From owner-chemistry@ccl.net  Sat Dec 16 12:33:28 1995
Received: from bedrock.ccl.net  for owner-chemistry@ccl.net
	by www.ccl.net (8.6.10/950822.1) id MAA05753; Sat, 16 Dec 1995 12:28:55 -0500
Received: from UMSLVMA.UMSL.EDU  for wwelsh@jinx.umsl.edu
	by bedrock.ccl.net (8.7.1/950822.1) id MAA17676; Sat, 16 Dec 1995 12:28:52 -0500 (EST)
Received: from admiral.umsl.edu by UMSLVMA.UMSL.EDU (IBM VM SMTP V2R2) with TCP;
   Sat, 16 Dec 95 11:30:24 CST
Received: from [134.124.76.179] by admiral.umsl.edu (5.4R3.10) id AA15583; Sat, 16 Dec 1995 11:28:50 -0600
Message-Id: <9512161728.AA15583@admiral.umsl.edu>
X-Sender: wwelsh@jinx.umsl.edu
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sat, 16 Dec 1995 11:28:35 -0600
To: chemistry@ccl.net
From: wwelsh@jinx.umsl.edu (William Welsh)
Subject: ESP versus NBO charges


Fellow Netters,

When NBO- and ESP-calculated partial charges disagree substantially, which
values are more "realistic" or acceptable?  Is there any known basis for
choosing one set over another? Are they conveying different information?  I
realize that 'partial charge' is not truly a measurable physical quantity
and that, therefore, these two approaches represent different paradigms for
dividing up the charge density.  

When I calculate the partial charge distribution of several organic esters 
(e.g., R-C(X)0R', where X = S or O) or of the corresponding anionic
tetrahedral intermediates (e.g., R-C(X)(OH)-OR') 6-31G*-optimized geometries
using ESP and NBO (in Spartan), sometimes I obtain vastly different charge
distributions.

Thanks for any replies ....

Bill Welsh
Dept. of Chem.
Univ. of Missouri-St. Louis