phenanthrene from the column using hexane, the eluent was
changed to mixed solutions consisting of hexane, chloroform and
ethanol, and the content of chloroform and ethanol was gradually
increased. Finally, the fractionalized effluents were completely
evaporated under reduced pressure, and the reaction products
were separately obtained. These products were recrystallized from
hexane, and the structures of the isolated products were identified
polar compounds observed as peaks in the HPLC chart are mostly
produced via 3 or 4.
When we carried out the reaction without supplying oxygen gas
practically no products were obtained. No products were obtained
when the reaction was carried out in water-free acetonitrile
bubbled with oxygen gas.
4
Lin and Valsaraj studied the oxidation of phenanthrene by a
1
13
as 2, 3 and 4 by H and C NMR using a JEOL JNM-AL400
2
TiO -photocatalyzed reaction in water in the presence of oxygen,
NMR spectrometer.{
and observed 2 and many other kinds of products, including 4,
before the mineralization. Kohtani et al. studied the reaction of
The peaks of the HPLC chart of the mixed products were
assigned to 1, 2, 3 and 4 by comparing their retention times with
those of authentic samples. The peaks due to the polar compounds
produced by the photocatalyzed reaction were not significant at its
early stages, suggesting that they were oxidation products of 2, 3
and 4. We did not identify these compounds because we were more
interested in those meaningful from the viewpoint of organic
synthesis.
phenanthrene on photoirradiated BiVO
8
in acetonitrile, and
4
obtained 2 and 4. However, these authors did not report the
production of 3, which was the main product in our study, carried
out in a mixed solvent of acetonitrile and water, and using TiO as
2
the photocatalyst.
In order to clarify the reaction paths of phenanthrene on
2
photoirradiated TiO in our mixed solvent, we carried out
photocatalyzed reactions using authentic samples of 2 and 4
under the same conditions as those used for the reaction of 1. The
results showed that 3 was produced from 2, whereas only polar
compounds were produced from 4. On the basis of these results
and other findings, we propose the reaction paths shown in
Scheme 1. Production of 4 is equivalent to production of the
dialdehyde from naphthalene, which was observed in our previous
The time profiles of the amounts of 2, 3 and 4 produced in the
test tube were different, as shown in Fig. 2. During the initial 80 h,
the amount of phenanthrene decreased linearly with reaction time,
suggesting that the reaction rate is determined by the supply of
2
photogenerated carriers to the surfaces of TiO particles. At the
very initial stage of the reaction, the rates of production of 3 and 4
were nearly the same. However, the amount of 4 soon levelled off
as the photoirradiation continued. The amount of 2 also levelled
off at the early stage, and its amount was much less than the
amounts of 3 and 4 throughout the reaction. The amounts of 2
and 4 were nearly constant during the period in which the amount
of phenanthrene decreased linearly. This suggests that 2 and 4
were rapidly converted to other species by successive photo-
catalyzed reactions. In contrast, the amount of 3 continued to
increase linearly during the period in which the amount of
phenanthrene decreased linearly. As a result, 3 became the
dominant product after photoirradiation for 10 h, and its yield
reached 45% at 100 h. After reaching a maxima, the amount of 3
started to decrease at a rate slower than the initial decay rate of 1.
The sum of the initial slopes of the rise of 3 and 4 is nearly equal to
the slope of the decay of 1, as seen in Fig. 2. This suggests that the
7
study. This path seems to be unique to the photocatalyzed
reactions of compounds with condensed aromatic rings. The
production of 2 from 1 is a common practice in the chemical
9
oxidation of 1; 2 is probably generated through the oxidation of 1
by positive holes on the surface of TiO
2
.
The photocatalyzed production of 3 from 2 is the key step in the
2
present system. This step was not driven by O , which was
2
?
supplied as a mixture of KO and 18-crown-6, or by OH radicals,
2
which were generated via the photo-Fenton reaction. Although we
do not know the details of this process, we tentatively propose that
2
is converted into 3 via acid anhydride compound 5, as shown in
Scheme 1. Presumably 5 is generated from 2 by surface Ti
peroxide, which is photogenerated in a manner similar to that of
1
0
the Baeyer–Villiger reaction, or by repeating the butterfly-type
11,12
reaction with the surface hydroperoxo species.
However, the
intermediate 5 was not isolated from the reaction solution,
probably because 5 can easily be hydrolyzed to a dicarboxylic
acid in a wet medium, such as the solvent used in the present study.
We speculate that 5 is immediately converted to 3 upon its
2
formation on the photoirradiated TiO .
The apparent quantum efficiency of the reaction, defined as the
ratio of the number of molecules of 3 produced compared to the
number of incident photon fluxes, was determined to be 1.7%
using light of around 365 nm wavelength, selected from the super-
high-pressure Hg lamp using a band-pass filter. The incident
photon flux was determined by the standard photochemical
actinometer method, based on the conversion of ferrioxalate ion
into Fe(II), using the same experimental setup as that used in the
photocatalytic reaction. No corrections were made for the losses of
incident photons due to reflection. By taking into account the fact
that 10 electrons must be removed from 1 to produce 3, the
apparent quantum efficiency of the whole photooxidation process
was estimated to be 17%.
Fig. 2 Time courses of the change in the amounts of 1 (#), 2 (e), 3 (n)
and 4 ($) included in the solution as the TiO
2
-photocatalyzed reaction
continued. The amounts were determined by HPLC. The solid and dashed
In conclusion, we have found that a coumarin compound is
synthesized from phenanthrene with considerably high efficiency
lines show the initial slopes of the rise of 3 and 4, respectively.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2804–2806 | 2805
Higashida, Suguru
