From:	Jeremy R Greenwood <jeremy <-at-> med.su.oz.au>
Date:	Tue, 18 Apr 1995 14:26:09 +1000 (EST)
Subject:	Summary: 13C shift and point charge correlation

Summary of responses to question concerning correlation between 13C nmr
shifts and point charge calculations.

Thanks to all who responded.

Original question:

--------------------------------------------------------------------------

Firstly, it occurs to me, that since the chemical shift of a proton or
carbon 13 nucleus depends on electronic shielding (in an nmr experiment),
there should be some (at least loose and qualitative) correlation between
point charges calculated by one or other theoretical method, and
chemical shift (not withstanding solvent effects).

I am attempting to identify the correct structures of some isomeric
heterocycles from their 1H and 13C spectra. Would you expect changes
in point charges to reflect changes in shift between equivalent
centres of the isomers? If so, calculated by which method? I'm using
semi-empirical (AM1, PM3), and HF calculations. My intuition is that
ESP-derived charges are more likely to reflect trends in chemical
shift.

--------------------------------------------------------------------------

Subject: Re: CCL:nmr & point charges / free-radical visualisation
To: jeremy \\at// med.su.oz.au (Jeremy R Greenwood)
Date: Wed, 12 Apr 95 12:04:31 EDT
From: Lisa Emily Chirlian 

	As a graduate student I investigated the relationship between
atomic charges (ab initio electrostatic potential derived) and C-13
NMR chemical shift.  While the trends were correctly predicted,
individual variation was great, so I think the answer to your question
is, no.

	If you had something you could make comparisons with (i.e. a
set of isomers identified some other way), you _might_ be able to make
some approximate correlations for your specific case.

	Lisa
--
-------------------------------------------------------------------------
 Lisa Chirlian					lchirlia (- at -) cc.brynmawr.edu
 Department of Chemistry, Bryn Mawr College, Bryn Mawr, PA 19010
-------------------------------------------------------------------------

From: snyder;at;euch4e.chem.emory.edu
Date: Wed, 5 Apr 1995 18:31:40 -0400 (EDT)
To: jeremy $#at#$ med.su.oz.au
Subject: POINT CHARGES AND NMR SHIFTS

Dear Dr. Greenwood:

    Recalling a discussion with an NMR spectroscopist, I know this to be
in the literature somewhere.  Sorry I don't have the refs.  I can
suggest, however, that you look in J.Am.Chem.Soc. and
Angew.Chem.Int.Edn.Engl. under the name of Paul Schleyer.  He and his
colleagues have lately been applying methods (in particular GIAO-MP2)
for the explicit calculation of NMR chemical shifts.  In some of these
papers the question of charge and shift are discussed.  In JACS 1994,
116, 6386 carbon charges are given for an interesting carbocation for which
both experimental and calculated 13C shifts are recorded.  JACS 1993, 115,
12385 focuses on 11B and 13C exp vs calculated shifts.

   Both of the latter paper point out the importance of including
correlation effects in the calculations.  This is interesting from the
point of view that most charges are derived by calculations that don't
include correlation; i.e. at the HF (Hartree Fock) level.  The
implication is that it may be very dangerous to go to far with the use of
charges as a guide to relative chemical shifts.  I would guess this to
work only for systems that are very close in structure.

   Please send me a summary of the responses your receive.

   Good luck.

     Jim Snyder

*********************************************************************

Dr. James P. Snyder
Emerson Center
Chemistry Department              Tel: (404)-727-2372 or 2380
Emory University                  Fax: (404) 727-6586
1515 Pierce Drive              E-Mail: snyder[ AT ]euch4e.chem.emory.edu
Atlanta, GA  30322 (USA)

*********************************************************************

From: BAELL "-at-" mel.dah.csiro.au (Jonathan Baell)
To: jeremy -AatT- med.su.oz.au (Jeremy R Greenwood)
Subject: Re: CCL:nmr & point charges / f
Date: Tue, 04 Apr 95 13:53

Dear Jeremy,

I'm sure there'll be published work to which you get plenty of
references, but just in case you don't, here's my 1000 lire worth.

I think charge would correlate in certain instances to chemical shift,
but I don't think some effects would be accommodated, particularly
conformationally-related ones.  I would be interested to know, for
example, whether the changes in point charges of the protons of a methyl
group in going from a strong aromatic deshielding zone to a shielding one
could explain the great difference in anticipated chemical shift.

And chemical shift changes between protons and attached carbons, of
course are not necessarily in proportion, connectivity versus field
effects having a great bearing on the ultimate resonance.

Re charge calculation methods: some of this has been worked over the list
in the last year or so.  I think MNDO ESP is the way to go.  Recent
improvements in available AMPAC/MOPAC versions have made the calculations
much faster.  Faster and even better methods have been published and I
think are in the pipeline for the software.  Two recent papers by Ford
and Wang represent breakthroughs (see Bingze Wang and George P. Ford
"Atomic Charges Derived from a Fast and Accurate Method for Electrostatic
Potentials Based on Modified AM1 Calculations, J. Comp. Chem. 15, 200-207
(1994); George P. Ford and Bingze Wang "New Approach to the Rapid
Semiempirical Calculation of Molecular Electrostatic Potentials Based on
the AM1 Wave Function: Comparison with Ab Initio HF/6-31G* Results" J.
Comp. Chem. 14, 1101-1111 (1993)).

Say hello to Hugh

Jonathan Ball

----------------------------------------------------------------------

From: Neil McKelvie 
Subject:      C-13 and other chemical shifts
To: "Jeremy.R.Greenwood" 
Status: OR

 In general, chemical shifts are NOT just dependent on electron densities
around the atom in question. They depend on the magnetic field strength at
this atom's nucleus. The first paper I know about dealing with this was by
H.S.Gutowsky over 30 years ago, on F-19 NMR. This must be in the old Pople,
Bernstein, and Schneider book, which of course is showing its age in many
ways but is still valuable. A simplified explanation is that a magnetic field
causes mixing in of the p-orbitals with m(l) + or -1 (which are complex) into
the real p-orbitals px and py. Now, if there is "p-orbital imbalance" either
from different electronegativities of attached atoms, or because one p-orbital
is involved in pi-bonding and the others in sigma-bonding, you get more of the
p(+1) mixed in, and so a considerable paramagnetic effect.  This is why the
C-13 shifts of carbonyl groups are so far downfield, not just because of any
partial (+) charge on the carbon. Attached protons are affected also - look at
the downfield shift of alkene hydrogens.
  (Prof.) Neil McKelvie; Chemistry Dept., The City College; NY,NY 10031 USA
(Voice) 212-650-6062 (Fax) 212-650-6107 NEMCC # - at - # CUNYVM.CUNY.EDU

----------------------------------------------------------------------------

Jeremy Greenwood
Adrien Albert Laboratory of Medicinal Chemistry
Department of Pharmacology
University of Sydney
NSW 2006 Australia


Similar Messages
07/07/1995:  Summary: Opinions on the quantum charges
05/01/1997:  calculation of chemical shifts (summary)
04/03/1995:  nmr & point charges / free-radical visualisation
08/01/1996:  Re: CCL:M:Heat of formation calculation using MOPAC.
10/02/1996:  SUMARY:PD atomic charges
02/21/1995:  charges again
08/01/1995:  Spin contamination, effect on energy and structure.
02/16/1995:  RE: charges schemes with CoMFA
07/31/1999:  Re: CCL:SUMMARY OF NMR SHIFTS, MP2 cf. HF
10/09/1991:  NMR chemical shifts from structure -"summary"


Raw Message Text