c

electrochemistry
www.elsevier.nl/locate/elecom communications
ELSEVIER Electrochemistry Communications 2 (2000) 207-210

Detection of active oxidative species in Ti02 photocatalysis using

the fluorescence technique
Ken-ichi Ishibashi a, Akira Fujishima a*b,Toshiya Watanabe a,c,Kazuhito Hashimoto a*c**
* Kanagawa Academy of Science and Technology, KSP Building, 3-2-I Sakado, Takatsrr-ku, Kawasaki, Kanagawa 213-0012, Japan
Department of Applied Chemistry, School of Engineering, The 1JniversiQ of Tokyo, 7-3-l Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
c Research Center for Advanced Science and Technology, The University of Tokyo. 4-6-l Komaba, Meguro-ku. Tokyo 153-8904. Japan

Received 13 December 1999: received in revised form 7 January 2000; accepted 10 January 2000

Abstract

The active oxidative species (mainly corresponding to hydroxyl radicals ( . OH) ) produced on TiOz film photocatalyst in aerated aqueous
solution have been quantitatively investigated by measuring the fluorescence derived by the reaction with either coumarin or terephthalic acid.
Upon UV light illumination, coumarin in aqueous solution and tcrephthalic acid in NaOH solution were converted on the TiO, film photocatalyst
to highly fluorescent hydroxy products. Analysis of the resultant hydroxy compounds provides sensitive and specific monitoring of the
generated active oxidative species. The rate of active species formation could be determined by the time dependence of the increase in
concentration of the resultant hydroxy products. Using this fluorescence method, the effect of light intensity on the generation efficiency of
active oxidative species was examined. The obtained results show that the efficiency of active oxidative species formation is constant at the
UV intensity from 50 pW cm- to 15 mW cm-, which is consistent with the results for . OH detection reported recently by the use of other
techniques. 02000 Elsevier Science S.A. All rights reserved.

Keywords; TiOZ photocatalysis; Hydroxyl radical; Fluorescence probe

1. Introduction to producehighly oxidizing .OH, which subsequentlyattacks

In recent yearsthe scientific andengineeringinterestin the
application of TiO, photocatalysis hasgrown exponentially
[ l-l 11. We have focusedour attention on developing TiO,-
coated materialswith self-cleaning, deodorizing and antibac-
terial functions using a weak UV light source in a living
environment [ 12-201. For mechanistic details of photoca-
talysis of TiO, materials, a number of researchershave
reported the detection and identification of various radicals
formed on TiOz during the photochemical reaction. Active
oxygen speciessuchashydroxyl radical ( *OH), superoxide
(0, - ) , and hydrogen peroxide (H,O,) have beenregarded
askey species.In fact we have shown recently by using the
chemiluminescencetechnique that the photogeneratedelec-
trons are trapped by adsorbedoxygen with a high quantum
yield, generating O,- [ 21,221. However, the detailed pro-
cessesof the consumptionof photogeneratedholesremain a
controversial issue,sofar.
In general, it is assumedthat photoinduced holes oxidize
hydroxide ions (or water molecules) adsorbedon the surface
* Corresponding author. Tel.: + 813-5452-5080;
e-mail: kazubito@fchem.chem.t.u-tokyo.ac.jp
an adsorbedpollutant molecule [ 23-271. The . OH radicals
producedareoften consideredto be the major speciesrespon-
sible for the photocatalytic oxidation reaction, although
detection is difficult becauseof its high reactivity. Some
evidenceexists to supportthe generationof .OH asthe active
species,that is, the formation of hydroxylated reaction inter-
mediates[ 28-301, distribution of the hydroxylation products
[ 31,321 and electron paramagneticresonance(EPR) detec-
tion of . OH using a spin trap agent [ 33-351.
In the present work, we have applied fluorescencetech-
niquesto the detection of .OH formed on a photo-illuminated
Ti02 surfaceusingeither coumarin or terephthalic acid which
readily reacts with . OH to produce highly fluorescentprod-
ucts, as shown in Fig. 1. This method hasbeen usedin radi-
ation chemistry, sonochemistry and biochemistry [ 36-471
for the detection of .OH generatedin water.
2. Experimental
fax: + 813-5452-5083; Coumarin, terephthalic acid and sodium hydroxide
( NaOH) of the purest gradewere purchasedfrom Wako Pure

