One-pot synthesis of Bi2O3/Bi2O4 p-n heterojunction for highly efficient photocatalytic removal of organic pollutants under visible light irradiation

Article abstract of DOI:10.1016/j.jpcs.2020.109376

Construction of p-n heterojunction is an effective strategy to obstacle the rapid recombination of photo-generated carriers for single photocatalyst. Herein, visible-light-active Bi₂O₃/Bi₂O₄ p-n junction was prepared by a one-pot hydrothermal method, using Bi(NO₃)₃·5H₂O, KCl, and NaBiO₃·2H₂O as precursors. The in-situ formed BiOCl reacts with NaBiO₃·2H₂O during the hydrothermal process, producing the target product of Bi₂O₃/Bi₂O₄ p-n heterojunction. As expected, the obtained Bi₂O₃/Bi₂O₄ heterojunction photocatalysts exhibited higher photocatalytic activity than Bi₂O₄ for the degradation of MO (methyl orange) and phenol under visible-light irradiation. The superior photocatalytic performance of Bi₂O₃/Bi₂O₄ hybrid is mainly attributed to the improved separation efficiency of e^--h⁺ pair induced by the internal electric field of p-n junction. The present work may provide a rational and facile strategy to construct a highly efficient Bi₂O₄-based p-n heterojunction photocatalyst toward pollutant removal.

Full text of DOI:10.1016/j.jpcs.2020.109376

Journal of Physics and Chemistry of Solids 140 (2020) 109376
Contents lists available at ScienceDirect
Journal of Physics and Chemistry of Solids
journal homepage: http://www.elsevier.com/locate/jpcs
One-pot synthesis of Bi2O3/Bi2O4 p-n heterojunction for highly efficient
photocatalytic removal of organic pollutants under visible light irradiation
,
,
,
,
b
,
,
,b
Chengyu Lia b, Yaya Maa b, Shizheng Zhenga b, Changyuan Hua *, Feng Qina b, Lin Weia
,
Cuiqing Zhanga b, Shuwang Duoa b, Quanhong Hua
,
,
,b
a Jiangxi Key Laboratory of Surface Engineering, Jiangxi Science and Technology Normal University, Nanchang, 330013, PR China
b School of Materials and Mechanical & Electrical Engineering, Jiangxi Science and Technology Normal University, Nanchang, 330013, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Bi2O4
Construction of p-n heterojunction is an effective strategy to obstacle the rapid recombination of photo-generated
carriers for single photocatalyst. Herein, visible-light-active Bi2O3/Bi2O4 p-n junction was prepared by a one-pot
hydrothermal method, using Bi(NO3)3⋅5H2O, KCl, and NaBiO3⋅2H2O as precursors. The in-situ formed BiOCl
reacts with NaBiO3⋅2H2O during the hydrothermal process, producing the target product of Bi2O3/Bi2O4 p-n
heterojunction. As expected, the obtained Bi2O3/Bi2O4 heterojunction photocatalysts exhibited higher photo-
catalytic activity than Bi2O4 for the degradation of MO (methyl orange) and phenol under visible-light irradi-
ation. The superior photocatalytic performance of Bi2O3/Bi2O4 hybrid is mainly attributed to the improved
separation efficiency of e--hþ pair induced by the internal electric field of p-n junction. The present work may
provide a rational and facile strategy to construct a highly efficient Bi2O4-based p-n heterojunction photocatalyst
toward pollutant removal.
Bi2O3
p-n junction
Photocatalysis
1. Introduction
restricts its photocatalytic performance [11–13]. Therefore, it is highly
desirable to enhance the separation efficiency of charge carriers, in
Photocatalysis has long been regarded as an effective method to
resolve environmental problems like air purification and wastewater
treatment [1–3]. As a type of potential visible-light-driven photo-
catalyst, bismuth-contained compounds have gained increasing atten-
tion in the application of photocatalysis [4]. Normally, bismuth in these
compounds exists in trivalent or pentavalent states [5–9]. Since both
trivalent and pentavalent bismuth-based semiconductor materials
possess favorable photocatalytic performance, it is questioned whether
there exists bismuth-based photocatalytic materials with Bi (III) and Bi
(V) mixed valence states simultaneously, which can promote the cata-
lytic activity. In fact, this material does exist. As a new member of
bismuth-based photocatalysts and an n-type semiconductor, monoclinic
dibismuth tetraoxide (Bi2O4) has received much attention, with mixed
valent states of Bi (III) and Bi (V) [7]. The narrow band gap of Bi2O4
(~2.0 eV) makes it absorb wavelengths up to 620 nm. Importantly,
Bi2O4 presents favorable photocatalytic activity for the degradation of
pollutants and the inactivation of bacteria under visible light [10]. Like
many other single component photocatalysts, however, the quick
recombination of photo-generated carriers of pure Bi2O4 seriously
order to fully excite the photocatalytic activity of Bi2O4 under a wide
range of visible light spectrum.
By virtue of the interfacial charge transfer effect, semiconductor
composites with heterostructured interface usually could convert solar
energy to chemical energy more effective compared with single-phase
semiconductor [14,15]. As a typical of heterojunction, p-n junction
can drive effectively the electrons moving to n-type semiconductor and
the holes transferring to p-type one, through the built-in electric field at
the interfaces of two semiconductors, leading to the promoted signifi-
cantly migration and separation efficiency of photoinduced carriers and
then increasing obviously photocatalytic behavior [16–18]. Conse-
quently, creating p-n junction is widely adopted to improve the activity
of photocatalyst. For instance, Yu et al. [19] used an ingenious method
to coat BiOI on the wall of a self-organized TiO2 nanotube array to
synthesize a p-n BiOI/TiO2 nanotube array. The prepared BiOI/TiO2
junction exhibited better photoelectrochemical activity than pure BiOI
and TiO2. Wang et al. [20] reported that Bi2O4/BiOCl p-n hetero-
structure displayed promoted photocatalytic performance for the
degradation of MO and tetracycline solution under visible-light
* Corresponding author. Jiangxi Key Laboratory of Surface Engineering, Jiangxi Science and Technology Normal University, Nanchang, 330013, PR China.
