Preparation and Performance of Visible-Light-Driven Bi2O3/ZnS Heterojunction Functionalized Porous CA Membranes for Effective Degradation of Rhodamine B

Article abstract of DOI:10.1002/pssa.201701061

Photocatalytic degradation is an effective method for fast removal of dye-contaminated water. In the present work, a bismuth oxide/zinc sulfide-cellulose acetate photocatalytic membrane (Bi₂O₃/ZnS-CA) is successfully fabricated via the phase-inversion method with a Bi₂O₃/ZnS heterojunction as the photocatalyst. The Bi₂O₃/ZnS-CA membranes shows high photocatalytic activity in degradation of Rhodamine B (RhB) under visible-light irradiation due to the Bi₂O₃/ZnS heterojunction structure contributing to the high efficiency of photogenerated charge separation. The photodegradation efficiency of Bi₂O₃/ZnS-CA membrane is significantly enhanced (about 30% improvement) and the photodegradation kinetic is four times faster than that of the Bi₂O₃-CA and the ZnS-CA membranes. Moreover, the Bi₂O₃/ZnS-CA membrane exhibits excellent permeation flux (649.7 L m^?2 h^?1) and high porosity (69.94%). The Bi₂O₃/ZnS-CA membrane also shows good adsorption performance of RhB, which could enhance the interaction between RhB and the catalyst. This work provides a new sight to construct highly effective visible-light-driven photocatalytic membranes for continuous organic dye treatment.

Full text of DOI:10.1002/pssa.201701061

a
ORIGINAL PAPER
Photocatalytic Membranes
www.pss-a.com
Preparation and Performance of Visible-Light-Driven
Bi2O3/ZnS Heterojunction Functionalized Porous CA
Membranes for Effective Degradation of Rhodamine B
Binrong Li, Jinyu Chu,* Yi Li, Minjia Meng,* Yanhua Cui, Qidi Li, and Yonghai Feng
rivers,[3,4] which causes great damage to the
Photocatalytic degradation is an effective method for fast removal of dye-
contaminated water. In the present work, a bismuth oxide/zinc sulfide-
cellulose acetate photocatalytic membrane (Bi2O3/ZnS-CA) is successfully
fabricated via the phase-inversion method with a Bi2O3/ZnS heterojunction
as the photocatalyst. The Bi2O3/ZnS-CA membranes shows high photo-
catalytic activity in degradation of Rhodamine B (RhB) under visible-light
irradiation due to the Bi2O3/ZnS heterojunction structure contributing to the
high efficiency of photogenerated charge separation. The photodegradation
efficiency of Bi2O3/ZnS-CA membrane is significantly enhanced (about 30%
improvement) and the photodegradation kinetic is four times faster than that
of the Bi2O3-CA and the ZnS-CA membranes. Moreover, the Bi2O3/ZnS-CA
membrane exhibits excellent permeation flux (649.7 L mꢀ2 hꢀ1) and high
porosity (69.94%). The Bi2O3/ZnS-CA membrane also shows good adsorption
performance of RhB, which could enhance the interaction between RhB and
the catalyst. This work provides a new sight to construct highly effective
visible-light-driven photocatalytic membranes for continuous organic dye
treatment.
ecosystem and even endangers human
health.[5–8] Therefore, it is urget to develop
effective way to remove RhB from the RhB-
contaminated water. Many methods, in-
cluding physical, chemical, and biologic
stratigies, have been used to deal with the
RhB contaminted water through adsrop-
tion or degradation.[6,9–11] From among
these methods, photocatalytic oxidation
attracts much attention due to the high
degradation efficiency, energy saving, and
environment-friendly properties.[12]
Recently, numerous semiconductor
materials have been applied as photo-
catalyst for the degradation of organic
dye into harmless chemicals, such as
TiO2,[13] Bi2O3,[14,15] C3N4,[16,17] CdS,[18,19]
etc. Among them, Bi2O3 stands out among
many attractive candidate semiconductor
materials because of its low cost, favorable
direct bandgap of 2.5–2.8 eV, and good
photostability in acidic conditions. Zhang
et al.[20] prepared nanosized Bi2O3 to
enhance the photocatalytic activity by a simple sonochemical
route. In addition, Liu et al.[21] showed the crystalline phase
evolution of Bi2O3 had an effect on photocatalytic performance.
Although the above methods can promote the photocatalytic
degradation of dyes, they are not effective in amelioration of
visible-light absorption[22] and photoinduced charge separation.
To extendthe lifespanofphotogeneratedelectrons and holes for
higher photocatalytic activity, various strategies have been
reported, such as doping with metal elements (Au-doped
Bi2O3[23]), nonmetal elements (S-doped Bi2O3[24]) and coupling
with other semiconductors to form heterojunction structures
(Bi2O3/Bi2MoO6,[25] Bi2O3/g-C3N4[26]). The heterojunction struc-
ture can not only combine at least two disparate functional
materials in one system but also reduce the recombination rate of
the electron–hole pairs. For instance, Juntrapirom et al.[27] found
that the novel SnS/BiOI heterostructures presented higher
photocatalytic performance and better stability than a single
semiconductor. Liu et al.[28] synthesized another BiVO4/WO3
heterojunction with an excellent visible-light response, a good
photostability and high photoelectrochemical activity. However, it
is difficult to separate it from treated water, which hinders its
practical application. The challenge is to develop an effective
method to improve the recovery and reuse of the catalyst.