1388-2481/00/$ - see front matter 02000 Elsevier Science S.A. All rights reserved
PIIS1388-2481(00)00006-0
208 K. lshibashi et al. /Electrochemistp
7-hydroxycoumarin
+-OH - (1)
Terephthalic acid 2-hydroxyterephthalic acid
COOH COOH
+.OH - (2)
6OOH Coon
Fig. 1. Formation of hydroxy products as the result of the reactions between
the probe molecules and .OH: ( 1) reaction between OH and coumarin;
(2) reaction between OH and terephthalic acid.
Chemicals. Coumarin was dissolved in pure water and used
ftir the experiments. For terephthalic acid, in order to ensure
solubility, solutions was prepared in dilute (2 X lo- M)
NaOH solution. The pure water used in this experiment was
distilled, deionized and then purified with the Milli-Q system
(Nippon Millipore Co., Tokyo).
Fluorescence spectra of generated 7-hydroxycoumarin and
2-hydroxyterephthalic acid were measured on a Hitachi F-
4500 fluorescence spectrophotometer.
Soda-lime glass coated with silica thin film (about 100
nm), were dip-coated using titanium isopropoxide solution
(Type NDH-520C, Nippon Soda). A withdrawal rate of 20
cm n-r& was used for all plates. Coated titanium isopro-
poxide was wiped off from one side of the plate using ethyl
acetate and the plates were dried in a furnace for 30 min at
120 C. The plates were subsequently calcined at 500 C for
1 h. The above dip-coating procedure was repeated four times
to produce TiO, film with anatase crystal structure on one
side of the plates.
A 9 X 9 mm TiO, film plate was placed at the side of the
bottom half of a 1 X 1 cm fused silica standard cell with 1 mL
of coumarin or terephthalic acid solution. TiOz film plate in
the solution was illuminated with a 200 W high-pressure
mercury lamp (Yamashita Denso, HYPERCURE 200)
through a UV filter (Toshiba, UV-D36B). The light intensity
was controlled in the range from 50 PW cm- to 15 mW
cm- using neutral density filters.
3. Results and discussion
TiO, film in coumarin 10 ~ M solution which was mag-
netically stirred was illuminated with UV light (1 mW
cm-). The fluorescence emission spectrum excited at 332
nm from the coumarin solution was measured every 5 min of
illumination. Fig. 2 (a) shows the induction of fluorescence
from lop3 M coumarin solution. Similar fluorescence meas-
urements were conducted with 3 15 nm excitation on 5 X 1O-4
M terephthalic acid solution in 2 X lo- M NaOH (Fig.
2(b)). As shown in the figures, gradual increases in the
fluorescence at about 460 nm and about 425 nm for coumarin
and terephthalic acid, respectively, were observed by the con-
tinuous illumination of UV light on the TiO, film. However,
Communications 2 (2000) 207-210
no increase of fluorescence was observed either when the
TiO, film had not been illuminated with UV light, or when
the substrate had not been coated with TiO, stirred in open
air. These results suggest that the fluorescence is caused by
chemical reactions of coumarin or terephthalic acid with spe-
cies formed on the TiO, via photocatalytic reactions.
Aqueous solutions of coumarin and terephthalic acid have
been used in the past as sensitive chemical dosimetry systems
in radiation chemistry [ 36-411 and sonochemistry [ 42-451.
When water is irradiated with ultrasound, . OH, HO?. , O2 -
and H202 are formed as a result of the high energies generated
during cavitational collapse. As an ionizing gamma or X-ray
photon travels through water, it also produces these reactive
intermediates either directly or indirectly by the splitting of
a water molecule. These generated radicals show remarkable
resemblance to the active oxygen species that have been pos-
tulated in Ti02 photocatalysis, e.g., . OH, OZ -- or H207_.
For the irradiation of gamma ray or ultrasound to coumarin
or terephthalic acid solutions, the induction of fluorescence
is known to be mediated specifically by the . OH to produce
fluorescent hydroxy products. Other primary water radicals
and secondary radicals, such as H. and HO,? , and also H207
by interaction of primary water radicals, do not significantly
influence the induced fluorescence. On the basis of these
results, the generation of fluorescent products in photo-illu-
minated TiOz systems would be ascribed to the reaction
between photogenerated . OH and coumarin or terephthalic
acid.
When reacted with . OH, coumarin forms only one mono-
hydroxylated isomer and, for terephthalic acid (Eq. ( 1) in
Fig. 1)) 2-hydroxyterephthalic acid is the only product having
5 1600
rd 1400
g 1200
E
2? 1000
.c8 800
5 600
P
E 400
s 200
G
0
Wavelength / nm
400 450 500 550
Wavelength / nm
Fig. 2. Fluorescence spectral changes observed during illumination of TiO,
film in (a) 1 X 10 M aqueous solution of coumarin (excitation at 332
nm) and (b) 4 X 10 ~ M NaOH solution of terephthalic acid (excitation at
315 nm). Each fluorescence spectrum was recorded every 5 min of UV
illumination.
 K. Ishibashi et al. /Electrochemistry Communications 2 (2000) 207-210 209

