Ultrathin and Small-Size Graphene Oxide as an Electron Mediator for Perovskite-Based Z-Scheme System to Significantly Enhance Photocatalytic CO2 Reduction

Article abstract of DOI:10.1002/smll.202002140

The judicious design of efficient electron mediators to accelerate the interfacial charge transfer in a Z-scheme system is one of the viable strategies to improve the performance of photocatalysts for artificial photosynthesis. Herein, ultrathin and small-size graphene oxide (USGO) nanosheets are constructed and employed as the electron mediator to elaborately exploit an efficient CsPbBr₃-based all-solid-state Z-scheme system in combination with α-Fe₂O₃ for visible-light-driven CO₂ reduction with water as the electron source. CsPbBr₃ and α-Fe₂O₃ can be closely anchored on USGO nanosheets, owing to the existence of interfacial strong chemical bonding behaviors, which can significantly accelerate the photogenerated carrier transfer between CsPbBr₃ and α-Fe₂O₃. The resultant improved charge separation efficiency endows the Z-scheme system exhibiting a record-high electron consumption rate of 147.6 μmol g^?1 h^?1 for photocatalytic CO₂-to-CO conversion concomitant with stoichiometric O₂ from water oxidation, which is over 19 and 12 times higher than that of pristine CsPbBr₃ nanocrystals and the mixture of CsPbBr₃ and α-Fe₂O₃, respectively. This work provides a novel and effective strategy for improving the catalytic activity of halide-perovskite-based photocatalysts, promoting their practical applications in the field of artificial photosynthesis.

Full text of DOI:10.1002/smll.202002140

Full PaPer
www.small-journal.com
Ultrathin and Small-Size Graphene Oxide as an Electron
Mediator for Perovskite-Based Z-Scheme System to
Significantly Enhance Photocatalytic CO2 Reduction
Yan-Fei Mu, Wen Zhang, Guang-Xing Dong, Ke Su, Min Zhang,* and Tong-Bu Lu
environmental crises.[1–10] The economic
The judicious design of efficient electron mediators to accelerate the interfacial
charge transfer in a Z-scheme system is one of the viable strategies to improve
the performance of photocatalysts for artificial photosynthesis. Herein,
ultrathin and small-size graphene oxide (USGO) nanosheets are constructed
and employed as the electron mediator to elaborately exploit an efficient
CsPbBr3-based all-solid-state Z-scheme system in combination with α-Fe2O3
for visible-light-driven CO2 reduction with water as the electron source.
CsPbBr3 and α-Fe2O3 can be closely anchored on USGO nanosheets, owing
to the existence of interfacial strong chemical bonding behaviors, which can
significantly accelerate the photogenerated carrier transfer between CsPbBr3
and α-Fe2O3. The resultant improved charge separation efficiency endows
the Z-scheme system exhibiting a record-high electron consumption rate of
147.6 µmol g−1 h−1 for photocatalytic CO2-to-CO conversion concomitant with
stoichiometric O2 from water oxidation, which is over 19 and 12 times higher
than that of pristine CsPbBr3 nanocrystals and the mixture of CsPbBr3 and
α-Fe2O3, respectively. This work provides a novel and effective strategy for
improving the catalytic activity of halide-perovskite-based photocatalysts, pro-
moting their practical applications in the field of artificial photosynthesis.
efficiency of this artificial photosynthesis
technology could be attained by resorting
to the development of cost-effective
photocatalytic systems. In this regard,
halide perovskite nanocrystals (NCs),
a
kind of fascinating light-harvesting
materials possessing low-cost fabrication
techniques and excellent optoelectronic
properties such as large extinction coef-
ficients and long photogenerated carrier
lifetime,[11–13] have been demonstrated to
be able to achieve visible-light-driven CO2
reduction concomitant with desirable
water oxidation recently,[14–16] exhibiting
great potential in the field of practical arti-
ficial photosynthesis. However, this kind
of semiconductor NCs in its pristine form
generally displays very low activity for
photocatalytic CO2 reduction, due to the
insufficient charge separation and the lack
of intrinsic catalytic sites.
Recently, continuous efforts have
been devoted to enhance the activity of
perovskite nanocrystals for photocatalytic CO2 reduction, such
1. Introduction
[18]
as loading perovskite NCs on graphene oxide[17] or g-C3N4
Inspired by natural photosynthesis, the use of sunlight-driven
CO2 reduction over photocatalysts with water as electron source
to produce fuels or high-value added chemical feedstocks is
one of promising approaches to surmount the energy and
to improve the photogenerated carrier separation, designing
appropriate heterojunction with other materials involving abun-
dant active sites,[14,15] or directly doping metal active sites on the
surface of perovskite NCs.[19] However, these strategies only
concern about the CO2 reduction half-reaction, and the H2O
oxidation half-reaction has attracted little attention, the latter
of which also plays a pivotal role in controlling the overall per-
formance. In this context, constructing appropriate Z-scheme
system[20–22] with halide perovskite NCs and another oxygen
evolution photocatalysts could be one of feasible strategies to
improve the photogenerated carrier separation and enhance the
water oxidation ability, which has not gained sufficient atten-
tion for perovskite NCs.[23]
The effectiveness of Z-scheme systems relies on the judi-
cious design of efficient electron mediators to achieve fast
interfacial charge transfer.[24−29] Herein, we have elaborately
designed ultrathin and small-size graphene oxide (USGO)
nanosheet and employed it as electron mediator to exploit an
efficient CsPbBr3-based all-solid-state Z-scheme system in com-
bination with hematite (coded as CsPbBr3/USGO/α-Fe2O3) for
visible-light-driven CO2 reduction. Our strategy is based on
the following important considerations: 1) From an economic
Y.-F. Mu, Dr. W. Zhang, G.-X. Dong, K. Su, Prof. M. Zhang, Prof. T.-B. Lu
MOE International Joint Laboratory of Materials Microstructure
Tianjin University of Technology
Tianjin 300384, China
E-mail: zm2016@email.tjut.edu.cn
Y.-F. Mu, Dr. W. Zhang, G.-X. Dong, K. Su, Prof. M. Zhang, Prof. T.-B. Lu
Institute for New Energy Materials and Low Carbon Technologies
School of Materials Science and Engineering
Tianjin University of Technology
Tianjin 300384, China
Prof. M. Zhang
Tianjin Key Laboratory of Organic Solar Cells and Photochemical
Conversion, School of Chemistry and Chemical Engineering
Tianjin University of Technology
Tianjin 300384, China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/smll.202002140.
DOI: 10.1002/smll.202002140
©
2002140 (1 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Small 2020, 2002140

www.advancedsciencenews.com
www.small-journal.com
viewpoint, α-Fe2O3 is an ideal material for photocatalytic water
oxidation, owing to its inexpensive fabrication techniques,
high chemical stability, and suitable band structure for visible-
light response and triggering water oxidation reaction.[30−34]
2) The 2D graphene material is a promising electron transfer
mediator owing to its excellent electron transport mobility,[35,36]
which can accelerate the charge transfer between CsPbBr3 and
α-Fe2O3, thus improving the photogenerated carrier separation
efficiency. 3) There are abundant functional groups such as car-
boxyl and hydroxyl in the surface of USGO,[37] which can form
strong chemical bonds with CsPbBr3,[38−40] facilitating CsPbBr3
loading and photogenerated electron transfer from CsPbBr3
to graphene oxide. As expected, this CsPbBr3/USGO/α-Fe2O3
Z-scheme photocatalyst system achieves efficient charge
transfer between CsPbBr3 and α-Fe2O3, exhibiting significantly
improved photocatalytic activity for CO2 conversion in a reac-
tion system with water as electron source, which is over 19
times higher than that of pristine CsPbBr3 nanocrystals.
Information) further demonstrates the formation of graphene
oxide with small size, exhibiting a weak and broad diffraction
peak at 2θ = 12.1° without the typical diffraction peak of native
graphite.[41]
Zeta potential (ζ) of as-prepared USGO dispersed in hexane
was further measured, being −67 mV, which is reversed with
those of α-Fe2O3 (30 mV) and CsPbBr3 (27 mV) in the same
condition (Figure 1c). The opposite charge nature of USGO
surface in comparison with those of α-Fe2O3 and CsPbBr3
should facilitate the spontaneous assembly of α-Fe2O3 and
CsPbBr3 on the surface of USGO. The prepared processes of
CsPbBr3/USGO/α-Fe2O3 Z-scheme are illustrated in Figure 1d,
and the detailed information regarding the preparation pro-
cesses is described in the Supporting Information. In brief, the
Fe2O3 semiconductor was firstly anchored on the as-prepared
USGO nanosheets through a hydrothermal process to oxidize
Fe3O4 nanoparticle precursor.[44] The XRD pattern (Figure 1e)
of as-prepared Fe2O3 anchoring on the USGO displays typical
diffraction peaks for α-Fe2O3 phase (JCPDS: 00-033-0664).[45]
The resulting USGO/α-Fe2O3 composite also exhibits negative
charge property with a ζ value of −32 mV (Figure 1c), which
was then mixed with as-prepared uniform cubic CsPbBr3
(Figure S3, Supporting Information) in hexane solvent to easily
generate the final CsPbBr3/USGO/α-Fe2O3 composite through
electrostatic self-assembly (Figure S4, Supporting Information).
XRD pattern of CsPbBr3/USGO/α-Fe2O3 (Figure 1e) demon-
strates the successful loading of CsPbBr3 on USGO/α-Fe2O3,
which contains the typical diffraction peaks of CsPbBr3 with
cubic phase (JCPDS: 00-054-0572).
2. Results and Discussion
The USGO nanosheets were synthesized with modified
methods of predecessors[41,42] by controlling the amount of oxi-
dants and oxidation time, and the details are described in the
Supporting Information. The mean size of as-prepared USGO
nanosheets can be obtained based on the analysis of atomic
force microscopy (AFM) image of USGO (Figure 1a), being
≈200–400 nm. Further analysis of the corresponding height
cross-sectional profile of USGO (Figure 1b) displays that the
mean height of USGO sheets is ≈0.75 nm, indicating that a
fully exfoliated ultrathin graphene oxide nanosheet was suc-
cessfully prepared.[43] The corresponding IR spectrum of as-
prepared USGO (Figure S1, Supporting Information) shows
the main characteristic absorption peaks of graphene oxide,
featuring abundant oxygen-containing functional groups. X-ray
diffraction (XRD) pattern of USGO (Figure S2, Supporting
The morphology and composition of CsPbBr3/USGO/α-
Fe2O3 were firstly inspected by transmission electron micros-
copy (TEM) measurement. As presented in Figure 2a, cubic
CsPbBr3 and rod-like α-Fe2O3 can be perceptibly observed on
the USGO nanosheet. It can be seen that α-Fe2O3 nanorods are
clustered on the surface USGO nanosheet, while CsPbBr3 nano-
cubes are homogeneous elsewhere on the surface of USGO
nanosheet, which can be further illustrated from the element
Figure 1. a) Typical tapping mode AFM image and b) the corresponding height cross-sectional profile of USGO deposited on the mica substrate.
The mean height of most USGO sheets is ≈0.75 nm. c) Zeta potentials of as-prepared CsPbBr3, α-Fe2O3, USGO, and USGO/α-Fe2O3 in hexane.
d) Schematic illustrations for the synthesis processes of CsPbBr3/USGO/α-Fe2O3. e) XRD patterns of as-prepared α-Fe2O3, USGO/α-Fe2O3,
CsPbBr3/USGO/α-Fe2O3, and CsPbBr3 powders.
©
Small 2020, 2002140
2002140 (2 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.small-journal.com
Figure 2. a) TEM image, b–h) elemental mappings, and i) HRTEM image with lattice spacing of as-prepared CsPbBr3/USGO/α-Fe2O3.
mappings of CsPbBr3/USGO/α-Fe2O3 composite (Figure 2b–h).
Further high-resolution TEM (HRTEM) measurement
(Figure 2i) confirmed the presence of α-Fe2O3 and CsPbBr3 on
the surface of USGO nanosheet, exhibiting clear lattice fringes
The detailed interaction behaviors between the compo-
nents in CsPbBr3/USGO/α-Fe2O3 composite were further
analyzed with the aid of high-resolution X-ray photoelectron
spectroscopy (XPS) measurements (Figure 3 and Figure S7,
Supporting Information). As depicted in Figure 3a, the Fe 2p
spectrum of pristine α-Fe2O3 exhibits two typical peaks at 710.8
[46]
of 0.25 nm for (110) plane of α-Fe2O3 and 0.58 nm for (100)
plane of cubic CsPbBr3,[47] respectively. The interaction behav-
iors of α-Fe2O3 and CsPbBr3 on the USGO nanosheets were
firstly inspected by measuring the Raman spectra of USGO
nanosheets before and after loading α-Fe2O3 and CsPbBr3. As
presented in Figure S5 (Supporting Information), there are two
characteristic peaks at 1343 and 1583 cm−1 for USGO, which can
be assigned to the corresponding D-band and G-band of typical
graphene oxide.[48] Note that the peak intensity ratio between
D- and G-bands for CsPbBr3/USGO/α-Fe2O3 is obviously larger
than that of pristine USGO, which reveals that the incorpora-
tion of α-Fe2O3 and CsPbBr3 can lead to a disordered carbon
structure in USGO. The UV−vis diffuse reflectance spectra of
CsPbBr3, α-Fe2O3, and CsPbBr3/USGO/α-Fe2O3 in Figure S6
(Supporting Information) show that compared with CsPbBr3
and α-Fe2O3, the light absorption of CsPbBr3/USGO/α-Fe2O3
composite possessed an obviously red absorption shift and
enhanced absorption intensity above 475 nm, further indicating
the strong interaction between CsPbBr3, USGO, and α-Fe2O3.
and 724.4 eV, which can be assigned to Fe 2p3/2 and Fe 2p1/2
,
respectively.[49] There are obvious positive shifts of binding
energy (≈0.5 eV) for both Fe 2p3/2 and Fe 2p1/2 (Figure 3a) in
USGO/α-Fe2O3, which could originate from the formation of
[36]
 
C O Fe bonds between α-Fe2O3 and USGO. As a result,
the C 1s peaks in USGO/α-Fe2O3 show perceptible shifting
to low energy in comparison with those of pristine USGO as
shown in Figure 3b. Incorporating the obvious positive shift
 
(≈0.3 eV) of the Fe O Fe bonds (at 530.0 eV) in O 1s peak
(Figure S7c, Supporting Information), the presence of strong
interaction between α-Fe2O3 and USGO is conducive to the
interfacial charge transfer between them. In addition, the
bonding energies of Br 3d5/2 (68.2 eV) and Br 3d3/2 (69.2 eV)
in CsPbBr3/USGO/α-Fe2O3 also present obvious positive shifts
(≈0.2 eV) with respect to those of pristine CsPbBr3 (Figure 3c),
and a decrease in the bonding energies for typical peaks of C
1s in CsPbBr3/USGO/α-Fe2O3 has been observed relative to
©
2002140 (3 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Small 2020, 2002140

www.advancedsciencenews.com
www.small-journal.com
Figure 3. High-resolution XPS spectra of as-prepared samples: a) Fe 2p for α-Fe2O3 and USGO/α-Fe2O3, b) C 1s for USGO and USGO/α-Fe2O3, c) Br
3d for CsPbBr3 and CsPbBr3/USGO/α-Fe2O3, and d) C 1s for USGO/α-Fe2O3 and CsPbBr3/USGO/α-Fe2O3.
