[CHM 371] NICS

[Andrew J. Pounds]

  I know that many of you are starting to work on NICS calculations.    
Some of you have also started to realize that there is a myriad of 
output from Gaussian that you have to sort through.   Here are some 
pieces of advice.

   1. Just as in any experiment work, you must demonstrate first that
      your theoretical method produces values in accordance with what
      has previously been done.  This means that you need to find a
      journal article (preferably the seminal article on the subject)
      and try to reproduce their results.
   2. In the case of NICS you must pay very close attention to the
      GEOMETRY, the THEORETICAL METHOD, and the BASIS SET.  It has been
      surmised that the actual NICS calculated value is less dependent
      on the basis set than the other factors, but that still does not
      change the fact that everything has to be the same to get the same
      result.  Remember what I said in CHM 331 -- a calculation done at
      the same geometry, basis set, and level of theory MUST produce the
      same value.
   3. Look very carefully at how the calculations were done.  In many of
      the papers I have read the geometry was optimized using one method
      and basis set and then the NICS calculation was done using another
      method and basis set using the previously optimized geometry. 
      That being said, if you are going to try and reproduce the work of
      one of these papers then you will have to follow the same
      procedure.  My recommendation is to optimize first and then to the
      optimized structure add the ghost atom for the NICS calculation.
   4. Make sure you specify your ghost atom with the symbol Bq.  The
      symbol X can be used to define dummy atoms for building
      Z-matrices, but Gaussian will throw away those coordinates (and no
      longer use that atom position) once the Z-matrix is complete.  By
      using Bq you are telling Gaussian that you want information about
      the electronic wavefunction as that particular point in the molecule.
   5. A Gaussian NMR calculation by default uses the Gauge-Independent
      Atomic Orbital (GIAO) method.  This is the NMR method used in many
      of the papers so you should not have to specify anything else. 
      What you do have to search for in the Gaussian Output file is the
      magnetic shilding values.  Do not concern yourself with the actual
      values found in the shielding tensor (XX, xY, etc,) but do look at
      the isotropic and anisotropic magnetic shieldings for your ghost
      atom.   You should only need one of these -- but I will let you
      figure that out based on the literature.


-- 
Andrew J. Pounds, Ph.D.  (pounds at theochem.mercer.edu)
Associate Professor of Chemistry and Computer Science
Mercer University,  Macon, GA 31207   (478) 301-5627

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://theochem.mercer.edu/pipermail/chm371/attachments/20111101/23d9289f/attachment.html