significant fluorescence (Eq. (2) in Fig. 1) . It can be seen

from Fig. 2(a) and (b) that the spectra of the generated
fluorescent products have the identical shape and wavelength
of maximum emission as that of 7-hydroxycoumarin or 2-
hydroxyterephthalic acid. Therefore, probably, the fluores-
cent products formed during TiO;, photocatalysis are due to
the reaction between . OH with coumarin or terephthalic acid.
There remains the possibility that photogenerated holes
directly react with coumarin or terephthalic acid to give a 5 10 15 20 25 30

hydroxy product via the radical cation [ 48-5 1] . However, in Illumination time / min

general, this reaction pathway proceeds preferentially when

the concentration of the substrate is high, e.g. lo- -10m2 M,
or when the substrate adsorbs strongly on the surface. It is
well known that aromatic compounds are oxidized by .OH
much faster than photogenerated holes in TiO, catalysis.
Moreover, when the concentration of aromatic compounds is
low (less than 10P3 M), they are mainly oxidized by . OH
[ 30,52,53]. Therefore, under the present experimental con-
ditions of presumably lo--lo- M probe molecules,
Illumination time I min
hydroxylation reactions of coumarin or terephthalic acidpro-
Fig. 3. Plots showing (a) the induced fluorescence intensity (460 nm)
teed mainly by . OH.
against the light illumination time for coumarin and (b) the induced fluo-
For more quantitative considerations, it is necessary to rescence intensity (425 nm) against light illumination time for terephthalic
know whether all the active oxidative species are trapped by acid.
probe molecules or not. We investigated the influence of
stirring the solution at different initial concentrations of the
probe molecules. For coumarin, below a concentration of
about 10e4 M, induction of fluorescence was increased by
stirring the solution magnetically during illumination. This
implies that there are insufficient coumarin molecules on the
TiO, film to react with all the active oxidative species. How-
ever, around the concentration of lo- M, no change in the
florescence intensity was observed even when the solution
0 2 4 6 8 10 14x10'
was stirred. When the concentration was raised significantly,
Light intensib / pLJ set-1 cm-2
e.g., 5 X 10d3, the fluorescence was reduced due toquenching
Fig. 4. Dependence of the initial formation rates of active oxidative species
and inner filter effects. Thus, there is a relatively narrow
on illuminated light intensity.
concentration region around 10e3 M which can be used for
the present purpose. In the case of terephthalic acid, the con- cm-* to 15 mW cm-*. This result is consistent with the
centration of 4 X 10e4 M gave optimum results. recently reported results which indicate the linear dependence
Fig. 3 (a) and (b) show the plots of increase in fluores- of . OH formation on the light intensity [ 54,551, suggesting
cence intensity against illumination time at 460 nm (cou-
the validity of the fluorescence method applied in this
marin) and 425 nm (terephthalic acid) respectively. The
experiment.
fluorescence intensity by UV light illumination in both cou-
marin and terephthalic acid solutions increased linearly
against time. Consequently, we can conclude that -OH rad-
icals formed at the TiO, interface are in proportional to the 4. Conclusions
light illumination time obeying zero-order reaction rate
kinetics. In this study, we have adapted the use of coumarin or
As was demonstrated above, the tentative detection of pho- terephthalic acid solution to measure . OH production in Ti02
togenerated . OH in Ti02 photocatalysis was suggestedusing photocatalysis. This method relies on the fluorescent signal
coumarin or terephthalic acid as a probe molecule. Therefore, generated by the hydroxylation of coumarin or terephthalic
we investigated the initial hydroxy product generation rate at acid. By using this technique, the tentative detection of . OH
different light intensities at the water ] TiO, interface applying generated at the water]TiO, interface was suggested. The
terephthalic acid as the probe molecule. In Fig. 4 is shown fluorescence method possesses the following attributes:
the UV light intensity dependence on the increase of fluores- rapid, sensitive, specific, needs only a simple standard instru-
cence intensity. This figure suggests that the efficiency of mentation and, once formed, the fluorescent product is stable
. OH formation is constant at the UV intensity from 50 FW and does not affect the normally occurring . OH reactions.
210 K. lshibashi et al. /Electrochemisty Communications 2 (2000) 207-210

References 1271 C.S. Turchi, D.F. Ollis, J. Catal. 122 (1990) 178.
[281 D.F. Ollis, C-Y. Hsiao, L. Budiman, C.-L. Lee, J. Catal. 88 (1984)
89.
[ll A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem. Rev. 95 (1995) 735. 1291 H. Al-Ekabi, N. Serpone, E. Pelizzetti, C. Minero, M.A. Fox, R.B.
[21 M.R. Hoffmann, S.T. Martin, W. Choi, D.W Bahnemann, Chem. Rev. Draper, Langmuir 5 (1989) 250.
95 (1995) 69. 1301 K. Okamoto, Y. Yamamoto, H. Tanaka, A. Itaya, Bull. Chem. Sot.
[31 D.F. Ollis, E. Pelizetti, N. Serpone, Environ, Sci. Technol. 25 ( 1991) Jpn. 58 (1985) 2015.
1523. 1311 M. Fujihira, Y. Satoh, T. Osa, Nature 293 ( 1981) 206.
r41 A. Heller, Act. Chem. Res. 28 (1995) 503. [=I C. Richard, J. Photochem. Photobiol. A 72 ( 1993) 179.
[51 M. Schiavello, Photocatalysis and Environment, Kluwer, Dordrecht, [331 C.D. Jaeger, A.J. Bard, J. Phys. Chem. 83 (1979) 3146.
1988. [341 J.R. Harbour, M.L. Hair, Adv. Colloid Interface Sci. 24 (1986) 103.
[61 N. Serpone, Res. Chem. Intermed. 20 (1994) 953. 1351 G. Riegel, J.R. Bolton, J. Phys. Chem. 99 (1995) 4215.
[71 A. Mills, R.H. Davies, D. Worsley, Chem. Sot. Rev. (1993) 417. [361 A.K. Collins, G.M. Makrigiorgos, G.K. Svensson, Med. Phys. 21
[81 D.F. Ollis, C. Turchi, Environ. Prog. 9 ( 1990) 229. (1994) 1741.
[91 H. Uchida, S. Itoh, H. Yoneyama, Chem. Lett. ( 1993) 1995. [371 H.M. Khan, M. Anwar, G. Ahmad, J. Radioanal. Nucl. Chem. Lett.
[lOI M.A. Fox, M.T. Dulay, Chem. Rev. 93 (1993) 341. 200 (1995) 521.
[II P. Pichat, Catal. Today 19 (1994) 313. [381 K. Gopakumar, U.R. Kini, S.C. Ashawa, N.S. Bhandari, G.U.
[I21 N. Negishi, T. Iyoda, K. Hashimoto, A. Fujishima, Chem. Lett. ( 1995) Krishnan, D. Krishnan, Radiat. Eff. 32 ( 1977) 199.
841. [391 R.W. Matthews, Radiat. Res. 83 ( 1980) 27.
H. Matsubara, M. Takada, S. Koyama, K. Hashimoto, A. Fujishima, [401 S.C. Ashawa, U.R. Kini, U. Madhvanath, Int. J. Appl. Radiat. Isot. 30
Chem. Lett. (1995) 767. (1979) 7.
K. Hashimoto, A. Fujishima, Denki Kagaku 60 (1992) 1057. [411 W.A. Armstrong, R.A. Facey, D.W. Grant, W.G. Humphreys, Can. I.
Chem. 41 ( 1963) 1575.
T. Watanabe, A. Kitamura, E. Kojima, C. Nakayama, K. Hashimoto,
A. Fujishima, in: D.E. Ollis, H. Al-Ekabi (Eds.), Photocatalytic Puri- [421 G.J. Price, E.J. Lenz, Ultrasonics 3 1 ( 1993) 45 1.
fication and Treatment of Water and Air, Elsevier, New York, 1993. [431 T.J. Mason, J.P. Lorimer, D.M. Bates, Y. Zhao, Ultrasonics Sono-
them. 1 (1994) S91.
Y. Kikuchi, K. Sunada, T. Iyoda, K. Hashimoto, A. Fujishima, J.
1441 X. Fang, G. Mark, C. Sonntag, Ultrasonics Sonochem. 3 ( 1996) 57.
Photochem. Photobioi. A: Chem. 106 ( 1997) 5 1.
[451 J.R. Mclean, A.J. Mortimer, Ultrasound Med. Biol. 14 ( 1988) 59.
I. Sopyan, S. Murasawa, K. Hashimoto, A. Fujishima, Chem. Lett.
(1994) 723. 1461 G.M. Makrigiorgos, J. Baranowska-Kortylewicz, E. Bump, S.K. Sahu,
R.M. Berman, AI. Kassis, Int. J. Radiat. Biol. 63 (1993) 445.
Y. Ohko, K. Hashimoto, A. Fujishima, J. Phys. Chem. A 101 (1997)
1471 J.C. Barreto, G.S. Smith, N.H.P. Strobel, P.A.McQuillin,T.A.Miller,
8057.
Life Sci. 56 ( 1995) 89.
[I91 P. Sawunyama, L. Jiang, K. Hashimoto, A. Fujishima, J. Phys. Chem.
[481 M.K. Eberhardt, J. Org. Chem. 42 (1977) 832.
B 101 (1997) 11000.
[491 M.K. Eberhardt, G. Ramirez, E.J. Ayala, J. Org. Chem. 54 ( 1989)
1201 K. Sunada, Y. Kikuchi, K. Hashimoto, A. Fujishima, Environ. Sci. 5922.
Technol. 32 (1998) 726.
[501 C. Walling. Act. Chem. Res. 8 ( 1975) 125.
Pll K. Ishibashi, Y. No&a, K. Hashimoto, A. Fujishima, J. Phys. Chem. [511 C. Walling, D.M. Camaioni, S.S. Kim, J. Am. Chem. Sot. 100 ( 197X)
102 (1998) 2117. 4814.
w-1 K. Ishibashi, K. Hashimoto, A. Fujishima, J. Phys. Chem., submitted 1521 S. Goldstein, G. Czapski, J. Rabani, J. Phys. Chem. 98 ( 1994) 6586.
for publication. 1531 J. Petal, J. Casado, J. Domenech, J. Photochem. Photobiol. A: Chem.
[=I R.I. Bickley, F.S. Stone, J. Catal. 3 1 (1973) 389. 44 (1988) 209.
1241 RI. Bickley, V. Vishwanathan, Nature 280 (1979) 306. [541 M.A. Grela, M.E.J. Coronel, A.J. Colussi, J. Phys. Chem. 100 (1996)
[251 H.V. Damme, W.K. Hall, J. Am. Chem. Sot. 101 (1979) 4373, 16940.
[261 P.F. Schwarz, N.J. Turro, S.H. Bossmann, A.M. Braun, A.-M.A. 1551 P.F. Schwarz, N.J. Turro, S.H. Bossmann, A.M. Braun, A.-M.A.A.
Abdel Wahab, H. Dim, J. Phys. Chem. B 101 (1997) 7127. Wahab, H. Diirr, J. Phys. Chem. B 101 (1997) 7127.