those in USGO/α-Fe2O3 (Figure 3d). Besides, there are negative
shifts (≈0.2 eV) of O 1s peak (Figure S7a, Supporting Informa-
tion) and positive shifts (≈0.3 eV) of Pb 4f peak (Figure S7b,
Supporting Information) in CsPbBr3/USGO composite with
respect to those in pristine USGO and CsPbBr3. These results
from corresponding Mott–Schottky plots with positive slopes
(Figure S8, Supporting Information). Moreover, the conduction
band position of α-Fe2O3 is even lower than the mid-position
between conduction and valence bands of CsPbBr3, indicating
that the Fermi energy level of CsPbBr3 is higher than that of
α-Fe2O3. Therefore, if CsPbBr3 and α-Fe2O3 simultaneously
contact the USGO electron mediator, the electrons in the sur-
face of CsPbBr3 tend to migrate to α-Fe2O3 via USGO to realize
the Femi energy level equilibrium of the system, resulting in a
built-in electric field in the interface of CsPbBr3/USGO/α-Fe2O3
pointing from CsPbBr3 to α-Fe2O3. This orientation of built-in
electric field can facilitate the photogenerated electron transfer
from α-Fe2O3 to CsPbBr3, suggesting that photogenerated car-
rier transfer between CsPbBr3 and α-Fe2O3 complies with a
Z-scheme pathway.
The time-resolved photoluminescence (TRPL) traces of sam-
ples were further recorded to check out the favorable influ-
ence of USGO on the photogenerated carrier transfer between
CsPbBr3 and α-Fe2O3. As depicted in Figure 4b, the PL decay
(red symbols) of CsPbBr3 is related to the radiative and non-
radiative deactivations of photogenerated excitons in CsPbBr3.
The mixture of CsPbBr3 and α-Fe2O3 (coded as CsPbBr3:α-
Fe2O3) exhibits just slightly accelerated PL decay (blue sym-
bols) in comparison with pristine CsPbBr3, indicating the
inferior charge transfer between CsPbBr3 and α-Fe2O3, which
may result from the electrostatic repulsion between CsPbBr3
and α-Fe2O3 owing to their same charge nature as presented
in Figure 1c. Note that loading CsPbBr3 on the surface of
indicate the existence of strong interaction between CsPbBr3
[38]
 
and USGO by forming Br O C bonds,
which can accel-
erate the interfacial electron transfer rate between CsPbBr3 and
USGO.
Apart from the connection pattern, it is known that the
transfer processes of photogenerated carriers between two
semiconductors practically associate with their energy offsets of
conduction and valence band edges, which have crucial influ-
ence on the ultimate photocatalytic activity. To obtain these
thermodynamic information, we first calculated the band-
gaps of CsPbBr3 and α-Fe2O3 by analyzing their Tauc plots
(Figure S8, Supporting Information) based on the UV−vis dif-
fuse reflectance spectra measurements (Figure S6, Supporting
Information), being 2.29 and 2.06 eV, respectively. The conduc-
tion band potentials of CsPbBr3 and α-Fe2O3 can be further
derived from their Mott–Schottky plots (Figure S8, Supporting
Information), being −1.07 and 0.24 V versus NHE as illus-
trated in Figure 4a, respectively. Thereby, the corresponding
valence band potentials of CsPbBr3 and α-Fe2O3 are 1.22 and
2.30 V versus NHE (Figure 4a), which are consistent with the
results of XPS valence spectra measurements (Figure S9, Sup-
porting Information). It is worth noting that both CsPbBr3
and α-Fe2O3 are n-type semiconductors, which can be inferred
©
Small 2020, 2002140
2002140 (4 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.small-journal.com
Figure 4. a) Schematic illustration of band structures for CsPbBr3 and α-Fe2O3 derived from the results of flat-band potential and UV–vis absorption
spectroscopy measurements. b) Time-resolved photoluminescence decay of as-prepared CsPbBr3, CsPbBr3:α-Fe2O3, CsPbBr3/USGO, and CsPbBr3/
USGO/α-Fe2O3. c) I–t curves of CsPbBr3, CsPbBr3:α-Fe2O3, CsPbBr3/USGO, and CsPbBr3/USGO/α-Fe2O3 plotted at a bias potential of −0.4 V (vs Ag/
AgCl) under light illumination. d) The yield of CO generated from photocatalytic CO2 reduction after 4 h of irradiation under 300 W Xe lamp, with the
light intensity of 100 mW cm−2.