E-mail address: hcy6257@163.com (C. Hu).
https://doi.org/10.1016/j.jpcs.2020.109376
Received 17 October 2019; Received in revised form 23 December 2019; Accepted 25 January 2020
Available online 27 January 2020
0022-3697/© 2020 Elsevier Ltd. All rights reserved.

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
irradiation. Nevertheless, the relatively complex preparation method of
the above-mentioned work still faces many problems. Undoubtedly,
there remains much room for further exploration to enhance the pho-
tocatalytic activity of p-n junction for potential applications.
Considering the structure of Bi2O4, if the rod-like Bi2O4 could load on
the sheet-like material by a one-pot method, this simple preparation
process can not only save a lot of time but also contribute to form a tight
interface contact between two materials, which can facilitate the sepa-
ration of photogenerated electron-hole pairs. As another member of the
bismuth oxide family, bismuth (III) oxide (Bi2O3) is a p-type semi-
conductor (Eg ¼ 2.3–2.8eV) with four main crystallographic poly-
morphs: α, β, γ, and δ phase [21]. Furthermore, it was reported Bi2O3 has
the visible light photocatalytic activity [22–24]. In this study, Bi2O3 has
been selected for coupling with Bi2O4. Since Bi2O4 and Bi2O3 are n-type,
p-type semiconductor, respectively, p-n heterojunction will be formed
when Bi2O4 is combined with Bi2O3 [25–27]. As a result, the separation
of photoinduced charge carriers of Bi2O4 can be accelerated by the
built-in electric field, which is in favor of improving the photocatalytic
performance of Bi2O4.
Fig. 1. XRD patterns of Bi2O3, Bi2O4, and Bi2O3/Bi2O4 p-n heterojunctions.
Considering that BiOCl can be easily obtained by simply stirring the
aqueous solution of Bi(NO3)3⋅5H2O and KCl [6], and Bi2O4 can be
facially produced through hydrothermal reaction, using NaBiO3⋅2H2O
as the sole raw materials [7]. The original aim of this work was to
synthesize BiOCl/Bi2O4 heterojunction via the one-pot method, using Bi
(NO3)3⋅5H2O, KCl, and NaBiO3⋅2H2O as the precursor. In the course of
the experiment, however, we found that the BiOCl generated by in-situ
reaction of Bi(NO3)3⋅5H2O and KCl would react with NaBiO3⋅2H2O to
form Bi2O3/Bi2O4 heterojunction rather than BiOCl/Bi2O4 hetero-
structure during the hydrothermal process. Therefore, the present work
is changed to constructing Bi2O3/Bi2O4 p-n junction to improve the
visible-light photocatalytic performance of Bi2O4. Interestingly,
as-prepared Bi2O3/Bi2O4 heterojunction presents promoted visible light
photocatalytic performance than Bi2O4. To identify the Bi2O3/Bi2O4 p-n
heterojunction has been successfully synthesized, a series of control
experiments were designed.
and KCl to 1.8, 2.4, 3.6 mmol, respectively. For comparison, pure Bi2O4
was synthesized by the same process without addition of Bi(NO3)3⋅5H2O
and KCl. Bare Bi2O3 could also be obtained by the similar method, just
increasing the amount of Bi(NO3)3⋅5H2O and KCl to 4.8 mmol.
2.1.2. Control experiment
The detailed process of control experiment could be found in the
Supporting Information.
2.2. Characterization
The detailed information on various characterization techniques was
shown in the Supporting Information.
2.3. Electrodes and electrochemical measurements
This paper reports the preparation of Bi2O3/Bi2O4 p-n hetero-
junctions with different molar ratios by a one-pot hydrothermal method,
using Bi(NO3)3⋅5H2O, KCl, and NaBiO3⋅2H2O as the raw materials. The
morphology and physicochemical properties of the obtained hetero-
junction were systematically characterized. The catalytic properties of
Bi2O3/Bi2O4 heterojunctions were investigated by photocatalytic
degradation of MO and phenol under visible light. The results indicate
that the enhanced activity of Bi2O3/Bi2O4 p-n heterojunction is mainly
attributed to the improved separation efficiency of electron-hole pair. In
addition, a possible photocatalytic decomposition mechanism over
Bi2O3/Bi2O4 heterojunction is also proposed. This work provides a
convenient method for in-situ preparation of Bi2O3/Bi2O4 p-n hetero-
junctions with improved visible-light catalytic activity.
The photocurrent and EIS measurements were conducted by elec-
trochemical workstation (Chenhua Instruments, CHI660D) with a stan-
dard three-electrode system. Platinum electrode, calomel electrode, and
indium-tin-oxide electrode coated with photocatalyst were applied as
the counter electrode, the reference electrode, and the working elec-
trode, respectively. 0.5 M Na2SO4 served as the electrolyte.
2.4. Photocatalytic activity measurements
The photocatalytic activity of Bi2O3/Bi2O4 samples was studied by
the degradation of MO and phenol. A 500 W xenon lamp with a 420 nm
cutoff filter was used as a visible-light source. In typical experiments, 24
mg photocatalyst was added to 58 mL MO solution (25 mg/L) or phenol
solution (15 mg/L). Before irradiation, the suspension was magnetically
stirred in the dark for 60 min to establish adsorption/desorption equi-
librium between the pollutant and the catalyst surface under normal
atmospheric conditions. At given irradiation time intervals, 3–4 mL
suspension was sampled and centrifuged. The concentration changes of
MO and phenol were analyzed by the Uv–vis spectrophotometer at 464
nm and 271 nm, respectively.
