Thin bismuth oxide films prepared through the sol-gel method as photocatalyst

Article abstract of DOI:10.1016/j.molcata.2006.08.016

In this paper, thin bismuth oxide films were prepared by the sol-gel method. The films were annealed at different temperatures, and then applied to degrade a kind of typical textile industry pollutant (Rhodamine B), respectively, in order to study the influence of bismuth oxide crystal phases on their photocatalytic properties. X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) techniques and a surface profiler were applied to characterize the thin films annealed at different temperatures. The results show that different annealed temperatures cause the transformation between monoclinic phase and tetragonal phase of bismuth oxides and bismuth oxide films annealed at 550 °C contain a higher proportion of the tetragonal phase of bismuth oxides than those annealed at other temperatures, which leads to higher electronic binding energy and photocatalytic properties for these oxides in the films.

Full text of DOI:10.1016/j.molcata.2006.08.016

Journal of Molecular Catalysis A: Chemical 261 (2007) 167–171
Thin bismuth oxide films prepared through the
sol–gel method as photocatalyst
∗
Wu Xiaohong , Qin Wei, He Weidong
Harbin Institute of Technology, Box 408, Harbin 150001, PR China
Received 17 June 2006; received in revised form 31 July 2006; accepted 1 August 2006
Available online 7 September 2006
Abstract
In this paper, thin bismuth oxide films were prepared by the sol–gel method. The films were annealed at different temperatures, and then
applied to degrade a kind of typical textile industry pollutant (Rhodamine B), respectively, in order to study the influence of bismuth oxide crystal
phases on their photocatalytic properties. X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM)
techniques and a surface profiler were applied to characterize the thin films annealed at different temperatures. The results show that different
annealed temperatures cause the transformation between monoclinic phase and tetragonal phase of bismuth oxides and bismuth oxide films annealed
◦
at 550 C contain a higher proportion of the tetragonal phase of bismuth oxides than those annealed at other temperatures, which leads to higher
electronic binding energy and photocatalytic properties for these oxides in the films.
©
2006 Elsevier B.V. All rights reserved.
Keywords: Sol–gel method; Bismuth oxide; Photocatalytic properties; Reaction constant
1
. Introduction
matic version of the spray pyrolysis process [9] or by CVD
method under atmospheric pressure [10]. But the above meth-
ods are expensive and inconvenient. Proper sol–gel methods
are proved to be good ways to get oxide films, for example,
TiO2 films can be obtained through sol–gel methods [11–14].
Taking into account that there are not so many reports about
preparing the bismuth oxide films produced through sol–gel
methods and the fact that the semiconductor of bismuth oxides
with an appropriate energy gap can be applied to deal with pol-
lutant, it is the goal of the present paper to produce bismuth
oxide films through the sol–gel method and study the photocat-
alytic properties of the bismuth oxides films annealed at different
temperatures.
Bismuth oxide Bi2O3 has a lot of merits, such as a wide
energy gap change (from 2 to 3.96 eV) [1], high refractive
index, dielectric permittivity, besides marked photoconductiv-
ity and photoluminescence [2]. It is known that bismuth oxides
have polymorphism: five modifications, which are ␣, ␤, ␥, ␦
and ␻-Bi2O3. Among them, the low-temperature ␣ and the
high-temperature ␦ phases are stable and the others are high-
temperature metastable [2]. Owing to their peculiar character-
istics, bismuth oxides have been studied to be used various
areas, such as sensor optical coatings [2], sensor technology
[
3], fine structure glasses [4] and electroreduction [5]. Recently,
bismuth oxide particles have been produced in many kinds of
methods, forexample, co-precipitation, atomization [6]andlow-
temperature electrodeposition [7]. However, the use of bismuth
oxide particles is often limited considering the difficulty of recy-
cling them. Bismuth oxide films which can be recycled easily
were produced through some methods in the past, for example,
the rf sputtering or vacuum evaporation method [8], the pneu-
2. Experimental
2.1. Film preparation
Thin bismuth oxide films were prepared with the following
method: 5 g Bi(NO ) ·5H O was dissolved in a nitric solu-
3
3
2
tion (1:8 HNO3:H2O). PEG 200 (HO(CH2CH2O)200H) and
citric acid was added in the above solution, while OP (t-Oct-
∗
6
4 2 2 x
C H (OCH CH ) OH, x = 9–10) was added as surfactant. The
Corresponding author. Tel.: +86 451 86402522; fax: +86 451 86402522.
E-mail address: wxhqw@263.net (W. Xiaohong).
above solution was beaten up for 3 h, and then sol solution was
1
381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.08.016

1
68
W. Xiaohong et al. / Journal of Molecular Catalysis A: Chemical 261 (2007) 167–171
produced. After that, pieces of simple glass were dipped into
the sol solution where they were kept for 3 min, and then with-
−
1
drawn at a rate of 5 cm min . These pieces of glass were dried
◦
up at 100 C in a baking oven. Then the sol solution was heated
to 60 C, where it was kept for 1 h to form a kind of yellow
◦
gel. Afterwards, the gel was subjected to differential thermal
analysis and thermogravimetric analysis (DTA/TGA), using a
comprehensive thermal analyzer (ZRY-2P from Shanghai Tian-
ping) in order to establish the thermal treatment schedule for the
◦
films. Twoexothermalpeaksat275and390 Cwerenoticeddur-
ing this analysis. To match the DTA/TGA results, all the films
◦
−1
were preliminarily treated with a heating rate of 1 C min to
◦
3
90 C, during the process of which two 30 min plateaus at 275
and 390 C were maintained, respectively. At the same time, to
get the films annealed at different temperatures, the films con-
◦
◦
tinued to be heated to 450, 500, 550 and 600 C, respectively, at
the same heating rate, where a 1 h plateau was maintained. The
films treated following the above different routes were labeled
as different samples.
2
.2. Film characterization
XRD with a Cu K␣ source (D/max-r B from Ricoh) was
applied to study the crystalline structure of the films with an
accelerating voltage and an applied current of 40 kV and 30 mA,
respectively. Also, XPS (Escalab-220IXL from VG Scientific
Ltd.) was used to study the surface composition of the prepared
samples. Besides the morphology of the produced films was
examined by SEM (S-570 from Hitachi), and a profiler (Dektek
6
M from VECCO) was used to measure the thickness of the
Fig. 1. (a) XRD spectra of the produced films annealed at different temperatures.
b) Proportion of intensity of T phase to the sum of those of T phase and M3
phase for bismuth oxide films annealed at different temperatures.
(
films.
2
.3. Photo-degradation experiments
3. Results
A cylindrical quartz cell with a size of 25 mm in diameter
and 50 mm in height, a 20 W quartz UV lamp (from Tianjin
Lvhuan Unique Lamp Company) with a maximum UV irradi-
ation peak of 365 nm were used to compose the bench-scale
photoreactor system. The schematic diagram of the photore-
actor had been shown before [15]. The photocatalytic prop-
erties of the films produced through the sol–gel method were
examined by measuring the removal of Rhodamine B dye
solution (10 mg L ). To evaluate the experiments of degrad-
ing Rhodamine B using the bismuth oxide films, a blank test
was done; the pollutant solution was photolyzed using a piece
of simple glass to examine the photostability of Rhodamine
3.1. Analysis of XRD results
Fig. 1a displays XRD spectra of the films. It can be seen from
the spectra that the bismuth oxide films annealed at different
temperatures are mainly composed of monoclinic and tetragonal
bismuth oxides, which are labeled as M and T, respectively. To
match the XRD spectra, a curve with the proportion of intensity
of T phase to the sum of those of T phase and M3 phase as verti-
cal ordinate and annealed temperature as abscissa, is established,
as shown in Fig. 1b. It can be seen from Fig. 1b that the pro-
−
1
◦
portion increases with the annealed temperature before 550 C,
◦
B under the irradiation mentioned above. Square samples of
and reaches the biggest at 550 C, which is 89.5%, 20.8% higher
2
◦
2
.25 cm were immersed into 10 mL aqueous Rhodamine B
than that at 450 C. However, the proportion decreases when the
◦
solution. The solution was stirred continuously and supplied
with air during the reaction process. The UV light irradiated
perpendicularly to the surface of the samples through the side-
wall of the cylindrical quartz cell and the distance from the
UV source to the films was 2 cm. The UV light irradiated
for 2 h, in the process of which the change of Rhodamine
B concentration with the irradiation time was measured by a
UV spectrophotometer (722N from Shanghai Fine Instrument
Company).
annealed temperature above 550 C continues to increase.
3.2. Results of XPS measurements
Fig. 2A displays the spectra of XPS, from which it can be
seen that the prepared bismuth oxide films mainly consist of
Bi, O and other elements such as C, Si, Na from the simple
glass substrates. The Bi4f7/2 XP spectra of bismuth oxide films
annealed at different temperatures are displayed in Fig. 2B, the

W. Xiaohong et al. / Journal of Molecular Catalysis A: Chemical 261 (2007) 167–171
169
◦
Fig. 3. SEM image of the produced film annealed at 550 C.
the Bi2O3 films under the UV-illumination, while the removals
of Rhodamine B using different bismuth oxide films are big-
ger than that using simple glass. It has been found that the
removal of Rhodamine B increases with the annealed tem-
◦
perature before 550 C and the removal for 2 h reaches 92%
◦
when the temperature reaches 550 C. However, as the temper-
Fig. 2. (A) XPS survey spectra for the surface of bismuth oxide films annealed
◦
◦
◦
◦
at different temperatures: (a) 450 C, (b) 500 C, (c) 550 C and (d) 600 C. (B)
Bi4f XP spectra of bismuth oxide films annealed at different temperatures (from
◦
top to bottom): 450, 500 and 550 C.
binding energies are 158.77, 159.01, 159.07 and 159.01 eV for
◦
4
50, 500, 550 and 600 C, respectively. It can be observed from
Fig. 2B that binding energy shifts to a higher value for Bi4f7/2
◦
as the annealed temperature increases from 450 to 550 C.
3
.3. Analysis of SEM results
The SEM microphotograph of the film annealed at 550 C
◦
magnified by 30k times is displayed in Fig. 3. It can be noticed
from the SEM microphotograph that the film is uniform and the
particles in the film are almost of the same size and shape.
3
.4. Results of the surface profiler measurement
The thickness of the films is 120 ± 10 nm for the four kinds of
samples according to the measured results of the surface profiler
measurement.
3
.5. Results of photo-degradation of Rhodamine B
The removals of Rhodamine B using and not using the pro-
duced films are shown in Fig. 4a. It can be noticed that the
removal speed of Rhodamine B is very low in the absence of
Fig. 4. (a) Removals of Rhodamine B with bismuth oxide films annealed at
different temperatures. (b) Kinetic linear simulation curve of Rhodamine B pho-
tocatalytic degradation with bismuth oxide films annealed at 450 C.
◦

1
70
W. Xiaohong et al. / Journal of Molecular Catalysis A: Chemical 261 (2007) 167–171
◦
ature above 550 C continues to increase, the removal decreases
instead.
The photogenerated hydroxyl radicals have high redox activ-
ities, which are able to decompose the pollutant. Apparently it
is more difficult for the bismuth oxides of a phase with a higher
energy gap than those with a lower band gap to be excited by the
irradiation of light with a long wavelength to produce conduc-
tion band electrons and valence band holes, which can incite the
4
. Discussion
4
.1. The reaction kinetics of photo-degradation of
Rhodamine B with the films
•
birth of high reactive radicals, such as OH, so as to facilitate
the degradation of Rhodamine B. Nonetheless, as irradiated by
the same light with energy big enough to activate two phases
of bismuth oxides, the bismuth oxides of a phase with a bigger
value of energy gap are superior to those with a smaller value,
because of the higher redox activities.
It has been well established [16,17] that photocatalysis exper-
iments follow the Langmuir–Hinshelwood model, where the
reaction rate R is proportional to the surface coverage θ (Eq.
(
1)):
dC
R = − dt = krθ =
krKC
(1)
4.3. The relationship of temperature, structure and
properties
1 + KC
where kr is the reaction rate constant, K the adsorption coefficient
of the reactant at the surface of the film and C its concentra-
tion. When C is very small, the product KC is negligible with
respect to unity so that Eq. (1) describes first-order kinetics. The
integration of Eq. (1) with the limit condition that at the start
of irradiation, t = 0, the concentration is the initial one, C = C0,
gives:
High electronic binding energy indicates a high band gap for
bismuth oxides which leads to high redox activities of photogen-
eratedholesandelectronsoftheseoxides. Theelectronicbinding
◦
energy of the bismuth oxides in the films annealed at 550 C is
the biggest of all for the four kinds of the films as showed in
Fig. 2, so these oxides are supposed to be the most photocatalyt-
icly active. Furthermore it can be observed from Fig. 1 that the
C
−
ln
= kappt = krKt
(2)
T phase is more dominant for the bismuth oxide films annealed
C0
◦
5
50 C than that for those annealed at other temperatures, while
where kapp is the apparent first-order reaction constant. Fur-
thermore, a kinetic linear simulation curve of Rhodamine B
photocatalytic degradation using bismuth oxide flims annealed
it is adverse for the M phase. So it can be induced that the biggest
electronic binding energy for these oxides is due to the increase
of bismuth oxides of T phase. Therefore, the bismuth oxides of
the T phase have higher photocatalytic activities than those of
M phase.
◦
at 450 C is shown in Fig. 4b. It is clear that the curve with time
(t) as abscissa and ln (C0/C) as vertical ordinate, is close to a
linear curve. It can be concluded that the photocatalytic degra-
dation of Rhodamine B using bismuth oxide films as catalyst fits
well with the first-order exponential decay curve, which follows
the first-order reaction kinetics. So a linear equation can be set
up, as Eq. (2) shows. In Eq. (2), kapp is the first-order reaction
constant, which signifies the efficiency of photo-degradation of
Rhodamine B. The values of kapp for the bismuth oxide films
The above analysis can be supported well by the results of
photo-degradation of Rhodamine B. It can be noticed in Fig. 4
that the removal for 2 h of Rhodamine B using the bismuth oxide
◦
films annealed at 550 C is as high as 92% and the value of kapp
−1
is as big as 0.022 min while the removal is only 81% and
−1
◦
the kapp value is only 0.015 min for those annealed at 450 C
which contain the smallest proportion of the T phase of bismuth
oxides and the biggest proportion of the M phase of bismuth
oxides.
◦
annealed at 500, 550 and 600 C are 0.015, 0.016, 0.022 and
−
1
0
.018 min , respectively. As a consequence, the bismuth oxide
◦
films annealed at 550 C have the best efficiency of all to degrade
Rhodamine B, while the films annealed at 500 C precede those
annealed 450 C to degrade Rhodamine B.
◦
5. Conclusion
◦
In conclusion, relatively uniform bismuth oxide films mainly
4
.2. The photocatalytic mechanism
composed of T phase and M phase can be prepared through
the sol–gel method annealed at different temperatures. Dif-
ferent temperatures result in the change of the proportion
of T phase intensity to M phase intensity, which leads to
the difference of the kinetics of photocatalytic degradation of
Rhodamine B using bismuth oxide films annealed different
It has been well known that semiconductors with a proper
energy gap can probably be used as catalysts, and bismuth oxides
are among them. Bismuth oxides can induce the forming of con-
duction band electrons and valence band holes when absorbing
radiation with energy equal to or bigger than their energy gap.
Then these electrons and holes have redox activities, and the
photogenerated holes can react with -OH around the catalyst to
◦
temperatures. It has been found that 550 C is prior to the
other annealed temperatures in the research to get high pro-
portion of the T phase bismuth oxides with high electronic
binding energy. The bismuth oxide films prepared by being
•
produce OH with high redox activities. The above description
can be illustrated by the following expressions [(1) and (2)]:
◦
annealed at 550 C show high photocatalytic properties, using
−1
+
−
which the first-order reaction constant is 0.022 min and the
removal of Rhodamine B photocatalyzed for 2 h using these
films is as big as 92% while the removal is 81% and the first-
Bi2O3 + hν ꢀ h + e
OH + h → OH
(1)
(2)
−
+
•

W. Xiaohong et al. / Journal of Molecular Catalysis A: Chemical 261 (2007) 167–171
171
−
1
[
[
8] K. Shimanoe, M. Suetsugu, N. Miura, N. Yamazoe, Solid State Ionics
113–115 (1998) 415–419.
9] O. Rico-Fuentes, E. Sa’nchez-Aguilera, C. Velasquez, R. Ortega-Alvarado,
J.C. Alonso, A. Ortiza, Thin Solid Films 478 (2005) 96–102.
order reaction constant is 0.015 min using those annealed at
◦
4
50 C.
References
[
10] T. Takeyamaa, N. Takahashib, T. Nakamurab, S. Itohc, Surf. Coat. Technol.
00 (2006) 4797–4801.
2
[
[
1] L. Leontie, M. Caraman, A. Visinoiu, G.I. Rusu, Thin Solid Films 473
2005) 230–235.
2] L. Leontie, M. Caraman, M. Alexe, C. Harnagea, Surf. Sci. 507–510 (2002)
80–485.
[11] J. Sabate, M.A. Anderson, H. Kikkawa, Q. Xu, S. CerveraMarch, C.G. Hill,
J. Catal. 134 (1992) 36.
[12] R.B. Revathi, M. Gr a¨ zel, J. Am. Chem. Soc. 79 (1996) 2185.
[13] N. Negishi, K. Takeuchi, Sol–gel. Sci. Technol. 22 (2001) 23.
[14] O. Regan, M. Gr a¨ zel, Nature 353 (1991) 737.
(
4
[
[
3] A.A. Tomchenko, Sens. Actuators B Chem. 68 (2000) 48–52.
4] Y. Shimizugawa, N. Sugimoto, K. Hirao, J. Non-Cryst. Solids 221 (1997)
[15] X. Wu, Z. Jang, H. Liu, S. Xin, X. Hu, Thin Solid Films 441 (2003)
130.
2
08–212.
[
[
5] M.A. Pe’rez, M. Lo’pez Teijelo, J. Electroanal. Chem. 583 (2005) 212–220.
6] C. Machado, S. Aidel, M. Elkhatib, H. Delalu, R. Metz, Solid State Ionics
[16] A.I. Kontos, I.M. Arabatzis, D.S. Tsoukleris, A.G. Kontos, M.C. Bernard,
D.E. Petrakis, P. Falaras, Catal. Today 101 (2005) 275–281.
[17] V.A. Sakkas, I.M. Arabatzis, I.K. Konstantinou, A.D. Dimou, T.A. Albanis,
P. Falaras, Appl. Catal. B: Environ. 49 (2004) 195–205.
1
49 (2002) 147–152.
[
7] E.W. Bohannan, C.C. Jaynes, M.G. Shumsky, J.K. Barton, J.A. Switzer,
Solid State Ionics 131 (2000) 97–107.