Direct Z-scheme Cs2O-Bi2O3-ZnO heterostructures for photocatalytic overall water splitting

Article abstract of DOI:10.1039/c8ta08033j

In this work, a direct Z-scheme Cs₂O-Bi₂O₃-ZnO heterostructure without any electron mediator is fabricated by a simple solution combustion route. Cs₂O is chosen as a sensitizer to expand the light absorption range, and in addition, its conduction band minimum (CBM) and valence band maximum (VBM) positions are suitable to construct a direct Z-scheme system with ZnO and Bi₂O₃. Structural and elemental analyses show clear evidence for heterostructure formation. The Z-scheme charge carrier migration pathway in Cs₂O-Bi₂O₃-ZnO is confirmed by high resolution XPS and ESR studies. The fabricated heterostructure exhibits a good ability to split water to H₂ and O₂ under simulated sunlight irradiation without any sacrificial agents or co-catalysts and has excellent photostability. The apparent quantum efficiency of the optimized Cs₂O-Bi₂O₃-ZnO heterostructure reaches up to 0.92% at 420 nm. The excellent efficiency of this fabricated heterostructure is attributed to the efficient charge carrier separation, the high redox potential of the CBM and VBM benefiting from a direct Z-scheme charge carrier migration pathway and the extended light absorption range.

Full text of DOI:10.1039/c8ta08033j

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Direct Z-scheme Cs O–Bi O –ZnO
2
2 3
heterostructures for photocatalytic overall water
splitting†
Cite this: J. Mater. Chem. A, 2018, 6,
1379
2
a
a
b
c
Abdo Hezam,
K. Namratha, Q. A. Drmosh, Deepalekshmi Ponnamma,
d
e
f
and K. Byrappa *ag
Adel Morshed Nagi Saeed, V. Ganesh, B. Neppolian
In this work, a direct Z-scheme Cs
fabricated by a simple solution combustion route. Cs
absorption range, and in addition, its conduction band minimum (CBM) and valence band maximum
VBM) positions are suitable to construct a direct Z-scheme system with ZnO and Bi . Structural and
elemental analyses show clear evidence for heterostructure formation. The Z-scheme charge carrier
migration pathway in Cs O–Bi –ZnO is confirmed by high resolution XPS and ESR studies. The
fabricated heterostructure exhibits a good ability to split water to H and O under simulated sunlight
irradiation without any sacrificial agents or co-catalysts and has excellent photostability. The apparent
quantum efficiency of the optimized Cs O–Bi –ZnO heterostructure reaches up to 0.92% at 420 nm.
2
O–Bi
2
O
3
–ZnO heterostructure without any electron mediator is
2
O is chosen as a sensitizer to expand the light
(
2 3
O
2
2 3
O
2
2
Received 18th August 2018
Accepted 5th October 2018
2
2 3
O
The excellent efficiency of this fabricated heterostructure is attributed to the efficient charge carrier
separation, the high redox potential of the CBM and VBM benefiting from a direct Z-scheme charge
carrier migration pathway and the extended light absorption range.
DOI: 10.1039/c8ta08033j
rsc.li/materials-a
photocatalysts usually have several issues including (i) rapid
Introduction
electron–hole (e/h) recombination, (ii) a wide bandgap, and (iii)
7,8
Replacing polluting energy sources (e.g. fossil fuels) with clean inappropriate conduction band or valence band positions.
and green renewable ones is essential for developing a sustain-
able ecology. Hydrogen (H ) is an ideal clean sustainable energy form a heterostructure can enhance charge separation and
Coupling a semiconductor with another semiconductor to
2
carrier because of its zero pollution and high combustion improve the H
2
production. If one of the coupled semi-
1,2
enthalpy features. Photocatalytic water splitting is a green conductors has a small band gap, the heterostructure produces
3
5
technique to generate H
2
from solar energy. However, photo-
2
H under sunlight/visible light irradiation. However, most re-
catalytic H
2
production is still in the laboratory research stage ported heterostructures require the addition of sacricial
because there are many technical or economic issues. Metal agents which react with holes leading to enhanced charge
4
oxide semiconductors are widely used to photocatalytically split separation and thereby improved H production efficiency. The
2
5,6
water into
H
2
and
O
2
.
However, pure metal oxide sacricial reaction is a downhill process which causes reduc-
tions in the Gibbs free energy, in such a way that the photon
9
a
energy is lost. Moreover, adding sacricial agents is not
Center for Materials Science and Technology, University of Mysore, Vijana Bhavana,
Manasagangothiri, Mysuru-570006, India. E-mail: kbyrappa@gmail.com
economical and they may not be environmentally friendly.
As a response to these issues, major research interest has
been focused on photocatalytic overall water splitting in which
the H and O are produced simultaneously using pure water
without any sacricial agents.
water splitting, the conduction band minimum (CBM) of the
b
Physics Department and Center of Research Excellence in Nanotechnology, King Fahd
University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
c
Center for Advanced Materials, Qatar University, P O Box 2713, Doha, Qatar
2
2
d
Department of Polymer Science and Technology, Sri Jayachamarajendra College of
9–11
For photocatalytic overall
Engineering, JSS Science & Technology University, Mysuru – 570006, India
e
Department of Physics & Nanotechnology, SRM Institute of Science and Technology,
+
photocatalyst should be more negative than that of H
(0–0.059 pH, V versus NHE) and the VB maximum (VBM) should
be more positive than that of O /H O (1.23–0.059 pH, V versus
2
/H
Chennai-603203, India
f
Energy and Environmental Remediation Lab, SRM-Research Institute, SRM Institute of
Science and Technology, Chennai-603203, India
g
2
2
Adichunchanagiri University, N.H.75, B. G. Nagara, Mandya District – 571448, India NHE).2,12 The heterostructure approach provides an excellent
†
Electronic supplementary information (ESI) available: PXRD spectra,
platform to optimize the CBM and VBM of the photocatalyst
and make it appropriate for overall water splitting. Both the
CBM and VBM can be easily adjusted by varying the compo-
electrochemical impedance measurements, photostability of the prepared
2
samples, dependence of the rate of H evolved on the wavelength of incident
light and summary of recent Z-scheme photocatalysts for overall water splitting.
See DOI: 10.1039/c8ta08033j
13
nents of the fabricated heterostructure.
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 21379–21388 | 21379

View Article Online
Journal of Materials Chemistry A
Paper
The charge migration pathway of the heterostructures
follows either a heterojunction approach or a Z-scheme
approach (ESI Scheme S1†). In the heterojunction (type II)
Experimental section
Materials
approach, the electrons migrate from the CB of the semi- Bi(NO ) $5H O (Sigma Aldrich, 98%), Zn(NO ) $6H O (Alfa
3
3
2
3 2
2
conductor with a higher CBM to the semiconductor with a lower Aesar, 99%), CsNO (Alfa Aesar, 99.9%), and glycine (Sigma-
3
CBM while holes migrate to the semiconductor with a higher Aldrich, 99%) were of analytical grade and used as received.
VBM. Accordingly, based on the heterojunction approach, the Ultrapure deionized water (18.2 MU cm) was used in the
coupling of semiconductors leads to a great reduction in redox experiments.
14,15
ability which reduces the H
2
production efficiency.
Hence,
many heterojunctions are not suitable for overall water splitting
and designing an appropriate heterojunction is rather neces-
Synthesis
3 3 2
About 70 ml 1.18 M Bi(NO ) $5H O was prepared using 1.2 M
sary for effective H production.
2
HNO
Zn(NO
3
as a solvent and was labeled BS. 70 ml 0.82 M
$6H O was prepared and labeled ZS. Equal volumes of
Z-Scheme heterostructures are excellent candidates for
overall water splitting as the electrons are accumulated in the
CB of the semiconductor with the highest CBM and holes are
accumulated in the VB of the semiconductor with the lowest
VBM which greatly enhances the redox ability and the charge
carrier separation efficiency of the heterostructure and hence
3
)
2
2
BS and ZS were mixed and added to 5 separate beakers.
Different amounts of CsNO were added to each beaker in such
a way as to obtain 0, 0.5, 1, 1.5, and 2 mol% Cs O and the
corresponding oxidizer samples were labeled CBZ0, CBZ0.5,
CBZ1, CBZ1.5, and CBZ2, respectively. The Bi to Zn ratio was
1 : 0.7 in all samples. The appropriate amount of glycine as
a fuel was added to each beaker in such a way that the fuel/
oxidizer valence ratio of the mixture was equal to one. The
mixture was magnetically stirred at 100 C until a sticky gel was
obtained. Then the stirring was stopped and the temperature
was raised to 320 C. Within a short time (<4 min) the gel started
foaming and then igniting with the production of a large
amount of gas. The color of the obtained powder was light
orange. Pristine Cs
same procedure. Cs
sphere at 180 C for 1 h to remove Cs
might have formed because of the reaction of Cs
3
2
improves its photocatalytic activity for splitting water into H
2
14,16–20
and O2.
Direct Z-scheme photocatalysts (the third gener-
ation of Z-scheme photocatalytic systems) need neither liquid
electron mediators which limit their usage nor noble metal
ꢀ
1
5,21
electron mediators which are expensive.
Recently direct Z-
22
scheme systems including Co
3
(PO
4
)
2
/a-Fe
2
O
3
,
black phos-
ꢀ
23
phorus/BiVO
overall water splitting.
A Bi O –ZnO heterostructure has been fabricated by several
4
,
and Bi
2
WO
6
–Cu
3
P were fabricated and used for
24
2
3
25–27
2
O, Bi
O powder was calcined in an argon atmo-
CO impurities which
2
O with CO .
2 3
O , and ZnO were prepared using the
research groups
and it has been reported that the Bi O –ZnO
2 3
2
exhibits excellent photocatalytic activity towards organic
ꢀ
28
29
2
3
pollutant degradation under UV, sunlight, and visible light
30
2
irradiation. The enhanced photocatalytic activity is attributed to
the improved charge carrier separation caused by electron and
Aer the calcination, Cs O was stored in a sealed quartz
ampoule lled with argon.
2
28,29
hole transfer between Bi
O
2 3
and ZnO.
To further enhance the
charge carrier separation and thereby the photocatalytic activity,
3
1
Bi O –ZnO was doped with Ag by Balachandran et al. and was Characterization
2
3
32
recently coupled with CeO by our research group. However, all
2
Scanning electron microscopy (SEM, Hitachi-S3400N) and high
resolution transmission electron microscopy (HR-TEM, Joel/
JEM 2100 model) were used to study the morphology and
structure of the sample. The crystal structures were investigated
using a powder X-ray diffractometer (PXRD, Rigaku smart Lab-
2 3 2 3
these reported Bi O –ZnO and Bi O –ZnO based heterostructures
are heterojunctions, in which the electrons migrate to the CB with
the lower redox potential, and are inappropriate for reducing
2
water and generating H . Hence, construction of a direct Z-
scheme Bi –ZnO- based system is much needed to overcome
2 3
O
˚
II, Cu Ka radiation, l ¼ 1.5414 A). The elemental analysis (EDX)
this limitation and enable it to produce H by water splitting. This
2
was carried out by a Thermo Scientic energy dispersive X-ray
task can be fullled by coupling Bi O –ZnO with a semiconductor
2
3
spectrometer attached to the SEM. X-ray photoelectron spectra
with an appropriate CBM and VBM.
(XPS) of the samples were obtained using an ESCALAB 250Xi,
Herein, the prior calculations of the CBM and VBM positions
of pristine Bi , ZnO, and Cs O indicate that Cs O can
construct a direct Z-scheme with both Bi and ZnO. This
motivated the authors to fabricate Cs O–Bi –ZnO hetero-
structures. Furthermore, Cs O acts as a sensitizer to increase
the visible light absorption. The small difference between the
Thermo Scientic. The diffuse reectance spectra (UV-Vis DRS)
of the samples were recorded in the solid form using a Thermo
Scientic evolution 201 spectrophotometer.
2
O
3
2
2
2 3
O
2
2 3
O
2
Water splitting experiments
positions of the ZnO VBM and the Bi O VBM prevents The photocatalytic activity towards the overall water splitting of
2
3
a remarkable reduction in the redox potential. Therefore, the pure water was examined under simulated solar irradiation
ꢁ
2
charge separation efficiency is signicantly enhanced and the (Xenon lamp, AM 1.5G, 100 mW cm ). The reaction took place
generated electrons and holes accumulate at energy bands with in a 135 ml quartz reactor. 5 mg of the sample was dispersed in
a high redox potential. Thus the fabricated Cs
shows enhanced photocatalytic activity and robust durability suspension was stirred in the dark for 30 min under nitrogen
towards overall water splitting. purging to remove the dissolved O . The H and O amounts
2 2 3
O–Bi O –ZnO 50 ml ultrapure deionized water. Prior to the reaction, the
2
2
2
21380 | J. Mater. Chem. A, 2018, 6, 21379–21388
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
were measured in regular time intervals using an off-line gas pyrroline N-oxide (DMPO) as the free radical trapping agent in
chromatograph (Shimadzu GC-2014) with a TCD detector and aqueous and methanol solution, respectively. 10 mg catalyst
nitrogen as carrier gas.
was dispersed in 0.5 ml water/methanol and 40 ml DMPO by
stirring for 10 min at room temperature. The signals were
recorded on a JES FA200 spectrometer before illumination (in
the dark) and aer 4 and 8 min of simulated sunlight
illumination.
Measurement of apparent quantum efficiency (AQE)
The apparent quantum efficiency (AQE) was measured using the
same experimental set-up and under the same conditions as the
overall water splitting experiments except for the use of band-
pass lters with different wavelengths (365, 420, 475, 525, 575, Results and discussion
and 625 nm). The power of incident light was measured using
a spectroradiometer (AvaSpec-2048-USB2). The number of
Morphological and structural studies
The morphology and microstructure of the prepared samples
were investigated by means of SEM and TEM techniques. Fig. 1a
and b show the SEM images of the as-synthesized Cs O–Bi
ZnO (sample CBZ1.5) heterostructure. As illustrated by the SEM
incident photons (Nphotons) was calculated from eqn (1).
N
photons ¼ Elt/hc
(1)
2
2 3
O –
where E is power of the incident light, l is light wavelength, t is images, a sponge-like structure with round macropores is
irradiation time, c is light speed, and h is Plank’s constant. observed (Fig. 1a and b). This macroporous nature is due to the
AQE% was calculated from eqn (2)
fast gas evolution taking place during the highly exothermic
combustion. In fact, the observed structure is characteristic of
metal oxides prepared by the solution combustion route using
AQE% ¼ (number of reacted electrons/Nphotons) ꢂ 100
33
¼
(2 ꢂ number of evolved H molecules/N
) ꢂ 100 (2) glycine as a fuel. Obviously, it is difficult to identify the exact
size and morphology of the individual nanoparticles from the
SEM images and therefore TEM (Fig. 1c) and HRTEM (Fig. 1d)
were used. While the TEM image shows agglomeration of the
CBZ nanoparticles of <10 nm size, the HRTEM image shows
2
photons
Electrochemical measurements
The photocurrent densities of the prepared samples were
measured using a CHI 608e electrochemical analyzer. Fluorine-
doped tin oxide (FTO) glass spin coated with the studied sample
was used as the working electrode. An Ag/AgCl electrode and
platinum wire were respectively used as reference and counter
electrodes. The three electrodes were immersed in 0.5 M Na SO
lattice fringes. The observed fringes with interlayer spacing of
0
.34, 0.32, and 0.52 in the HRTEM respectively correspond to
34
2
the (012) plane of Cs O (JCPDS card no. 09-0104), the (121)
35
plane of Bi
ZnO (JCPDS card no. 36-1451). This conrms the formation
2 3
of the heterostructured Cs O–Bi O –ZnO, and these results are
2 3
O (JCPDS card no. 71-2274), and the (002) plane of
36–38
2
4
2
electrolyte during the photocurrent measurements. The electro-
lyte was purged with argon for 30 min before starting the
measurements. The working electrode was irradiated at 100 mW
in good agreement with the PXRD results (ESI Fig. S1†).
ꢁ2
EDX analysis
cm using a 250 W xenon lamp and the photocurrent was
measured at 0 V during light off–on cycling. The working elec- To further conrm the formation of the Bi
trodes were prepared as follows. 5 mg of the sample was dispersed Bi –ZnO heterostructures and their purity, EDX analysis was
in 2 ml isopropanol. 4 ml Naon solution (Sigma Aldrich, 5 wt%) employed. Fig. S2† shows the spectra of Bi –ZnO and the
was added and the solution was sonicated for 20 minutes. Then optimized Cs O–Bi –ZnO. Only Bi, Zn and O are present in
the resultant solution was spin coated on a cleaned FTO glass the spectrum of Bi –ZnO and Bi, Zn, Cs, and O in the spec-
–ZnO conrming the purity of the prepared
2 3 2
O –ZnO and Cs O–
2 3
O
2 3
O
2
2 3
O
O
2 3
(
1 cm ꢂ 2 cm). Aerwards, the electrodes were dried in air fol- trum of Cs
O–Bi O
2 2 3
ꢀ
lowed by sintering at 400 C for 40 min to enhance adhesion.
heterostructures. The results of the elemental analysis were
Electrochemical impedance was measured using the same summarized in Table S1.†
5
set up (open circuit, frequency: 1 to 10 Hz, AC voltage: 5 mV
amplitude). For Mott–Schottky measurements, the working XPS analysis
electrodes were prepared following the same procedure as
mentioned above. The counter and reference electrodes were
To investigate the surface composition of the Bi O –ZnO and
2
3
Cs O–Bi O –ZnO heterostructures, XPS studies were carried
2
2 3
respectively a platinum wire and an Ag/AgCl electrode. The
measurements were carried out at room temperature in the
dark at 100 Hz with an amplitude of 5 mV. The obtained
potentials were converted to normal hydrogen electrode (NHE)
scale.
out. As observed in Fig. 2, the fully scanned spectra indicate the
presence of elements Cs, Bi, Zn, O, and C in the sample CBZ1.5
and only Bi, Zn, O, and C in the sample CBZ0 conrming the
successful formation of Cs O–Bi O –ZnO and Bi O –ZnO het-
2 2 3 2 3
erostructures. The C 1s peak is from the carbon tape used in the
XPS sample holder.
Mechanism evidence
To investigate the charge carrier migration, high resolution
To conrm the Z-scheme mechanism, spin-trapping electron XPS analysis was employed for Zn 2p, Bi 4f and Cs 3d. The Zn
spin resonance (ESR) was used which detected the hydroxyl 2p3/2 and Zn 2p1/2 of sample CBZ0 are located respectively at
ꢁ
(cOH) and superoxide (cO
2
) radicals using 5,5-dimethyl 1023 eV and 1047.02 eV (Fig. 3a), with both showing positive
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 21379–21388 | 21381

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 1 (a, b) SEM images of the optimized Cs
2 2
O–Bi O
3
–ZnO heterostructure (sample CBZ1.5) at two different points; (c) TEM, and (d) HRTEM
images of sample CBZ1.5.
4f5/2 peak positions of sample CBZ1.5 are respectively at 159.2
and 164.46 eV and positive shis occur in their positions.
Moreover, Fig. 3c shows the Cs 3d5/2 and Cs 3d3/2 positions at
724.2 eV and 738.6 eV respectively, with a negative shi
observed in the positions of Cs 3d and Cs 3d3/2 compared to
5
/2
43
the reported ones.
The negative shi in the positions of Bi 4f7/2 and Bi 4f5/2 in
the sample CBZ0 conrm the electron migration from ZnO to
Bi
Bi
2
2
O
O
3
indicating the formation of a junction between ZnO and
(see Scheme 1). However, aer coupling Bi –ZnO with
3
2 3
O
Cs
electrons of both ZnO and Bi
by the negative shi in the positions of the Cs 3d peaks and the
2
O, the charge carrier migration pathway changes and the
2
O
3
transfer to Cs O as conrmed
2
15
positive shi in the positions of the Bi 4f and Zn 2p peaks. As
the position of the Cs O CB is higher than that of the ZnO and
Fig. 2 XPS survey spectra of the optimized Cs
ostructure (sample CBZ1.5) and the Bi
sample CBZ0).
2
O–Bi
2
O
3
–ZnO heter-
O
2 3
–ZnO heterostructure
2
(
Bi
2
O
3
, based on the heterojunction approach, the electrons are
with a lower CBM.
supposed to transfer to the CB of Bi
2 3
O
Hence the electron transfer pathway cannot be explained by the
heterojunction mechanism, but by the Z-scheme mechanism
(Scheme 1).
High resolution O 1s spectra of the samples CBZ0 and
CBZ1.5 are shown in Fig. 3d and e which were tted into three
shis compared to the standard positions of Zn 2p and Zn
3
/2
43
40,41
2
p
1/2 (ref. 39) and the reported results of our group.
Aer coupling the Bi –ZnO with Cs O, both the Zn 2p3/2
and Zn 2p1/2 peaks show further positive shis compared to
those of Bi –ZnO. These positive shis can be attributed to
the transfer of electrons from ZnO to Bi in sample CBZ0
Bi O –ZnO), while in the case of sample CBZ1.5, the positive
2
O
3
2
Gaussian peaks O , O , and O located at 530.1, 531.7, and
2
O
3
I
II
III
532.3 eV, respectively for the sample CBZ0 and at 530.1, 531.4,
2 3
O
and 533 eV for the sample CBZ1.5. The component O is mainly
I
(
2
3
associated with the lattice oxygen in the heterostructure. The
shis indicate the electron transfer from ZnO to either Bi O or
2
3
ꢁ
components OII and OIII are respectively ascribed to OH and
Cs O.
2
44–46
2
adsorbed H O.
The exact migration pathway of ZnO electrons can be
conrmed from the high resolution XPS analysis of both Bi 4f
and Cs 3d. Fig. 3b shows that the Bi 4f7/2 and Bi 4f5/2 of sample Overall water splitting
CBZ0 are located at 158.12 and 163.36 eV, respectively, and
The photocatalytic activity of the Cs O–Bi O –ZnO hetero-
structures towards water splitting was tested under simulated
solar irradiation in pure water without using any sacricial
2
2 3
there are negative shis in their locations compared to the re-
42
ported pure Bi O . Fig. 3b also shows that the Bi 4f7/2 and Bi
2
3
21382 | J. Mater. Chem. A, 2018, 6, 21379–21388
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 3 High resolution XPS spectra of (a) Zn 2p and (b) Bi 4f of sample CBZ0 and sample CBZ 1.5, (c) Cs 3d of sample CBZ1.5, (d) O 1s of sample
CBZ0, and (e) O 1s of sample CBZ 1.5.
2 3 2 2 3
Scheme 1 Comparison of the charge carrier migration pathway in the Bi O –ZnO and Cs O–Bi O –ZnO heterostructures based on XPS
analysis.
agents. Fig. 4a shows the H
time for all prepared samples. Notably, H
2
and O
2
evolution as a function of aer 3 cycles (Fig. 4b). The stability of the heterostructure was
and O gases were further conrmed by measuring the PXRD spectrum of the
2
2
produced simultaneously in a stoichiometric ratio of 2 : 1 for sample CBZ1.5 before and aer the photocatalytic activity test
the Cs O–Bi O –ZnO samples (CBZ0.5, CBZ1, CBZ1.5, and (ESI Fig. S4†). It was found that the crystal structure of the used
2
2 3
CBZ2). Moreover, no water splitting takes place in the absence CBZ1.5 did not change during the photocatalytic reaction,
of Cs O (sample CBZ0) (see ESI Table S2†). The H and O verifying the stability of the photocatalyst. Therefore, these
evolution rates increase with increasing amount of Cs O to results reveal that Cs O–Bi –ZnO is a promising long-term
) when the stable and efficient photocatalyst for overall water splitting.
O is 1.5 mol% (see Fig. 4a), while no H was The AQE of H production through overall water splitting by
detected in the absence of Cs O. Further increase of the the optimized Cs O–Bi –ZnO (sample CBZ1.5) measured at
concentration of Cs O beyond 1.5 mol% causes the photo- 365, 420, 475, 525, 575, and 625 nm shows the values 1.68%,
catalytic activity to decrease. Meanwhile, the Cs O–Bi O –ZnO 0.92%, 0.43%, 0.12%, 0.027%, and 0.008%, respectively. It is
2
2
2
2
2
2 3
O
ꢁ1
ꢁ1
reach the highest rates (149.5 and 73.2 mmol g
concentration of Cs
h
2
2
2
2
2
2 3
O
2
2
2 3
sustains high O and H evolution rates with excellent stability noted that the Cs O–Bi O –ZnO exhibits a lower AQE at 420 nm
2
2
2
2 3
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 21379–21388 | 21383

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 4 (a) H
2 2 2 2
and O evolution from pure water per hour of simulated sunlight irradiation (AM 1.5 G) of the various samples, (b) typical H and O
evolution from pure water of the sample CBZ1.5 under simulated sunlight irradiation (AM 1.5 G), (c) photocurrent response of the various
ꢁ
2
2 4
samples; the working electrode was irradiated at 100 mW cm using a 250 W xenon lamp in aqueous 0.5 M Na SO , and (d) UV-Vis DRS spectra.
compared to those of Z-scheme systems with co-catalysts and/or This is attributed to the increase in charge transfer resistance
51,52
electron mediators (ESI Table S3† entries 1, 2, 6, and 8). But the caused by excess Cs
2
O.
The drastic increase in the impedance
Cs O–Bi O –ZnO shows a comparable AQE or activity to those of radius of the sample CBZ0 reects the low charge transfer effi-
2
2 3
direct Z-scheme systems (ESI Table S3† entries 9–12).
ciency of the Bi O –ZnO heterostructure when compared with the
2 3
Cs O–Bi O –ZnO heterostructures. The above results also support
2
2 3
the impact of the Cs
Bi O –ZnO heterostructure.
2 3
2 2
O in the photocatalytic activity of the Cs O–
Photocurrent and electrochemical impedance studies
Charge carrier separation is a main factor affecting the photo-
47
catalytic activity of a photocatalyst. To investigate the charge
UV-Vis DRS analysis
carrier separation efficiency, the photocurrent densities of the
various samples under simulated solar light were measured and To gain more insight into the enhanced photocatalytic activity
are presented in Fig. 4c. Compared to Bi –ZnO, the Cs O– of the Cs O–Bi –ZnO, the optical absorption spectra of the
Bi –ZnO heterostructures (samples CBZ0.5, CBZ1, CBZ1.5, prepared samples were obtained. As illustrated in Fig. 4d, the
CBZ2) exhibit a remarkable enhancement in photocurrent absorption band edge of the Cs O–Bi –ZnO heterostructure
2
O
3
2
2
2 3
O
2 3
O
2
2 3
O
density, reecting the role of Cs O in improving the charge clearly shis to a longer wavelength compared to that of the
2
separation. The highest photocurrent density is obtained for the Bi O –ZnO heterostructure, reecting the impact of the Cs O in
2
3
2
sample CBZ1.5 and conrms that it has the highest charge expanding the absorption of light to the visible region. It is also
48
carrier separation. The lowest e/h recombination rate of the noted that the sample with 1.5 mol% of Cs
2
O (CBZ1.5) exhibits
sample CBZ1.5 is a key reason behind it having the highest the highest light absorption in both the UV and visible light
4
9
photocatalytic activity.
regions. Thus the optimal concentration of Cs
2
O is identied as
O beyond
Nyquist impedance plots of the different samples are depicted 1.5 mol%. It can be seen that a further increase in Cs
2
in ESI Fig. S5.† As shown in the plots, the impedance arc radius 1.5 mol% leads to a decrease in the photocurrent density, light
decreases with increasing concentration of Cs O and CBZ1.5 absorption and photocatalytic overall water splitting efficiency.
2
exhibits the smallest impedance arc radius revealing that it has The reason for this observation is attributed to the “shielding
50
the highest efficiency of photogenerated charge carrier transfer.
effect” in which the excess Cs
2
O partially shields the light
However, beyond 1.5 mol%, the impedance arc radius increases. absorption of Bi
2
O
3
and ZnO. In addition, the excess Cs O
2
21384 | J. Mater. Chem. A, 2018, 6, 21379–21388
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
obstructs the electron and hole injection for water splitting, values are 3.89, 5.79, and 5.99 eV for Cs
2
O, ZnO, and Bi
In addition, the VBM was measured experi-
mentally using valence band XPS (VB-XPS) analysis (Fig. 5d–f)
and found to be 0.41, 2.82, and 2.89 eV for Cs O, ZnO, and
Bi , respectively. These values were close to those theoreti-
cally estimated (ESI Table S4†).
2 3
O ,
51–54
32,56
thus increasing the charge recombination rate.
respectively.
Band position determination
2
2 3
O
Estimation of the CBM and VBM is essential to evaluate the
potential of the photocatalysts for water reduction and oxida-
The CBM was further measured using Mott–Schottky
measurements. Fig. S6† shows the Mott–Schottky plots of
tion. To measure the band gaps of pristine Cs
Bi , their UV-Vis DRS spectra were recorded as displayed in
Fig. 5a. Fig. 5b and c show Tauc plots of Cs O, ZnO and Bi
derived from the UV-DRS spectra according to eqn (3).
2
O, ZnO, and
2 3
O
pristine ZnO, Cs
2 2 3
O, and Bi O . All of the slopes were positive
2
2 3
O
indicating that ZnO, Cs O, and Bi O are n-type semi-
55
2
2 3
conductors. These ndings are in agreement with other reports
5
7–63
1
/m
in the literature.
conductors is strongly related to the CBM.
where the ‘m’ values of 2 and 0.5 are assigned, respectively, to band potentials of ZnO, Cs O, and Bi
indirect band (Bi ) and direct band (Cs
conductors. Also, h represents Plank’s constant, v is light tively ꢁ0.23 and 2.87 eV for ZnO, ꢁ1.52 and 0.44 eV for Cs
frequency, E
denotes the band gap, and K is the constant of and 0.18 and 2.92 eV for Bi , and are close to those obtained
The at-band potential for n-type semi-
(
hna) ¼ K(hn ꢁ Eg)
(3)
64–66
Here, the at-
2
2
O
3
are respectively
O
3
2
O and ZnO) semi- ꢁ0.23, ꢁ1.52, and 0.18 V. Thus the CBM and VBM are respec-
2
2
O,
2 3
O
g
proportionality. The band gaps estimated as depicted in Fig. 5b from the Tauc plot and VB-XPS analyses (see Table S4–S6†).
and c were 1.96, 3.1, and 2.74 eV for Cs O, ZnO, and Bi O ,
2
2 3
respectively. The VBM and CBM of three semiconductors were,
respectively, calculated using eqn (4) and (5) and are shown in
Table S4.†
Mechanism of overall water splitting
The photocurrent responses as well as the electrochemical
impedance measurements conrm the enhanced charge sepa-
ration and charge transport in the Cs O–Bi O –ZnO hetero-
structure, however, these results are not enough to prove the
Z-scheme mechanism indicated by the XPS studies. To further
2
2 3
VBM ¼ c ꢁ Eef + 0.5Eg
CBM ¼ E ꢁ VBM
(4)
g
(5)
verify the charge carrier migration in the Cs
2 2 3
O–Bi O –ZnO
where c denotes the electronegativity of the semiconductor and heterostructure, spin-trapping ESR analysis was performed in
55
ꢁ
2
E
ef is the energy of free electrons on the hydrogen scale.
c
which the cOH and cO
were detected using DMPO as the free
Fig. 5 (a) UV-Vis DRS spectra of pristine Cs
of pristine (d) ZnO, (e) Cs O, and (f) Bi
2 2 3 2 2 3
O, Bi O , and ZnO. Tauc plot of (b) pristine Cs O and ZnO, and (c) pristine Bi O , and VB-XPS spectra
2
2 3
O .
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 21379–21388 | 21385

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 6 ESR spectra of sample CBZ1.5 with DMPO obtained under both dark and simulated sunlight irradiation (a) in aqueous suspension to detect
ꢁ
DMPO–cOH radical species, and (b) in methanol to detect DMPO–cO
2
radical species.
radical trapping agent in aqueous and methanol solution, ZnO) will absorb photons generating e/h pairs (Scheme 2). The
67–69
respectively.
sunlight irradiation and in the dark are shown in Fig. 6. No then combine with the photogenerated holes in the VB of
peaks are observed in the dark while the characteristic peaks of Cs O. As Cs O has a small band gap, the electrons in the VB of
2 3
The ESR signals of CBZ1.5 under simulated electrons migrate from the CB of Bi O and the CB of ZnO and
2
2
DMPO–cOH appear under simulated sunlight irradiation with Cs O can easily be excited by simulated sunlight irradiation
2
the intensity increasing with irradiation time (Fig. 6a). The and accumulate in its CB.
ꢁ
characteristic signals of cO2 are also detected in a DMPO
The difference in the positions of the Bi O VB and the ZnO
2 3
system under simulated sunlight irradiation (Fig. 6b). Thus, VB results in the migration of holes to the ZnO VB. In this way,
ꢁ
cOH and cO
2
are produced under simulated sunlight con- the photogenerated electrons and holes are separated effi-
O has a potential greater than
O, and cO /H O and the ZnO VB is
Based on the heterojunction approach, the electrons more positive than the redox potential of O /H O and cOH/OH
should accumulate in the CB of Bi O which has the lowest or cOH/H O. Hence, the overall water splitting and spin-

rming that the charge migration pathway does not follow the ciently. Moreover, the CB of Cs
2
heterojunction approach but follows the Z-scheme approach. the redox potential of H /H
2
2
2
2
2
2
2
3
2
CBM among the three metal oxides Cs O, Bi O , and ZnO. trapping ESR results can be explained by the Z-scheme mech-
2
2 3
However, the potential of the CBM is not sufficient to produce anism. Therefore the results strongly indicate that the Cs O–
2
ꢁ
cO
2
or even produce H
2
(Scheme 2). Thus the heterojunction Bi
2
O
3
–ZnO heterostructure is a direct Z-scheme system rather
mechanism fails to explain the overall water splitting and the than a heterojunction. Moreover, it can be noted that the ZnO
ꢁ
production of cO
2
2 3
conrmed by the spin-trapping ESR VB is very near to the Bi O VB, thus the migration of photo-
experiments. In contrast, based on the Z-scheme approach, generated holes to the ZnO VB does not lead to a remarkable
when the Cs O–Bi –ZnO is illuminated with simulated reduction in the redox potential while the charge separation
sunlight irradiation, all three metal oxides (Cs O, Bi O , and efficiency is greatly enhanced.
2
2 3
O
2
2 3
Scheme 2 Comparison of the charge carrier migration pathway in the Cs
approaches. The schematic diagram also shows that the Z-scheme approach is the appropriate approach to explain the mechanism of pho-
tocatalytic overall water splitting over Cs O–Bi –ZnO.
2 2 3
O–Bi O –ZnO based on the heterojunction and Z-scheme
2
2 3
O
21386 | J. Mater. Chem. A, 2018, 6, 21379–21388
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
Finally, the photostability of the Cs
2
O–Bi
2
O
3
–ZnO can be
9 S. Chen, T. Takata and K. Domen, Nat. Rev. Mater., 2017, 2,
attributed to the direct Z-scheme charge carrier migration
1–17.
21
pathway; that is, the photoinduced electrons in ZnO and Bi O
10 G. Zhang, Z. A. Lan and X. Wang, Chem. Sci., 2017, 8, 5261–
5274.
11 J. Ran, J. Zhang, J. Yu, M. Jaroniec and S. Z. Qiao, Chem. Soc.
Rev., 2014, 43, 7787–7812.
2
3
are consumed by recombination with the photoinduced holes
in the VB of Cs O.
2
1
1
2 A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
3 R. Marschall, Adv. Funct. Mater., 2014, 24, 2421–2440.
Conclusions
A novel Cs
a simple solution combustion method and conrmed by PXRD,
HR-TEM, and XPS analyses. The fabricated heterostructure 15 J. Low, C. Jiang, B. Cheng, S. Wageh, A. A. Al-Ghamdi and
exhibited excellent catalytic activity towards overall water J. Yu, Small Methods, 2017, 1, 1700080.
splitting under simulated sunlight irradiation with good pho- 16 K. Maeda, ACS Catal., 2013, 3, 1486–1503.
tostability. The optimum H and O evolution rates on Cs O– 17 H. Li, W. Tu, Y. Zhou and Z. Zou, Adv. Sci., 2016, 3, 1500389.
Bi –ZnO were ca. 149.5 and ca.73.2 mmol g under 18 Q. Yuan, D. Liu, N. Zhang, W. Ye, H. Ju, L. Shi, R. Long, J. Zhu
simulated sunlight irradiation without using any sacricial and Y. Xiong, Angew. Chem., 2017, 129, 4206–4210.
agents or co-catalysts. High resolution XPS conrmed that the 19 S. Sun, T. Hisatomi, Q. Wang, S. Chen, G. Ma, J. Liu,
2
O–Bi
2
O
3
–ZnO heterostructure was fabricated using 14 P. Zhou, J. Yu and M. Jaroniec, Adv. Mater., 2014, 26, 4920–
4935.
2
2
2
ꢁ1
ꢁ1
2
O
3
h
charge carrier migration pathway followed the heterojunction
S. Nandy, T. Minegishi, M. Katayama and K. Domen, ACS
Catal., 2018, 8, 1690–1696.
approach in the Bi O –ZnO heterostructure and the Z-scheme
2
3
approach in the Cs O–Bi O –ZnO heterostructure. ESR studies 20 Y. Wang, H. Suzuki, J. Xie, O. Tomita, D. J. Martin,
2
2 3
further conrmed the direct Z-scheme charge carrier migration
pathway in the Cs O–Bi –ZnO heterostructure. The remark-
ably high photocatalytic activity of the Cs O–Bi
attributed to the efficient separation of e/h pairs, the efficient
charge transfer occurring at the intimate contact interface 22 M. Han, L. Hu, Y. Zhou, S. Zhao, L. Bai, Y. Sun, H. Huang,
among the three metal oxides, the suitable positions of the CBM
Y. Liu and Z. Kang, Catal. Sci. Technol., 2018, 8, 840–846.
and VBM of the heterostructure, and the extended light 23 M. Zhu, Z. Sun, M. Fujitsuka and T. Majima, Angew. Chem.,
M. Higashi, D. Kong, R. Abe and J. Tang, Chem. Rev., 2018,
2
2 3
O
118, 5201–5241.
2
2
O
3
–ZnO is 21 Q. Xu, L. Zhang, J. Yu, S. Wageh, A. A. Al-ghamdi and M. Jaroniec,
Mater. Today, 2018, DOI: 10.1016/j.mattod.2018.04.008.
absorption range induced by the presence of Cs O.
Int. Ed., 2018, 57, 2160–2164.
2
2
2
2
2
4 A. Rauf, M. Ma, S. Kim, M. S. A. S. Shah, C. H. Chung,
J. H. Park and P. J. Yoo, Nanoscale, 2018, 10, 3026–3036.
5 C. Xu, K. Rho, J. Chun and D. E. Kim, Nanotechnology, 2005,
17(1), 60.
6 B. Ling, X. W. Sun, Y. Q. Shen and Z. L. Dong, Appl. Phys. A:
Mater. Sci. Process., 2010, 98, 91–96.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
7 C. Li, J. Zhang, J. Yang, T. Wang, X. Lv and Z. Tang, Appl.
Catal., A, 2011, 402, 80–86.
Parts of this work were supported by UGC JRF (Award No F.19-1/
2
0
013(SA-I)) and the DST-SERB, Govt of India grant no. YSS/2015/ 28 S. Balachandran and M. Swaminathan, J. Phys. Chem. C,
00651 awarded under the young scientist scheme.
2012, 116, 26306–26312.
2
3
9 X. Wang, P. Ren and H. Fan, Mater. Res. Bull., 2015, 64, 82–87.
0 S. Yi, X. Yue, D. Xu, Z. Liu, F. Zhao, D. Wang and Y. Lin, New
J. Chem., 2015, 39, 2917–2924.
References
1
2
3
4
5
6
7
8
X. Li, J. Yu, J. Low, Y. Fang, J. Xiao and X. Chen, J. Mater. 31 S. Balachandran, N. Prakash and M. Swaminathan, RSC Adv.,
Chem. A, 2015, 3, 2485–2534.
2016, 6, 20247–20257.
S. S. Mao, S. Shen and L. Guo, Prog. Nat. Sci.: Mater. Int., 32 A. Hezam, K. Namratha, Q. A. Drmosh, Z. H. Yamani and
012, 22, 522–534. K. Byrappa, Ceram. Int., 2017, 43, 5292–5301.
T. Huang, X. Lin, J. Xing, W. Wang, Z. Shan and F. Huang, 33 T.-S. Nguyen, G. Postole, S. Loridant, F. Bosselet, L. Burel,
2
Mater. Sci. Eng., B, 2007, 141, 49–54.
M. Aouine, L. Massin, P. G ´e lin, F. Morn and L. Piccolo,
J. Mater. Chem. A, 2014, 2, 19822–19832.
34 A. Band, A. Albu-Yaron, T. Livneh, H. Cohen, Y. Feldman,
L. Shimon, R. Popovitz-Biro, V. Lyahovitskaya and
R. Tenne, J. Phys. Chem. B, 2004, 108, 12360–12367.
M. Ge, Q. Li, C. Cao, J. Huang, S. Li, S. Zhang, Z. Chen,
K. Zhang, S. S. Al-Deyab and Y. Lai, Adv. Sci., 2017, 4, 1–31.
M. Reza Gholipour, C.-T. Dinh, F. B ´e land and T.-O. Do,
Nanoscale, 2015, 7, 8187–8208.
T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 35 S. Sood, A. Umar, S. Kumar Mehta and S. Kumar Kansal,
3, 7520–7535. Ceram. Int., 2015, 41, 3355–3364.
K. Qi, B. Cheng, J. Yu and W. Ho, Chin. J. Catal., 2017, 38, 36 Y. Liu, Z. H. Kang, Z. H. Chen, I. Shaq, J. A. Zapien and
936–1955. I. Bello, Cryst. Growth Des., 2009, 9, 3222–3227.
S. J. a. Moniz, S. a. Shevlin, D. J. Martin, Z.-X. Guo and 37 M. Zare, K. Namratha, M. S. Thakur and K. Byrappa, Mater.
J. Tang, Energy Environ. Sci., 2015, 00, 1–29. Res. Bull., 2019, 109, 49–59.
4
1
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 21379–21388 | 21387

View Article Online
Journal of Materials Chemistry A
Paper
3
3
4
8 M. Zare, K. Namratha, M. S. Thakur, S. Yallappa and 53 Y. J. Yuan, D. Chen, S. Yang, L. X. Yang, J. J. Wang, D. Cao,
K. Byrappa, Mater. Chem. Phys., 2018, 215, 148–156.
9 S. Yi, X. Yue, D. Xu, Z. Liu, F. Zhao, D. Wang and Y. Lin, New
J. Chem., 2015, 39, 2917–2924.
W. Tu, Z. T. Yu and Z. G. Zou, J. Mater. Chem. A, 2017, 5,
21205–21213.
54 Q. Li, B. Guo, J. Yu, J. Ran, B. Zhang, H. Yan and J. R. Gong, J.
Am. Chem. Soc., 2011, 133, 10878–10884.
0 A. Hezam, K. Namratha, Q. A. Drmosh, B. N. Chandrashekar,
G. K. Jayaprakash, C. Cheng, S. S. Swamy and K. Byrappa, 55 H. Tian, F. Teng, J. Xu, S. Lou, N. Li, Y. Zhao and M. Chen,
Ceram. Int., 2018, 44, 7202–7208. Sci. Rep., 2015, 5, 7770.
1 A. Hezam, K. Namratha, Q. A. Drmosh, B. N. Chandrashekar, 56 S. I. Lopatin, S. M. Shugurov and G. A. Semenov, Inorg.
4
K. K. Sadasivuni, Z. H. Yamani, C. Cheng and K. Byrappa,
CrystEngComm, 2017, 19, 3299–3312.
Mater., 2005, 41(12), 1340–1344.
57 J. Anderson and G. V. de W. Chris, Rep. Prog. Phys., 2009, 72,
4
4
2 S. Yi, X. Yue, D. Xu, Z. Liu, F. Zhao, D. Wang and Y. Lin, New
J. Chem., 2015, 39, 2917–2924.
3 T. Chukeaw, A. Seubsai, P. Phon-in, K. Charoen, T. Witoon,
126501.
58 H. Lu, Y. Guo and J. Robertson, J. Appl. Phys., 2017, 121(22),
224304.
W. Donphai, P. Parpainainar, M. Chareonpanich, D. Noon, 59 M. Imran, A. Bin Yousaf, M. Farooq and P. Kasak, Int. J.
B. Zohour and S. Senkan, RSC Adv., 2016, 6, 56116–56126. Hydrogen Energy, 2018, 43, 4256–4264.
4 J.-C. Wang, L. Zhang, W.-X. Fang, J. Ren, Y.-Y. Li, H.-C. Yao, 60 Y. Cui, X. Zhang, R. Guo, H. Zhang, B. Li, M. Xie, Q. Cheng
4
J.-S. Wang and Z.-J. Li, ACS Appl. Mater. Interfaces, 2015, 7,
631–8639.
5 Z. Zhang, D. Jiang, C. Xing, L. Chen, M. Chen and M. He,
Dalton Trans., 2015, 44, 11582–11591.
6 C. Ren, W. Qiu, H. Zhang, Z. He and Y. Chen, J. Mol. Catal. A:
Chem., 2015, 398, 215–222.
and X. Cheng, Sep. Purif. Technol., 2018, 203, 301–309.
61 P. Hajra, S. Shyamal, H. Mandal, P. Fageria, S. Pande and
C. Bhattacharya, Electrochim. Acta, 2014, 123, 494–500.
62 J. Ke, J. Liu, H. Sun, H. Zhang, X. Duan, P. Liang, X. Li,
M. O. Tade, S. Liu and S. Wang, Appl. Catal., B, 2017, 200,
47–55.
8
4
4
4
4
4
7 Z. Zhang, J. Huang, Y. Fang, M. Zhang, K. Liu and B. Dong, 63 M. Ahila and E. Subramanian, J. Electroanal. Chem., 2017,
Adv. Mater., 2017, 29, 1606688. 805, 146–158.
8 D. Liu, Y. Lv, M. Zhang, Y. Liu, Y. Zhu, R. Zong and Y. Zhu, J. 64 P. Xia, B. Zhu, B. Cheng, J. Yu and J. Xu, ACS Sustainable
Mater. Chem. A, 2014, 2, 15377–15388. Chem. Eng., 2017, 6(1), 965–973.
9 H. He, J. Lin, W. Fu, X. Wang, H. Wang, Q. Zeng, Q. Gu, Y. Li, 65 X. Li, K. Xie, L. Song, M. Zhao and Z. Zhang, ACS Appl. Mater.
C. Yan, B. K. Tay, C. Xue, X. Hu, S. T. Pantelides, W. Zhou and
Z. Liu, Adv. Energy Mater., 2016, 6, 1–7.
0 W. Che, W. Cheng, T. Yao, F. Tang, W. Liu, H. Su, Y. Huang,
Interfaces, 2017, 9, 24577–24583.
66 J. Hou, C. Yang, Z. Wang, W. Zhou, S. Jiao and H. Zhu, Appl.
Catal., B, 2013, 142–143, 504–511.
5
Q. Liu, J. Liu, F. Hu, Z. Pan, Z. Sun and S. Wei, J. Am. Chem. 67 Y. Nosaka and A. Y. Nosaka, Chem. Rev., 2017, 117, 11302–
Soc., 2017, 139, 3021–3026. 11336.
1 X. Yang, H. Huang, B. Jin, J. Luo and X. Zhou, RSC Adv., 2016, 68 Z.-R. Tang, Y. Zhang and Y.-J. Xu, RSC Adv., 2011, 1, 1772.
5
5
6
, 107075–107080.
69 C. Zhou, C. Lai, P. Xu, G. Zeng, D. Huang, C. Zhang,
M. Cheng, L. Hu, J. Wan, Y. Liu, W. Xiong, Y. Deng and
M. Wen, ACS Sustainable Chem. Eng., 2018, 6, 4174–4184.
2 X. L. Yin, L. L. Li, W. J. Jiang, Y. Zhang, X. Zhang, L. J. Wan
and J. S. Hu, ACS Appl. Mater. Interfaces, 2016, 8, 15258–
15266.
21388 | J. Mater. Chem. A, 2018, 6, 21379–21388
This journal is © The Royal Society of Chemistry 2018