Photooxidation of a conjugated diene by an exciplex mechanism: Amplification via radical chain reactions in the perylene diimide-photosensitized oxidation of α-terpinene

Article abstract of DOI:10.1021/jp983346t

Irradiation of the perylene diimide (1) or 9,10-dicyanoanthracene (DCA) in the presence of α-terpinene (2-HH) in the presence of molecular oxygen leads to moderately efficient oxidation of 2-HH to p-cymene (2). Although 1 might be expected to photosensitize oxidations by the conventional "singlet oxygen pathways, spectroscopic studies indicate that while oxygen can quench the fluorescent singlet of 1, no singlet oxygen is produced. 2-HH is also an efficient quencher of the fluorescent singlets of 1 and DCA and, for nonpolar solvents such as methylene chloride, in each case the quenching results in formation of an exciplex or contact radical ion pair. Under conditions where quenching by 2-HH to form the exciplex is complete, maximum quantum yields of 2 are obtained, thus indicating that the exciplex is the precursor to its formation. Nonproductive decay of the exciplex to starting materials is its major fate, thus the moderately high quantum efficiencies for formation of 2 require a mechanism involving amplification. Spin-trapping experiments suggest the role of hydroperoxy radicals and amplification by a radical chain mechanism involving these radicals and the intermediate 2-H^. is proposed. Possible paths for reaching these radicals from the exciplex are considered; either oxygen quenching of the exciplex or proton transfer within the exciplex followed by oxygen interception of the semireduced perylene diimide appear viable. For the reaction of DCA with 2-HH and oxygen, it is found that the much longer-lived exciplex undergoes quenching by oxygen.

Full text of DOI:10.1021/jp983346t

J. Phys. Chem. A 1998, 102, 9095-9098
9095
Photooxidation of a Conjugated Diene by an Exciplex Mechanism: Amplification via
Radical Chain Reactions in the Perylene Diimide-Photosensitized Oxidation of r-Terpinene
Liaohai Chen,‡ Lucian A. Lucia,§ E. R. Gaillard,† and David G. Whitten*,‡
Department of Chemistry and NSF Center for Photoinduced Charge Transfer, UniVersity of Rochester,
Rochester, New York 14627
H. Icil
Department of Chemistry, Eastern Mediterranean UniVersity, Famagusta, N. Cyprus, Mersin 10, Turkey
S. Icli
Department of Chemistry, Faculty of Science, Ege UniVersity, 35100 BornoVa, Izmir, Turkey
ReceiVed: August 11, 1998
Irradiation of the perylene diimide (1) or 9,10-dicyanoanthracene (DCA) in the presence of R-terpinene (2-
HH) in the presence of molecular oxygen leads to moderately efficient oxidation of 2-HH to p-cymene (2).
Although 1 might be expected to photosensitize oxidations by the conventional “singlet oxygen” pathways,
spectroscopic studies indicate that while oxygen can quench the fluorescent singlet of 1, no singlet oxygen
is produced. 2-HH is also an efficient quencher of the fluorescent singlets of 1 and DCA and, for nonpolar
solvents such as methylene chloride, in each case the quenching results in formation of an exciplex or contact
radical ion pair. Under conditions where quenching by 2-HH to form the exciplex is complete, maximum
quantum yields of 2 are obtained, thus indicating that the exciplex is the precursor to its formation.
Nonproductive decay of the exciplex to starting materials is its major fate, thus the moderately high quantum
efficiencies for formation of 2 require a mechanism involving amplification. Spin-trapping experiments suggest
the role of hydroperoxy radicals and amplification by a radical chain mechanism involving these radicals and
the intermediate 2-H• is proposed. Possible paths for reaching these radicals from the exciplex are considered;
either oxygen quenching of the exciplex or proton transfer within the exciplex followed by oxygen interception
of the semireduced perylene diimide appear viable. For the reaction of DCA with 2-HH and oxygen, it is
found that the much longer-lived exciplex undergoes quenching by oxygen.
Introduction
R-terpinene by 1. The results of this study indicate a novel
photooxidation that proceeds by an initial electron transfer
quenching of the dye singlet by the diene, followed by a reaction
of this exciplex (or products derived from it) with oxygen. The
subsequent amplification through a radical chain sequence
allows the overall reaction to occur with moderate efficiency
despite what must be a highly inefficient reaction from the
exciplex.