USGO can bring forth a significantly accelerated PL decay
(purple symbols) with respect to pristine CsPbBr3, indicating
the occurrence of efficient electron transfer between CsPbBr3
and USGO due to the formation of strong chemical bonds
in the interface of CsPbBr3/USGO. Moreover, there is a fur-
ther acceleration of PL decay (olive symbols) if CsPbBr3 and
α-Fe2O3 were simultaneously incorporated on the surface of
USGO (CsPbBr3/USGO/α-Fe2O3), which can be ascribed to the
excellent electron transport ability of USGO and the favorable
energy offset between CsPbBr3 and α-Fe2O3. All these PL evolu-
tion traces can be well fitted with a multiexponential function,
and the corresponding values of amplitude-averaged lifetimes
are tabulated in Table S1 (Supporting Information), being 31.3,
29.1, 20.6, and 13.9 ns for CsPbBr3, CsPbBr3:α-Fe2O3, CsPbBr3/
USGO, and CsPbBr3/USGO/α-Fe2O3, respectively.
Photocurrent response and electrochemical impedance
spectroscopy (EIS) measurements were further carried out to
give evidence of the benefit of USGO on the charge separa-
tion between CsPbBr3 and α-Fe2O3. As presented in Figure 4c,
mixing CsPbBr3 with α-Fe2O3 in the absence of USGO just
bring forth a slightly higher photocurrent intensity with
respect to pristine CsPbBr3, which is even lower than that
of CsPbBr3/USGO, highlighting the pivotal role of USGO
in charge transfer. The photocurrent intensity of CsPbBr3/
USGO/α-Fe2O3 further proves the facilitation of USGO to
the interfacial charge separation, exhibiting a significant
improvement of photocurrent intensity compared to pristine
CsPbBr3 as depicted in Figure 4c. Furthermore, the values of
semicircular radius in Nyquist plots for these samples are in
the order: CsPbBr3 > CsPbBr3:α-Fe2O3 > CsPbBr3/USGO >
CsPbBr3/USGO/α-Fe2O3 (Figure S10, Supporting Informa-
tion), indicating that the charge-transfer resistance obviously
decreased for CsPbBr3/USGO/α-Fe2O3, as evidenced by the
corresponding fitting results (22 614 Ω for CsPbBr3, 12 627 Ω
for CsPbBr3:α-Fe2O3, 8679 Ω for CsPbBr3/USGO, and 4573 Ω
for CsPbBr3/USGO/α-Fe2O3). These results are consistent with
the photocurrent response measurements, once again demon-
strating the favorable influence of USGO on the interfacial
charge transfer.
The photocatalytic performances of CsPbBr3, α-Fe2O3,
CsPbBr3/USGO, and CsPbBr3/USGO/α-Fe2O3 for CO2 reduc-
tion were evaluated in CO2-saturated acetonitrile/water solution
under simulated sunlight irradiation with Xe lamp coupled
with a 400 nm filter, and the light intensity was adjusted to
100 mW cm−2 by calibrating with a NREL-calibrated Si solar
cell. Herein, there is only very small amount of water in
the reaction system, because of the instability of halide perov-
skite in the reaction system with a large amount of water. For
these CsPbBr3-based photocatalysts, the results of chromato-
graphic analyses show that the main reduction product is CO,
and only negligible CH4 and H2 can be detected. Although
the CO2-to-CH4 conversion by the photogenerated electrons
in these catalysts is thermodynamically permissible, the unfa-
vorable evolution of CH4 may originate from the rapid des-
orption of the reaction intermediate, that is CO,[50] from the
surface of the catalyst. This phenomenon is consistent with
©
2002140 (5 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Small 2020, 2002140

www.advancedsciencenews.com
www.small-journal.com
the common observation in the photocatalytic systems with
CsPbBr3 as photocatalyst.[16−19] As depicted in Figure 4d, the
yield of CO is only 3.7 µmol g−1 h−1 with pristine CsPbBr3 as
photocatalyst, owing to the insufficient charge separation. It
is noted that no reduction product can be detected with pure
α-Fe2O3 as photocatalyst, because the conduction band poten-
tial (0.24 V vs NHE) of α-Fe2O3 is not high enough to trigger
CO2 reduction reaction. Under the same condition, loading
CsPbBr3 on USGO can bring forth an obvious improvement
of photocatalytic activity as presented in Figure 4d. An obvi-
ously enhanced yield of 14.6 µmol g−1 h−1 can be achieved for
CO2-to-CO conversion, owing to the strong interaction between
CsPbBr3 and USGO.
13CH4 with m/z value of 17 (Figure S14, Supporting Informa-
tion) can be clearly observed by the gas chromatography/mass
spectrometry analysis. Moreover, oxidation product O2 can also
be detected in the gas products, and the ratio of the amounts
of CO and O2 is close to 2:1 (Figure S15, Supporting Informa-
tion), indicating that the photogenerated holes in CsPbBr3/
USGO/α-Fe2O3 are mainly consumed by the water oxidation
reaction. Further H218O isotope labeling experiment demon-
strated that O2 is the product of water oxidation, where the
oxidation production 18O2 (m/z = 36) can be clearly observed in
the result of gas chromatography/mass spectrometry analysis
(Figure S16, Supporting Information).
