;density,2150.2,type=spin;individual} > > {cube,spindensity; > density,2150.2,type=spin; > }> >