1
,4-dimethylnaphthalene (9) and 2-methylnaphthalene (12).
acridinium state.13 Fukuzumi and co-workers did not detect
singlet oxygen emission during the photolysis of 1 in
acetonitrile in the presence of 9,10-dimethylanthracene and
thus excluded a singlet oxygen (type II) path for the
formation of the 9,10-dimethylanthracene endoperoxide.5
Under the various reaction conditions used by us for the
chemical detection of singlet oxygen, all substrates with high
oxidation potentials such as 2-methyl-2-butene (2) with an
The former is known to add singlet oxygen in a [4+2]
cycloaddition to give the endoperoxide 10. Under electron-
transfer conditions (e.g., DCA in CH CN), the naphthalene
3
aldehyde 11 is formed as result of arene oxidation, depro-
tonation from the methyl group, and oxygen addition (type
I mechanism). In the less polar solvent methylene chloride,
however, predominately the type II product 10 was observed.
With 1 in CH
2
14
2
Cl
2
, the aldehyde 11 was the major product
Eox of 1.81 V (vs SCE) as well as the regioselectivity probes
(
Scheme 4). More significant were the results from
1-methylcyclohexene (6, Eox ) 1.88 V vs SCE)14 and
limonene (8) did result solely in ene products typical for
type II photooxygenation. The stereochemistry probe li-
monene 8 showed a product composition that was nearly
congruent with results obtained with typical singlet oxygen
sensitizers (such as tetraphenylporphyrin, TPP, in nonpolar
solvents, and rose bengale, RB, in polar solvents). This probe
is highly sensitive for electron-transfer or radical-induced
oxygenation, leading to a different regio- and diastereo-
Scheme 4
1
0
isomeric product composition. Concerning the photocata-
lytic activity of 1, turnover numbers (TON) of 100 were
easily achieved, accounting for efficient energy transfer
sensitizing. This result is in excellent agreement with the
results of the singlet oxygen detection described by Benniston
1
2
et al. The tetrasubstituted alkene 4 with an Eox of 1.50 V
showed only traces of an additional product which might
originate from an electron-transfer-induced oxygenation.
These traces were, however, also detected with RB in
acetonitrile and thus might arise from a radical-type back-
ground reaction. Substrates with oxidation potentials lower
than 1.5 V do, however, show a divergent oxygenation
pattern; i.e., they reveal type II as well as electron-transfer
reactivity. The naphthalene derivatives 9 and 12, respectively,
have oxidation potentials of 1.05 and 1.22 V (vs SCE).
These results indicate that the photocatalytic reactive state
of 1 is also capable of oxidizing substrates with moderate to
low oxidation potentials in competition with singlet oxygen
generation.
2
-methylnaphthalene (12), a substrate that does not react with
singlet oxygen (i.e., no conversion under irradiation condi-
tions in the presence of TPP). Both DCA and 1 gave the
aldehyde 13 and naphthoquinone 14 as major oxidation
products. The efficiency of these two catalysts, however, was
clearly different: whereas the DCA photolyses always
resulted in relatively low conversions, the reactions with 1
as a photocatalyst were complete after standard reaction
times. The solvent effect for the acridinium catalyst 1 was
reversed compared with DCA: efficient oxygenation was
only observed in methylene chloride, whereas the reaction
in acetonitrile was strongly retardedsa phenomenon also
described for the photooxygenation of 4,4′-dimethylbi-
1
5
16
Acknowledgment. Financial support by the Deutsche
Forschungsgemeinschaft (DFG) is acknowledged. The acri-
dinium salt 1 was a gift from Prof. S. Fukuzumi, University
of Osaka.
Supporting Information Available: 1H and 13C NMR
spectra of the photooxygenation experiments with 2, 4, 6,
1
1
phenyl.
There is a lively scientific dispute on whether the remark-
8
, 9, and 11 under singlet oxygen conditions (RB/CH
3
CN
ably long-lived photocatalytically reactive state of 1 is a
charge-transfer state with approximately 2.4 eV pair energy
or the 2 eV acridinium triplet state.12 Verhoeven and co-
workers have detected singlet oxygen by time-resolved
luminescence and report a quantum yield of 42% in aceto-
nitrile, a fact accounting for an energy-transfer active triplet
or TPP/CH Cl ) and with 9-mesityl-10-methylacridinium
2
2
3
perchlorate 1 as the photocatalyst. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL0628661
(13) Benniston, A. C.; Harriman, A.; Li, P.; Rostron, J. P.; van
Ramesdonk, H. J.; Groeneveld, M. M.; Zhang, H.; Verhoeven, J. W. J.
Am. Chem. Soc. 2005, 127, 16054.
(14) Patz, M.; Mayr, H.; Maruta, J.; Fukuzumi, S. Angew. Chem., Int.
Ed. 1995, 34, 1225.
(
11) Suga, K.; Ohkubo, K.; Fukuzumi, S. J. Phys. Chem. A 2005, 109,
0168.
12) Benniston, A. C.; Harriman, A.; Li, P.; Rostron, J. P.; Verhoeven,
1
(
J. W. Chem. Commun. 2005, 2701. Verhoeven, J. W.; van Ramesdonk, H.
J.; Zhang, H.; Groeneveld, M. M.; Benniston, A. C.; Harriman, A. Int. J.
Photochem. 2005, 7, 103.
(15) Santamaria, J.; Ouchabane, R. Tetrahedron 1986, 42, 5559.
(16) For 12, 1.63 V has been reported recently: Sakamoto, M.; Cai, X.;
Hara, M.; Fujitsuka, M.; Majima, T. J. Am. Chem. Soc. 2004, 126, 9709.
Org. Lett., Vol. 9, No. 4, 2007
613