1. Introduction
Recently, the water environments are facing the serious problem
of dye contamination as the textile industries and the printing
and dyeing factories are scaling up and the treatment of dye
wastewater is very difficult due to their nonbiodegradable in
nature.[1,2] Rhodamine B (RhB) is one of the most commonly
used dyes and is usually reported to contaminate lakes and
Dr. B. Li, Prof. J. Chu, Q. Li
School of Environment and Safety Engineering
Jiangsu University
Zhenjiang 212013, China
E-mail: jychu@ujs.edu.cn
Dr. B. Li, Y. Li, Prof. M. Meng, Dr. Y. Cui
School of Chemistry and Chemical Engineering
Jiangsu University
Zhenjiang 212013, China
E-mail: mmj@ujs.edu.cn
Dr. Y. Feng
School of Institute for Advanced Materials
Jiangsu University
Zhenjiang 212013, China
DOI: 10.1002/pssa.201701061
Phys. Status Solidi A 2018, 1701061
1701061 (1 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Photocatalytic membranes have shown great potential for use
and recovery in energy-efficient wastewater treatment since they
combine the physical separation and the organic degradation
achieved by photocatalysis in a single system.[29] The most
common method for fabricating the photocatalytic membranes
is to blend the catalyst into a membrane matrix. The common
membrane matrix includes polyvinylidenefluoride (PVDF),[30]
polysulfone (PSf),[31] cellulose acetate (CA), etc. Among these,
CA-based membranes have been widely utilized as support
membranes due to the advantages of biodegradability,[32]
nontoxicity and low cost.[33] CA-based photocatalytic membranes
also provided sufficient interaction between RhB and the catalyst
because of high water flux, hydrophilicity and high porosity.[34]
In this work, the Bi2O3/ZnS heterojunction, which has been
rarely reported before, was successfully synthesized by a one-
step hydrothermal method. Meanwhile, the environment-
friendly CA membranes that were developed to immobilize
Bi2O3/ZnS nanocomposite (Bi2O3/ZnS-CA membranes) were
prepared via facile phase inversion for RhB degradation under
visible-light irradiation. The structure and properties of the
prepared materials can be obtained by relevant characterization
by means of field emission scanning electron microscopy
(FESEM), transmission electron microscopy (TEM), X-ray
diffraction (XRD-6100) at a scanning speed of 7ꢁ minꢀ1, X-ray
photoelectron spectroscopy (XPS), UV–vis absorption/diffuse
spectra (DRS), electron spin resonance (ESR), DCF fluorescence
measurements, and the catalytic performance of the Bi2O3/ZnS-
CA membranes was determined by a UV–vis spectrophotometer.
This work provided an insight into the design of photocatalytic
membranes and improvement of photocatalytic activity.
H2O followed by stirring for 10 min. Then, the prepared Bi2O3
powder was added into the mixed solution as a certain
proportion (for instance, Bi2O3: ZnS ¼ 1:2 referred to add
5.0 mmol of Bi2O3). Finally, the mixture was stirred for 10 min
again and transferred into a 100-mL Teflon-lined autoclave that
was maintained at 170 ꢁC for 5.0 h. The product was collected
after centrifugation, washed by deionized water and dried at
60 ꢁC for 12 h. For comparison, ZnS was prepared by a
hydrothermal
method.
2.2 g
(10 mmol)
of
Zn-
(COOH)2 ꢂ 2H2O and 3.8 g (50 mmol) of thiourea were dissolved
in 70 mL of H2O followed by stirring for 10 min. Then, the
mixture was transferred into a 100-mLTeflon-lined autoclave and
maintained at 170 ꢁC for 5.0 h. After cooling down to room
temperature, the product was sequentially washed with
deionized water and absolute ethanol several times and then
was dried at 60 ꢁC for 12 h.
2.3. Preparation of Bi2O3/ZnS-CA Photocatalytic
Membranes
The CA-based photocatalytic membranes were prepared by
incorporating the Bi2O3/ZnS heterojunction via the phase-
inversion method. Briefly, the Bi2O3/ZnS heterojunction powder
and CA particles were dissolved in DMSO solution under
agitation to form a casting solution. After stirring at 40 ꢁC for
4.0 h, the casting solution was pumped into a vacuum for 60 min
in order to remove the bubbles. Then, the casting solution was
cast onto a clean glass plate and horizontally transferred to a
water coagulation bath for 30 min to remove the residual solvent.
Upon complete coagulation, the resultant membranes were
peeled and soaked in deionized water for further utilization. For
comparison, original CA membranes, Bi2O3-CA membranes
and ZnS-CA membranes were prepared with the same method
by the following recipes in Table 1.
2. Experimental Section
2.1. Materials
Cellulose acetate (CA), dimethyl sulfoxide (DMSO), thiourea,
bismuth nitrate pentahydrate (Bi(NO3)3 ꢂ 5H2O) powder, and
zinc acetate dehydrate (Zn(COOH)2 ꢂ 2H2O) were purchased
from Sinopharm Chemical Reagent Co., Ltd. Ethanol (C2H5OH,
99.5%, Aladdin), methyl alcohol (CH3OH, 99.5%, Aladdin), 5,5-
dimethyl-1-pyrroline N-oxide (DMPO, C6H11NO, 97%, Aladdin),
2,7-dichlorofluorescein diacetate (DCFH-DA) and Rhodamine B
were supplied by Shanghai Macklin Biochemical Co., Ltd.
2.4. Characterization
The morphology and microstructure of the obtained samples
were characterized by field emission scanning electron
microscopy (SEM, JSM-6010PLUS/LA) and transmission elec-
tron microscopy (TEM, JEM-2100 (HR)). X-ray diffraction was
carried out on thꢀe1X-ray diffractometer (XRD-6100) at a scanning
speed of 7ꢁ min in the range from 10 to 80ꢁ. The chemical
composition was analyzed by X-ray photoelectron spectroscopy
(XPS, Thermo SCIENTIFIC ESCALAB 250Xi) with Al Kα
excitation radiation (1486.8 eV). UV–vis absorption spectra were
2.2. Synthesis of Bi2O3/ZnS Heterojunction Structure
The Bi2O3 crystal was synthesized by a typical method.[15] First,
9.7 g of Bi(NO3)3 ꢂ 5H2O was dissolved into 200 mL of deionized
water and stirred for 10 min. Then, the pH value of the solution
was adjusted to 8.0 by adding NaOH (0.2 M). The suspension
was sonicated for 60 min at 80 ꢁC and then centrifuged, washed
three times with distilled water and dried at 60 ꢁC overnight.
Finally, the light-yellow precipitate was calcined at 400 ꢁC for
2.0 h and ground to obtain the Bi2O3 powder.
Table 1. Compositions of casting solution.
Additives
Membrane
CA (wt.%)
Materials
Amounts (wt %)
DMSO (wt.%)
CA
13
13
13
13
–
–
2
2
2
87
85
85
85
Bi2O3-CA
ZnS-CA
Bi2O3
The Bi2O3/ZnS heterojunction was prepared by using a one-
step hydrothermal method. 2.2 g (10 mmol) of Zn(COOH)2, 3.8 g
(50 mmol) of thiourea, and 0.5 g of PEGwere added into 70 mL of
ZnS
Bi2O3/ZnS-CA
Bi2O3/ZnS
Phys. Status Solidi A 2018, 1701061
1701061 (2 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
recorded on a UV–vis–NIR spectrophotometer to analyze the
light absorption of samples. The surface hydrophilicity of the
membranes was measured by a contact-angle measuring
instrument (KSV CM200).
2.7. ESR experiments DCF Fluorescence Measurements
In ESR experiments, the •OH and •O2ꢀ species were trapped by
the DMPO. 10 mg of the as-prepared samples were dissolved in
0.5 mL of deionized water (•OH) or methanol (•O2ꢀ), respec-
tively. Then, 20 mL of DMPO was added in 70 mL of the above
suspension followed by ultrasonic dispersion for 5.0 min.
In the DCF fluorescence measurement, 0.004 g of Bi2O3, ZnS
and Bi2O3/ZnS heterojunction were diluted in 2 mL of deionized
water followed by ultrasonic dispersion for 10 min. After light
illumination for 15 min, 20 mL of DCFH-DA solution (20 mM)
was added to the solution rapidly. Next, the mixture was allowed
to react sufficiently for 1.0 h at 37 ꢁC and then diluted 1500 times.
Finally, the fluorescence intensity was measured by a fluores-
cence spectrophotometer (Cary Eclipse) with excitation and
emission at 488 nm and 525 nm, respectively. The volume ratio
of the catalyst solution and DCFH-DA solution (20 mM) was 100:
1. For comparison, a blank sample was prepared by the same
method without a catalyst.
2.5. Permeation Properties and Porosity of Membranes
A pure-water flux measurement of photocatalytic membranes
was conducted on a dead-end membrane system with an
effective area of 3.14 cm2. The water flux J (L mꢀ2 hꢀ1) was
measured at 0.08 MPa and calculated by Eq. (1).
Q
A ꢃ T
J ¼
ð1Þ
where Q is the quantity of permeate (L), A is the effective
membranes area (m2), and T is the time (h) to obtain the quantity
of Q.
The porosity of the prepared membranes was determined by
using a gravimetric method as expressed in Eq. (2).[35]
3. Results and Discussion
ðWwꢀWdÞ
ρH
eð%Þ ¼
ð2Þ
3.1. Characterization of Bi2O3/ZnS Heterojunction
ðWwꢀWdÞ
Wd
ρc
þ
ρH
Figure 1 shows the XRD patterns of Bi2O3, ZnS, and the Bi2O3/
ZnS heterojunction, respectively. As shown in Figure 1a, the
XRD pattern of Bi2O3 still shows identifiable characteristic
peaks at 27.5, 33.3, 46.4ꢁ, unambiguously ascribed to the (120),
(200) and (041) planes of Bi2O3 (JCPDS 41–1449), respec-
tively.[36] The small peaks at around 19.7ꢁ and 21.8ꢁ were
indexed to the (–111) and (020) diffraction planes of Bi2O3,
respectively. The sharp, strong and well-separated diffraction
peaks indicated that the samples were well-crystallized after
being calcining at 400 ꢁC. As for the ZnS sample, three major
diffraction peaks at 28.7, 48.2, and 57.2ꢁ were observed, which
corresponded to the (111), (200) and (311) planes of the cubic
zinc blende (JCPDS 39–1363).[37] In the XRD patterns of the
Bi2O3/ZnS heterojunction, the characteristic peaks of Bi2O3 at
27.5 and 33.3ꢁ can be clearly found with sharp peaks. However,
the intensity of these peaks was obviously weakened in
comparison with Bi2O3, which may be due to the distortion
of the Bi2O3 crystal phase and the change of crystal structure
during the hydrothermal process. From the high-resolution
XRD patterns of Bi2O3/ZnS (Figure 1b), it was observed that
the dominant peak at 28.7ꢁ due to the ZnS (111) facet shifted to
28.5ꢁ and the width of the peak was narrowed, which might
originate from the distortion of the ZnS crystal phase. In the
XRD pattern of Bi2O3/ZnS samples (Figure 1c), the same
reason can also explain that the peak at 57.2ꢁ attributed to ZnS
(311) shifted to 56.8ꢁ, whereas the peak at 47.7ꢁ disappeared,
which was due to the distortion of the Bi2O3 crystal phase after
introduction of ZnS. The results above suggested that the
Bi2O3/ZnS heterojunction was successfully fabricated.