From the thermodynamic point of view, the result of signifi-
cantly enhanced performance of CsPbBr3/USGO/α-Fe2O3 indi-
cates that the photogenerated carrier separation in CsPbBr3/
USGO/α-Fe2O3 is Z-scheme mechanism. If the pathway of
interfacial charge transfer between CsPbBr3 and α-Fe2O3 com-
plies with traditional type-II heterojunction, the photogenerated
electrons in CsPbBr3 should transfer to α-Fe2O3 via USGO
mediator. As shown in Figure 4a, the conduction band potential
of α-Fe2O3 is lower than the reduction potential of CO2-to-CO
conversion (−0.52 V vs NHE). The photogenerated electrons
in α-Fe2O3 have no driving force to trigger the reduction reac-
tion of CO2, which contrasts with the results of photocatalytic
measurements. To further confirm the Z-scheme mechanism
in CsPbBr3/USGO/α-Fe2O3 system, we resorted to in situ
irradiated XPS (ISI-XPS) measurement[51,52] to inspect the
orientation of interfacial charger transfer between CsPbBr3
and α-Fe2O3. As presented in Figure 5a, without light irra-
diation, the Br 3d spectrum of CsPbBr3/USGO/α-Fe2O3 could
be resolved into two typical peaks corresponding to Br 3d5/2
(68.2 eV) and Br 3d3/2 (69.2 eV). In addition, Pb 4f spectrum
of CsPbBr3/USGO/α-Fe2O3 exhibits two characteristic peaks at
138.3 and 143.2 eV (Figure 5b), which can be assigned to Pb
4f7/2 and Pb 4f5/2, respectively. Light irradiation brings forth an
obvious negative shift (≈0.3 eV) of binding energies for both Br
3d and Pb 4f, indicating an increased electron density around
the Br and Pb nucleus by accumulating electrons. On the
other hand, a characteristic peak at 711.3 eV for Fe 2p3/2 can
be clearly observed in Fe 2p spectrum of CsPbBr3/USGO/α-
Fe2O3 without light irradiation (Figure 5c). Meanwhile, the
characteristic peak position of Fe 2p1/2 at 725.3 eV can also be
obtained by fitting the Fe 2p spectrum with considering the
interference of Cs 3d signal (Figure 5d). Upon light irradia-
tion, both Fe 2p1/2 and Fe 2p3/2 exhibit positive shifted binding
energies, being 711.6 and 725.5 eV, respectively, which suggests
the occurrence of holes accumulation in α-Fe2O3. These phe-
nomena of binding energy shifting induced by light irradia-
tion confirm that photogenerated electrons in α-Fe2O3 transfer
to CsPbBr3 through USGO electron mediator, which is in line
with Z-scheme mechanism.
More importantly, if CsPbBr3 and α-Fe2O3 were simulta-
neously incorporated onto the USGO nanosheet, the activity
of photocatalytic CO2 reduction can be further significantly
enhanced as shown in Figure 4d. CsPbBr3/USGO/α-Fe2O3 com-
posite achieves an exciting yield of 73.8 µmol g−1 h−1 for CO2-to-
CO conversion, which is over 19 times higher than that of pure
CsPbBr3. To the best of our knowledge, the electron consumption
rate (Relectron = 147.6 µmol g−1 h−1) with CsPbBr3/USGO/α-Fe2O3
as photocatalyst is well above that of reported metal-halide
perovskite materials photocatalysts with desirable water as elec-
tron source and under 1 sun light irradiation (Table S2, Sup-
porting Information). In addition, to evaluate the superiority
of USGO for accelerating the charge transfer between CsPbBr3
and α-Fe2O3, the photocatalytic activity of CsPbBr3:α-Fe2O3
composite has also been assessed under the same condition. As
shown in Figure 4d, CsPbBr3:α-Fe2O3 composite just exhibits
a slightly improved photocatalytic activity (5.8 µmol g−1 h−1)
compared to pristine CsPbBr3, highlighting the important role
of USGO in accelerating the interfacial charge transfer rate
by providing an electron transport freeway. Besides, the time-
dependent data for the evolution of CO (Figure S11, Supporting
Information) and the XRD patterns of CsPbBr3/USGO/α-Fe2O3
after photoreduction of CO2 (Figure S12, Supporting Informa-
tion) suggest that the catalytic activity and crystal structures of
CsPbBr3 and α-Fe2O3 were basically maintained in CsPbBr3/
USGO/α-Fe2O3 after photocatalytic reaction for 8 h. The
slightly increased diffraction peaks of CsPbBr3/USGO/α-Fe2O3
may result from the occurrence of a certain degree of aggre-
gation after reaction. Furthermore, the reusability of CsPbBr3/
USGO/α-Fe2O3 composite was also tested with two consecutive
cycles of the photocatalytic experiments. As shown in Figure S13
(Supporting Information), after two cycles, the activity of
CsPbBr3/USGO/α-Fe2O3 was remained for 90%, indicating the
relatively stability of CsPbBr3/USGO/α-Fe2O3 composite in the
reaction system containing small amount of water.
The origin of reduction product CO was further identified
by performing a series of reference experiments with CsPbBr3/
USGO/α-Fe2O3 as photocatalyst. There is negligible light
response in the absence of light irradiation or photocatalyst in
the reaction system, suggesting that CO2-to-CO conversion is
triggered by the photogenerated carriers in CsPbBr3/USGO/α-
Fe2O3. In addition, replacing CO2 with N2 also results in negli-
gible light response, indicating that CO is produced from CO2
reduction. This speculation has been further confirmed by iso-
tope tracking experiment with 13CO2 instead of CO2, where the
major reaction product 13CO with m/z value of 29 and negligible
3. Conclusion
In summary, we have successfully constructed an efficient
halide-perovskite based all-solid-state Z-scheme system using
USGO as electron mediator, which presents an exciting electron
consumption rate of 147.6 µmol g−1 h−1 for the conversion of
©
Small 2020, 2002140
2002140 (6 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.small-journal.com
Figure 5. High-resolution XPS spectra of CsPbBr3/USGO/α-Fe2O3 in the dark or under 300 W Xe lamp irradiation with a 400 nm filter: a) Br 3d, b) Pb
4f, c) Fe 2p3/2, and d) the full-spectrum of Fe 2p.
photocatalytic CO2-to-CO with water as the sacrificial reductant.
XPS measurements revealed that α-Fe2O3 and CsPbBr3 display
strong interaction with USGO by forming chemical bonds.
Conflict of Interest
The authors declare no conflict of interest.
With respect to the mixture of α-Fe2O3 and CsPbBr3, the intro-
duction of USGO can bring forth a significant improvement
Keywords
of charge separation efficiency between α-Fe2O3 and CsPbBr3
as demonstrated by the photophysical and electrochemical
measurements. Further ISI-XPS characterization confirmed
the pathway of interfacial charge transfer between CsPbBr3
and α-Fe2O3, and gave an obvious evidence of the formation of
Z-scheme heterojunction between CsPbBr3 and α-Fe2O3. This
work provides an important inspiration for the utilization of
halide perovskite as well as various low-cost semiconductors in
the field of artificial photosynthesis.
charge transfer, CO2 reduction, perovskite nanocrystals, photocatalysis,
Z-scheme systems
Received: April 2, 2020
Revised: April 29, 2020
Published online:
[1] X. Li, J. Yu, M. Jaroniec, X. Chen, Chem. Rev. 2019, 119, 3962.
[2] H.-L. Wu, X.-B. Li, C.-H. Tung, L.-Z. Wu, Adv. Mater. 2019, 31,
1900709.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[3] J. Ran, M. Jaroniec, S.-Z. Qiao, Adv. Mater. 2018, 30, 1704649.
[4] J. H. Montoya, L. C. Seitz, P. Chakthranont, A. Vojvodic,
T. F. Jaramillo, J. K. Nørskov, Nat. Mater. 2017, 16, 70.
[5] X. Liu, S. Inagaki, J. Gong, Angew. Chem., Int. Ed. 2016, 55, 14924.
[6] K. K. Sakimoto, A. B. Wong, P. Yang, Science 2016, 351, 74.
[7] J. L. White, M. F. Baruch, J. E. Pander III, Y. Hu, I. C. Fortmeyer,
J. E. Park, T. Zhang, K. Liao, J. Gu, Y. Yan, T. W. Shaw, E. Abelev,
A. B. Bocarsly, Chem. Rev. 2015, 115, 12888.
Acknowledgements
Y.-F.M. and W.Z. contributed equally to this work. This work was
financially supported by Natural Science Foundation of Tianjin City
(17JCJQJC43800), National Key R&D Program of China (2017YFA0700104),
NSFC (21790052 and 21702146), and the 111 Project (D17003).
[8] S. Berardi, S. Drouet, L. Francàs, C. Gimbert-Suriñach,
M. Guttentag, C. Richmond, T. Stoll, A. Llobet, Chem. Soc. Rev.
2014, 43, 7501.
©
2002140 (7 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Small 2020, 2002140

www.advancedsciencenews.com
www.small-journal.com
[9] S. N. Habisreutinger, L. Schmidt-Mende, J. K. Stolarczyk, Angew.
Chem., Int. Ed. 2013, 52, 7372.
[10] S. C. Roy, O. K. Varghese, M. Paulose, C. A. Grimes, ACS Nano
2010, 4, 1259.
[31] M. Mishra, D.-M. Chun, Appl. Catal., A 2015, 498, 126.
[32] A. J. Cowan, C. J. Barnett, S. R. Pendlebury, M. Barroso, K. Sivula,
M. Grätzel, J. R. Durrant, D. R. Klug, J. Am. Chem. Soc. 2011, 133,
10134.
[11] Q. A. Akkerman, G. Rainò, M. V. Kovalenko, L. Manna, Nat. Mater.
2018, 17, 394.
[12] M. V. Kovalenko, L. Protesescu, M. I. Bodnarchuk, Science 2017,
358, 745.
[13] X. Li, F. Cao, D. Yu, J. Chen, Z. Sun, Y. Shen, Y. Zhu, L. Wang, Y. Wei,
Y. Wu, H. Zeng, Small 2017, 13, 1603996.
[14] L.-Y. Wu, Y.-F. Mu, X.-X. Guo, W. Zhang, Z.-M. Zhang, M. Zhang,
T.-B. Lu, Angew. Chem., Int. Ed. 2019, 58, 9491.
[33] K. Sivula, F. L. Formal, M. Grätzel, ChemSusChem 2011, 4, 432.
[34] Y. Lin, G. Yuan, S. Sheehan, S. Zhou, D. Wang, Energy Environ. Sci.
2011, 4, 4862.
[35] M.-Q. Yang, N. Zhang, M. Pagliaro, Y.-J. Xu, Chem. Soc. Rev. 2014,
43, 8240.
[36] Z. Pan, G. Zhang, X. Wang, Angew. Chem., Int. Ed. 2019, 58, 7102.
[37] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev.
2010, 39, 228.
[15] Z.-C. Kong, J.-F. Liao, Y.-J. Dong, Y.-F. Xu, H.-Y. Chen, D.-B. Kuang,
C.-Y. Su, ACS Energy Lett. 2018, 3, 2656.
[38] X.-X. Guo, S.-F. Tang, Y.-F. Mu, L.-Y. Wu, G.-X. Dong, M. Zhang, RSC
Adv. 2019, 9, 34342.
[16] J. Hou, S. Cao, Y. Wu, Z. Gao, F. Liang, Y. Sun, Z. Lin, L. Sun, Chem.
- Eur. J. 2017, 23, 9481.
[17] Y.-F. Xu, M.-Z. Yang, B.-X. Chen, X.-D. Wang, H.-Y. Chen,
D.-B. Kuang, C.-Y. Su, J. Am. Chem. Soc. 2017, 139, 5660.
[18] M. Ou, W. Tu, S. Yin, W. Xing, S. Wu, H. Wang, S. Wan, Q. Zhong,
R. Xu, Angew. Chem., Int. Ed. 2018, 57, 13570.
[39] X. Li, Y. Wang, H. Sun, H. Zeng, Adv. Mater. 2017, 29, 1701185.
[40] J. De Roo, M. Ibáñez, P. Geiregat, G. Nedelcu, W. Walravens,
J. Maes, J. C. Martins, I. Van Driessche, M. V. Kovalenko, Z. Hens,
ACS Nano 2016, 10, 2071.
[41] L. Zhang, J. Liang, Y. Huang, Y. Ma, Y. Wang, Y. Chen, Carbon 2009,
47, 3365.
[19] Y.-F. Mu, W. Zhang, X.-X. Guo, G.-X. Dong, M. Zhang, T.-B. Lu,
ChemSusChem 2019, 12, 4769.
[42] L. Zhang, X. Li, Y. Huang, Y. Ma, X. Wan, Y. Chen, Carbon 2010, 48,
2361.
[20] A. J. Bard, J. Photochem. 1979, 10, 59.
[21] H. Tada, T. Mitsui, T. Kiyonaga, T. Akita, K. Tanaka, Nat. Mater.
2006, 5, 782.
[43] S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas,
E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen, R. S. Ruoff,
Nature 2006, 442, 282.
[22] J. Yu, S. Wang, J. Low, W. Xiao, Phys. Chem. Chem. Phys. 2013, 15,
16883.
[44] H. Zhang, A. Xie, C. Wang, H. Wang, Y. Shen, X. Tian, J. Mater.
Chem. A 2013, 1, 8547.
[23] Y. Jiang, J.-F. Liao, H.-Y. Chen, H.-H. Zhang, J.-Y. Li, X.-D. Wang,
D.-B. Kuang, Chem 2020, 6, 766.
[45] T. Zhang, C. Zhu, Y. Shi, Y. Li, S. Zhu, D. Zhang, Mater. Lett. 2017,
205, 10.
[24] Y. Wang, H. Suzuki, J. Xie, O. Tomita, D. J. Martin, M. Higashi,
D. Kong, R. Abe, J. Tang, Chem. Rev. 2018, 118, 5201.
[25] Q. Wang, T. Hisatomi, Y. Suzuki, Z. Pan, J. Seo, M. Katayama,
T. Minegishi, H. Nishiyama, T. Takata, K. Seki, A. Kudo, T. Yamada,
K. Domen, J. Am. Chem. Soc. 2017, 139, 1675.
[46] D. Li, R. Xu, Y. Jia, P. Ning, K. Li, Chem. Phys. Lett. 2019, 731,
136623.
[47] J. Liu, B. Shabbir, C. Wang, T. Wan, Q. Ou, P. Yu, A. Tadich, X. Jiao,
D. Chu, D. Qi, D. Li, R. Kan, Y. Huang, Y. Dong, J. Jasieniak,
Y. Zhang, Q. Bao, Adv. Mater. 2019, 31, 1901644.
[48] Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts, R. S. Ruoff, Adv.
Mater. 2010, 22, 3906.
[26] A. