UVA self-photosensitized oxygenation of β-ionone

Article abstract of DOI:10.1111/j.1751-1097.2007.00147.x

The steady-state UVA (350 nm) photolysis of (E)-β-ionone (1) in aerated toluene solutions was studied by ¹H NMR spectroscopy. The formation of the 1,2,4-trioxane (2) and 5,8-endoperoxide (5) derivatives in the ratio of 4:1 was observed. Time-resolved laser induced experiments at 355 nm, such as laser-flash photolysis, photoacoustic and singlet oxygen ¹O₂ phosphorescence detection, confirmed the formation of the excited triplet state of 1 with a quantum yield Φ_T = 0.50 as the precursor for the generation of singlet oxygen ¹O₂ (Φ_Δ = 0.16) and the isomeric α-pyran derivative (3), which was a reaction intermediate detected by NMR. In turn, the reaction of ¹O₂ with 1 and 3 occurred with rate constants of 1.0 × 10⁶ and 2.5 × 10⁸ m^-1s ^-1 to yield the oxygenated products 5 and 2, respectively, indicating the relevance of the fixed s-cis configuration in the α-pyran ring in the concerted [2+4] cycloaddition of ¹O₂.

Full text of DOI:10.1111/j.1751-1097.2007.00147.x

Photochemistry and Photobiology, 2007, 83: 1313–1318
UVA Self-Photosensitized Oxygenation of b-Ionone
1
Claudio D. Borsarelli* , Mirta Mischne , Agustina La Venia and Faustino E. Moran Vieyra
2
2
1
´
1
´
Instituto de Ciencias Quımicas, Facultad de Agronomıa y Agroindustrias, Universidad Nacional de Santiago del Estero,
Santiago del Estero, Argentina
´
2
´
´
Instituto de Quımica Organica de Sıntesis, Universidad Nacional de Rosario, Rosario, Santa Fe, Argentina
´
Received 5 March 2007; accepted 30 March 2007; DOI: 10.1111 ⁄ j.1751-1097.2007.00147.x
(E)-b-ionone (1) is a natural short-chain analog of the
ABSTRACT
polyenes retinoid and carotenoids, hence studies on its
phototransformations have attracted considerable attention
(7,8). In addition, ionones and their derivatives are highly
valued fragrance constituents and important building blocks in
organic synthesis (9–11). In this context, in previous work we
reported that UVA (350 nm) irradiation of 1 under aerobic
conditions resulted in the formation of the 1,2,4-trioxane
derivative 2, which is a 6-member heterocycle containing three
oxygen atoms (Scheme 1) (12,13). To the best of our
knowledge it was the first report on the spontaneous light-
induced addition of O2 in the ionone series without the
presence of a dye-sensitizer molecule, uncovering a new aspect
in the area of the ionone photochemistry.
The steady-state UVA (350 nm) photolysis of (E)-b-ionone (1) in
aerated toluene solutions was studied by 1H NMR spectroscopy.
The formation of the 1,2,4-trioxane (2) and 5,8-endoperoxide (5)
derivatives in the ratio of 4:1 was observed. Time-resolved laser
induced experiments at 355 nm, such as laser-flash photolysis,
photoacoustic and singlet oxygen 1O2 phosphorescence detec-
tion, confirmed the formation of the excited triplet state of 1 with
a quantum yield FT = 0.50 as the precursor for the generation
of singlet oxygen 1O2 (FD = 0.16) and the isomeric a-pyran
derivative (3), which was a reaction intermediate detected by
1
NMR. In turn, the reaction of O2 with 1 and 3 occurred with
rate constants of 1.0 · 106 and 2.5 · 108 )1s)1 to yield the
M
oxygenated products 5 and 2, respectively, indicating the
On the basis of these previous reports, in this work we
present a kinetic study of the UVA-photoinduced transforma-
1
tion of 1 in aerated toluene solutions monitored by H NMR
relevance of the fixed s-cis configuration in the a-pyran ring in
the concerted [2+4] cycloaddition of O2.
1
spectroscopy, and also by performing time-resolved laser-
induced experiments, such as transient absorption and photo-
acoustic spectroscopies, and NIR phosphorescence detection
INTRODUCTION
1
The reaction of singlet molecular oxygen 1O2 with olefinic
compounds is one of the general photochemical oxidations (1).
These processes, classified as Type II photooxygenations, are
highly selective showing often the characteristics of pericyclic
pathways. Depending on the nature of the olefinic target,
valuable intermediates such as allylic hydroperoxides, endop-
eroxides (1,2-dioxenes) and 1,2-dioxetanes result from the
reactions involving 1O2 via three principal classes—ene reac-
tion, [2+2] and [2+4] cycloadditions (2,3). These transforma-
tions represent powerful protocols for introducing molecular
oxygen in a highly specific fashion into organic compounds
with wide applications in organic synthesis (2–4).
of O2.
MATERIALS AND METHODS
(E)-b-ionone (1), 2-hydroxybenzophenone (HBP), 1,4-diazabicy-
clo[2.2.2]octane (DABCO) and perinaphthenone (PN) were from
Sigma (St. Louis, MO), 9,10-dimethylanthracene (DMA) was from
Aldrich (Milwaukee, WI). High purity argon was from Indura
(Tucuman, Argentina). Toluene (HPLC grade) was from Sintorgan
(Buenos Aires, Argentina). All chemicals were used as supplied.
Typically, a 10 mM 1 toluene solution was placed in a Pyrex vessel
(cutoff at 310 nm) and irradiated under aerobic conditions at room
temperature in a Rayonet RPR 100 photochemical reactor with a
1
Usually, O2 is generated by an energy-transfer mechanism
after photoexcitation of a sensitizer dye molecule in the
presence of ground-state triplet oxygen, 3O2 (1–4). The
reaction of 1O2 with the singlet ground state of organic
molecules is so specialized that examples of spontaneous
addition of 3O2 are rare. However, certain types of olefinic
compounds undergo light-induced oxidation without dye
sensitizer, and in those cases self-photosensitized oxygenation
mechanisms have been proposed to rationalize the formation
of the oxygenated products (5,6).
*Corresponding author email: cborsa@unse.edu.ar, cborsarelli@yahoo.com.ar
(Claudio D. Borsarelli)
Scheme 1. Isomers and photooxidation products of (E)-b-ionone.
ꢀ 2007 The Authors. JournalCompilation. The AmericanSocietyofPhotobiology 0031-8655/07
1313

1314 Claudio D. Borsarelli et al.
RPR-3500 UVA lamp (maximum output at 350 nm). The 1H spectra
were recorded at 300 and 75 MHz on a Bruker AVANCE NMR
spectrometer in CDCl3 solutions. Chemical shifts are given in d using
Me4Si as internal standard (12,13). The resonance signals of the
components of the reaction mixture were well separated in the olefinic
region allowing their analysis by 1H-NMR spectroscopy. Periodically a
10 mL aliquot of the reaction mixture was removed, and after
evaporation of the solvent, the crude materials were used to record
the NMR spectra. The component distribution was estimated by the
integration of the NMR peaks of the starting substrate and the
products in the vinylic region (4.5–7.5 p.p.m.).
5.68 p.p.m. (J = 5.8 Hz). The doublet signals at 4.59 and
5.75 p.p.m. (J = 5.8 Hz) were attributed to endoperoxide 5,
according to reported data (17). The inset of Fig. 1 shows the
kinetic curves for the photolysis of 1, indicating that the
complete depletion of 1 was accompanied by the simultaneous
formation of the trioxane 2 (ca 80%) and the endoperoxide 5
(ca 20%). The concentration of the isomeric a-pyran derivative
3 was increased in small proportion at the beginning of the
photolysis and afterward depleted, indicating that the a-pyran
was a reaction intermediate. The formation of 3 takes place via
a hetero-electrocyclization equilibrium from the cis-isomer
(Z)-b-ionone 4, which in turn is generated by isomerization
from the excited triplet state of 1 (7,18–20). Under our
experimental conditions, the cis isomer 4 was not detected in
the NMR experiments. This result is expected because of the
large shift of the isomerization equilibrium to the formation of
3 from 4 at room temperature (K ꢀ 10 at 18ꢁC) (21).
Conversely, the 1H NMR analysis of the reaction mixture
irradiated in toluene solutions saturated with DABCO, a well-
UV–Vis absorption and fluorescence spectra of 1 were recorded
with an HP 8453 photodiode array spectrophotometer and a Hitachi
F-2500 spectrofluorometer, respectively. All time-resolved laser-
induced experiments were performed with an Nd-YAG laser (Con-
tinuum Minilite II) as excitation source generating 355 nm pulses of
5 mJ and 10 ns pulse width. Transient absorption spectra were
recorded with a Luzchem m-LFP 112 laser-flash photolysis system.
Time-resolved photoacoustic experiments were completed with a
home-made system with an acoustic transit time of 500 ns as described
before (14,15). To avoid degradation of the sample less than 10 single
laser shots of <100 lJ were averaged. The calorimetric reference HBP
was used and the absorbance at 355 nm of both reference and sample
solutions was matched within 2%. Time-resolved phosphorescence
1
known quencher of O2 (22), mainly showed the presence of 1
and 3, indicating the key role of 1O2 on the formation of 2 and
5 (data not shown).
1
detection of O2 was done using a home-made system (16), composed
of a Peltier cooled Ge photodiode (Judson J16TE2-66 G) placed at
right angle to the excitation laser pulse. Spurious light was filtered with
a 1270 nm band pass filter (Spectrogon BP-1260). In all cases, the
transient signals were fed to a Tektronix TDS3032B digital oscillo-
scope linked to an on-line PC for data transfer and analysis. The
experiments were performed in duplicate under controlled temperature
of 25 1ꢁC.
Time-resolved laser-induced experiments
Figure 2 shows the transient absorption spectrum of the triplet
state of b-ionone (31*) obtained by laser-flash photolysis upon
excitation at 355 nm of 3 mM 1 in argon-saturated toluene
solution. The transient absorption maximum at 335 nm and
lifetime of 132 ns for 31* were similar to those reported in
hexane (23) and benzene (24). The triplet quantum yield (FT)
RESULTS AND DISCUSSION
Photo-stationary experiments
1
Figure 1 shows the H NMR spectrum after 60 min of UVA
3
of 1* was determined by time-resolved photoacoustic experi-
irradiation of 1 in toluene solutions under aerobic conditions.
Because of their typical vinylic hydrogen resonances (12,13),
the formation of 2 with two doublets at d 6.42 and 6.53 p.p.m.
(J = 8.5 Hz) and the presence of 1 with two doublets at 6.11
and 7.27 p.p.m. (J = 16.4 Hz) was assigned. In addition, the
doublets attributed to a-pyran 3 were found at 5.02 and
ments. Figure 3 shows the photoacoustics for both the
calorimetric reference HBP and the sample 1 obtained after
355 nm laser excitation of argon-saturated toluene solutions.
In time-resolved photoacoustic experiments, the observed
sample signal S(t), Eq. (1), is the convolution of the instru-
mental function R(t) (the signal of the calorimetric reference
Figure 1. 1H NMR spectrum (300 MHz, Cl3CD) of 1 (10 mM) in air-
saturated toluene solution after 60 min of UVA irradiation. Inset:
NMR kinetic curves for the components of the reaction mixture: ( ) 1,
(h) 2, (d) 3 and (s) 5.
Figure 2. Transient absorption spectrum of 1 (3 mM) in argon-
saturated toluene solutions at different times after the laser pulse at
355 nm. Inset: 31* decay signals at 340 nm in argon- and air-saturated
toluene solutions.

Photochemistry and Photobiology, 2007, 83 1315
photocalorimetric method (15,26). Therefore, the fitted ampli-
tude factor for the fast component of the acoustic wave
ufast = 0.69 represents the fraction of the laser energy per
absorbed photon at 355 nm (E355 = 337 kJ mol)1) that is
released as prompt heat in the formation of 31*. The measured
fluorescence quantum yield of 1 was very low (<10)4) and
consequently the fraction of the absorbed energy dissipated as
light was negligible. Thus, the amount of the energy ‘‘stored’’
during the formation of 31* is given by Eq. (4), where ET is the
3
energy content of 1* (=207 kJ mol)1) (23).
UTET ¼ E355ð1 ꢂ ufastÞ
ð4Þ
The FT = 0.50 ( 0.02) was calculated from Eq. (4), a
value in accord with that obtained by laser-flash photolysis in
benzene solutions (24). On the other hand, the slower
component uslow, contains information of the decay of 31*,
as the fitted lifetime of the acoustic signal sslow = 140 ns was
similar to the triplet lifetime obtained by laser-flash photolysis
3
(see above). However, 1* is the precursor of the cis isomer 4,
which in turn is equilibrated with the corresponding hetero-
cyclic isomer 3 (19,21). The lifetime of both forward and
reverse isomerization reactions is in the range of hundreds of
seconds (21), exceeding by far the upper time resolution of the
photoacoustic experiment, i.e. <2 ls. Therefore, it is expected
that the slower photoacoustic component uslow = 0.38 reflects
3
the 1* fi 4 isomerization process in addition to the relax-
ation process to the ground state of 1. Therefore, it is
inaccurate to assume that DVstr ꢀ 0 during the decay of 31*, as
isomerization processes in organic solvents are accompanied
by important DVstr contributions because of solvent reorgan-
ization effects (27). Therefore, solvent-dependent protocols or
theoretical calculations are needed to separate the heat and
structural volume contributions of the acoustic wave (27),
procedures that are out of the scope of this report.
Figure 3. Laser-induced photoacoustics for the calorimetric reference
HBP and sample 1 (1 mM) in argon-saturated toluene solutions,
together with the two-sequential exponential fitting simulation (dashed
line).
As shown in the inset of Fig. 2, in air-equilibrated toluene
solutions the lifetime of 31* was shortened to 64 ns, indicating
that the fraction of triplet states quenched by ground state
that releases all the absorbed energy as heat) and the sample
function H(t), Eq. (2), with ui and si as the amplitude factor
and lifetime for the i-th transient, respectively (25).
molecular oxygen O2, gTS = [1 ) (sTAir ⁄ sTAr)] = 0.52. Con-
3
3
sidering that in air-saturated organic solvents the O2 concen-
tration is ca 2 mM,
M
a quenching rate constant of ca
SðtÞ ¼ RðtÞ ꢁ HðtÞ
ð1Þ
ð2Þ
4 · 109
the quenching of the triplet state of organic molecules by O2
)1s)1 was calculated, which is a typical value for
3
1
(1). The quenching process produces O2, as confirmed by the
1
direct detection of the phosphorescence decay signal of O2 at
ꢁ
ꢀ ꢂ
X
u
HðtÞ ¼
i exp ꢂt si
si
i
1270 nm after laser excitation of air-saturated toluene solu-
tions of 1 (Fig. 4).
1
Satisfactory fits were obtained with a sum of two exponen-
tial decays in the fitting function (dashed line in Fig. 3). The
lifetime of the fast component of the acoustic wave was fixed at
sfast £ 10 ns because of the duration of the excitation laser
pulse. Therefore, this component considers all the relaxation
processes involved in the formation of 31* during the laser
pulse, Eq. (3).
The quantum yield of formation of O2, FD, produced by 1
was determined by extrapolation of the initial intensity (I0) of
1
the O2 signal by the first-order fitting of the phosphorescence
decay, as the I0 value is proportional to FD, Eq. (5) (28).
I0 ¼ jkphUDð1 ꢂ 10ꢂAÞEl
ð5Þ
j is a proportionality instrumental constant, kph is the
solvent-specific radiative decay rate constant for O2, El is the
1
hm
UT
3
1
1 ꢂ! 1 ꢃ ꢂ! 1ꢃ
ð3Þ
incident laser energy and A is the absorbance at the excitation
wavelength. Thus, the FD value for 1 can be obtained by simple
comparison of the slopes of the linear plot of I0 vs El for the
sample and reference compounds under identical experimental
conditions. In this case, PN (FD = 0.92) was used as reference
compound (28). The top inset of Fig. 4 shows the I0 vs El plots
Because the contribution of the structural molecular chan-
ges (i.e. volume changes, DVstr) to the acoustic wave in the
formation of 1* is expected to be negligible, as no intramo-
lecular charge transfer or solvent hydrogen bonding interac-
tions occur, the photoacoustic technique behaves as
3
a

1316 Claudio D. Borsarelli et al.
Figure 5. First-order plots for the 1O2-mediated oxidation of 1, DMA,
and 3 in air-saturated toluene solutions. Inset: UV–Vis absorption
changes of 3 (every 45 s) produced during the PN-photosensitization at
k > 420 nm (Schott cutoff filter KV418).
Figure 4. Phosphorescence decay signals of 1O2 at 1270 nm obtained
by photosensitization at 355 nm of both the reference PN (50 lM) and
sample 1 (5 mM) in aerated toluene solutions together with the
exponential fitting of the decay portion. Top inset: Laser energy
dependence of the zero time intensity of the 1O2 signal, I0, Eq. (5).
Bottom inset: dependence of the observed decay rate constant of 1O2,
kD, vs the concentration of 1, Eq. (9).
yielding kt,1 = 1.2 · 106
quencher of O2 as compared with other related molecules in
M
)1s)1, indicating that 1 is a modest
1
similar solvents, such as b-carotene (kt = 1.1 · 1010
M )
)1s)1
(22). This is a consequence of the much higher energy level for
the triplet state of 1 (ET = 207 kJ mol)1) (23) than for
b-carotene (ET = 82 kJ mol)1) (29), excluding the possibility
to quench 1O2 (ED = 94.5 kJ mol)1) via a downhill energy-
transfer mechanism (physical interaction). Consequently the
chemical quenching process must be the major component of
obtained for 1 and PN in air-saturated toluene solutions with
A355 = 0.7 in both cases. The linearity of the plots with zero
intercepts rules out undesirable phenomena such as multi-
photonic absorption or ground state depletion, and the slope
comparison allowed the calculation of FD = 0.16 ( 0.02) for
1. In addition, FD is related with all the sequential processes
1
kt,1. In fact, the reactive quenching rate constant of O2 by 1
M
)1s)1 by means of an
was calculated as kr,1 = 1.0 · 106
1
involved in the production of O2, Eq. (6).
actinometrical procedure (30), using PN as photosensitizer at
>420 nm and DMA as reference compound, whose reactive
quenching rate constant with 1O2 is well known,
UD ¼ UTgTRSD
ð6Þ
1
kr,DMA = 4.7 · 107
M
)1s)1 (22) (Fig. 5).
SD is the fraction of quenching events that yields free O2.
According to Eq. (6) and the calculated gTS and FT values,
SD = 0.62 was obtained, indicating that ca 60% of the
It is interesting to note that the kr,1 value is similar to that
obtained for the reaction of 1O2 with b-carotene in AOT
reverse micelle solution with rose bengal as photosensitizer
(31). In that case, a cyclic 5,8-endoperoxide b-carotene
derivative is formed as the primary photooxygenation product
3
1
quenched 1* produces O2.
1
The lifetime of O2 observed upon photosensitization of 1
was shorter and depending on its concentration
(3 mM < [1] < 9 mM) than that observed for the sensitizer
1
by [2+4] cycloaddition of O2 (31,32). The formation of 5 is
0
PN (ca 50 lM), with sD = 27 ls. This indicates the presence
produced by the RB-photosensitized oxygenation of 1 in
methanol solutions (17). In addition, the formation of
5,8-endoperoxide derivatives is produced by self-photosensiti-
zation of other retinoids, such as retinal (33) and the
pyridinium retinoid analog A1E (6). Therefore, on the basis
of the above NMR results (Fig. 1) and the reported data for
of both parallel physical and chemical quenching reactions of
1O2 by 1, with rate constants kq,1 and kr,1, respectively, Eqs. (7)
and (8).
kq;1
3
1O2 þ 1 ꢂ! O2 þ 1
ð7Þ
ð8Þ
1
the reaction of O2 with retinoid-like compounds, the forma-
kr;1
tion of the 5,8-endoperoxide 5 by direct [2+4] cycloaddition
reaction of 1O2 to 1, Eq. (8), can be assumed. However, the
results of Fig. 1 show that the trioxane 2 was the major
photooxidation product (ca 80%). The formation of 2 can be
expected by reaction of 1O2 with the a-pyran 3 also by a [2+4]
cycloaddition step, Eq. (10), as reported by Etoh et al. in RB-
photosensitized reactions in methanol solutions (34).
1O2 þ 1 ꢂ! Oxidation products
The kt,1 = kq,1 + kr,1 is the total rate constant for the whole
quenching process of 1O2 by 1, which was obtained by analyzing
the observed rate constants for the phosphorescence decay of
1O2, kD, as a function of the concentration of 1, Eq. (9),
kD ¼ k0D þ kt;1½1ꢄ
kD = (27 ls))1 is the observed decay rate constant of O2
without 1 that was obtained using PN as photosensitizer. The
bottom inset of Fig. 4 shows the linear fitting of Eq. (9)
ð9Þ
kr;3
1O2 þ 3 ꢂ! 2
ð10Þ
0
1
In our case, the rate constant for reaction (10),
kr,3 = 2.5 · 108 )1s)1 was obtained by comparison with
M

Photochemistry and Photobiology, 2007, 83 1317
3. Clennan, E. L. and A. Pace (2005) Advances in singlet oxygen
chemistry. Tetrahedron 61, 6665–6691.
4. Bartoschek, A., T. T. El-Idreesy, A. G. Griesbeck, L.-O. Hoinck,
J. Lex, C. Miara and J. M. Neudorfl (2005) A family of new 1,2,4-
trioxanes by photooxygenation of allylic alcohols in sensitizer-
doped polymers and secondary reactions. Synthesis 14, 2433–2444.
5. Hakimelahi, G. H., M. L. Jain, T. Wei Ly, I.-C. Chen, K. S.
Ethiraj, J. R. Hwu and A. A. Moshfegh (2001) Self-sensitized
photooxygenation of 3,4-dialkoxyfurans to vitamin C or its
derivatives. J. Org. Chem. 66, 7067–7071.
6. Jochkush, S., R. X. Ren, Y. P. Jang, Y. Itagaki, H. R. Vollmer-
Snarr, J. R. Sparrow, K. Nakanishi and N. J. Turro (2004) Pho-
tochemistry of A1E. A retinoid with a pyridinium moiety: Com-
petition between pericyclic photooxygenation and pericyclization.
J. Am. Chem. Soc. 126, 4646–4652.
7. Polyakov, N. E., A. I. Kruppa, V. S. Bashurova, R. N. Musin,
T. V. Leshina and L. D. Kispert (1999) Single electron transfer in
the phototransformations of b-ionone in the presence of electron
acceptor. J. Photochem. Photobiol. A, Chem. 128, 65–74.
8. Polyakov, N. E., A. I. Kruppa, V. S. Bashurova, T. V. Leshina
and L. D. Kispert (2002) The role of radicals in the phototrans-
formations of b-ionone in the presence of electron donors in
solution. J. Photochem. Photobiol. A, Chem. 153, 113–119.
9. Abarbri, M., J.-L. Parrain, A. Ducheˆ ne and J. Thibonnet (2006)
Stereoselective synthesis of trienoic acids: Synthesis of retinoic
acids and analogues. Synthesis 17, 2951–2970.
10. Yang, M., Q.-R. Peng, J.-B. Lan, G.-F. Song and R.-G. Xie (2006)
CaCl2- or MgCl2-catalyzed allylic oxidations of ionone-like dienes.
Synlett 16, 2617–2620.
´
11. Frater, G., J. A. Bajgrowicz and P. Kraft (1998) Fragrance
chemistry. Tetrahedron 54, 7633–7703.
12. Huber, S. N. and M. P. Mischne (1995) A simple and efficient
approach for the synthesis of a 1,2,4-trioxane related to the
artemisinin. Nat. Prod. Lett. 7, 43–46.
13. Mischne, M. P., S. N. Huber and J. Zinckzuk (1999) Photo-
chemical reaction of b-ionone derivatives in the presence of oxy-
gen. A convenient synthesis of substituted 1,2,4-trioxanes. Can. J.
Chem. 77, 237–241.
Scheme 2. Reaction mechanism proposed for the self-photosensitized
oxygenation of (E)-b-ionone.
the actinometer DMA as mentioned before, inset of Fig. 5.
The a-pyran 3 was previously formed by 350 nm irradiation
of 1 in Ar-saturated solutions, as reported by Buchi and
Yang (18). The kr,3 value is approximately 1 order of
magnitude higher than for 1,3-cyclopentane or 1,3-cyclohex-
ane rings (35), indicating that the electron donating effect of
the substituents at the end of the diene unit of 3 could
activate the electrophilic attack by 1O2 (3). As compared
with 1, the 250-times higher reactivity of 3 can be assigned
to the fixed s-cis-1,3-diene configuration of the a-pyran ring,
indicating the favorable role played by entropic factors in
[2+4] cycloadditions.
CONCLUSIONS
14. Borsarelli, C. D., S. G. Bertolotti and C. M. Previtali (2002)
Thermodynamic changes in the photoinduced proton-transfer
reaction of the triplet state of safranine-T. Photochem. Photobiol.
Sci. 1, 574–580.
15. Rios, A. O., A. Z. Mercadante and C. D. Borsarelli (2007) Triplet
state energy of the carotenoid bixin determined by photoacoustic
calorimetry. Dyes Pigments. 74, 561–565.
16. Montenegro, M. A., I. L. Nunes, A. Z. Mercadante and C. D.
Borsarelli (2007) Photoprotection of vitamins in skimmed milk by
aqueous soluble lycopene—gum arabic microcapsule. J. Agric.
Food Chem. 55, 323–329.
The UVA self-photosensitized oxygenation mechanism of
b-ionone in aerated toluene solutions is summarized in
Scheme 2. The relevance of this reaction in synthetic routes
of stable trioxane precursors, such as 2, is based on the
photosensitized formation of 1O2 from 1 and the enhanced
1
reactivity toward O2 of the heterocyclic isomer 3, which was
also formed by photoisomerization steps from the triplet
state of 1. Thus, the reaction sequence for the formation of 2
involves a one-pot three steps domino process (isomerization,
hetero-electrocyclization and [2+4] Diels-Alder type oxygen
addition) (36), representing an improved procedure for the
preparation of 1,2,4-trioxane-type compounds. The forma-
tion in less proportion of the 5,8-endoperoxide 5 by direct
reaction of 1 with 1O2 was also observed, and the relative
distribution of oxygenated products [2] ⁄ [5] is controlled by
the ratio kr,3[3] ⁄ kr,1[1].
17. Mousseron-Canet, M., J.-C. Mani and J.-P. Dalle (1967) Photo-
oxydation sensilibilisee dans la serie de la b-ionone. Bull. Soc.
´ ´ ´
Chim. 2, 608–612.
18. Buchi, G. and N. C. Yang (1957) Light-catalyzed organic reac-
tions. VI. The isomerization of some dienones. J. Am. Chem. Soc.,
79, 2318–2323.
19. van Wageningen, A., H. Cerfontain and J. A. J. Geenevasen
(1975) Photochemistry of non-conjugated dienones. Part V. Pho-
tolysis of (E)-b-ionone and its isomeric a-pyran. J. Chem. Soc.
Perkin II 12, 1283–1286.
20. Cerfontain, H., J. A. J. Geenevasen and P. C. M. van Noort (1980)
Photochemistry of dienones. Part 7. On the photosensitized
isomerization of (E)-b-ionone and its isomeric a-pyran. Evidence
for exciplex formation between the a-pyran and Fluoren-9-one.
J. Chem. Soc. Perkin II 7, 1057–1062.
21. Marvell, E. N., G. Caple, T. A. Gosink and G. Zimmer (1966)
Valence isomerization of a cis-dienone to an a-pyran. J. Am.
Chem. Soc. 88, 619–620.
Acknowledgements—The authors thank the Argentinian Funding
Agencies CONICET, ANCPyT (PICTO 764), CICyT-UNSE and
Fundacion Josefina Pratts for financial support. F.E.M.V. and A.L.