2. Experimental
2.1. Sample preparation
2.1.1. Bi2O3/Bi2O4 p-n heterojunction
Bi2O3/Bi2O4 p-n heterojunctions were synthesized by a hydrother-
mal method. First, 3 mmol Bi(NO3)3⋅5H2O and 3 mmol KCl were added
to 60 mL deionized water, and the mixture was stirred at room tem-
perature for 30 min. Then 1 mol/L of NaOH was added to the mixture,
and the pH was adjusted to 7.0. After stirring for another 30 min at room
temperature, 6 mmol NaBiO3⋅2H2O was poured into the above solution
under vigorous stirring. The autoclave was heated at 160 �C for 12 h at
autoclaved pressure and then cooled to room temperature. The collected
precipitates were washed thoroughly with ethanol and deionized water
and dried in air at 60 �C. The obtained sample was denoted as Bi2O3/
Bi2O4-3. Bi2O3/Bi2O4-1, Bi2O3/Bi2O4-2, Bi2O3/Bi2O4-4 could be pre-
pared by the same method, just changing the amount of Bi(NO3)3⋅5H2O
2.5. Active species trapping experiments
Active species trapping experiments were conducted by adding
various scavengers to MO solution (25 mg/L, 58 mL) before the pho-
tocatalytic tests, including 2.0 mM isopropanol (IPA, a quencher of
�OH), 2.0 mM Na2C2O4 (a quencher of hþ) and 0.2 mM p-benzoquinone
(BQ, a quencher of �OÀ2 ). The other procedure of experiment is the same
as photocatalytic activity measurement.
2

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
Scheme 1. Schematic diagram of the preparation process of Bi2O3/Bi2O4 p-n heterojunctions.
3. Results and discussion
Teflon-lined stainless steel autoclave and then heated at 160 �C for 12 h
under autogenous pressure. As shown in Fig. S3, all the diffraction peaks
of obtained composites could be indexed to tetragonal β-Bi2O3 (JCPDS
card No.27–0050) and monoclinic Bi2O4 (JCPDS card No.50–0864) [7,
24], accompanied with the complete disappearance of characteristic
peaks of BiOCl and NaBiO3⋅2H2O. Interestingly, when the amount of
BiOCl was up to 0.5g, relatively pure tetragonal β-Bi2O3 could be pro-
duced. It seems that BiOCl may react with NaBiO3�2H2O during the
hydrothermal process in our case. As reported by the previous work,
NaBiO3�2H2O could transform into Bi2O4 during the hydrothermal re-
action at 160 �C [7,15]. Bearing this in mind, one can raise another
question. Namely, BiOCl reacts directly with the raw materials of
NaBiO3�2H2O or with the in-situ formed Bi2O4 during the hydrothermal
reaction. To resolve this issue, at last, the forth control experiment was
implemented. The suspension of BiOCl and Bi2O4 was loaded into a
Teflon-lined stainless steel autoclave and maintained at 160 �C for 12 h.
As can be seen from Fig. S4, all collected composites are just composed
of BiOCl and Bi2O4, implying that BiOCl does not react with Bi2O4
during the hydrothermal process. Based on the above results, it can be
rationally concluded that in-situ formed BiOCl could react with
NaBiO3�2H2O to produce Bi2O3/Bi2O4 junction during hydrothermal
reaction process (160 �C, 12 h) if suitable quantities of Bi(NO3)3⋅5H2O,
KCl and NaBiO3�2H2O were used as raw materials. At present, however,
the detailed reaction mechanism between BiOCl and NaBiO3�2H2O is
not clear for us during the hydrothermal reaction.
3.1. Crystalline characterizations
Fig. 1 shows the XRD patterns of Bi2O3, Bi2O4, and the Bi2O3/Bi2O4
p-n heterojunctions. The observed diffraction peaks of pure Bi2O4 are in
good agreement with the monoclinic phase of Bi2O4 (JCPDS No.
50–0864) [7]. For the bare Bi2O3 sample, most of the diffraction peaks
can be attributed to the tetragonal phase of Bi2O3 (JCPDS card
No.27–0050), and only a few Bi2O4 phase structure is found in its XRD
patterns [24]. Therefore, relatively pure Bi2O3 can be obtained by a
hydrothermal method when the ratio of Bi(NO3)3⋅5H2O, KCl, and
NaBiO3⋅2H2O is optimized. As for the Bi2O3/Bi2O4 heterojunctions, the
diffraction peaks of both Bi2O3 and Bi2O4 are found, and no other im-
purity phase is observed, demonstrating the successful formation of
Bi2O3/Bi2O4 heterojunction. Besides, the diffraction peaks of BiOCl
cannot be detected in all samples, illustrating there is no BiOCl/Bi2O4
junction in our sample.
