Reinaldo,

in reply to your question, there are a few points to make. The following
is,
of course, my view only, and I am not fully sure that I am capable of
providing the entire picture.

(1) The NMR shielding tensor is, from a theoretical point, a purely
electronic
property and not a nuclear property. This means, for instance, that you
can
calculate the shielding at any point in space as is exploited in
Schleyer's
'NICS' method. (Also implemented in ADF.) For your problem, it means
that the
NMR shielding is independent on whether you have proton or deuterium:
the
nuclear charge is the same, so the electronic structure is the same,
too.
Thus, putting a deuterium into your calculation should give you the
exact
same result.
(2) How, then, does the deuterium shift arise? Well, the nuclear
shielding is
extremely sensitive to, among other thigs, the molecular geometry.
Further,
what you measure at a given temperature is not the shielding at the
stationary
point but rather that at the thermally averaged structure. Hydrogen,
being
very light, is the one nucleus where these effects are far more
pronounced
in the shielding than for any other nucleus. Deuterium, due to its mass,
results in slightly different structures which results in a change in
shielding.
(3) So what does this mean for your problem? My take is that you have at
least two options: Either you try to model the anharmonic vibrations
which
gives you a handle on the finite-temperature geometry. Or you try to get
an accurate experimental geometry for both, the original and the deuterated
system, and you calculate the shieldings based on these structures.

I hope that this helps!

Best regards, Georg

> Dear Adf users,
>
> A friend of mine is studying (from the experimental side) the shift in
> the isotropic shieldings that occurs in some organic species when some
> hydrogens are replaced by deuterium atoms.
> I know that ADF allows to change the atomic mass of a given element at
> the "create" level, so I can define a deuterium atom with no problems.
> But,
> can I expect some appreciable changes in the isotropic shieldings
> obtained
> through an NMR job using that approach to simulate isotopic effects?
> Thanks in advance.
> Regards,
>
> Reinaldo

--
==========================================================
Dr. H. Georg Schreckenbach
Department of Chemistry and Biochemistry, Concordia University
Montreal, Quebec, Canada, H3G 1M8
phone: (+514) 848-3352;  FAX: (+514) 848-2868
schrecke_at_alcor.concordia.ca
http://www.geocities.com/gschreckenbach/
==========================================================