where Ww is the weight of wet membranes and Wd is the weight
of the dry membranes (g). ρH and ρc are the density of water
(0.998 g cmꢀ3) and cellulose (1.5 g cmꢀ3), respectively. The mean
pore radius rm was calculated accordingly to the Guerout–
Elford–Ferry equation (Eq. (3))
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð2:9 ꢀ 1:7eÞ ꢃ 8ηlQ
rm
¼
ð3Þ
e ꢃ A ꢃ ΔP
where η is the water viscosity (8.9 ꢃ 10ꢀ4 Pa s), l is the
membranes thickness (m), Q is the volume of the permeate
water per unit time (m3 sꢀ1), and ΔP is the load pressure (Pa).
The water flux (J) of photocatalytic membranes was determined
according to (Eq. (1)).
2.6. Photocatalytic Property of Membranes
The photocatalytic activity of Bi2O3/ZnS-CA photocatalytic
membranes was evaluated by degrading Rhodamine B under
visible-light irradiation (λ ꢄ 420 nm). The photocatalytic mem-
branes loaded with different content (0.05 g, 0.1 g, 0.2 g) of
photocatalyst (the photocatalytic activity of Bi2O3/ZnS hetero-
junction climbed to a maximum when the ratio of Bi2O3: ZnS
reached 1:2) was placed in a quartz cup that contained 50 mL of
5.0 mg Lꢀ1 RhB solution. The membranes were first placed in
the dark for 0.5 h to reach adsorption equilibrium. Then, the
solution was irradiated by a xenon light source with the power of
200 W. 2.0-mL samples were withdrawn from the quartz cup at
30 min intervals. The solutions were kept in the dark before
measuring the concentrations. The RhB concentration was
measured by an ultraviolet–visible (UV–vis) spectrophotometer
(Cary 8454) at 554 nm.
The morphology of the composites was observed by SEM and
TEM. The SEM image of the Bi2O3 (Figure 2a) displayed rod-
like particles with average lengths of 15 mm, which was
consistent with the previous literature.[38] In Figure 2b, the
SEM images showed that particles of pure ZnS adopted a
Phys. Status Solidi A 2018, 1701061
1701061 (3 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 1. XRD patterns (a) and partially enlarged patterns (b, c) prepared based on different samples.
spherical morphology with an average diameter arranged from
1 to 2 mm. In TEM images (Figures 2c and d), however, the size
of the rod-like Bi2O3 was extremely reduced and the sheet-like
structure appeared, indicating the recrystallization of Bi2O3
during the hydrothermal reaction. Meanwhile, it can be
observed that the irregular ZnS nanoparticles adhered to the
surface of Bi2O3 rods. Therefore, the nanosized ZnS was
immobilized on the surface of the Bi2O3 rods, forming a
relatively uniform Bi2O3/ZnS heterojunction.
To analyze the chemical composition and investigate the
surface valence state of these elements in the Bi2O3/ZnS
heterojunction, the X-ray photoelectron spectroscopy (XPS)
analysis has been carried out, in which Bi, O, S, Zn elements can
be observed in Figure 3a. In the high-resolution XPS spectrum of
Bi 4f (Figure 3b), the two strong peaks at 158.3 and 163.5 eV were
attributed to Bi 4f7/2 and Bi 4f5/2 of Bi3þ in the sample,
respectively.[39] The peaks appearing at 161.8 and 162.7 eV were
assigned to S 2p3/2 and S 2p1/2, which suggested the existence of
S2ꢀ in the final product. The Zn 2p peaks located at 1022.8 and
1046.1 eV could be attributed to Zn 2p3/2 and Zn 2p1/2
,
respectively (Figure 3c).[40] Therefore, this strongly indicated
the existence of ZnS. Furthermore, Figure 3d showed the
obvious peak at 529.5 eV, which was assigned to the Bi─O bonds
in the sample of the Bi2O3/ZnS heterojunction.[41] A small peak
at 531.3 eV was observed, which was due to the irreversibly
adsorbed water molecules on the surface of the sample. It can be
concluded that the Bi2O3/ZnS heterojunction has been
successfully fabricated on the basis of the above analysis results
of XRD, SEM, TEM, and XPS.
Light absorption is one of the crucial factors for photo-
catalysis, especially the visible-light-driven photocatalysis. The
UV–vis diffuse reflectance spectra of the samples were recorded
and the results are shown in Figure 4. In the case of pure ZnS,
the absorption edge was around 400 nm, which indicated that it
can absorb only UV light. In comparison with ZnS, the Bi2O3
exhibited a certain photoresponse ability to the visible light.[42]
However, in the case of the Bi2O3/ZnS heterojunction, an
obvious enhancement of the absorption intensity in the visible
region was clearly observed and absorption range was extended
to the whole range of visible-light spectra, which indicated that it
Figure 2. SEM images of Bi2O3 (a) and ZnS (b) and TEM images of the
Bi2O3/ZnS heterojunction (c, d).
Phys. Status Solidi A 2018, 1701061
1701061 (4 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 3. XPS spectra of the Bi2O3/ZnS heterojunction: survey (a), Bi4f and S2p (b), Zn2p (c) and O1s(d).
could absorb the whole range of light. The results of UV–vis
absorption spectra demonstrated that the Bi2O3/ZnS hetero-
junction can improve the absorption of visible light and provide
proper reaction conditions for photocatalytic degradation.
planes. For ZnS-CA membranes, the main diffraction peaks at
28.7, 48.2, and 57.2ꢁ were assigned to the (111), (200), and (311)
crystal planes, respectively. The peaks of Bi2O3 can be seen
clearly in Bi2O3-CA membranes. Meanwhile, the characteristic
peaks of Bi2O3/ZnS were also observed in the XRD pattern of
Bi2O3/ZnS-CA membranes. Therefore, the catalyst was immo-
bilized into the membranes successfully.
3.2. Morphology and Structure of Membranes
The top and cross-sectional views of membranes are
compared in Figure 6, as observed by SEM images. Compared
with the original CA membranes, the aggregation of hybrid
membranes could be observed on the surface after the different
catalysts were added.[31] Moreover, the holes on various
The XRD patterns of the membranes are shown in Figure 5. As
for CA membranes, the characteristic peak at 22.6ꢁ was
consistent with previous reports.[43] In blending membranes,
the characteristic peaks were attributed to different crystal
Figure 4. UV–vis absorption spectra of the as-prepared Bi2O3, ZnS and
Bi2O3/ZnS heterojunction.
Figure 5. XRD patterns of membranes.
Phys. Status Solidi A 2018, 1701061
1701061 (5 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 6. SEM images of top surface and cross-sectional morphology: CA, Bi2O3-CA, ZnS-CA and Bi2O3/ZnS-CA membranes.
membranes surface were reduced or disappeared, which may be
due to the fact that the hydrophilic composites with higher
affinity toward water could migrate to the membrane surface
during the phase-inversion process. In particular, the surface of
the Bi2O3-CA membranes with no holes may be due to the larger
size of the Bi2O3 particles, causing the mass aggregation of the
membranes matrix on the surface of membranes and blocking
the surface holes. In the cross-sectional view, however, the
photocatalytic membranes formed more regular and wider
channels than those of pristine CA membranes. For Bi2O3/ZnS-
CA membranes, the amelioration in cross-sectional structure
could be noted clearly from the SEM images. This phenomenon
could indicate the stronger hydrophilicity of Bi2O3/ZnS than that
of Bi2O3 and ZnS particles.[44] These results were also consistent
with the porosity and pore size data of membranes in Table 2.
This showed that the Bi2O3/ZnS-CA membranes exhibited
bigger pores and higher porosity than the other three
membranes.
3.3. Hydrophilicity of Membranes
The surface hydrophilicity of the photocatalytic membranes was
an important factor in measuring the permeability, which can be
Table 2. Porosity and surface mean pore size of membranes.
Membrane
Porosity (%)
Mean pore size (nm)
CA
54.50
65.84
67.73
69.94
113.27
121.21
130.66
140.17
Bi2O3-CA
ZnS-CA
Bi2O3/ZnS-CA
Figure 7. Water contact angle of various membranes.
Phys. Status Solidi A 2018, 1701061
1701061 (6 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
analyzed by the sessile drop technique. In general, smaller water
contact angle means higher hydrophilicity.[44] As shown in
Figure 7, the original CA membranes presented the mean water
contact angle of 72ꢁ, which was lower than Bi2O3-CA (81ꢁ) and
ZnS-CA (79ꢁ) membranes. This phenomenon can be explained
by the fact that pure CA membranes presented high permeation,
which was consistent with the results of water flux (to be shown
in the next section). However, it should be noted that the Bi2O3/
ZnS-CA membranes exhibited excellent hydrophilicity with
lower contact angle (60ꢁ). This amelioration of hydrophilicity
may be attributed to the migration of hydrophilic nanomaterials
moving toward the membrane/water interface, which can
decrease the interface energy during the phase-inversion
process.[45,46]
Figure 8. Pure-water flux of various membranes.
3.4. Permeation Properties of Membranes
The results of membrane performance concerning the water
permeation flux are illustrated in Figure 8. The water flux of the
pristine CA membranes (573.2 L mꢀ2 hꢀ1) was higher than that
of Bi2O3-CA membranes (439.5 L mꢀ2 hꢀ1) and ZnS-CA mem-
branes (535 L mꢀ2 hꢀ1). This interesting point may be explained
by the fact that the CA membrane surface had more holes. The
Bi2O3/ZnS-CA membranes had the higher water flux
(649.7 L mꢀ2 hꢀ1) when adding the Bi2O3/ZnS heterojunction
into membranes, which displayed more hydrophilicity and
facilitated the water molecules passing through the membranes.
In addition, the higher water permeability of Bi2O3/ZnS-CA
membranes indicated that the more regular channels and higher
hydrophilicity were dominant in the infiltration process on a
dead-end membranes system under 0.08 MPa. Thus, the
increase of hydrophilicity, together with the higher porosity
and bigger channels made the Bi2O3/ZnS-CA membranes
possess superior water flux over the other three membranes.
3.5. Photocatalytic Property of Membranes
RhB was chosen as a model organic pollutant to investigate the
photocatalytic activities of the Bi2O3/ZnS-CA membranes
(Figure 9). Figure 9a displays the degradation ratio of RhB
under different conditions, where C was the concentration of
RhB after visible-light irradiation and C0 was the concentration
of RhB after adsorption for 30 min in the dark. After 30 min, the
Figure 9. Various membranes (a) and Bi2O3/ZnS-CA membranes loading different content (b) for degradation of RhB under visible-light irradiation; the
corresponding kinetic linear simulation curves of various membranes (c); Recyclability of the Bi2O3/ZnS-CA membranes (d).
Phys. Status Solidi A 2018, 1701061
1701061 (7 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
concentration of RhB showed no obviously change. This
demonstrated that the CA membranes had a certain adsorption
capacity but with no photocatalytic activity. Similarly, the Bi2O3-
CA and ZnS-CA membranes also displayed low photocatalytic
activities under the visible-light irradiation. On the contrary, the
degradation rate on the Bi2O3/ZnS-CA membrane was higher
than that of Bi2O3-CA (35.7%) and ZnS-CA membranes (39.4%),
which was attributed to the high photocatalytic efficiency of
Bi2O3/ZnS among them. Degradation experiments of Bi2O3/
ZnS-CA membranes with different concentrations of catalysts
are shown in Figure 9b. This demonstrated that the degradation
efficiency was enhanced as the content of catalyst increased. To
further investigate photocatalytic performance, pseudofirst-
order kinetic fitting is shown in Figure 9c. The estimated rate
after 5 cycles. Therefore, all of the above results indicated that the
fabrication of heterostructures (Bi2O3/ZnS) was an effective way
to enhance the photocatalytic activity of membranes.
3.6. Research on Photocatalytic Mechanism
In order to detect the active oxygen species formed over the
Bi2O3/ZnS heterojunction during degradation process, the
electron spin resonance (ESR) spin-trap technique was used.
As shown in Figure 10, there was no characteristic ESR signal
in the dark. The Bi2O3/ZnS had the characteristic peaks of
DMPO-•OH adduct signals but its intensity was extremely low
(Figure 10a). On the contrary, the DMPO-•O2ꢀ was successfully
detected and the signal strength wasꢀstrong under the light,
constant (k) on the optimal membranes was 0.00439 minꢀ1
,
•
which was approximately 4.14 and 4.03 times that of ZnS-CA
and Bi2O3-CA membranes, respectively. To study the stability of
Bi2O3/ZnS-CA membranes, the used membranes were col-
lected, and the reusability was further examined in five RhB
degradation experiments. As shown in Figure 9d, to a certain
extent, the removal rate of the membranes showed a decay witꢀh
which further confirmed that the O2 was the main reactive
species in the process of photocatalytic degradation (Figure 10b).
ꢀ
Therefore, it was clearly shown that •O2 radicals played an
important role in the degradation process, while a few •OH
radicals hardly participated in the photocatalytic reaction.
To further investigate the concentration of reactive oxygen
radicals, reactive oxygen (ROS) was detected by using fluores-
cence of 2, 7-dichlorofluorescein diacetate (DCFH-DA), which is
•
five cycles. The reason for this may be that the generated O2
attacked the membrane structure under long-time visible-light
irradiation, resulting in the catalyst falling off. However, the
removal rate of the catalytic membrane can still reach about 70%
Figure 10. DMPO spin-trapping ESR spectra for Bi2O3/ZnS: (a) in
Figure 11. DCF fluorescence intensity of catalyst suspension under light
illumination (a), and plot of DCF fluorescence intensity as a function of
the different catalyst under light or in the dark (b).
aqueous dispersion for DMPO-•OH and (b) in methanol dispersion for
DMPO-•O2
.
ꢀ
Phys. Status Solidi A 2018, 1701061
1701061 (8 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 12. The photocatalytic mechanism of RhB degradation over Bi2O3/ZnS-CA membranes.
a nonfluorescent dye. The DCFH-DA could react with free 4. Conclusions
radicals and produce a fluorescent dye dichlorofluorescein
(DCF)[47] and the fluorescence intensity of DCF was correlated
In summary, the effective Bi2O3/ZnS-CA membranes were
successfully fabricated via the phase-inversion technique by
supplementing a Bi2O3/ZnS heterojunction into the PVDF
matrix. As the functional material, the Bi2O3/ZnS heterojunc-
tion ameliorated the visible-light response capability. Moreover,
the Bi2O3/ZnS-CA membranes exhibited significant improve-
ment in porosity, hydrophilicity and permeability, which can
facilitate the interaction between the target pollutant and the
catalyst. Compared with Bi2O3-CA and ZnS-CA membranes, the
Bi2O3/ZnS-CA membranes showed a 35.7 and 39.4% increase in
degradation efficiency and 302 and 314% increase in photo-
degradation kinetics toward RhB, respectively. Owing to the
attractive photocatalytic efficiency and photogenerated charge
separation resulting from the unique heterostructure between
Bi2O3 and ZnS, the Bi2O3/ZnS-CA membranes exhibited high
performance for the removal of RhB. All these advantages shown
by Bi2O3/ZnS-CA membranes demonstrate their great potential
for further organic dye treatment.
