High-performance photooxidation of phenol with singlet oxygen in an ionic liquid

Article abstract of DOI:10.1246/cl.2009.358

Photooxidation of phenol using singlet oxygen generated in ionic liquid was reported. Singlet oxygen is produced via photosensitized reaction of 5,10,15,20-tetrakis(4-sulfophenyl)por-phine in ethylammonium nitrate. The production rate of singlet oxygen wa

Full text of DOI:10.1246/cl.2009.358

358
Chemistry Letters Vol.38, No.4 (2009)
High-performance Photooxidation of Phenol with Singlet Oxygen in an Ionic Liquid
Nobuhiro Inoue, Toshio Ishioka, and Akira HarataÃ
Department of Molecular and Material Sciences, Interdisciplinary Graduate School of Engineering Sciences,
Kyushu University, Kasuga 816-8580
(Received January 6, 2009; CL-090034; E-mail: harata@mm.kyushu-u.ac.jp)
Photooxidation of phenol using singlet oxygen generated in
can be easily supplied from the atmosphere without any special
treatment generating environmental pollution. However, none
have tried to oxidize phenols by singlet oxygen produced in ionic
liquids. In this report, singlet oxygen produced in ethylammoni-
um nitrate (EAN) was used to oxidize phenols aiming to a high
yield and as an approach to green chemistry.
ionic liquid was reported. Singlet oxygen is produced via photo-
sensitized reaction of 5,10,15,20-tetrakis(4-sulfophenyl)por-
phine in ethylammonium nitrate. The production rate of singlet
oxygen was 2.2 times faster than that in aqueous media contain-
ing the same. The resulting oxidized product from phenol was
pure benzoquinone with no by-product like quinhydrone usually
found in aqueous media. The advantage of this method is high
yield of benzoquinone with no harmful agents like metal cataly-
sis or oxidizer as well as easy product separation.
EAN was synthesized by mixing equimolar amounts of
EtNH2 and HNO3 in an ice bath. EAN was dried at 70 ꢀC for
3 h on a rotary evaporator under reduced pressure before use.
Singlet oxygen produced by photosensitized reaction was meas-
ured with UV–visible absorption spectroscopy using trapping
agent 9,10-diphenylanthracene (DPA) without phenol in solu-
tion. DPA is known to react quantitatively with singlet oxygen
producing 9,10-diphenylanthracene endoperoxide. The absorp-
tion maximum of DPA is around 355 nm while DPA-endoperox-
ide has no absorption band in the range 300–500 nm. Decrease
in the absorbance value at 355 nm can be used to calculate the
amount of DPA-endoperoxide formed, which is proportional to
the amount of singlet oxygen produced.6 The measurement
was carried out after irradiation with a Xe lamp (StockerYale,
150W) to a sample solution (2.5 mL) under ambient atmosphere.
Initial concentration of photosensitizer 5,10,15,20-tetrakis(4-
sulfophenyl)porphine (TSPP) was 1:0 Â 10À5 M and that of trap-
ping agent DPA was 1:0 Â 10À4 M, which is enough to trap all of
the dissolved oxygen.
The rate of singlet oxygen production can be calculated by
the initial slope of curves in Figure 1 that shows the decrease
of the absorption at 355 nm with the period of light irradiation
time. The rate is faster in EAN than in nonionic solvents, H2O
and CH3CN. The factors of the rate acceleration/deceleration
are (1) initial concentration of oxygen in solution, (2) efficiency
of light absorption by the photosensitizer, (3) viscosity of the
solvent, and (4) lifetime of transient species. Initial oxygen con-
centration can be estimated as the saturated solubility for each
solvent. The data are available for H2O and CH3CN and the
values are 2:69 Â 10À4 and 1:99 Â 10À3 M, respectively, but
In recent decades, ionic liquids have attracted considerable
attention as a medium for organic synthesis,1 solvation of bioma-
cromolecules,2 and enzyme reactions3 because solvation scheme
is quite different in ionic liquids compared to nonionic solvent,
promoting enhanced reactivity or producing other final products.
In photochemistry, several studies utilizing ionic liquids have
been reported. One study used various kinds of ionic liquids as
a medium for photochemical reactions and found an enhance-
ment of triplet excited states and elongation of lifetimes of radi-
cal ions.4 Another report concentrated on the photoisomerization
of cis- and trans-stilbene and found the mechanism involves the
process of electron-transfer reactions from excited stilbene to
cation, which is different from reactions in traditional solvents.5
Longer lifetime of the triplet excited state in ionic liquids
means that there are increased chances of energy transfer with
a reactant. Singlet oxygen can be produced in such a process be-
cause it is usually generated via energy transfer from an excited
photosensitizer to oxygen in the ground state (Scheme 1). Singlet
oxygen is a good oxidizer because its energy lies 94 kJ molÀ1
above the ground state and it can oxidize a variety of species
without any harmful agents. As for photochemical reactions in
ionic liquids, highly efficient production of singlet oxygen is
expected and it has a potential for industrial use as an attractive
oxidizer.
In this report, phenol was selected as a reactant oxidized by
singlet oxygen since phenols are known to resist from biodegra-
dation and have attracted particular interest in wastewater treat-
ment. Oxidized product 1,4-benzoquinone is a precursor of Vita-
min E and has various industrial applications. If the oxidation
reaction proceeds efficiently, this method can be an ideal clean
method for oxidation. It is because ionic liquids have little or
no vapor pressure, and oxygen as the source of singlet oxygen
1
0.8
0.6
0.4
0.2
0
1S(Sensitizer)
1S*
1S*
+ h
ν
3S*
0
10
20
30
40
50
60
+ 3O2
3S*
+ 1O2
1S
Irradiation time / min
Scheme 1. Singlet oxygen generation via photosensitized reac-
tion.
Figure 1. Temporal change in absorbance at 355 nm of DPA,
for ( ) TSPP in EAN, ( ) TSPP in CH3CN, ( ) TSPP in H2O.
Copyright Ó 2009 The Chemical Society of Japan

Chemistry Letters Vol.38, No.4 (2009)
359
Table 1. BQ yield of EAN, H2O, and CH3CN solvent
Solvent
Yield of BQa/%
By-product
—
Quinhydrone
—
EAN
H2O
CH3CN
10.0
8.9
—
aYield of BQ was determined by HPLC system (Irradia-
tion time: 5 h, PhOH: 0.1 M).
326
346
Magnetic field / mT
100
80
60
40
20
0
Figure 2. ESR spectrum of EAN solution including Mn marker
) before irradiation and ( ) under irradiation.
(
not for EAN. When DPA is 5:0 Â 10À4 M, the absorption max-
imum of DPA is not disappeared completely. From the oxygen
consumption behavior in EAN, the value can be roughly estimat-
ed around 5 Â 10À4 M. Initial concentration of oxygen is not the
origin of the rate acceleration in EAN. The light absorption effi-
ciency can be estimated from absorption spectrum of photosen-
sitizer in each solvent. The light absorption efficiency is in the
order CH3CN > H2O > EAN, and there is a discrepancy with
the results of singlet oxygen production. The viscosity of the sol-
vent is in the order of EAN > H2O > CH3CN and molecular
diffusion in the solutions containing singlet oxygen formation
are assumed in the reversed order. The reaction rate controlled
by diffusion should be faster in CH3CN than in EAN. This con-
flicts with the experimental data. From above discussions, longer
lifetime of the transient species can only explain the faster pro-
duction of singlet oxygen in EAN. To prove the longer lifetime
of transient species, electron spin resonance spectroscopy in the
presence of sensitizer under irradiation was measured and the
results are shown in Figure 2. In EAN a broad signal appeared
between two strong marker signals (MnO). The broad signal
was observed only in EAN, not observed in H2O and CH3CN.
These results suggest that the broad ESR signal likely to origi-
nate from triplet TSPP. In ionic liquid, lifetime of triplet sensi-
tizer was observed to have longer lifetime. However, the pro-
longed lifetime of triplet sensitizer is at most twice than in other
solvent.7 So it is considered that the transient species is exciplex
consist of triplet TSPP derivative. The transient species is not as-
signed clearly but exciplex seems to be the most likely candidate
because the broad signal is not observed in O2-saturated EAN
solution containing no sensitizer under irradiation.
0
5
10
15
20
Irradiation time / h
Figure 3. Irradiation time dependence of BQ yield in EAN.
of benzoquinone is much lower by conventional methods. For
high yield and selective oxidation, almost or even more than a
stoichiometric amount of catalyst is required.8 Figure 3 shows
the benzoquinone yield using singlet oxygen in EAN as a func-
tion of the irradiation time. The yield of benzoquinone increased
linearly with the irradiation time. This suggests that photooxida-
tion of phenol is possible until the phenol is completely con-
sumed. Furthermore, it is possible to increase the yield by using
a high power light source.
In conclusion, singlet oxygen generation and photooxidation
of phenol were carried out in an ionic liquid, ethylammonium
nitrate. Generation of singlet oxygen and its transient reactant
was confirmed during the photooxidation reaction. Photooxida-
tion of phenol by singlet oxygen in ionic liquids can be an effi-
cient method because of the characteristics of high yield and
high selectivity. A simple extraction treatment was used in this
letter for separating the product, however, there is a possibility
of autoseparation becoming unnecessary because the solubility
in ionic liquids can be easily controlled by changing the compo-
nent of either ion.
Phenol was photooxidized for 5 h under the same irradiation
conditions. Initial concentration of phenol and photosensitizer
TSPP were 0.1 and 10 Â 10À5 M, respectively. After irradiation
the products were extracted repeatedly with hexane and the ex-
tracted part was characterized with NMR spectroscopy. The
quantity of each product was analyzed with HPLC and a UV–
vis detector. Results are listed in Table 1. Benzoquinone was
obtained with no impurity in EAN, while quinhydrone was ob-
tained as a by-product in H2O and no benzoquinone was detected
in CH3CN. In EAN, ESR spectrum in the presence of phenol was
similar to that observed without phenol and its intensity gradual-
ly decreased with time. This shows that transient species were
continuously produced during the reaction and that it was con-
sumed by photooxidation.
References
1
2
J. D. Holbrey, K. R. Seddon, Clean Prod. Proc. 1999, 1, 223.
R. P. Swatloski, S. K. Spear, J. D. Holbrey, R. D. Rogers,
J. Am. Chem. Soc. 2002, 124, 4974.
3
4
5
6
7
8
T. Itoh, E. Akasaki, K. Kudo, S. Shirakami, Chem. Lett.
2001, 262.
M. Alvaro, B. Ferrer, H. Garcia, M. Narayana, Chem. Phys.
Lett. 2002, 362, 435.
C. Lee, G. Mamantov, R. M. Pagni, J. Chem. Res., Synop.
2002, 122.
M. J. Steinbeck, A. U. Khan, M. J. Karnovsky, J. Biol. Chem.
1993, 268, 15649.
J. R. Herance, B. Ferrer, J. L. Bourdelande, J. Marquet,
H. Garcis, Chem.—Eur. J. 2006, 12, 3890.
Eur. Pat. Appl. (BASF AG). EP 475,272 A2, 1992.
The aerobic oxidation of phenol in the presence of heavy
metal compounds as catalysis has some drawbacks. The yield
Published on the web (Advance View) March 14, 2009; doi:10.1246/cl.2009.358