The preparation process of Bi2O3/Bi2O4 p-n junction was shown in
Scheme 1. To explore how the Bi2O3/Bi2O4 junction is formed during the
one-pot reaction process, using Bi(NO3)3⋅5H2O, KCl, and NaBiO3⋅2H2O
as raw materials, a series of control experiments are designed. Firstly, as
shown in Fig. S1, all the diffraction peaks of the sample agree well with
the JCPDS card No.06–0249 of BiOCl [6]. Obviously, BiOCl can be easily
synthesized just by stirring the mixture aqueous solution of Bi
(NO3)3⋅5H2O and KCl for 60 min under neutral conditions at room
temperature. Secondly, Bi(NO3)3⋅5H2O and KCl were added to deionized
water, followed by tuning the pH value to 7, and then NaBiO3⋅2H2O was
poured into the above mixture solution. The obtained suspension was
kept vigorously stirring for 12 h at room temperature. The as-prepared
sample is identified as the composite of BiOCl and NaBiO3⋅2H2O since
only these phase structures of BiOCl and NaBiO3⋅2H2O are found in the
XRD patterns (Fig. S2) [6,7]. The result means in-situ formed BiOCl
would not react with NaBiO3�2H2O at room temperature. Is this the case
during the hydrothermal process? To answer this question, thirdly,
NaBiO3⋅2H2O and pre-prepared BiOCl were directly mixed in
3.2. Morphology characterization
Fig. 2a–c shows the typical SEM images of resulting Bi2O4, Bi2O3, and
Bi2O3/Bi2O4-3 p-n heterojunction. As-prepared Bi2O4 (Fig. 2a) exhibits
irregular submicrorods structure with a length of about 2–4 μm. For pure
Bi2O3 (Fig. 2b), a sheet-shaped structure with a diameter between 0.5
and 3 μm can be observed. As far as Bi2O3/Bi2O4-3 composite (Fig. 2c),
both nanoparticles (NPs) and rod-like shaped Bi2O4 are attached firmly
on the surface of Bi2O3, forming a heterostructured interface between n-
type Bi2O4 and p-type Bi2O3, which implies that the Bi2O3/Bi2O4 p-n
3

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
Fig. 2. SEM image of Bi2O4 (a), Bi2O3 (b), and Bi2O3/Bi2O4-3 composite (c); TEM image of Bi2O3/Bi2O4-3 composite (d, e).
heterojunction is successfully constructed via the one-pot hydrothermal
process. The formation of Bi2O3/Bi2O4 p-n heterojunction can be further
proved by the result of TEM characterization. As shown in Fig. 2d–e, it
can be clearly seen that many granular and short-rod-shaped Bi2O4 are
dispersed on the Bi2O3 surface to produce Bi2O3/Bi2O4 heterojunction,
consistent with the result of SEM characterization. It is expected that this
structure with intimate interface contact between Bi2O3 and Bi2O4 could
accelerate the interfacial charge carrier migration and separation and
thus improve the photocatalytic activity [28].
transition of the semiconductor (for a direct transition, n ¼ 1; for an
indirect transition, n ¼ 4). Here, n is equal to 4 for Bi2O4 and 1 for Bi2O3
[7,30]. The energy of the band gap is calculated by extrapolating the
straight line to the abscissa axis. As shown in Fig. 3b, the band gap of
Bi2O4 and Bi2O3 is estimated to be 2.0 and 2.7 eV, respectively. These
results are in good agreement with previous reports [7,31–33]. The
potential of valence band (VB) and conduction band (CB) of photo-
catalyst plays an important role in photocatalysis reaction. The VB and
CB of semiconductor are calculated from the following empirical equa-
tion: EVB ¼ X À Ee þ 0.5Eg, ECB ¼ EVB À Eg, where X is the geometric
mean of the absolute electronegativity of atoms in semiconductor
(~5.13 eV for Bi2O4 and 6.12 eV for Bi2O3) [7,34]; Ee is the energy of
free electrons on the hydrogen scale (about 4.5 eV); EVB and ECB are VB
and CB edge potential of semiconductor, respectively; and Eg is semi-
conductor energy band gap determined by the UV–vis DRS results
[35–37]. Therefore, the VB values of Bi2O3 and Bi2O4 are 2.97 eV and
1.63 eV, respectively. Accordingly, their CB values are 0.27 eV and
À 0.37 eV, respectively.
3.3. DRS characterization
Fig. 3a shows the UV–visible diffuse reflectance spectra of the syn-
thesized Bi2O4, Bi2O3, and Bi2O3/Bi2O4 composites. For pure Bi2O4, the
absorption threshold wavelength is about 620 nm, and strong absorp-
tion is observed in the visible light region. In contrast, the absorption
cutoff wavelength of Bi2O3 is about 455 nm, indicating that it can only
absorb a part of visible light. However, the Bi2O3/Bi2O4 composite ex-
hibits significant red-shift absorption as compared to Bi2O4, and the
absorption edge is up to 697 nm, indicating that the Bi2O3/Bi2O4 sample
could absorb a wider range of visible light. The similar phenomenon was
reported in previous work of the CuI–BiOI/Cu film [3]. For semi-
conductors, an optical band gap can be calculated from the absorption
3.4. XPS analysis
To determine the surface elemental compositions and their chemical
states of the Bi2O4 and Bi2O3/Bi2O4-3 heterojunction, X-ray photoelec-
tron spectroscopy (XPS) measurement was performed, and the results
are shown in Fig. 4. As can be seen, in the full spectrum of Bi2O4 and
Bi2O3/Bi2O4-3 (Fig. 4a), the presence of bismuth (Bi4p, Bi4d, Bi4f,
spectrum using the equation
α
hν
¼ A(h
ν
-Eg)n/2, where
α, h, ν, A, and Eg
are absorption coefficients, Planck constant, light frequency, constant,
and band gap [29]. The n-value is determined by the type of optical
4

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
Fig. 3. UV–vis diffuse reflectance spectra (DRS) of Bi2O4, Bi2O3, and Bi2O3/
Bi2O4 composites (a); the plot of (
α
h
ν)
n/2 vs. h
ν of the samples (n ¼ 4 for Bi2O4
and n ¼ 1 for Bi2O3) (b).
Bi5d), oxygen (O1s), and carbon (C1s) is obvious. The C1s signal is
derived from a carbon-based contaminant in the atmosphere or vacuum
system [15]. Nevertheless, the Cl2s could not be detected in the full
spectrum of Bi2O3/Bi2O4-3 heterojunction, further illustrating BiOCl is
not exist in the Bi2O3/Bi2O4 p-n heterojunction. As shown in Fig. 4b, the
high-resolution spectra shows that the Bi 4f 7/2 (or Bi 4f 5/2) peak in
Bi2O4 could be deconvoluted well into two bimodal peaks at binding
energies of 158.8 eV and 159.6 eV (or 164.2 eV and 164.9 eV), which is
consistent with the reported references on Bi2O4 [38]. Compared with
pure Bi2O4, the Bi4f 7/2 (or 4f 5/2) peak of the Bi2O3/Bi2O4-3 hetero-
junction shifts to lower binding energy, which is the result of strong
interfacial interaction between Bi2O4 and Bi2O3 [20]. It is worth noting
that the Bi3þ/Bi5þ atomic ratio of Bi2O4 is about 0.97, which is very
close to the stoichiometric formula of Bi2O4 (Bi3þBi5þO4). While the
stoichiometric ratio of Bi2O3/Bi2O4-3 is as high as 1.12, which means
that the content of Bi3þ is higher. The increase in the Bi3þ species con-
tent in the Bi2O3/Bi2O4-3 sample should be derived from Bi2O3, which
further confirms the formation of the Bi2O3/Bi2O4 p-n heterojunction.
The asymmetric O 1s peak of Bi2O4 (Fig. 4c) can be deconvolved into
three peaks at 529.9 eV, 531.2 eV, and 532.8 eV. The peak at 529.9 eV
corresponds to the lattice oxygen in the Bi–O bond. The higher binding
energies of 531.2 eV and 532.8 eV are attributed to chemisorption and
physical adsorption of oxygen on the surface of the photocatalyst,
respectively [39–41]. The O 1s peak of the Bi2O3/Bi2O4-3 hetero-
junction can also be assigned to three peaks, namely lattice oxygen
(529.3 eV), chemisorbed oxygen (531.1 eV) and physically adsorbed
oxygen (532.9 eV). The peak of lattice oxygen is obviously shifted to low
Fig. 4. Full survey XPS spectra of samples (a), high-resolution XPS Bi4f (b), and
O1s spectra (c) of samples.
binding energy, which may be also related to the strong interaction
between Bi2O3 and Bi2O4. XPS results were consistent with XRD, SEM,
and TEM results.
3.5. Nitrogen adsorption analysis
The nitrogen adsorption-desorption isotherms and specific surface
5

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
3.6. Photocatalytic performance
The photocatalytic degradation activity of as-prepared samples was
investigated using MO and phenol as model pollutants under visible
light irradiation. To achieve an adsorption-desorption equilibrium be-
tween the contaminant and the photocatalyst, the reaction mixture was
stirred in the dark for 60 min in advance. Fig. 6a shows the adsorption
and photocatalytic degradation of MO over the obtained photocatalysts.
It is clear that Bi2O3/Bi2O4 heterostructures show a slightly higher
adsorption capability for MO molecule compared with pure Bi2O4. After
60 min of visible light irradiation, Bi2O3 exhibits relatively poor pho-
tocatalytic activity and the removal rate of MO is only 26.87%. In
contrast, Bi2O4 has preferable photocatalytic activity, which can remove
about 60.1% of MO. When the Bi2O3/Bi2O4 p-n heterojunction was
constructed, all of heterojunctions exhibit higher photocatalytic activity
under the same conditions. As the content of Bi2O3 increases, the pho-
tocatalytic activity of the p-n heterojunction increases first and then
decreases. Among all the heterojunctions, the Bi2O3/Bi2O4-3 has the
highest activity and can degrade 97.7% of MO, indicating the optimum
content of Bi2O3. To quantify the photocatalytic performance, the ki-
netics of MO degradation reaction on the sample was analyzed using a
pseudo first-order model [42]. The apparent rate constant k can be ob-
tained from a fitted line graph of -ln(C/C1) versus t and is shown in
Fig. 6b. As can be seen from Fig. 6b, Bi2O3/Bi2O4 composites have
higher photocatalytic degradation rate constant (k), in comparison with
Bi2O4. The k value of optimal Bi2O3/Bi2O4-3 sample is as high as
0.03623/min, which is about 2.4 times higher as that of Bi2O4
(0.01483/min).
Fig. 5. Nitrogen adsorption-desorption isotherms and BET surface areas (SBET
)
(inserted table) of Bi2O3, Bi2O4, and Bi2O3/Bi2O4-3 composite.
areas of Bi2O4, Bi2O3, and Bi2O3/Bi2O4 composites were determined by
nitrogen adsorption-desorption experiments(Fig. 5). The pure Bi2O4 and
Bi2O3 without hysteresis loops are obviously a typical type II curve,
indicating that they are non-porous structures. The same results were
observed on the Bi2O3/Bi2O4 p-n heterojunction. The BET surface area
(SBET) was listed in the insert table. It is well known that a higher BET
specific surface area can provide more active centers and absorb more
active species. The SBET values of Bi2O4, Bi2O3, and Bi2O3/Bi2O4 com-
posites are 2.6, 4.5, and 5.9 m2/g, respectively, indicating that the
specific surface area of the composite did not increase significantly. The
results show that the specific surface area is not the main factor affecting
the photocatalytic performance of Bi2O3/Bi2O4 p-n heterojunction.
In addition to organic dyes, other colorless chemical contaminants
like phenol are also used as target contaminants to evaluate the visible-
light activity of Bi2O3/Bi2O4 heterojunctions. As shown in Fig. 6c, Bi2O4
and all heterojunctions present insignificant adsorption behavior for
phenol except Bi2O3/Bi2O4-4 sample, indicating that phenol is difficult
Fig. 6. Adsorption and photocatalysis of MO (a) and phenol (c) over samples under visible light (λ > 420 nm) irradiation. The preudo-first-order reaction kinetics of
the prepared samples for MO (b) and phenol (d) degradation.