with the content of reactive oxygen radicals. As shown in
Figure 11a, the fluorescence intensity of DCF in a blank sample
(it was prepared by adding the DCFH-DA into deionized water
without any photocatalyst) was extremely low. However, the peak
intensity of the Bi2O3/ZnS heterojunction suspension with
DCFH-DA reached a maximum after illumination for 15 min,
demonstrating that the heterojunction enhanced the photo-
catalytic activity. The fluorescence intensity of Bi2O3 was close to
that of the ZnS, which indicated the similar photocatalytic
activity of these two semiconductors. Figure 11b presents the
fluorescence intensity of photocatalyst suspension with DCFH-
DA under dark and visible-light conditions. It is clearly seen that
the fluorescence intensity extremely increased after light
illumination. Therefore, the fBi2O3/ZnS heterojunction can
enhance the production of free radicals and photocatalytic
performance.
On the basis of the above results, the possible photocatalytic
mechanism for RhB degradation is described in detail in
Figure 12. When the Bi2O3/ZnS heterojunction system was
irradiated by visible light, the Bi2O3 with a narrow bandgap
was excited to produce photogenerated electrons (eꢀ) and
photoinduced holes (hþ). Then, the electrons in the conduc-
tion band (CB) of ZnS transferred quickly to the CB of Bi2O3,
whereas the photoinduced holes in the valence band (VB) of
Bi2O3 transferred quickly to VB of ZnS. This is the key to
reduce the recombination rate of the electron–hole pairs. The
active species (•O2ꢀ) and the photoinduced holes reacted
directly with RhB to yield degradation products.[30] However,
the CA membranes played a substrate support and absorption
role with no photocatalytic activity. In summary, this Bi2O3/
ZnS heterojunction structure can effectively separate elec-
Acknowledgements
This work was financially supported by the National Natural Science
Foundation of China (Nos. 21406085 and 21676127), the Ph.D.
Programs Foundation of Ministry of Education of China (No.
20123227120015), the Programs of Senior Talent Foundation of Jiangsu
University (No. 15JDG024), the Natural Science Fund for Colleges and
Universities in Jiangsu Province (No. 15KJB550003, 14KJB150005), the
Post Doctoral Fund of Jiangsu Province (No. 1501104B), the Natural
Science Foundation of Jiangsu Province (BK20150483), and the Special
Financial Grant from the China Postdoctoral Science Foundation
(2014T70488).
tron–hole pairs, which made more •O2 and holes to
ꢀ
Conflict of Interest
participate in oxidation reactions and improve the photo-
catalytic activity.
The authors declare no conflict of interest.
Phys. Status Solidi A 2018, 1701061
1701061 (9 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Keywords
[21] G. Liu, S. Li, Y. Lu, J. Zhang, Z. Feng, C. Li, J. Alloys Compd. 2016, 689,
787.
[22] L. Rimoldi, D. Meroni, G. Cappelletti, S. Ardizzone, Catal. Today
2017, 281, 38.
Bi2O3, heterojunction, photocatalytic membranes, Rhodamine B, ZnS
Received: December 21, 2017
Revised: February 3, 2018
Published online:
[23] H. Y. Jiang, K. Cheng, J. Lin, Phys. Chem. Chem. Phys. 2012, 14,
12114.
[24] S. Jiang, L. Wang, W. Hao, W. Li, H. Xin, W. Wang, T. Wang, J. Phys.
Chem. C 2015, 119, 14094.
[25] Q. He, Y. Ni, S. Ye, RSC Adv. 2015, 7, 27089.
[26] H. Jiang, G. Liu, T. Wang, P. Li, J. Lin, J. Ye, RSC Adv. 2015, 5, 92963.
[27] S. Juntrapirom, D. Tantraviwat, S. Suntalelat, O. Thongsook,
S. Phanichphant, B. Inceesungvorn, J. Colloid Interface Sci. 2017,
504, 711.
[28] C. Liu, Y. Yang, J. Li, S. Chen, W. Li, X. Tang, Chem. Eng. J. 2017, 326,
603.
[29] S. Leong, A. Razmjou, K. Wang, K. Hapgood, X. Zhang, H. Wang, J.
Membr. Sci. 2014, 472, 167.
[30] C. Wang, Y. Wu, J. Lu, J. Zhao, J. Cui, X. Wu, Y. Yan, P. Huo, ACS Appl.
Mater. Interfaces 2017, 9, 23687.
[31] H.Wu,Y.Liu,L.Mao,C.Jiang,J.Ang,X.Lu,J.Membr.Sci.2017,532,20.
[32] V. Ghaffarian, S. M. Mousavi, M. Bahreini, N. S. Parchin, Int. J.
Environ. Sci. Technol. 2017, 14, 1.
[1] M. A. Alvi, A. A. Al-Ghamdi, M. ShaheerAkhtar, Mater. Lett. 2017,
204, 12.
[2] A. Houas, H. Lachheb, M. Ksibi, E. Elaloui, C. Guillard,
J. M. Herrmann, Appl. Catal. B 2001, 31, 145.
[3] O. Mekasuwandumrong, P. Pawinrat, P. Praserthdam, J. Panpranot,
Chem. Eng. J. 2010, 164, 77.
[4] M. M. Nassar, Y. H. Magdy, Chem. Eng. J. 1997, 66, 223.
[5] L. Y. Yang, S. Y. Dong, J. H. Sun, J. L. Feng, Q. H. Wu, S. P. Sun, J.
Hazard. Mater. 2010, 179, 438.
[6] R. Jain, M. Mathur, S. Sikarwar, A. Mittal, J. Environ. Manage. 2007,
85, 956.
[7] J. Rochat, P. Demenge, J. C. Rerat, Toxicological European Research
ꢀ
Recherche Europeenne En Toxicologie. 1978, 1, 23.
[8] J. Sharma, Sukriti, P. Anand, V. Pruthi, A. S. Chaddha, J. Bhatia,
B. S. Kaith, Mater. Chem. Phys. 2017, 196, 270.
[33] X. Sun, C. Lu, W. Zhang, D. Tian, X. Zhang, Carbohydr. Polym. 2013,
98, 405.
[9] B. Lin, G. Yang, B. Yang, Y. Zhao, Appl. Catal. B 2016, 198, 276.
[10] R. Wang, L. Cao, J. Alloys Compd. 2017, 722, 445.
[11] M. Zambianchi, M. Durso, A. Liscio, E. Treossi, C. Bettini,
M. L. Capobianco, A. Aluigi, A. Kovtun, G. Ruani, F. Corticelli,
Chem. Eng. J. 2017, 326, 130.
[34] F. E. Ahmed, B. S. Lalia, R. Hashaikeh, Desalination 2015, 356, 15.
[35] N. M. Mokhtar, W. J. Lau, A. F. Ismail, B. C. Ng, RSC Adv. 2014, 4,
63367.
[36] L. Shan, G. Wang, L. Liu, Z. Wu, J. Mol. Catal. A: Chem. 2015, 406,
145.
ꢀ
[12] S. Mozia, D. Darowna, A. Orecki, R. Wrobel, K. Wilpiszewska,
[37] C.-J. Chang, K.-W. Chu, Int. J. Hydrogen Energy 2016, 41, 21764.
[38] W. Sun, H. Zhang, J. Lin, J. Phys. Chem. C 2014, 118, 17626.
[39] Y. Cong, Y. Ji, Y. Ge, H. Jin, Y. Zhang, Q. Wang, Chem. Eng. J. 2017,
307, 572.
A. W. Morawski, J. Membr. Sci. 2014, 470, 356.
[13] X. Yan, K. Ye, T. Zhang, C. Xue, D. Zhang, C. Ma, J. Wei, G. Yang, New
J. Chem. 2017, 41, 8482.
[14] H. Lu, B. Dong, S. Wang, L. Zhao, Z. Xun, L. Wan, J. Li, Phys. Status
Solidi A 2012, 209, 2157.
[40] L. Q. Shao, D. L. Jiang, P. Xiao, L. M. Zhu, S. C. Meng, M. Chen, Appl.
Catal., B 2016, 198, 200.
[15] J. Ke, J. Liu, H. Sun, H. Zhang, X. Duan, P. Liang, X. Li, M. O. Tade,
S. Liu, S. Wang, Appl. Catal. B 2017, 200, 47.
[16] B. Lin, H. An, X. Yan, T. Zhang, J. Wei, G. Yang, Appl. Catal. B 2017,
210, 173.
[17] B. Lin, H. Li, H. An, W. Hao, J. Wei, Y. Dai, C. Ma, G. Yang, Appl.
Catal. B 2018, 220, 542.
[18] C. Xue, X. Yan, H. An, H. Li, J. Wei, G. Yang, Appl. Catal. B 2018, 222,
157.
[19] C. Xue, H. An, X. Yan, J. Li, B. Yang, J. Wei, G. Yang, Nano Energy
2017, 39, 513.
[41] M. Ge, C. Cao, S. Li, S. Zhang, S. Deng, J. Huang, Q. Li, K. Zhang,
S. S. Al-Deyab, Y. Lai, Nanoscale 2015, 7, 11552.
[42] X. Wang, P. Ren, H. Fan, Mater. Res. Bull. 2015, 64, 82.
[43] H. Zhao, S. Chen, X. Quan, H. Yu, H. Zhao, Appl. Catal., B 2016, 194,
134.
[44] Z. Xu, T. Wu, J. Shi, K. Teng, W. Wang, M. Ma, J. Li, X. Qian, C. Li,
J. Fan, J. Membr. Sci. 2016, 520, 281.
[45] H. Rabiee, V. Vatanpour, J. Membr. Sci. 2014, 472, 185.
[46] H. Zhao, L. Wu, Z. Zhou, L. Zhang, H. Chen, Phys. Chem. Chem.
Phys. 2013, 15, 9084.
[20] L. Zhang, W. Wang, J. Yang, Z. Chen, W. Zhang, L. Zhou, S. Liu, Appl.
Catal. A 2006, 308, 105.
[47] G. Yang, L. Liu, J. Wang, C. Bortolini, M. Dong, J. Colloid Interface Sci.
2017, 495, 37.
Phys. Status Solidi A 2018, 1701061
1701061 (10 of 10)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim