Nitrogen-doped SrTiO3/TiO2 composite photocatalysts for hydrogen production under visible light irradiation

Article abstract of DOI:10.1016/j.jallcom.2008.04.078

Nitrogen-doped SrTiO₃ powders were prepared by solid phase method, and further combined with TiO₂ by sol-gel method. Then Pt particles were deposited to prepare Pt-loaded composite powders by hydrogen reduction method. These obtained

Full text of DOI:10.1016/j.jallcom.2008.04.078

Journal of Alloys and Compounds 472 (2009) 429–433
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Nitrogen-doped SrTiO3/TiO2 composite photocatalysts for hydrogen
production under visible light irradiation
Yan Jian-Huia,b, Zhu Yi-Ronga,b,∗, Tang You-Gena, Zheng Shu-Qina,b
a Department of Chemistry and Chemical Engineering, Hunan Institute of Science and Technology, Yueyang 414000, China
b School of Chemistry and Chemical Engineering, Central South University, Changsha 410083, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Nitrogen-doped SrTiO3 powders were prepared by solid phase method, and further combined with TiO2
by sol–gel method. Then Pt particles were deposited to prepare Pt-loaded composite powders by hydro-
gen reduction method. These obtained powders were characterized by X-ray diffraction (XRD), scanning
electron microscopy (SEM), Fourier transform infrared spectra, thermogravimetric–differential thermal,
ultraviolet (UV)–vis diffuse reflectance spectra and fluorescence spectra techniques. Photocatalytic pro-
duction of hydrogen from oxalic acid solution was used as the probe reaction to evaluate the photocatalytic
activity of hydrogen production under visible light in detail from aspects of different calcination tempera-
tures, and different contents of metal Pt. The result demonstrates that when the calcination temperature is
400 ◦C, the optimized photocatalytic activity of hydrogen production under visible light can be achieved.
Pt depositing greatly improves the photocatalytic activity, and the average hydrogen production rate is
up to 5.1 mmol g cat−1 h−1 with 2 wt.% loaded Pt.
Received 21 October 2007
Received in revised form 19 April 2008
Accepted 24 April 2008
Available online 10 June 2008
Keywords:
Composite catalyst
Sol–gel
Photocatalytic
Nitrogen-doped
Hydrogen production
© 2008 Elsevier B.V. All rights reserved.
1. Introduction
were further deposited to prepare Pt-loaded composite powders
by hydrogen reduction method, whose photocatalytic activity for
Energy and environment have become two important problems
with the rapid development of industry. Photocatalytic reactions of
semiconductor such as decomposition of water have been received
great attention around the world [1,2]. SrTiO3, one of the important
photocatalysts, has been used for water splitting and mineralization
of organic pollutants under ultraviolet (UV) irradiation, whereas it
is active only in the UV region because of its wide band gap (3.2 eV).
Therefore, its photocatalytic properties in UV light range have been
investigated for several decades [3,4]. Nowadays, the research on
photocatalysis in the field of visible light has attracted many peo-
ple’s attention [5–9], and it has become a target of researchers to
improve the photocatalytic activity of SrTiO3 and extend its optical
absorption edge towards the visible light range. To our knowledge,
researches of SrTiO3 on the photocatalytic hydrogen production
are still scarce. We have not found any report on nitrogen-doped
SrTiO3/TiO2 photocatalyst under visible light irradiation.
hydrogen production from oxalic acid solution was investigated
under visible light irradiation. The effects of several factors such as
the calcination temperatures of composite powders, and the con-
tents of metal Pt on the rate of hydrogen production were studied.
2. Experimental
2.1. Preparation of photocatalysts
According to our previous research [10], the nitrogen-doped SrTiO3 powders
were prepared as follows: hexamethylenetetramine (HMT) was mixed with a per-
ovskite SrTiO3 powder and grinded thoroughly in an agate mortar. The mass ratio
of hexamethylenetetramine to SrTiO3 powders was 3 to 1. The mixed powder was
packed in a lidded double alumina crucible and calcined at 450 ◦C under aerated
conditions for 1 h. Nitrogen-doped SrTiO3 powders obtained was grinded in mortar
after calcination.
Nitrogen-doped SrTiO3/TiO2 composite powders were prepared as follows:
tetra-butyl titanate solution (10 ml) was added into anhydrous ethanol solution
(20 ml) at room temperature under stirring, and the apparent slight yellowish solu-
tion was designated as A; the double distilled water (6 ml) and anhydrous ethanol
solution (30 ml) were added into concentrated nitric acid (1.5 ml) under stirring, and
the solution was designated as B. These two solutions were both continuously stirred
for 15 min, then A solution was added dropwise (1 ml min−1) into B solution under
vigorous stirring. After stirring for 2 h, nitrogen-doped SrTiO3 powders were added
(ensuring that the content of nitrogen-doped SrTiO3 powders in composite catalysts
is 30 wt.%) and stirred continuously for some time until the gel formed, which was
allowed to stand at room temperature for a further 24 h to complete peptization. It
was dried in vacuum conditions for 12 h, and then calcined at different temperatures
for 1 h.
In this paper, nitrogen-doped SrTiO3 powders were prepared
by solid phase method, and nitrogen-doped SrTiO3/TiO2 compos-
ite powders were prepared by sol–gel method and then Pt powders
∗
Corresponding author at: Department of Chemistry and Chemical Engineering,
Hunan Institute of Science and Technology, Yueyang 414000, China.
Tel.: +86 730 8648503; fax: +86 730 8640436.
E-mail address: zhuyirong2004@163.com (Y.-R. Zhu).
0925-8388/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2008.04.078

430
J.-H. Yan et al. / Journal of Alloys and Compounds 472 (2009) 429–433
Different contents of Pt-loaded nitrogen-doped SrTiO3/TiO2 composite pow-
ders were prepared by impregnation hydrogen reduction method, and the required
weight of nitrogen-doped SrTiO3/TiO2 powders through ultrasonic treatment were
added to different concentrations of chloroplatinic acid hexahydrate solution. This
mixture solution obtained was stirred at an ambient temperature overnight, and
dried in water at 373 K for several hours, and then the dried powders were trans-
ferred to tubular furnace. The samples were heated from room temperature to 673 K
in a nitrogen atmosphere, and then maintained at 673 K for 2 h after the nitro-
gen atmosphere was replaced by hydrogen atmosphere, and finally cooled to room
temperature under nitrogen atmosphere protection.
2.2. Characterization of photocatalysts
The X-ray diffraction (XRD) patterns of the samples were obtained on a Rigaku
D/max 2550 VB+ 18 kW X-ray diffractometer with Cu K␣ radiation at a scanning
rate of 0.1◦ 2ꢀ s−1. The accelerating voltage and the applied current were 40 kV and
300 mA, respectively. The average crystallite sizes were determined according to the
Scherrer equation. The surface morphologies of the samples were obtained from
JEOL JSM-5600LV scanning electron microscopy (SEM). The chemical structure of
the samples was examined using the Fourier transform infrared spectrophotome-
ter (FT-IR, AVTATAR, 370) in the 400–4000 cm−1 frequency range with KBr as a
reference. Thermogravimetric and differential thermal (TG–DTA) patterns of the
samples were recorded by using thermogravimetric–differential thermal analyzer
to determine the temperature of possible decomposition and phase changes and the
samples were heated at the rate of 10 ◦C min−1 from room temperature to 1000 ◦C.
The UV–vis spectroscopy of the samples was recorded on a UV–vis spectropho-
tometer (type 100–60) for the UV–vis absorption studies with wavelength range of
200–700 nm. Fluorescence spectrum (FS) was obtained with an LS-55 fluorescence
spectrum spectrometer.
Fig. 1. XRD patterns of the samples: (a) SrTiO3 and (b) N–SrTiO3.
SrTiO3 are of small spherical shape. They have homogeneous dis-
tributions and only a fraction of agglomeration, and the surface
morphologies are similar to those of pure SrTiO3 powders. As
shown in Fig. 3(b), the as-prepared composite catalyst is irregu-
larly spherical and its particle sizes are not identical. Some large
particles are nitrogen-doped SrTiO3 whose surfaces were covered
by tiny TiO2 particles to form a heterogeneous core-shell structure,
furthermore, the surface of particles is not smooth, which offers a
good site for the photocatalytic reaction. The surface morphology
of Pt-loaded composite catalyst is similar to that of unloaded one,
and no Pt particles are observed mainly because of low Pt content.
Fig. 4 displays the Fourier transform infrared spectra of TiO2
(400 ◦C), nitrogen-doped SrTiO3 and nitrogen-doped SrTiO3/TiO2
(600 ◦C). From the FT-IR spectrum of TiO2, the peaks of 3420 and
1630 cm−1 correspond to the stretching vibrations of O H and
bending vibrations of adsorbed water molecules, respectively, indi-
cating the existence of O H on the surface of TiO2. The stretching
vibrations of Ti O are not observed due to the coverage of strong
absorption band in the range of 750–450 cm−1. From the FT-IR spec-
trum of nitrogen-doped SrTiO3, the peak of 3440 cm−1 corresponds
to the stretching vibrations of adsorbed water O H, and the absorp-
tion peak of 1458 cm−1 is probably caused by the residual organic
compounds coming from the process of nitrogen doping. The strong
2.3. Photocatalytic experiments
The photocatalytic reaction was carried out in self-made tube-shaped quartz
reactor including three layers connected with gas collecting devices, and water was
used as the external circulation cooling and wind as the internal cooling. The reac-
tion temperature was kept at 25 0.2 ◦C by controlling the external circulation water
in the water jacket of the reactor during the entire experiment. The photocatalyst
powder (0.6 g) was dispersed by a magnetic stirrer in 600 ml aqueous solution con-
taining 0.05 mol L−1 oxalic acid as the sacrificial reagent. A 250-W xenon lamp was
used as the light source. Before the irradiation, the reactor was deaerated with nitro-
gen for about 30 min, and the catalyst was kept in suspension state by a magnetic
stirrer. The gas evolved was gathered by drainage method and analyzed by GC using
a molecular sieve 13× column and nitrogen as the carrier gas.
3. Results and discussion
3.1. Characterization of photocatalysts
XRD is used to investigate the phase structures of the samples.
The XRD patterns of SrTiO3, N–SrTiO3, N–SrTiO3/TiO2 (350–600 ◦C)
and Pt/N–SrTiO3–TiO2(400 ◦C, 2 wt.%) are shown, respectively in
Figs. 1(a) and (b) and 2(c)–(g) and (h). The average crystal size of
SrTiO3 was determined to be 38.8 nm and nitrogen-doped SrTiO3
samples was 38.4 nm. The average crystal size of the TiO2 sin-
gle crystal in composite catalyst calcined at 350, 400, 450, 500
and 600 ◦C was determined to be 8.5, 9.1, 9.5, 11.8 and 12.6 nm,
respectively. The size of the particles becomes bigger with the tem-
perature rising, and this is confirmed by the XRD patterns in which
the width of the diffraction peak is narrower and the shape of peak
is sharper as the temperature rises. The results obtained were also
agrees with those observed by SEM. The composite catalysts mainly
consist of anatase TiO2 and SrTiO3. From Fig. 2(g) we can observe
that no peak of rutile phase TiO2 is observed after being calcined
at 600 ◦C (g), which illustrates that the combination of nitrogen-
doped SrTiO3 with TiO2 inhibits the crystal transformation of TiO2
in composite catalyst to some extent. In addition, the peaks of nitro-
gen and platinum are not observed either from Figs. 1 and 2, and
this may be due to the low dopant content of nitrogen and platinum
beyond the minimum detection limit of XRD.
The SEM micrographs of the nitrogen-doped SrTiO3 and Pt-
loaded nitrogen-doped SrTiO3/TiO2 (400 ◦C, 2 wt.%) are shown in
Fig. 3(a) and (b). From Fig. 3(a), the particles of nitrogen-doped
Fig. 2. XRD patterns of the samples: (c–g) N–SrTiO3/TiO2 and (h) Pt/N–SrTiO3–TiO2.

J.-H. Yan et al. / Journal of Alloys and Compounds 472 (2009) 429–433
431
Fig. 5. TG–DTA curve of N–SrTiO3/TiO2 gel.
At the same time, the absorption peak near 1400 cm−1 is due to
the Ti Ti vibration [11], which further shows the interaction of
O
the two different semiconductors after combination. However, this
peak is not observed on the single sample of nitrogen-doped SrTiO3
and TiO2 from the FT-IR spectra.