6

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
Fig. 8. Effects of various scavengers on the photocatalytic efficiency of Bi2O3/
Bi2O4-3 sample for the degradation of MO under visible light irradiation.
heterojunction can greatly increase the photocurrent density. Under
visible-light irradiation, The electrons generated on the VB of Bi2O3 and
Bi2O4 would transfer onto these CB, leaving photo-induced holes on its
VB. Driven by the built-in electric field, the holes on the VB of Bi2O4 are
transferred to the VB of Bi2O3, and the electrons on the CB of Bi2O3 are
transferred to the CB of Bi2O4. Therefore, the formation of the Bi2O3/-
Bi2O4 p-n heterojunction helps to separate carriers as much as possible
by the internal electric field [45].
Apart from the photocurrent response, the charge carrier migration
can be investigated by electrochemical impedance spectroscopy (EIS).
As shown in Fig. 7b, Bi2O3/Bi2O4-3 shows a much smaller semicircle
radius of EIS Nyquist plots than pristine Bi2O4, suggesting the lower
charge transfer resistance and thus more efficient interfacial charge
transfer across the contact surface at the p-n heterojunction. The result
of EIS is consistent with that of photocurrent in Fig. 7a.
Fig. 7. Photocurrent (a) and electrochemical impedance spectra (EIS) (b) of
samples under the visible light irradiation at 0 V vs.SEC. [Na2SO4] ¼ 0.5 M.
to be adsorbed onto the surface of photocatalyst. Nevertheless, all
Bi2O3/Bi2O4 heterojunctions have better visible light photocatalytic
activity for phenol degradation than pure Bi2O4. Similarly, the Bi2O3/
Bi2O4-3 sample has the best catalytic activity, with the highest rate
constant (0.00386/min), which is 1.49 times higher as that of Bi2O4
(0.00259/min). Obviously, the proper amount of Bi2O3 is the key to
achieve the best photocatalytic efficiency of Bi2O3/Bi2O4 p-n
heterojunction.
3.8. Roles of reactive species
In order to understand the mechanism of photocatalytic process, it is
necessary to study the main active species formed in the Bi2O3/Bi2O4 p-n
heterojunction system. Therefore, active species capture experiments
were conducted by adding different scavengers to the photocatalytic
system. As displayed in Fig. 8, the addition of sodium oxalate (hole (hþ)
scavenger) almost completely inhibits the degradation of MO, demon-
strating that hþ is essential for the degradation of MO molecules in the
photocatalytic process [46]. The addition of BQ (⋅OÀ2 free radicals
scavenger) also has a significant inhibitory effect on the degradation
efficiency, indicating that ⋅OÀ2 plays an important role in the photo-
catalytic process as well [47]. When isopropanol (IPA) (⋅OHÀ radicals
scavenger), the photodegradation efficiency of MO remained un-
changed, indicating that ⋅OHÀ radical is not the main active species [7].
The active species capture experiments indicate that the hþ and ⋅O2À
radicals are dominant active species in the photocatalytic reaction
process.
The photostability and reusability of photocatalytic materials are
prerequisites for its practical application. Thus, the photochemical sta-
bility of Bi2O3/Bi2O4-3 junction was implemented via three runs pho-
tocatalytic degradation of MO under visible light irradiation. As
depicted in Fig. S5, the Bi2O3/Bi2O4-3 sample displays stability to some
extent in the first two recycle test. However, the decay of photocatalytic
activity is very obvious after the third cycling run. Hence, the stability of
Bi2O3/Bi2O4 junction needs to be further improved.
3.7. Photoelectrochemical properties
Photocurrent measurement has generally known as an effective
method for studying photoinduced electron-hole separation and
migration in photocatalysts [43,44]. Transient photocurrent responses
of pure Bi2O4 and Bi2O3/Bi2O4-3 p-n heterojunctions (Fig. 7a) under
several 30s on-off cycles of irradiation were investigated. Because of the
generation of photogenerated charge carriers, it can be seen that the
photocurrent intensity increases quickly with the opening of light. After
turning off the light, the photocurrent was suddenly restored to the
original dark current state. Compared with pure Bi2O4, the Bi2O3/-
Bi2O4-3 heterojunction has a higher current density, about 5.5 times
higher than that of Bi2O4. Obviously, the formation of Bi2O3/Bi2O4 p-n
3.9. Photodegradation mechanisms
According to the above analysis, the possible photocatalytic mech-
anism for Bi2O3/Bi2O4 heterojunctions was proposed in Fig. 9. Based on
the previous reports, the type of semiconductor for Bi2O3 and Bi2O4 is p-
type and n-type, respectively [7,23,24]. Therefore, the electron con-
centration of Bi2O3 is lower than that of Bi2O4 and the Fermi level (Ef) of
Bi2O4 is close to its CB, while the Ef of Bi2O3 is near to its VB. Before
contact, the CB and VB of Bi2O3 are lower than those of Bi2O4 [20]. After
contact, the Ef of p-type Bi2O3 is moved up and the Ef of n-type Bi2O4 is
7

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
Fig. 9. A possible photocatalytic degradation mechanism of pollutant over Bi2O3/Bi2O4 p-n heterojunctions under visible light irradiation.
moved down until an equilibrium state is formed. When the Bi2O3/Bi2O4
p-n heterojunction is irradiated under visible light, both Bi2O3 and Bi2O4
could be excited to generate photo-generated carriers. Due to the role of
the internal field with the direction from n-type Bi2O4 to p-type Bi2O3,
the photogenerated holes on the VB of Bi2O4 tend to transfer to the VB of
Bi2O3. The accumulated holes on the VB of Bi2O3 can directly degrade
pollutants under visible-light irradiation since hþ is the most important
oxidative species confirmed by the trapping experiments of active spe-
cies [48]. Meanwhile, the staggered energy band structure favors the
rapid transfer of the photogenerated electrons on the CB of Bi2O3 to the
CB of Bi2O4. The photoinduced electrons on the CB bottom of n-type
Bi2O4 can react with dissolved O2 to generate reactive ⋅OÀ2 radicals,
which also plays an important role in the degradation process of pol-
lutants. Thus, the photo-generated carriers are effectively separated by
the p-n heterojunction interface formed in the Bi2O3/Bi2O4 hetero-
junctions. Therefore, the photocatalytic activity of Bi2O3/Bi2O4 heter-
ostructure is superior to pure Bi2O4 for the degradation of organic
pollutants under visible-light irradiation.
