CCL:G: Dipole moment of charged species in Gaussian

On Wed, May 8, 2013 at 4:03 PM, Herbert Fruchtl herbert.fruchtl,+,"> <> wrote:

Sent to CCL by: Herbert Fruchtl []
Intuitively, I would center any multipole analysis at the centre of charge. The centre of atomic (not just nuclear!) charges, calculated in any partition scheme (Mulliken, Hirshfeld, Bader, ...) should be a reasonable approximation of this.


This can only work when you have an overall charged molecule, as for the calculation of the origin for all of these quantities (nuclear charge, mass, charge etc) one divides by the total quantity, which is zero for an uncharged molecule. In the case of an ion, 'the centre of charge' incidentally is defined as that origin, for which the total dipole is zero (see Stone, The Theory of Intermolecular Forces, p. 33)



On 08/05/13 12:58, Reis Heribert wrote:
Hello John,

In my opinion the choice of the origin for multipoles is most of the time just a
matter of what one is trying to investigate. So a choice like the one you
propose may be interesting in questions related to, say, chemical reactivity.

But apart from that, it is not quite clear to me how you would go about
computing such an origin. Computing the electron density due to the valence
orbitals on a grid? Or do you have another idea in mind? Maybe something related
to Hirshfelds' free atom densities?


On Tue, May 7, 2013 at 8:53 PM, John McKelvey
<> <owner-chemistry!^!

<mailto:owner-chemistry!^!>> wrote:


    Perhaps there is another way to look at it, and it is a subset of looking at
    the center of nuclear charge.  Perhaps it is more sensitive (?) to look at
    the center of  constituent free atom valence electrons?

    John McKelvey

    On Tue, May 7, 2013 at 12:26 PM, Reis Heribert
    <> <owner-chemistry/./

    <mailto:owner-chemistry/./>> wrote:


        The definition of the dipole moment of a charged species is well
        defined. The only problem with it (and with all higher multipoles) is,
        that its value depends on the origin of the coordinate system. Thus for
        comparison purposes the origin of the coordinate system of different
        molecules should be comparable. A very common choice is the centre of
        mass; another possibility is the center of nuclear charge, which is the
        standard origin in Gaussian09, if one does not specify the 'Nosymm'
        keyword. You can also turn the problem on its head and define the origin
        as that point where the dipole is zero...

        I'm assuming that you want to compare the absolute value of the dipole
        between different molecules. In case you also want to compare the
        components, you also should use comparable orientations of the
        coordinate systems (e.g the coordinate system where e.g. the inertia (or
        polarizability or ...) tensor is diagonal). This, however, is
        independent of being charged or uncharged.


    John McKelvey
    10819 Middleford Pl
    Ft Wayne, IN 46818
    260-489-2160 <tel:260-489-2160>
    jmmckel/./ <mailto:jmmckel/./" target="_blank">>

Herbert Fruchtl
Senior Scientific Computing Officer
School of Chemistry, School of Mathematics and Statistics
University of St Andrews
The University of St Andrews is a charity registered in Scotland:
No SC013532

-= This is automatically added to each message by the mailing script =-
E-mail to subscribers: or use:

E-mail to administrators: or use

Before posting, check wait time at:

Job: Conferences:

Search Messages: