Summary - How to find the molecular multiplicity?



  Hello! CCLers
   Earlier this week, I posted a question about molecular multiplicity.
 Original question:
 >   I am using Gaussian98W and have basic question of quantum chemistry.
 >
 >   Starting with the G98 calculation, we must set the Charge and
 Multiplicity,
 > but I don't know the practical method to find the each molecular
 multiplicity.
 >
 >   I have read books and chemical articles which are describing the
 multiplicity
 > and its numerical expression. But I can't find the other methods except the
 > numerical expression.
 >
 >   Does any chemical researcher always calculate the molecular multiplicity
 with
 > such numerical expression in his brain?  Or is there any other method?
 >
 >   I will summarize the answers for novice quantum chemistry researcher,
 just like me!
  I have received very useful replies and summarize them. Reading these replies,
 I aware
 that there are something easy rules to find the molecular multiplicity. If you
 like to get in
 touch with replyers, please let me know.
  I hope this summary will be helpful to some others.
  Thanks a lot!
 **** 1) from Alan Shusterman ****
 In most cases the multiplicity can be set according to the following rules:
 All electrons in pairs --> singlet (multiplicity = 1)
 One unpaired electron (free radical) --> doublet (multiplicity = 2)
 Two unpaired electrons (diradical) -->
 and unpaired electrons have opposite spin --> singlet (multiplicity = 1)
 or
 and unpaired electrons have same spin --> triplet (multiplicity = 3)
 Multiplicity is the number of degenerate spin eigenfunctions that can be
 attached
 to the spatial part of the wavefunction.
 **** 2) from David Shobe ****
 First of all, for closed shell molecules--including most of the isolable
 organic molecules--the multiplicity is 1, no further questions needed.
 Essentially "closed shell" means that the molecule follows the octet
 rule.
 Molecular fragments such as radicals and carbenes may have multiplicity >1.
 Organic radicals can be assumed to have multiplicity 2 (with very rare
 exceptions).  Carbenes can be either 1 or 3 (singlet or triplet), as can
 biradicals such as tetramethylene and "trimethylenemethane"
 (2-methylene-1,3-propanediyl).  If the multiplicity is unknown, you have to
 try both.
 The d-block elements and some of the f-block elements are trickiest.  You
 may recall the "high-spin" and "low-spin" states from ligand
 field theory.
 Even worse, the multiplicity tends to become blurred when relativistic
 effects (spin-orbit coupling) are taken into account.  Again, in cases of
 unknown multiplicity, one has to try two or more options, each of which is a
 separate calculation.
 Generally, instead of using the formula in the quantum textbooks, I just add
 1 to the number of unpaired electrons (singly-occupied orbitals).  The
 textbook formula is very useful, however, in determining what S**2 should
 be.  This is important as a multiplicity=3  calculation is not guaranteed to
 yield the correct S**2 value of 2.000.  This is known as spin contamination,
 as it involves mixing-in of higher spin states.
 **** 3) from Mary O'Connor ****
 Hello--
      A general guideline (or so I have read), is that most organic molecules in
 the
 ground states are closed shells, and thus, have a multiplicity of one and no
 charge. Since multiplicity only means the number of ways that an electron can
 orient itself to a magnetic field, no unpaired electrons means it can only be in
 one orientation. I thonk the formula is 2J + 1 for spin states, where J is +/-
 1/2.
 For closed shells this sums to 1. Then feed any unpaired electrons into the
 formula. I suppose this could be done by drawing Lewis structures to find the
 number of unshared electrons. Sometimes, knowing the charge will preset your
 options for multiplicity in Gaussian. My first instinct is to go with the lowest
 multiplicity choice offered to you, as this will probably be either the correct
 one, or the first excited state. I'm fairly new with Gaussian myself, but these
 are
 the ideas I keep in mind. I will be interested in the other replies that you
 get.
   Good luck!
 **** 4) from Wang Dongqi ****
 Dear Telkuni,
 There is an easy way to find the multiplicity. Just to find the number of
 the unpair electron, so you will get the multiplicity=2S+1.(S=n*1/2, n is
 the number of unpair electron), that is, to molecules and ions, S=0, while
 to radical, S=n*1/2. OK?
 Wish this is useful for you.
 ---------------------------------------------------
        Telkuni Tsuru     telkuni \\at// venus.dti.ne.jp