CRediT authorship contribution statement
Chengyu Li: Writing - original draft, Writing - review & editing.
Yaya Ma: Visualization. Shizheng Zheng: Validation. Changyuan Hu:
Funding acquisition, Investigation. Feng Qin: Resources, Software. Lin
Wei: Resources, Software. Cuiqing Zhang: Resources, Software. Shu-
wang Duo: Resources, Software. Quanhong Hu: Resources, Software.
Acknowledgments
Financial support from the National Natural Science Foundation of
China (No.21663012), the Natural Science Foundation of Jiangxi
Province (20181BAB203009), and the Scientific &Technological Project
of Jiangxi Science and Technology Normal University (No.
2015CXTD003), P.R. China is gratefully acknowledged.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jpcs.2020.109376.
4. Conclusions
The Bi2O3/Bi2O4 p-n heterojunctions were successfully prepared via
a facile one-pot hydrothermal method and applied to photocatalytic
removal of MO and phenol under visible light illumination. The optimal
Bi2O3/Bi2O4-3 sample shows a rate constant of 0.03623/min for MO,
which is about 2.4 times higher than that of Bi2O4, and a rate constant of
0.00386/min for phenol, which is about 1.49 times higher than that of
Bi2O4. On the basis of a series of characterizations, the enhanced pho-
tocatalytic activity of Bi2O3/Bi2O4 is primarily attributed to the forma-
tion of p-n heterojunction, which promotes the charge carrier
transportation and separation through the built-in electric field. Our
work provides a simple and effective method to enhance the photo-
catalytic performance of Bi2O4.
References
[1] Kavita Sahu, Aditi Bisht, Sini Kuriakose, Satyabrata Mohapatra, J. Phys. Chem.
Solid. 137 (2020) 109223.
[2] H.L. Jiang, M.L. Li, J. Liu, X.Q. Li, L. Tian, P.H. Chen, Ceram. Int. 44 (2018)
2709–2717.
[3] Y.Q. Zhang, C. Lin, Q. Lin, Y.B. Jin, Y.H. Wang, Z.Z. Zhang, H.X. Lin, J.L. Long, X.
X. Wang, Appl. Catal. B Environ. 235 (2018) 238–245.
[4] Kavita Sahu, Sini kuriakose, Jaspal Singh, Biswarup Satpati, Satyabrata Mohapatr,
J. Phys. Chem. Solid. 121 (2018) 186–195.
[5] S. Mishra, M. Priyadarshinee, A.K. Debnath, K.P. Muthe, B.C. Mallick, N. Das,
P. Parhi, J. Phys. Chem. Solid. 137 (2020) 109211.
[6] L.M. Sun, L. Xiang, X. Zhao, C.J. Jia, J. Yang, Z. Jin, X.F. Cheng, W.L. Fan, ACS
Catal. 5 (2015) 3540–3551.
[7] W.J. Wang, X.Q. Chen, G. Liu, Z.R. Shen, D.H. Xia, P.K. Wong, J.C. Yu, Appl. Catal.
B Environ. 176–177 (2015) 444–453.
Declaration of competing interest
[8] R.A. He, S.W. Cao, P. Zhou, J.G. Yu, Chin. J. Catal. 35 (2014) 989–1007.
[9] D. Xia, I.M. Lo, Water Res. 100 (2016) 393.
[10] X.F. Chang, J. Huang, C. Cheng, W. Sha, X. Li, G.B. Ji, S. Deng, G. Yu, J. Hazard
Mater. 173 (2010) 765–772.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[11] Abdul Hameed, Muhammad Aslama, M. Iqbal, I. Ismaila, Numan Salahd, Paolo,
Appl. Catal. B Environ. 163 (2015) 444–451.
[12] Y. Hua, D.Z. Li, F.Q. Sun, Y.Q. Weng, S.Y. You, Y. Shao, J. Hazard Mater. 301
(2016) 362–370.
[13] H.L. Jiang, X.Q. Li, M.L. Li, P.P. Niu, T. Wang, D.Z. Chen, P.H. Chen, J.P. Zou,
Chem. Eng. J. 358 (2019) 58–66.
8

C. Li et al.
Journal of Physics and Chemistry of Solids 140 (2020) 109376
[14] T.H. Yang, Y.W. Harn, L.D. Huang, M.Y. Pan, W.C. Yen, M.C. Chen, C.C. Lin, P.
K. Wei, Y.L. Chueh, J.M. Wu, J. Catal. 329 (2015) 167–176.
[15] Y.Y. Ma, C.Q. Zhang, C.Y. Li, F. Qin, L. Wei, C.Y. Hu, Q.H. Hu, S.W. Duo, Colloids
Surf., A 580 (2019) 123757.
[31] A. Hameed, T. Montini, V. Gombac, P. Fornasiero, J. Am. Chem. Soc. 130 (2008)
9658–9659.
[32] Y. Bian, Y. Ma, Y. Shang, P. Tan, J. Pan, Appl. Surf. Sci. 430 (2018) 613–624.
[33] R. Hu, X. Xiao, S. Tu, X. Zuo, J. Nan, Appl. Catal. B Environ. 163 (2015) 510–519.
[34] Y. Huang, W. Fan, B. Long, H. Li, F. Zhao, Z. Liu, Y. Tong, H. Ji, Appl. Catal. B
Environ. 185 (2016) 68–76.
[16] F. Mian, G. Bottaro, M. Rancan, L. Pezzato, V. Gombac, P. Fornasiero, L. Armelao,
Omega 2 (2017) 6298–6308.
[17] F. Schuster, B. Laumer, R.R. Zamani, C. Mag_n, J.R. Morante, J. Arbiol,
M. Stutzmann, ACS Nano 8 (2014) 4376–4384.
[35] C. Yu, P. Yang, L. Tie, S. Yang, S. Dong, J. Sun, J. Sun, Appl. Surf. Sci. 455 (2018)
8–17.
[18] Z.W. Wang, M. Chen, D.L. Huang, G.M. Zeng, P. Xu, C.Y. Zhou, C. Lai, H. Wang,
M. Cheng, W.J. Wang, Chem. Eng. J. 374 (2019) 1025–1045.
[19] G.P. Dai, J.G. Yu, G. Liu, J. Phys. Chem. C 115 (2011) 7339–7346.
[20] J.X. Wang, Z.Z. Zhang, X. Wang, Y. Shen, Y.F. Guo, P.K. Wong, R.B. Bai, Chin. J.
Catal. 39 (2018) 1792–1803.
[36] B.Y. Xu, Y. An, Y.Y. Liu, X.Y. Qin, X.Y. Zhang, Y. Dai, Z.Y. Wang, P. Wang, M.
H. Whangbo, B.B. Huang, J. Mater. Chem. 5 (2017) 14406–14414.
[37] G. Zhu, Y. Liu, M. Hojamberdiev, J. Han, J. Rodríguez, S.A. Bilmes, P. Liu, New J.
Chem. 39 (2015) 9557–9568.
[38] C. Zhang, X. Li, S. Zheng, Y. Ma, C. Hu, C. Li, Colloids Surf., A 548 (2018) 150–157.
[39] C. Yu, W. Zhou, L. Zhu, G. Li, K. Yang, R. Jin, Appl. Catal. B Environ. 184 (2016) 1.
[40] Y. Ding, G. Zhang, X. Wang, L. Zhu, H. Tang, Appl. Catal. B Environ. 202 (2017)
528.
[21] Y. Sang, X. Cao, G.D. Dai, L.X. Wang, Y. Peng, B.Y. Geng, J. Hazard Mater. 381
(2020) 120942.
[22] Y. Bian, Y. Ma, Y. Shang, P. Tan, J. Pan, Appl. Surf. Sci. 430 (2018) 613–624.
[23] L. Zhang, G. Wang, Z. Xiong, H. Tang, C. Jiang, Appl. Surf. Sci. 436 (2018)
162–171.
[41] T. Xiong, M. Wen, F. Dong, J. Yu, L. Han, B. Lei, Y. Zhang, X. Tang, Z. Zang, Appl.
Catal. B Environ. 199 (2016) 87.
[24] X. Xiao, S. Tu, C. Zheng, H. Zhong, X. Zuo, J. Nan, RSC Adv. 5 (2015)
74977–74985.
[42] Z.S. Liu, B.T. Wu, J.N. Niu, P.Z. Feng, Y.B. Zhu, Mater. Res. Bull. 63 (2015)
187–193.
[25] M.L. Guan, D.K. Ma, S.W. Hu, Y.J. Chen, S. Huang, Inorg. Chem. 50 (2011)
800–805.
[43] Sini Kuriakose, Biswarup Satpati, Satyabrata Mohapatra Phys, Chem. Chem. Phys.
17 (2015) 25172–25181.
[26] K.H. Reddy, K. Parida, P.K. Satapathy, J. Mater. Chem. A 5 (2017) 20359–20373.
[27] Z. Deng, T. Liu, T. Chen, J. Jiang, W. Yang, J. Guo, J. Zhao, H. Wang, L. Gao, ACS
Appl. Mater. Interfaces 9 (2017) 12469–12477.
[44] Y. Liu, C.S. Xie, J. Li, T. Zou, D.W. Zeng, Appl. Catal., A 433–434 (2012) 81–87.
[45] P. Suyana, Priyanka Ganguly, Balagopal N. Nair, Environ. Sci.: Nano 4 (2017)
212–221.
[28] Sini Kuriakose, Kavita Sahu, Saif A. Khan, A. Tripathi, D.K. Avasthi,
Satyabrata Mohapatra, Opt. Mater. 64 (2017) 47–52.
[46] T. Takei, R. Haramoto, Q. Dong, N. Kumada, Y. Yonesaki, N. Kinomura, T. Mano,
S. Nishimoto, Y. Kameshima, M. Miyake, J. Solid State Chem. 184 (2011)
2017–2022.
[29] H.J. Lu, L.L. Xu, B. Wei, M.Y. Zhang, H. Gao, W.J. Sun, Appl. Surf. Sci. 303 (2014)
360–366.
[47] Y.B. Ding, F. Yang, L.H. Zhu, Appl. Catal. B Environ. 164 (2015) 151–158.
[48] W. Li, X.K. Jia, P.T. Li, B.L. Zhang, H.P. Zhang, W.C. Geng, Q.Y. Zhang, Sustain.
Chem. Eng. 3 (2015) 1101–1110.
[30] S. Bai, J. Jiang, Q. Zhang, Y.J. Xiong, Chem. Soc. Rev. 44 (2015) 2893–2939.
9