´
thank CONICET for a fellowship grant.
22. Wilkinson, F. and J.G. Brummer (1981) Rate constants for the
decay and reactions of the lowest electronically excited state of
molecular oxygen in solution. J. Phys. Chem. Ref. Data. 10,
809–999.
23. Becker, R. S., R. V. Bensasson, J. Lafferty, T. G. Truscott and
E. J. Land (1978) Triplet excited states of carbonyl-containing
polyenes. J. Chem. Soc. Faraday Trans. 2. 74, 2246–2255.
REFERENCES
1. Turro, N. J. (1978) Modern Molecular Photochemistry. Benja-
min ⁄ Cummings, Menlo Park, CA.
2. Greer, A. (2006) Christopher Foote’s discovery of the role of
singlet oxygen [1O2(1Dg)] in photosensitized oxidation reactions.
Acc. Chem. Res. 39, 797–804.

1318 Claudio D. Borsarelli et al.
24. Chattopadhyay, S. K., C. V. Kumar and P. K. Das (1985) Triplet
excitation transfer involving b-ionone. A kinetic study by laser
flash photolysis. Photochem. Photobiol. 42, 17–24.
25. Rudzki, J. E., J. L. Goodman and K.S. Peters (1985) Simultaneous
determination of photoreaction dynamics and energetics using
pulsed, time-resolved photoacoustic calorimetry. J. Am. Chem.
Soc. 107, 7849–7854.
naringin and its chalcone. J. Photochem. Photobiol. A, Chem. 186,
47–56.
31. Montenegro, M. A., M. A. Nazareno, E. N. Durantini and C. D.
Borsarelli (2002) Singlet oxygen quenching ability of carotenoids
in a reverse micelle membrane mimetic system. Photochem. Pho-
tobiol. 75, 353–362.
32. Stratton, S. P., W. H. Schaeffer and D. C. Liebler (1993) Isolation
and identification of singlet oxygen oxidation products of
b-carotene. Chem. Res. Toxicol. 6, 542–547.
33. Baron, M. H., M. J. Coulange, C. Coupry, D. Baron, J. Favrot
and M. M. Abo Aly (1989) All-trans retinal photoisomerization
and photooxidation from UV laser radiation. Vibrational assign-
ment of all-trans 5,8-peroxyretinal. Photochem. Photobiol. 49,
739–743.
26. Nonell, S., C. Martı, I. Garcıa-Moreno, A. Costela and R. Sastre
´ ´
(2001) Opto-acoustic study of tinuvin-P and rhodamine 6 G in
solid polymeric matrices. Appl. Phys. B 72, 355–360.
27. Feiss, A., B. Wegewijs, W. Gartner and S. E. Braslasvky (1997)
The role of triplet state in retinal photoisomerisation as studied by
laser induced optoacoustic spectroscopy. J. Phys. Chem. 101,
7620–7627.
28. Martı
´
, C., O. Jurgens, O. Cuenca, M. Casals and S. Nonell (1996)
34. Etoh, H., K. Ina and M. Iguchi (1973) Photosensitized oxygen-
ation of a-pyran derived from b-ionone. Agric. Biol. Chem. 37,
2241–2244.
35. Adam, W. and A. G. Griesbeck (1995) Photooxygenation of
1,3-dienes. In CRC Handbook of Organic Photochemistry and
Photobiology, Chapter 25 (Edited by W. M. Horspool and P.-S.
Song), pp. 311–324. CRC Press, Boca Raton, FL.
36. Tietze, L. F. (1996) Domino reactions in organic synthesis. Chem.
Rev. 96, 115–136.
Aromatic ketones as standards for singlet molecular oxygen
O2(1Dg) photosensitization. Time-resolved photoacoustic and
near-IR emission studies. J. Photochem. Photobiol. A, Chem. 97,
11–18.
29. Lambert, C. and R. W. Redmond (1994) Triplet energy level of
b-carotene. Chem. Phys. Lett. 228, 495–498.
30. Montenegro, M. A., M. A. Nazareno and C. D. Borsarelli (2007)
Kinetic of the photosensitized oxygenation of the flavanone