Photooxidations proceeding by dye sensitization of singlet
oxygen are regarded as almost ubiquitous for aerated organic
solutions containing dyes and electron-rich substrates such as
alkenes, dienes, and amines.1-4 We have recently found that
irradiation of aerated acetonitrile or methylene chloride solutions
of perylene diimides, such as 1 with the conjugated dienes
abietic acid and R-terpinene (2-HH), results in formation of
product distributions, for the former, similar to those obtained
when the same substrate is irradiated in the presence of well-
known singlet oxygen sensitizers such as methylene blue or
eosin.5 While it was anticipated that the perylene diimides
should be potential photosensitizers for singlet oxygen and a
“conventional” singlet oxygen mechanism seemed most likely
for these photooxidations, we were surprised to find that no
Experimental Section
Reagents were generally purchased from Aldrich Chemical
Co. and were used as received unless noted otherwise. R-Ter-
pinene was purified by three separate distillations prior to use.
9,10-Dicyanoanthracene (DCA), a gift from Dr. Samir Farid at
Eastman Kodak, was used as received. The heterocyclic fulgide
actinometer (Aberchrome 540) was purchased from Aberchro-
mics Ltd. The synthesis of perylene bis(2,5-di-phenyl)imide
was previously described.6
1
evidence could be obtained for the sensitization of O2* by 1
or related imides, despite the moderately high quantum efficien-
cies for the above reactions. In the present paper we report a
detailed study of the sensitization of the photooxidation of
The degassed samples were prepared by six cycles of freeze-
pump-thaw degassing and sealed under high vacuum (ca. 1 ×
10-5 Torr). A Hewlett-Packard 8453 diode array spectropho-
tometer and a Spex 111A Fluorolog 2 spectrofluorimeter were
used to record the UV-vis and fluorescence spectra, respec-
tively. GC-MS experiments were carried out on a HP 5890
† Current address: Department of Chemistry, Northern Illinois University,
DeKalb, IL 60115-2862.
‡ Current address: CST-1 MS J565; Los Alamos National Laboratory,
Los Alamos, NM 87545.
§ Current address: Institute of Paper Science and Technology, Georgia
Institute of Technology, Atlanta, GA 30318-5794.
10.1021/jp983346t CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/21/1998

9096 J. Phys. Chem. A, Vol. 102, No. 45, 1998
Chen et al.
series gas chromatograph equipped with a 30 m HP-5 high-
performance capillary column (cross-linked 5% Ph Me Silicone)
and an HP 5970 series mass-selective detector. A Bruker ESR
spectrometer was used for the ESR spin-trapping experiments.
The ESR spectrometer was operated at 100 kHz modulation,
the microwave power was 20 mW, and the modulation
amplitude was 0.5 G. 5,5-dimethyl-1-pyrroline-N-oxide (DMPO)
was used as the spin-trapping reagent. Hyperfine coupling
constants were obtained by simulating and optimizing the
experimental spectra using a commercial ESR data program
(Keithley Instrument, Inc.). High-performance liquid chroma-
tography was used to follow the photooxidation reaction for
R-terpinene with the perylene diimide. HPLC analysis was
performed using a HP 1050 series liquid chromatograph system
with a pump system, an autosampler, and a diode-array detector.
A Waters Nova-Pak C18 column was used with acetonitrile as
the mobile phase.
Steady-state quantum yield measurements for the formation
of p-cymene were carried out by HPLC analysis. Typically, a
solution of 1 × 10-5 M perylene and 0.31 M R-terpinene in
acetonitrile (in which case, 95% of the fluorescence from
perylene diimide was quenched by R-terpinene) was irradiated
at room temperature using a 200 W high-pressure mercury lamp
with a Corning 3-71 cutoff filter (λ > 460 nm) and a 509 nm
interference filter. The progress of the reaction was followed
by monitoring the p-cymene peak with HPLC. A correlation
plot for the peak area of pure p-cymene as a function of
concentration was made so that the formation of product could
be monitored quantitatively. Irradiation time was controlled by
periodically performing HPLC analysis to achieve ca. 1-5%
conversion of the starting material. Heterocyclic fulgide 540
actinometry was used to determined the excitation light intensi-
ties.
tions using phenalenone10 as a sensitizer quite clearly demon-
strated the presence of 1O2* from its phosphorescence. In
addition, laser flash interrogation of nondegassed methylene
chloride solutions of 1 showed no evidence for the triplet-
triplet absorption of 1 at 500 nm which should have arisen if
the singlet sensitization of oxygen had occurred.
As indicated above, irradiation of 1 in the presence of oxygen
and the above-mentioned dienes does lead to the formation of
typical photooxidation products. In the case of R-terpinene, the
chief oxidation product detected under irradiation of 1 at 509
nm is the corresponding aromatic hydrocarbon, p-cymene (2).11
HPLC product studies and steady-state quantum yield measure-
ments showed that the overall transformation to 2 occurs only
in the presence of oxygen and with no consumption of 1. In
degassed solution no products corresponding to 2 were observed.
Clearly oxygen serves to oxidize 2-HH to 2 in a fairly efficient
net two-electron process that is driven by formation of the
aromatic product. Interestingly, 2-HH and related dienes have
been found to form mostly endoperoxide adducts under sensitiz-
ing conditions with other dyes known to produce singlet
oxygen,12,13 whereas our results indicate a clean, relatively high
yield of the aromatic product 2 with a quantum yield of 30-
40%. Since it had been shown several years ago that conjugated
dienes such as 2-HH quench the fluorescence of aromatic
hydrocarbons by an “exciplex” mechanism,14-16 it appeared
possible that the initial step in the sensitized reaction might be
a fluorescence quenching of 1 by 2-HH. The quenching
constant for the fluorescence of 1 by 2-HH in methylene
chloride was found to be 1.3 × 1010 M-1 s-1. Since 2-HH is
a liquid and infinitely soluble in methylene chloride, it was easy
to conduct experiments in which all of the excited 1 is quenched
by 2-HH to the exclusion of all other potential excited-state
pathways for the photooxidation. Under these conditions, with
oxygen-saturated solutions, a limiting high quantum yield for
the formation of 2, 0.4, was obtained. Since the fluorescence
quenching should produce a contact radical ion-pair or exciplex
([1•-, 2-HH•+], ∆G ∼ -0.1 V), it was logical to interrogate
the system with laser flash photolysis.
Picosecond transient absorption spectra of both degassed and
nondegassed methylene chloride solutions of 1 and 2-HH
display the same moderately strong signal centered at 705 nm
corresponding to absorption of the radical anion of 1 (decay
constant 1.5 × 1010 s-1). The singlet-singlet absorption of 1
is also in the same region, but excess 2-HH (4.2 M) quenches
the lifetime of 1 to essentially within the instrumental response
function (<40 ps).
From the above results it is possible to outline possible
mechanistic paths for the overall reaction. The involvement of
1O2* can be completely eliminated from the lack of phospho-
rescence and the conditions (high concentration of 2-HH,
rendering all quenching of 1* via the exciplex). Thus, exciplex
(or contact radical ion pair) formation by electron-transfer
quenching (eq 1) and subsequent reactions of the exciplex must
Singlet oxygen phosphorescence was probed by laser flash
photolysis.7 The emission from singlet oxygen phosphoresence
was monitored at 1270 nm with the detector perpendicular to
the laser excitation using a Ge diode detector. The output was
digitized with a Tektronix TDS 620 digitizing oscilloscope and
then analyzed by computer. The transient absorption spectra
were obtained by picosecond time-resolved laser flash photoly-
sis. The instrumental setup has been described in the literature8
and the fast kinetic decay analysis was treated by home-built
user software (SIMPLEX, Version 3) in which a first-order
exponential fit was applied to the data in conjunction with the
deconvolution of the instrument response.
Results and Discussion
The perylene diimides such as 1 show strong absorption in
the visible and fluorescence quantum yields near unity; the high
fluorescence efficiency correlates with a relatively large S1-
T1 gap (27 kcal/mol) which should be sufficient to promote the
3
1
spin conversion, O2 to O2*, which requires 22 kcal/mol.9
Although oxygen quenches the fluorescence of 1 at a near
1* + 2-HH f 1•-,2-HH+•
1•-,2-HH+• f 1•- + 2-HH•+
1•-,2-HH•+ f 1-H• + 2-H•
(1)
(2)
(3)
1•-;2-HH•+ + O2 f 1,2-HH•+,O2•- f 1 + 2-H• + O2H•
(4)
diffusion-controlled rate (kq ) 6 × 109 M-1 s-1 in acetonitrile),
no singlet oxygen phosphorescence can be detected under these
conditions, whereas control experiments under the same condi-
be responsible for the eventual formation of product 2 in
moderately high efficiency. Some reasonable possibilities for

Photooxidation of a Conjugated Diene
J. Phys. Chem. A, Vol. 102, No. 45, 1998 9097
reaction of the exciplex, [1•-., 2-HH•+], leading toward product
include dissociation or cage escape, eq 2; proton transfer within
the exciplex, eq 3; or reaction with oxygen, eq 4. However,
each or all of these reactions must proceed with very low
efficiency since no intermediates other than the exciplex are
detectable by laser flash photolysis and the rapid decay of the
exciplex (τ ) 67 ps), even in the presence of oxygen, results
mostly in regeneration of the ground states of 1 and 2-HH.
Due to the relatively high efficiencies of formation of 2 (0.3-
0.4), there must be some kind of amplification mechanism which
can be initiated by one or more of the above steps. In fact, the
subsequent reaction of intermediates generated in eqs 2-4 can
lead to the observed products in moderate efficiencies since for
each of these reactions there exist paths to radicals which may
participate in chain reactions providing such an amplification
mechanism. Thus cage escape (eq 2), followed by oxygen
interception of the perylene anion radical 1•-, can lead to
superoxide (eq 5), which, together with the radical cation
H + O2
+ O2H•
(9)
H
•
H
∆Hr = –31.9 kcal/mol
+ O2H•
H + H2O2
(10)
H
•
H
∆Hr = –18.8 kcal/mol
amplification that allows the reaction to proceed with moderate
efficiency. The major mechanistic question is which of the paths
from the exciplex to the pair of radicals is most probable. The
rate constant for cage escape for oppositely charged ions from
a solvent-separated radical ion pair in methylene chloride is
estimated to be 3 × 108 s-1 28-30
.
Considering the measured
rate for exciplex decay, this decay path should provide for at
most 1-2%. However, since the mechanism involving cage
escape and subsequent capture of 1•- by oxygen31 involves
additional steps, most likely involving bimolecular reaction
between intermediates present in very low concentrations (e.g.,
O2•- and 2-HH•+) to reach the chain carrying radicals, we think
it unlikely to be a major reaction path.
•-
1
•- + O2 f 1 + O2
(5)
2-HH•+, could lead to the formation of chain-carrying radicals
2-H• and OOH•. The same radicals may be generated directly,
if oxygen reacts with a semireduced perylene radical formed
from proton transfer within the exciplex (eqs 3 and 6) or from
The second possible path between the exciplex and the chain-
carrying radicals involves proton transfer within the exciplex
(eq 3) and subsequent interception of 1-H• by oxygen. This
path appears reasonable, since the cation radical 2-HH•+ and
anion radical 1•- should be anticipated to be strong acids and
bases, respectively.32,33 However, we have no estimate for the
rate of eq 3 and hence its importance as a decay path for the
exciplex. The fact that the perylene functions as a catalyst and
no products from its degradation are produced indicates that if
1-H• is generated, its capture by oxygen must be very efficient.
The third possible path between the exciplex and the radicals
involved in eqs 7 and 8 involves direct oxygen quenching of
the exciplex. However, as the laser flash photolysis results
indicate, the very short exciplex lifetime is the same, within
experimental error, for aerated and degassed solutions. It is
not surprising that oxygen does not ostensibly affect the decay
rate of the exciplex/ion-pair (lifetime ∼67 ps) since if it assumed
to quench the ion-pair according to eq 4 at an upper limit
diffusional quenching rate (ca 1 × 1010 M-1 s-1), the effective
quenching is limited to no more than 0.6%, well within
experimental error.34
Based on the evidence available both the second and third
mechanisms described above are reasonable and for the present
system there is no basis for determining whether either or both
are predominant. However for a related system, it is possible
to obtain evidence supporting a potential role of the third
mechanism. Since it is known that irradiation of DCA/
diphenylethylene in aerated solutions gives rise to superoxide,
presumably by oxygen quenching of the exciplex,35 we inves-
tigated the possibility that a similar reaction to that sensitized
by 1 might occur with enhanced quantum efficiencies when
solutions containing DCA and 2-HH are irradiated. Indeed,
laser flash excitation of DCA in the presence of 2-HH under
complete quenching conditions in methylene chloride led to
detection of a transient assigned to the DCA,2-HH exciplex
which has a lifetime of 500 ps in degassed solutions. In air-
saturated solution the same transient was observed but with a
reduced lifetime (300 ps), which is consistent with a Stern-
Volmer quenching constant of 70, assuming diffusion controlled
kinetics with oxygen. 2 is the only photoproduct detected under
these conditions, and its quantum yield, measured upon steady-
1-H• + O2 f 1 + OOH•
(6)
decay of the exciplex-oxygen quenching complex (eq 4).
That OOH• and 2-H• are involved in the reaction seems clear
from the following. ESR studies were carried out with 1, 2-HH,
and excess DMPO (5,5-dimethyl-1-pyrroline-N-oxide) as a spin
trap. Very strong ESR signals were recorded which could be
attributed to either the superoxide-DMPO adduct or the
protonated superoxide-DMPO adduct (g-value ) 2.0085, aN )
14.5, aHB ) 11.6, aHγ ) 1.2 and aN/aHγ ) 1.25).17-19 No signals
were observed for corresponding degassed solutions. Whether
the species trapped by DMPO is superoxide or the hydroperoxyl
radical formed via eqs 3 and 4 (or direct scavenging of a radical
by atom transfer) cannot be distinguished; however, the trapping
to produce only the single species corresponding to an oxygen
adduct provides support for a mechanism such as that shown
(eqs 7 and 8). Hydrogen peroxide, the other final product
2-H• + O2 f 2 + OOH•
(7)
(8)
OOH• + 2-HH f H2O2 + 2-H•
besides 2, according to the sequence proposed, was also detected
by the iodide-thiosulfate procedure.20,21
The proposed chain carrying steps which could provide the
amplification necessary to reach the moderately high observed
quantum yield are shown in eqs 7 and 8. Approximate heats
of reaction for these reactions can be obtained by reference to
eqs 9 and 10, which estimate heats of reaction for propagation
steps, based on the transformation of 1,3-cyclohexadiene to
benzene.22-25 These heats of reactions are sufficiently exo-
thermic to lend support for the sequence proposed. The chain
propagation steps may be primarily limited by the rate and
efficiency of eq 9 which proceeds at a rate of 8 × 105 s-1 for
the cyclohexadiene-benzene case.26,27
From a mechanistic perspective, it thus seems reasonable that
an initial exciplex formation can be followed by generation of
the chain-carrying radicals 2-H• and OOH• and that their
reaction through propagation steps 7 and 8 provides the

9098 J. Phys. Chem. A, Vol. 102, No. 45, 1998
Chen et al.
(15) Stephenson, L. M.; Whitten, D. G.; Hammond, G. S. Chem. Ioniz.
Excitation 1967, 7, 695.
(16) Penner, T. L.; Whitten, D. G.; Hammond, G. S. J. Am. Chem. Soc.
state irradiation of the air-saturated solutions, is 0.9, a value 3
times higher than that observed for the system 1-2-HH. Here
again, the relatively high quantum yield under conditions where
only a moderate fraction of the initially excited molecules can
be intercepted by oxygen points to amplification by a chain
process.
1970, 92, 2861.
(17) Reszka, K.; Bilski, P.; Chignell, C. F. Free Radicals Res. Commun.
1992, 17, 377.
(18) Saito, I.; Matsuura, T.; Inoue, K. J. Am. Chem. Soc. 1983, 105,
3200.
The results reported here suggest that related mechanisms
involving coupling of radical chain amplification with inefficient
generation of superoxide or hydroperoxy radicals may be
important in a number of other reactions involving dye fading
or other photoredox reactions where production of singlet
oxygen is unlikely or impossible.
(19) Harbour, J. R.; Hair, M. L. J. Phys. Chem. 1978, 82, 1397.
(20) Kokatnur, V. R.; Jelling, M. J. Am. Chem. Soc. 1941, 63, 1432.
(21) Russell, G. A. J. Am. Chem. Soc. 1957, 79, 3871.
(22) The addition of para-alkyl substituents will impart even greater heats
of reaction than those predicted in eqs 6 and 7 since the products will be
more stabilized than the parent benzene.
(23) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 71st
ed.; CRC Press: Boca Raton, FL, 1990.
(24) O’Neal, H. E.; Benson, S. W. In Free Radicals; Kochi, J. K., Ed.,
1973; p 275.
(25) Shum, L. G. S.; Benson, S. W. J. Phys. Chem. 1983, 87, 3479.
(26) Pan, X.-M.; Schuchmann, M. N.; Sonntag, C. von, J. Chem. Soc.,
Perkin Trans. 2 1993, 1021.
(27) The hydroperoxyl radical has been shown to be a major chain carrier
in related reactions: Howard, J. A.; Ingold, K. U. Can. J. Chem. 1967, 46,
785.
Acknowledgment. We thank the US Department of Energy
(Grant DE-FG02-86ER13504) and NATO Scientific Affairs
Division (Grant CRG 950257) for support of this research. We
thank Dr. Samir Farid of Eastman Kodak for helpful suggestions.
References and Notes
(28) Farid, S. Private communication.
(29) Arnold, B. R.; Farid, S.; Goodman, J. L.; Gould, I. R. J. Am. Chem.
Soc. 1996, 118, 5482.
(30) Arnold, B. R.; Noukakis, D.; Farid, S.; Goodman, J. L.; Gould, I.
R. J. Am. Chem. Soc. 1995, 117, 4399.
(1) For reviews see refs 2-4.
(2) Griesbeck, A. G. In CRC Handbook of Organic Photochemistry
and Photobiology; Horspool, W. M., Song, P.-S., Eds.; CRC Press: Boca
Raton, FL, 1994; pp 301-310.
(3) Adam, W.; Griesbeck, A. G. In CRC Handbook of Organic
Photochemistry and Photobiology; Horspool, W. M., Song, P.-S., Eds.; CRC
Press: Boca Raton, FL, 1994; pp 311-324.
(31) The reduction potential of 1 is ∼ -0.85 V vs SCE, thus the electron
transfer from 1•- to O2 may be slightly uphill (∆G ∼+0.07 V), but
considering the relative ratios of reactants, the transfer is feasible. In
nanosecond laser flash photolysis experiments we irradiated (532 nm) a
perylene diimide in acetonitrile with quenching by pentamethyl benzene
(0.7 M; a concentration quenching > 90% of the perylene excited states).
Under these conditions, in degassed solutions, a second-order decay
attributed to ion-radical recombination following cage escape (efficiency
∼ 0.6) is observed extending over 800 us. This decay is accelerated when
oxygen is added [oxygen saturated; ([O2] ∼ 0.01 M)]; from the differences
in decay in oxygen saturated vs degassed solutions, we estimate the rate
constant for the reaction of 1•- with O2 to be k ∼ 1 × 105 M-1 s-1. In
none of the experiments where we observe the anion radical of 1 do we
detect any spectral shifts which would be consistent with its decay by
processes other than return electron transfer.
(4) Rigaudy, J. In CRC Handbook of Organic Photochemistry and
Photobiology; Horspool, W. M., Song, P.-S., Eds.; CRC Press: Boca Raton,
FL, 1994; pp 325-334.
(5) Krohn, K.; Budianto, E.; Florke, U.; Hausen, B. M. Liebigs Ann.
Chem. 1992, 3, 911-919.
(6) Icli, S.; Uysal, F. Tr. J. Chem. 1992, 16, 307.
(7) Rodgers, M. A.; Snowden, P. T. J. Am. Chem. Soc. 1982, 104,
5541.
(8) Chen, H.; Farahat, M.; Law, K. Y.; Perlstein, J.; Whitten, D. G. J.
Am. Chem. Soc. 1996, 118, 2586.
(9) Ford, W. E.; Kamat, P. V. J. Phys. Chem. 1987, 91, 6373.
(10) Schmidt, R.; Tanielian, C.; Dunsbach, R.; Wolff, C. J. Photochem.
Photobiol. A 1994, 79, 11.
(11) Both GC-MS and HPLC studies show the occurrence of three
products; the major one (ca. 90%) was confirmed to be 2 by NMR. On the
basis of the literature (Pan, X.; Schuchmann, M.; Sonntag, C. J. Chem.
Soc., Perkin Trans. II 1993, 289) we assume the minor products may be
the corresponding 1,4-cyclohexadiene and an endoperoxide.
(12) Schenck, G. O.; Ziegler, K. Naturwissenshaften 1944, 32, 157.
(13) Matusch, R.; Schenck, G. O. Angew. Chem. 1988, 100, 729.
(14) Stephenson, L. M.; Whitten, D. G.; Vesley, G. F.; Hammond, G.
S. J. Am. Chem. Soc. 1966, 88, 3665.
(32) Nicholas, A. M.; Arnold, D. R. Can. J. Chem. 1982, 60, 2165.
(33) Ci, X.; Whitten, D. G. J. Am. Chem. Soc. 1987, 109, 7215.
•-
•-
(34) If O2 were formed from 1•-, following cage escape, or if O2
escaped from the cage before reaction 3 occurred, the corresponding out-
of-cage reaction should be very slow since, under steady-state conditions,
the concentrations of the two intermediates should be very low. However,
other more complex sequences for proton transfer by a “relay” process might
•
•
permit the generation of O2H and 2-H.
(35) Eriksen, J.; Foote, C. S. J. Am. Chem. Soc. 1980, 102, 6083.