is consistent with an enol-mediated reaction; we were
surprised by the quality of the correlation, since the fraction
of enol formed must change with the solvent. Most likely
3
absence and presence of 1% acetic acid (CH COOD when
appropriate), as shown in Figure 3. The large isotope effect
the enol concentration also correlates reasonably well with
H
2
the hydrogen bond acceptor properties of the solvent (â ).
The rate constants for nitroxyl radical reaction with 1 in
light hydrocarbons are about four times what could be
predicted by extrapolating the data in other solvents, when
H
2
the data are plotted against â . This may be due in part to
the trapping of the radical by 3 in nonpolar solvents,
according to Scheme 4. While our evidence for 7 is
Scheme 4. Hydrogen Abstraction from 1/5 by the
Prefluorescent Probe 3
Figure 3. Reaction of 3 with 1 in methanol (4), methanol-O-d
(b), in methanol containing 1% acetic acid (red O), and in
3
methanol-O-d containing 1% CH COOD (red b). Plotted according
to eq 4.
(
∼10, see Table 1) and the absence of any significant effect
11,17
by addition of acetic acid
is consistent with a mechanism
inconclusive, its formation in nonpolar solvents may explain
the increase in rate constant. It should also be noted that
while the hydrocarbon solvents are not hydrogen bond
acceptors, the probe itself may play to some degree this role,
despite its low concentration. 14 Nitroxyl radicals also show
significant solvent effects, at least in their trapping of carbon-
centered radicals.15
involving simple hydrogen abstraction from enol 5.
Some earlier observations of substituent dependence of
the reaction of alkoxyl radicals with lactones may now be
5
easier to explain. For example, replacement of the aryl
substituent on the lactone ring leads to a significant increase
in the rate constant for hydrogen abstraction, which was not
easily explained before; it now appears likely that the
observed increase in rate constant may result from changes
in the abundance of the enol form.
In conclusion, our results show that the dominant hydrogen
donor in the case of 1 is not the lactone commercialized as
HP-136, but rather the enol form in equilibrium with the
lactone, in this way making it analogous to many phenolic
antioxidants where the H-donor is the OH group. In effect,
the lactone form is not an antioxidant, but rather its enol is.
Despite this, once the hydrogen transfer has occurred, the
radical (2) behaves predominantly as a carbon centered
radical, but with little or no reactivity toward oxygen.
The fast exchange observed in polar solvents is consistent
with the presence of significant enol concentration in these
solvents which is reflected in the rate constant (major
interaction with HBA solvents). The strong solvent effect
dependence among the solvents employed shows that the
enol form is the reactive species in agreement with the
3,16
behavior observed for phenolic compounds.
The conclusion that the hydrogen donor is 5 rather than 1
led us to explore if hydrogen transfer could be mediated by
electron transfer (i.e., electron transfer from the enol anion
followed by protonation), as has been established to occur
11
with DPPH (another common probe) in alcoholic solvents.
This was done by monitoring the progress of reaction 5
according to eq 4, for methanol and methanol-O-d in the
Acknowledgment. We acknowledge the generous finan-
cial support from the Natural Sciences and Engineering
Research Council of Canada. A.A. thanks the Fundaci o´ n
Andes (Chile). We thank Dr. K. U. Ingold for valuable
comments on earlier versions of this contribution.
(12) This corresponds to an H/D isotope effect of e1.5. The reactive
species clearly is formed without requiring that the C-H/C-D exchange
be complete. Unfortunately, it is virtually impossible to carry out laser
experiments in less than 30 s following D2O addition.
(
13) Frenette, M.; Aliaga, C.; Font-Sanchis, E.; Scaiano, J. C. Org. Lett.
004, 6, 2579-2582.
14) Product 7 is expected to be rather unstable and if formed does not
Supporting Information Available: Experimental pro-
cedures and additional details of kinetic studies by laser flash
photolysis. This material is available free of charge via the
Internet at http://pubs.acs.org.
2
(
survive chromatographic analysis. Further, the concentrations of probe used
are too low for NMR analysis. At high concentration dimer 6 is the final
product.
(
15) Beckwith, A. L. J.; Bowry, V. W.; Ingold, K. U. J. Am. Chem. Soc.
992, 114, 4983-4992.
16) Wayner, D. D. M.; Lusztyk, E.; Page, D.; Ingold, K. U.; Mulder,
P.; Laarhoven, L. J. J.; Aldrich, H. S. J. Am. Chem. Soc. 1995, 117, 8737-
744.
OL050839P
1
(
(17) Foti, M. C.; Daquino, C.; Geraci, C. J. Org. Chem. 2004, 69, 2309-
2314; Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888-5896.
8
3668
Org. Lett., Vol. 7, No. 17, 2005