Shape- and size-selective photocatalytic reactions by layered titanic acid powder suspended in deaerated aqueous alcohol solutions

Article abstract of DOI:10.1039/b006497l

Protonic titanate of H(x)Ti(2-x/4)Z(x/4)O₄ · H₂O (HLe; x ? 0.7, Z = vacancy) powder of a layered crystal structure was prepared, suspended in aqueous alcoholic solutions and photoirradiated by light of wavelength > 300 nm at room temperature under a deaerated atmosphere. HLe showed photocatalytic activity of H₂ production, similar to ordinary titanium(IV) oxide powders. The rate and time course of H₂ evolution by HLe depended strongly on the kind of alcohol. Detailed investigation revealed that the photocatalytic activity was regulated by the shape and size of the alcohol, due to the selection of adsorption at the interlayer of HLe.

Full text of DOI:10.1039/b006497l

View Article Online / Journal Homepage / Table of Contents for this issue
Shape- and size-selective photocatalytic reactions by layered titanic
acid powder suspended in deaerated aqueous alcohol solutions
Bunsho Ohtani,*a Shigeru Ikeda,a Hideaki Nakayamab and Sei-ichi Nishimotob
a Catalysis Research Center, Hokkaido University, Sapporo, 060-0811, Japan.
E-mail: ohtani=cat.hokudai.ac.jp
b Division of Energy and Hydrocarbon Chemistry, Kyoto University, Sakyo-ku, Kyoto 060-8501,
Japan
Received 8th August 2000, Accepted 19th September 2000
First published as an Advance Article on the web 24th October 2000
Protonic titanate of H Ti
Z
O É H O (HLe; x ^ 0.7, Z \ vacancy) powder of a layered crystal structure
x
2~x@4 x@4 4
2
was prepared, suspended in aqueous alcoholic solutions and photoirradiated by light of wavelength [300 nm
at room temperature under a deaerated atmosphere. HLe showed photocatalytic activity of H production,
2
similar to ordinary titanium(IV) oxide powders. The rate and time course of H evolution by HLe depended
2
strongly on the kind of alcohol. Detailed investigation revealed that the photocatalytic activity was regulated
by the shape and size of the alcohol, due to the selection of adsorption at the interlayer of HLe.
H O (HLe, x ^ 0.7, Z \ vacancy), has been synthesized.14h16
1. Introduction
2
Preliminary experiments by DomenÏs group have shown that
Photocatalytic reactions occurring on the surface of photoir-
radiated semiconductors have attracted much attention by
scientists. Their possible environmental applications,1h3 such
as photoinduced removal of contaminated harmful chemicals
in indoor and outdoor atmospheres, are now in the stage of
practical testing by many companies, especially those in
Japan.3 Although, in such decomposition of organic materials,
their chemical bonds are cleaved non-selectively mainly by
positive holes that are formed in the valence band of the semi-
conducting materials as a result of photoexcitation of elec-
trons into the conduction band, it has been clariÐed that
under selected reaction conditions, e.g., in the absence of
molecular oxygen, one can expect selective organic reac-
tions.4h7 For example, we have shown that an amino acid,
lysine, undergoes selective deaminocyclization into piperidine-
the platinized HLe has sufficient photocatalytic activity when
irradiated in methanolÈwater mixed solvent to liberate molec-
ular hydrogen (H ) and that the rate of H evolution from
2
2
C ÈC straight-chain alcohols decreased along with the chain
1
4
length.17 Assuming that the photocatalytic reaction of these
alcohols occurred at the inner surface between the TiO layers
x
of HLe, it can be expected that such a tendency for the reac-
tion rate to depend on the size of the substrate can be used for
selection of molecular size or shape by the entrance of the
inner part of the photocatalyst. In this study, we examined the
photocatalytic reaction of HLe with a wide range of substrates
under various reaction conditions, and we obtained evidence
of size- and shape-selective photocatalytic reaction by HLe.
2. Experimental
2-carboxylic acid (pipecolinic acid) by aqueous suspensions of
titanium(IV) oxide (TiO )8 or cadmium(II) sulÐde (CdS).9 The
2.1. Materials
2
use of platinized TiO enabled us to synthesize optically active
2
HLe powder was synthesized according to the previously
reported methods.14h16,18 A brief summary of the procedure
is as follows. A mixture of anatase TiO powder (Merck) and
caesium carbonate (Cs CO , Wako Pure Chemical Industries)
was heated at 1073 K for 40 h. The product, Cs Ti
pipecolinic acid selectively and efficiently.9
In order to realize further useful applications of photo-
catalytic reactions to organic synthesis, the addition of selec-
tivity for speciÐc reactions or products is necessary. One
possible strategy to give size and/or shape selectivity to the
photocatalysts is the incorporation of a photocatalyst (e.g.,
2
2
3
Z
O
x
2~x@4 x@4 4
(
CsLe; Z \ vacancy), was suspended in ca. 1 mol dm~3
hydrochloric acid for 72 h to give HLe. Degussa P-25 TiO ,
TiO ) into porous materials (e.g., zeolites)10h12 and modiÐ-
2
2
which does not have a porous structure, was used as a repre-
cation of the photocatalyst surface with organic compounds.
sentative photocatalyst. Platinization (0.5 wt.%) of P-25 TiO
However, there are the problems that the porous materials
absorb or scatter the light for excitation of photocatalysts and
that the surface-attached compounds undergo photocatalytic
decomposition. One possible way to overcome these problems
is to use a photocatalyst whose structure has an inner surface,
i.e., a photocatalyst with a layered, tunnel or porous crystal
structure. We have reported the photocatalytic reaction of
TiO crystallites with a tunnel structure, TiO (B), but the size
2
was achieved by impregnation followed by H reduction, as
2
reported previously.19 Methanol, ethanol, propan-2-ol,
propan-1-ol, butan-1-ol, butan-2-ol and tert-butyl alcohol, all
of the highest available grades, were purchased from Wako
(with the exception of tert-butyl alcohol; Nakalai Tesque) and
used without further puriÐcation. Hydrated hexachloro-
platinic acid (H PtCl É xH O) was supplied from Nakalai.
2
6
2
2
2
of the tunnel was too small (ca. 0.4È0.7 nm in inner diameter)
to induce selective photocatalytic reaction,13 though we could
observe its photocatalytic activity.
2.2. X-ray di†raction analysis
A Rigaku GeigerÑex 2013 di†ractometer (CuKa, Ni Ðlter, 35
kV, 20 mA) was used to record di†ractograms at a scanning
rate of 1 deg min~1 (ref. 19).
As
a
crystal structural modiÐcation of titanic acid,
O É
TiO É nH O, lepidocrocite-like titanic acid, H Ti
Z
2
2
x
2~x@4 x@4 4
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313
This journal is ( The Owner Societies 2000
5
308
DOI: 10.1039/b006497l

View Article Online
2
.3. Transmission electron microscopy
that the ratio was ca. 0.7 as reported previously.14h16 Fig. 1(a)
shows a TEM picture of HLe.
A Hitachi H-800 electron microscope operated at 200 kV
acceleration voltage was used. The procedure for preparing a
TEM sample on microgrids has been reported previously.8
Electron di†raction patterns were recorded in the apparatus
with a camera length of 0.75 m.
Flat plates of size ranging from a few tens of nanometres to
a few micrometres were seen in all of the views. Detailed
inspection revealed that there were many steps, as indicated
by di†erent degrees of darkness of the image, at which the
density of electrons in the sample changed. This fact conÐrms
that HLe had a layered structure. Evidence of the HLe struc-
ture was also obtained from electron di†raction patterns of
the sample powders. A representative pattern is shown in Fig.
2.4. Photocatalytic reaction
A 50 mg portion of photocatalyst power, HLe or TiO , was
suspended in a mixture of alcohol and water (in most cases
2
1(b). An expected reciprocal lattice drawn by assuming that
the HLe crystallite has its (010) plane perpendicular to the
electron beam is also shown in Fig. 1(c), and it is obvious that
the apparent electron di†raction pattern is reasonably inter-
preted by the layered structure of HLe. The results of TEM
observation of platinized HLe are discussed in a later section.
5
0/50 vol.% except for butan-1-ol and butan-2-ol, see text)
and deaerated by bubbling of argon (Ar) for at least 15 min.
To avoid volatilization of alcohols during the bubbling, the
suspensions were chilled in an ice bath. The deaerated suspen-
sion in a sealed glass tube that transmits light of wavelength
[
300 nm was irradiated by a 400 W mercury arc (Eiko-sha)
at 298 K under magnetic stirring (1000 rpm).
3.2. Photocatalytic reaction of HLe with methanol and
propan-2-ol
2.5. Product analyses
HLe was suspended in 50 vol.% aqueous solution of methanol
or propan-2-ol and irradiated at 298 K under an Ar atmo-
sphere. A small portion of hexachloroplatinic acid, containing
Pt atoms corresponding to 0.5 wt.% for HLe, was added to be
reduced in situ to platinum metal deposits. The time course of
the reaction is shown in Fig. 2(a).
Products in the gas phase, molecular hydrogen (H ), were
analyzed using a Shimadzu GC-8A gas chromatograph (GC)
2
equipped with a molecular sieve 5A column and a thermal
conductivity detector (TCD). A portion of the gas phase of the
reaction vessel was withdrawn with an airtight syringe and
injected into the GC during and after the photoirradiation.
For the product from aqueous propan-2-ol solution, a sample
obtained after centrifugation to remove the photocatalysts
was analyzed using a Shimadzu GC-7A gas chromatograph
equipped with a PEG20M column and a Ñame ionization
detector (FID).
Just after commencement of photoirradiation, liberation of
H
in the methanol system could be seen and its rate was
2
almost constant during the 9 h period of irradiation, while an
appreciable induction period, ca. 3 h, was observed for the
propan-2-ol system and the rate of H liberation after the
2
induction period was much smaller, ca. 1/10, than that in the
methanol system. In both systems, H evolution could be seen
2
only under the condition of photoirradiation, indicating that
3
. Results and discussion
this reaction is ““photocatalyticÏÏ. In the control experiment in
which hexachloroplatinic acid was not added, the bare HLe
gave a negligible amount of H . Consequently, in situ depos-
3.1. Preparation of HLe
2
X-ray di†raction (XRD) patterns of CsLe and HLe were in
good agreement with those reported previously.14h16 The
molar ratio of vacancies, x, was not determined in this work,
but, based on the coincidence of XRD patterns we assumed
ited Pt enhances the H evolution, as commonly seen for TiO
2
2
photocatalytic reactions.8,20h22 On the basis of these results,
one possible interpretation of the induction period in the
propan-2-ol system is that the period corresponds to the time
Fig. 1 (a) A TEM picture of HLe. (b) An electron di†raction pattern of an HLe particle. (c) An expected reciprocal lattice obtained on the basis
of the assumption that the HLe crystallite has its (010) plane perpendicular to the electron beam in TEM. (d, e) TEM pictures of HLe with (d) in
situ photodeposited and (e) impregnated Pt particles.
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313
5309

View Article Online
Fig. 3 Photocatalytic activity of in situ platinized HLe suspended in
aqueous (50 vol.%) methanol solution as a function of the amount of
Pt loading.
employed for platinization of HLe throughout this study. Fig.
3
shows the dependence of photocatalytic activity in the
aqueous methanol solution on the amount of Pt loaded by
photodeposition.
The H evolution was negligible without Pt loading, and
the rate increased with Pt loading up to 0.2È0.5%. Further
Fig. 2 (a) Time courses of photocatalytic H production from
2
2
aqueous (50 vol.%) methanol (open circles) and propan-2-ol (closed
loading at [1 wt.% reduced the activity, maybe due to the
enhancement of electronÈpositive hole recombination. Similar
behavior has been observed for an ordinary TiO2 photo-
catalyst.8 In the subsequent experiments, HLe platinized at 0.5
wt.% in situ was used.
If we assume that the rate of reaction (1) is much slower
when propan-2-ol is employed, i.e., the oxidation of propan-2-
ol by positive holes as a counter-reaction to Pt deposition is
slower than that of methanol, the presence of an induction
period as shown in Fig. 2(a) can be rationalized. However, the
following results made us consider another reason. Fig. 2(b)
circles) in the presence of suspended HLe particles. Hexachloro-
platinic acid solution containing 0.5 wt.% Pt was added before irra-
diation to be deposited during the photoirradiation. (b) Platinized
HLe powders were recovered from the experiment in part (a), dried
and used again in this experiment. Symbols are the same as those in
(
a).
for the complete reduction of hexachloroplatinic acid by the
following reaction:
H PtCl ] 2R1R2CHOH \ Pt ] 2R1R2CO ] 6HCl (1)
2
6
shows the time course of photocatalytic H evolution from
2
(
2
R1 \ R2 \ H for methanol and R1 \ R2 \ CH for propan-
methanol and propan-2-ol solutions, in the absence of hexa-
3
-ol) along with the dehydrogenation of alcohols to liberate
chloroplatinic acid, with a suspension of HLe that was pre-
loaded with Pt before the commencement of photoirradiation.
A powder of HLe recovered from each photoirradiated (9 h)
suspension and dried was used in these experiments. The
methanol system showed almost the same time course as that
in Fig. 2(a), suggesting that the time for Pt deposition is short
enough in this case. On the other hand, the induction period
for the propan-2-ol system was shortened by the use of a Pt-
preloaded HLe photocatalyst. These results suggest that the
induction period seen in Fig. 2(a) was due mainly to the time
needed for the in situ Pt deposition, but another reason still
existed to give an appreciable induction period in Fig. 2(b).
H (refs. 20, 22):
2
R1R2CHOH \ R1R2CO ] H
(2)
2
In this study, we detected almost the stoichiometric amount of
acetone in the propan-2-ol system, but we did not try to
analyze formaldehyde. Previous experiments using TiO as a
2
photocatalyst have shown the stoichiometric, i.e., 1 : 1, forma-
tion of formaldehyde with H from methanol.20
2
Fig. 1(d) shows a TEM picture of a Pt-deposited HLe
photocatalyst. Small Pt deposits of a few nanometres in size
were observed. Careful inspection of the TEM picture showed
that most of these Pt deposits were located along the step
lines, i.e., at the edge of the sheet or layer, rather than on the
Ñat terrace. It is presumed that such deposition proceeds via
two steps: reduction of hexachloroplatinic acid into Pt atoms
by photoexcited electrons in HLe, and coagulation of the
atoms into Pt deposits of nanometre size. However, in a
photocatalytic reaction, the deposited Pt can be a site for the
reduction. Therefore, we presumed that the reduction of hexa-
chloroplatinic acid proceeds at the step edge rather than on
the terrace. Previous reports have suggested that the interlayer
of HLe crystallite acts as a cation exchange material16 and,
thereby, hexachloroplatinic acid, an anion, may not be incorp-
orated into HLe. This is another reason for the Pt deposition
at the step edge. Platinization of HLe was also performed by
Fig. 4 shows the time courses of H evolution from meth-
2
anol and propan-2-ol solutions with three kinds of Pt-loaded
HLe powders: Pt photodeposited from methanol by 4 h irra-
diation (M-4) and from propan-2-ol by 2 h irradiation (during
the induction period) and by 14 h irradiation (at constant
photocatalytic activity). Again, the methanol system showed
almost the same time-course curves regardless of the kind of
photocatalyst, and the propan-2-ol system gave curves that
were slightly di†erent from each other. It is obvious that the
rates of H evolution, when they are compared at the time
2
after the induction period, were not so di†erent if the same
alcohol was used; the rate was predominantly controlled by
the kind of alcohol used, but not by the kind of photocatalyst.
The data in this Ðgure also indicate that the in situ Pt deposi-
tion might have been completed in the propan-2-ol system
even during the induction period, since the P-2 catalyst
showed comparable or even higher photocatalytic activity in
both systems. This also suggests that at least a part of the
induction period was due to a factor other than the time for
impregnationÈH reduction. As also shown in Fig. 1(e), this
2
procedure tended to produce aggregates of Pt particles, and
lower photocatalytic activity (ca. 3 lmol h~1) was observed,
presumably due to insufficient Pt loading; only a part of HLe
was platinized. Therefore, in situ photodeposition was
5310
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313

View Article Online
was added, and the suspension was purged of air by Ar bub-
bling for 15 min before irradiation. (c) HLe powder, an
aqueous solution of alcohol and hexachloroplatinic acid were
mixed, and the resulting suspension was purged of air for 15
min before irradiation (standard procedure).
The methanol system was also insensitive to these reaction
conditions and gave almost the same time-course curves in all
cases. On the other hand, the time of contact had an inÑuence
on the photocatalytic reaction of propan-2-ol; contact of the
HLe photocatalyst with propan-2-ol for a long time shortened
the induction period a little, but the steady state rate of H2
evolution was not greatly a†ected, as was seen after ca. 6 h
irradiation. These Ðndings suggest that the photocatalytic
reaction of propan-2-ol proceeds at the interlayer of the HLe
photocatalyst and that a considerably long contact time is
necessary to obtain a steady-state reaction rate. To exclude
the induction time due to the Pt photodeposition, the Pt-
loaded HLe powders were recovered, washed with water,
dried, and re-used in the photocatalytic reactions. Fig. 5(b)
shows the results. As expected, the methanol system did not
show any marked di†erence and the long-time contact was
Fig. 4 Time courses of photocatalytic H production from aqueous
2
(
50 vol.%) methanol (open symbols) and propan-2-ol (closed symbols)
in the presence of HLe particles with ex situ deposited Pt. Circles:
HLe recovered from 4 h irradiated suspension of methanol. Triangles:
HLe recovered from
2 h irradiated suspension of propan-2-ol.
Squares: HLe recovered from 14 h irradiated suspension of propan-2-
ol. Photodeposition of Pt was carried out under the same conditions
as in Fig. 2.
necessary for the photocatalytic H production from propan-
2
2
-ol, though the induction period was greatly shortened by
pre-deposition of Pt.
At present, we have no direct experimental evidence of the
reaction mechanism, including the incorporation of alcohol
molecules into the interlayer of the HLe photocatalyst.
However, if we assume that methanol molecules, which are
smaller than propan-2-ol molecules, can easily access the
interlayer but that propan-2-ol molecules cannot, the sensi-
tivity of the reaction rate toward the reaction conditions must
be more marked in the propan-2-ol system than in the meth-
anol system, as described above.
Pt deposition, e.g., time for incorporation of alcohol molecules
into the interlayer spaces in the HLe photocatalyst.
Fig. 5 shows the inÑuence of time of contact before the
commencement of photoirradiation. The following three
samples were prepared. (a) HLe powder was suspended in an
aqueous solution of alcohol and kept in the dark under an Ar
atmosphere for 42 h, and a solution of hexachloroplatinic acid
was added just before irradiation. (b) HLe powder was sus-
pended in an aqueous solution of hexachloroplatinic acid and
kept in the dark under an Ar atmosphere for 42 h. Alcohol
3
.3. Photocatalytic H evolution from various alcohols
2
Several aliphatic alcohols were used as substrates for the
photocatalytic reaction of HLe. Fig. 6 shows the steady-state
rates of H evolution by HLe with in situ and ex situ depos-
2
ited Pt. Except for methanol and ethanol, the time courses of
H evolution from these alcohols showed an induction period
2
Fig. 5 (a) Time courses of photocatalytic H production from
2
aqueous (50 vol.%) methanol (open symbols) and propan-2-ol (closed
symbols) in the presence of HLe particles. Circles: HLe and aqueous
alcoholic solution were mixed for 42 h before irradiation. Triangles:
HLe and hexachloroplatinic acid solution were mixed for 42 h before
the irradiation. Squares: HLe, alcohol, and hexachloroplatinic acid
solution were mixed just before photoirradiation. (b) HLe powders
were recovered from the reaction mixture of the experiment in part (a)
and used. Symbols are the same as those in (a).
Fig. 6 Steady-state rate of photocatalytic
H
production from
2
aqueous alcohol solutions by in situ (left, full bars) and ex situ (right,
open bars) platinized (0.5 wt.%) HLe particles (a) and ex situ platin-
ized (0.5 wt.%) P-25 particles (b). The concentration of alcohol was 50
vol.%, except for butan-1-ol (5.8 vol.%) and butan-2-ol (6.7 vol.%).
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313
5311

View Article Online
Table 1 Photocatalytic H production from aqueous alcohol solutions by in situ (0.5 wt.%) platinized HLe particles
2
Concentration
(%)
H
productiona
Induction period
/h
2
Alcohol
Abbrev.
/lmol h~1
Methanol
Ethanol
MeOH
EtOH
50
50
50
50
5.8
6.7
50
11.0
7.2
1.4
1.2
0.2
0.5
0.6
D0
D0
1.0
1.5
2.8
4.0
3.5
Propan-1-ol
Propan-2-ol
Butan-1-ol
Butan-2-ol
tert-Butyl alcohol
Pr-1-OH
Pr-2-OH
Bu-1-OH
Bu-2-OH
t-BuOH
a Quantum efficiency of H evolution from methanol solution was calculated to be ca. 0.4% from the results of previous experiments23 using the
2
same photoirradiation setup for TiO .
2
(
as observed and shown in the previous section for the
of the non-porous TiO photocatalyst, suggesting that size
and shape selection are achieved in HLe. This is the Ðrst
reported example of a selective photocatalytic reaction, except
for the preliminary results by DomenÏs group.17 Investigation
of extensions of the reaction substrate, e.g., amines or carbox-
ylic acids, is now under way.
2
propan-2-ol system), the period of which was increased with
the order shown in Table 1.
The smaller activity of HLe with pre-deposited Pt com-
pared with the in situ platinized one was commonly seen for
these alcohols; however, both platinized HLe powders showed
similar tendencies in the rate depending on the kind of
alcohol. The order obeyed the general rules of: (1) a higher
rate with a small number of carbon atoms, and (2) a higher
rate of straight chains than branched alkyl groups among
isomers of the same molecular weight. The exceptions of
butan-1-ol and butan-2-ol may be due to their lower concen-
Acknowledgements
Professor Kazunari Domen (Tokyo Institute of Technology) is
gratefully acknowledged for his helpful suggestions and stimu-
lating discussion throughout this study. The authors thank
Professor Masashi Inoue (Kyoto University) for his help in
transmission electron microscopic measurements and Dr
Masao Ohashi (Tokuyama College of Technology) for his
helpful discussion. This work was partly supported by a
Grant-in-Aid for ScientiÐc Research in Priority Area of
““Electrochemistry of Ordered InterfacesÏÏ (No. 11118208) from
the Ministry of Education, Science, Sports, and Culture,
Japan.
trations in the reaction mixture. In these experiments, H2
evolution is attributable to the reaction (2) (intramolecular
dehydrogenation), except for tert-butyl alcohol, which under-
goes intermolecular dehydrogenation and oxidative scission to
give acetone.21
Similar photocatalytic reactions were carried out by using
platinized P-25 TiO , which is known to consist of non-
porous spherical primary particles without size-selective pores
or spaces. As shown in Fig. 6(b), the activity was more than
one order of magnitude higher than that of HLe. Notably,
ethanol, but not methanol, gave the maximum rate in the case
2
of P-25 TiO , and reduction in the rate due to the alkyl chain
2
length and branching was not so obvious. (Relatively lower
References
activities for butan-1-ol and butan-2-ol were also observed.)
These di†erent behaviors of HLe and the non-porous P-25
support the hypothesis of a photocatalytic reaction mecha-
nism of HLe, including penetration of the substrate alcohol
into the inner space between the layers and oxidation by pho-
togenerated positive holes. This is also supported by the fact
that P-25 TiO2 gave negligible induction period for all the
kinds of alcohol used in this study. The shapes and sizes of the
alcohols to be adsorbed in the HLe controlled the rate of
photocatalytic reaction. Thus, a mechanism of shape- and
size-selective photocatalytic reaction has been proved as
1
2
3
L. Amalric, C. Guillard, N. Serpone and P. Pichat, J. Environ.
Sci. Health, A, 1993, 28, 1393.
R. Ho†mann, S. T. Martin, W. Choi and D. W. Bahnemann,
Chem. Rev., 1995, 95, 69.
A. Fujishima, K. Hashimoto and T. Watanabe, T iO
2
PhotocatalysisÈFundamentals and Applications, BKC, Tokyo,
1
999, p. 176.
4
5
6
M. A. Fox and M. T. Dulay, Chem. Rev., 1993, 93, 341.
H. Kisch, J. Prakt. Chem., 1994, 336, 635.
B. Ohtani, in New Challenges in Organic Electrochemistry, ed.
T. Osa, Gordon and Breach Science Publishers, Amsterdam,
1998, p. 55.
follows. The HLe catalyst consisted of TiO layers, as was
directly observed by TEM, and in situ photodeposition of Pt
7
B. Ohtani and T. Ohno, in PhotocatalysisÈFundamentals and
Applications, ed. M. Kaneko and I. Okura, Kodansha ScientiÐc,
Tokyo, 2000, in the press.
x
occurred at the step edge of layers, since an anionic PtCl 2~
6
cannot be adsorbed in the interlayer of anionic TiO . The
x
8
9
B. Ohtani, K. Iwai, S.-i. Nishimoto and S. Sato, J. Phys. Chem. B,
1997, 101, 3349.
substrate alcohol penetrates into HLe by exchange of as-
B. Ohtani, S. Tsuru, S.-i. Nishimoto, T. Kagiya and K. Izawa, J.
Org. Chem., 1990, 55, 5551.
incorporated water molecule. The rate and saturated amount
of adsorption of each alcohol in HLe, though they were not
directly measured, depended on the size and shape of each
alcohol.
10 H. Yoneyama and S. Nippa, Chem. L ett., 1988, 1807.
11
12
13
14
H. Yoneyama, S. Haga and S. Yamanaka, J. Phys. Chem., 1989,
9
3, 4833.
O. Beaune, A. Finiels, P. Geneste, P. Graffin, J.-L. Olive and A.
Saaedan, J. Chem. Soc., Chem. Commun., 1992, 1649.
B. Ohtani, K. Tennou, S.-i. Nishimoto and T. Inui, J. Photosci.,
4. Conclusion
1
995, 2, 7.
T. Sasaki, Y. Komatsu and Y. Fujiki, J. Chem. Soc., Chem.
Commun., 1991, 817.
It has been proved that HLe, having a layered crystal struc-
ture, shows sufficient photocatalytic activity for H2 pro-
15 T. Sasaki, M. Watanabe, Y. Michiue, Y. Komatsu, F. Izumi and
duction, along with oxidation of alcohol, from aqueous
S. Takenouchi, Chem. Mater., 1995, 7, 1001.
T. Sasaki, M. Watanabe, H. Hashizume, H. Yamada and H.
Nakazawa, J. Am. Chem. Soc., 1996, 118, 8329.
alcoholic solutions. The H production rate depended strong-
1
6
2
ly on the size and shape of the alcohol, unlike the dependence
5312
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313

View Article Online
1
7
M. Nakamura, Y. Sawada, S. Shibata, K. Shinohara, J. Nomura,
K. Domen, C. Hirose, T. Sasaki and M. Watanabe, T he 67th
Annual Meeting of Chemical Society of Japan, 1994, Tokyo,
20 S.-i. Nishimoto, B. Ohtani and T. Kagiya, J. Chem. Soc., Faraday
T rans. 1, 1985, 81, 2467.
21 S.-i. Nishimoto, B. Ohtani, H. Shirai and T. Kagiya, J. Chem.
Soc., Perkin T rans. 2, 1986, 661.
3
M110.
1
8
9
I. E. Grey, I. C. Madsen, J. A. Watts, L. A. Bursill and J. Kwiat-
kowska, J. Solid State Chem., 1987, 66, 7.
22 B. Ohtani, M. Kakimoto, S.-i. Nishimoto and T. Kagiya, J.
Photochem. Photobiol., A: Chem., 1993, 70, 265.
23 H. Kominami, T. Matsuura, K. Iwai, B. Ohtani, S.-i. Nishimoto
and Y. Kera, Chem. L ett., 1995, 693.
1
B. Ohtani, Y. Ogawa and S.-i. Nishimoto, J. Phys. Chem. B, 1997,
101, 3746.
Phys. Chem. Chem. Phys., 2000, 2, 5308È5313
5313