Fig. 5 shows the TG–DTA curve of nitrogen-doped SrTiO3/TiO2
dry gel. From the differential thermal curve, the strong endothermic
peak at about 100 ◦C is due to the free adsorbed water of the gel and
the volatilization of organic solvents such as ethanol and acetic acid
[12], meanwhile the corresponding weight loss is observed from
the thermogravimetric curve. The first exothermic peak at 152 ◦C
is due to the decomposition of the residual organic compounds;
the second exothermic peak at 358 ◦C corresponds to the crystal-
lization of the amorphous phase TiO2 into the anatase phase. In
addition, the appearance of another endothermic peak and slight
weight loss of the thermogravimetric curve at about 578 ◦C are per-
haps owing to the transformation of the anatase phase TiO2 into
the anatase phase, besides, the possibility of the so-called hetero-
junction energy effects caused by the combination of two different
oxides could not be ruled out.
Fig. 3. SEM patterns of the samples: (a) N–SrTiO3 and (b) Pt/N–SrTiO3/TiO2.
Fig.
6 displays the UV–vis absorption spectra of SrTiO3,
broad absorption peak in the range of 3100–3450 cm−1 corresponds
to the O H stretching vibrations of the nitrogen-doped SrTiO3/TiO2
composite samples, which indicates that the two different semi-
conductors interact with each other after combination, resulting
in existence of more O H on the surface of composite catalysts.
N–SrTiO3, N–SrTiO3/TiO2 (400 ◦C) and Pt/N–SrTiO3/TiO2 (400 ◦C,
2 wt.%). From Fig. 6, the absorption intensity of the samples is
almost the same in UV light range, whereas it is evidently differ-
Fig. 6. UV–vis diffuse reflectance spectra of the samples: (a) SrTiO3, (b) N–SrTiO3,
Fig. 4. Infrared spectra of the samples: (a) N–SrTiO3, (b) TiO2 and (c) N–SrTiO3/TiO2.
(c) N–SrTiO3/TiO2 and (d) Pt/N–SrTiO3/TiO2.

432
J.-H. Yan et al. / Journal of Alloys and Compounds 472 (2009) 429–433
ent in visible light range. The diffuse reflectance rate of SrTiO3 is
appropriate to 100% in the range of 400–700 nm, namely, it shows
no absorption for the visible light. While the diffuse reflectance rate
of modified SrTiO3 evidently decreased, and it shows rather strong
absorption for the visible light. After SrTiO3 was doped by nitro-
gen, the replacing O2− with N3− would result in the formation of
anion defects for the charge compensation, and the anion defects
seemed to lead to high visible light absorption ability of the sample,
according to Justicia et al.’s work [13]. When the nitrogen-doped
SrTiO3 combines with TiO2, the interaction occurs between the
two different semiconductor oxides, and the visible light absorption
ability is further enhanced compared to nitrogen-doped SrTiO3 so
that it is favorable for the improvement of the photocatalytic activ-
ity. UV–vis diffuse reflectance spectrum is also a common method
used to track the formation of metal nanoparticles. The visible light
absorption of the composite catalysts loaded by Pt is significantly
enhanced. According to literature [14], it results from the contri-
bution of metal nanoparticles plasma absorption peak, whereas
the different plasma absorption peaks are caused by many reasons,
mainly including the size and shape of deposited metal particles,
surface electron density of the metal particles and property of
medium for metal deposition. It is also demonstrated in experiment
that nitrogen doping, combination of different semiconductors and
deposition of noble metal all contribute to the enhancement of
absorption for the visible light.
Fluorescence spectrum is an effective method for researching
on the electronic structure and optical properties of the semi-
conductor nanoparticles as well as showing the information of
oxygen vacancy or defects of material surface and the separation
and recombination of photo-generated carriers [15,16]. The fluores-
cence spectrum intensity determines the number of electron–hole
recombination. A large number of electrons and holes generate
after the excitement of photocatalysts under visible light irradia-
tion and only a fraction of them recombine rapidly. Their energy is
released in the form of fluorescence, and high fluorescence spec-
trum intensity means high electron–hole pair recombination rate
and poor photocatalytic activity. Fig. 7 displays the fluorescence
spectra of SrTiO3, N–SrTiO3, TiO2 (400 ◦C), N–SrTiO3/TiO2 (400 ◦C),
and Pt/N–SrTiO3/TiO2 (400 ◦C, 2 wt.%). As shown in Fig. 7, the flu-
orescence spectrum intensities of SrTiO3 and TiO2 are relatively
higher, whichillustratesthat thephoto-induced electronsandholes
are easy to recombine together. The fluorescence spectrum inten-
sity of nitrogen-doped SrTiO3 decreases slightly, showing that the
appropriate content of doped nitrogen facilitates the effective sep-
Fig. 8. Photocatalytic hydrogen production activity of N–SrTiO3/TiO2 calcined at
different temperatures.
aration of photo-generated electron–hole pair since the replacing
O2− with N3− will result in the formation of oxygen vacancies, and
the appropriate oxygen vacancies can inhibit the recombination
of photo-generated charge to some extent [17]. The fluorescence
spectrum intensity of nitrogen-doped SrTiO3/TiO2 decrease greatly,
illustrating that the combination of nitrogen-doped SrTiO3 with
TiO2 forms the heterojunction structure, which becomes the con-
venient channel for the transfer of photo-generated electrons and
holes in the process of photocatalytic reaction because of its spe-
cial energy band structure and transport characteristics of carriers,
leading to the effective separation of photo-generated carriers,
the reduction of electron–hole recombination, and thus the pho-
tocatalytic activity improves [18,19]. The fluorescence spectrum
intensity further decreases after depositing Pt, which indicates
that the deposition of Pt plays a competitive role in capturing the
photoelectron, and results in the separation of photo-generated
electron–hole pairs, the reduction of photo-generated carriers
recombination, therefore, the photocatalytic activity enhances. The
order of fluorescence spectrum intensity agrees with that of pho-
tocatalytic activity obtained by our experiments.
3.2. Photocatalytic activity of photocatalysts
Photocatalytic hydrogen production activity of nitrogen-doped
SrTiO3/TiO2 composite samples calcined at different temperatures
under visible light irradiation is shown in Fig. 8. From Fig. 8, we can
see that the photocatalytic hydrogen production activity assumes
the law of increasing first and decreasing last with calcination
temperature rising, and the highest photocatalytic hydrogen pro-
duction activity is achieved when the calcination temperature is
400 ◦C. The reason is that the amorphous TiO2 with low activ-
ity is not completely transformed into the anatase TiO2 with high
activity when the calcination temperature is below 400 ◦C, which
is confirmed by the TG–DTA and XRD analysis. In addition, a cer-
tain amount of organic substance carbon remained in the sample
leads to the reduction of the photocatalytic activity. When the cal-
cination temperature is above 400 ◦C, nitrogen element is oxidized
into nitrogen oxides gas which releases into air with the rising
of calcination temperature, and thus the nitrogen content doped
in the sample decreases, weakening the effect caused by visible
light irradiation. On the other hand, as the calcination temperature
rises, the size of particles becomes larger, and the specific surface
area becomes smaller, which are unfavorable for the generation
Fig. 7. Fluorescence spectra of the samples: (a) SrTiO3, (b) N–SrTiO3, (c) TiO2, (d)
N–SrTiO3/TiO2 and (e) Pt/N–SrTiO3/TiO2.

J.-H. Yan et al. / Journal of Alloys and Compounds 472 (2009) 429–433
433
Pt particles covered on the surface of catalyst become more, and
owing to the aggregation of Pt particles, larger size Pt clusters form,
which can scatter the visible light, leading to the failure to trigger
the composite catalyst covered by Pt effectively under visible light
irradiation and the decreasing of electrons and holes, as a result,
the photocatalytic activity decreases [20,21].
4. Conclusions
No hydrogen was produced for single SrTiO3 and TiO2 under
visible light irradiation. SrTiO3 was doped nitrogen by solid phase
method and further combined with TiO2 by sol–gel method. The
photocatalytic hydrogen production activity was greatly improved
under visible light irradiation. When the temperature of calcination
is 400 ◦C, the optimized photocatalytic activity of hydrogen produc-
tion under visible light irradiation can be achieved. Pt depositing
greatly improved the photocatalytic activity, and the average hydro-
gen production rate is up to 5.1 mmol g cat−1 h−1 with 2 wt.% loaded
Pt.
Fig. 9. Photocatalytic hydrogen production activity of different contents of Pt
deposited N–SrTiO3/TiO2.
Acknowledgement
of photo-generated carriers, adsorption of sacrificial agent on the
surface of catalyst and absorption of visible light, therefore, the
photocatalytic activity decreases.
This research was financially supported by National Nature
Science Foundation of China (No. 20876039), and the Scientific
Research Foundation of Hunan Provincial Education Deparment,
China (No. 08A026).
Fig. 9 illustrates the photocatalytic hydrogen production activ-
ity of nitrogen-doped SrTiO3/TiO2 (400 ◦C) samples deposited with
different contents of Pt under visible light irradiation. As shown in
Fig. 9, the photocatalytic hydrogen production activity assumes the
law of increasing first and decreasing last with loaded Pt increas-
ing, and the highest photocatalytic hydrogen production activity is
achieved when the amount of loaded Pt is 2 wt.%. The reason for
this is as follows: when the composite semiconductor is loaded by
a certain amount of Pt, and if the content of Pt is controlled within a
suitable range, these metal Pt crystal nucleus and composite semi-
conductor systems can be regarded as an ultramicroclosed-circuit
photoelectrochemical cell, and the electrons and holes generated
by light irradiation are, respectively localized at metal Pt and com-
posite semiconductor. The enrichment of electrons on metal Pt, and
generation of holes on the surface of composite semiconductors
leads to the separation of the electrons and holes and the declining
probability of electron–hole pair recombination, consequently the
photocatalytic activity of photocatalyst improves. When the con-
tent of loaded Pt is too low, the size of Pt particles covered on the
surface of catalyst is small and Pt is uniformly dispersed, however,
the small amount of loaded Pt may cause the failure of effective
electron–hole pair separation and thus lead to the low photocat-
alytic activity; when the content of loaded Pt is too high, loaded
References
[1] A. Fujishima, K. Honda, Nature 238 (1972) 37.
[2] A. Linsebigler, G. Lu, J.T. Yates, Chem. Rev. 95 (1995) 735.
[3] G.B. Wang, Y.J. Wang, B.J. Song, Y.Q. Xu, Chin. J. Inorg. Chem. 19 (2003) 988.
[4] M. Avudaithai, T.R.N. Kutty, Mater. Res. Bull. 22 (1987) 641.
[5] H.T. Zhang, X.B. Wu, Y.M. Wang, J. Phys. Chem. Solids 68 (2007) 280.
[6] T. Ohno, T. Tsubota, Y. Nakamura, K. Sayama, Appl. Catal. A: Gen. 288 (2005) 74.
[7] J.S. Wang, S. Yin, M. Komatsu, Q.W. Zhang, F. Saito, T. Sato, J. Photochem. Pho-
tobiol. A: Chem. 165 (2004) 149.
[8] T. Lshii, H. Kato, A. Kudo, J. Photochem. Photobiol. A: Chem. 163 (2004) 181.
[9] M. Miyauchi, M. Takashio, H. Tobimatsu, Langmuir 20 (2004) 232.
[10] Y.R. Zhu, Y.G. Tang, J.H. Yan, Q. Liu. J. Inorg. Mater. 23 (2008) (in press).
[11] S.H. Zhong, J.H. Wang, Petrochem. Technol. 22 (1993) 307.
[12] J.G. Yu, X.J. Zhao, J. Inorg. Mater. 15 (2000) 347.
[13] I. Justicia, P. Ordejon, G. Canto, J.L. Mozos, J. Frasedas, G.A. Battiston, R. Gerbasi,
A. Figueras, Adv. Mater. 19 (2002) 1399.
[14] A. Henglein, J. Phys. Chem. 97 (1993) 5457.
[15] X.Z. Li, F.B. Li, C.L. Yang, J. Photochem. Photobiol. A: Chem. 141 (2001) 209.
[16] L. Foress, M. Schubnnell, J. Appl. Phys. B 56 (1993) 363.
[17] H. Irie, Y. Watanabe, K. Hashimoto, J. Phys. Chem. B 107 (2003) 548.
[18] M. Ueda, S. Otsuka-Yao-Matsuo, Sci. Technol. Adv. Mater. 5 (2004) 187.
[19] T. Omata, S. Otsuka-Yao-Matsuo, J. Photochem. Photobiol. A: Chem. 156 (2003)
243.
[20] Q.H. Zhang, L. Gao, Acta Chim. Sin. 63 (2005) 65.
[21] D.S. Geng, G.X. Lv, Y.S. Bi, Y.P. Bi, Acta Chim. Sin. 63 (2005) 658.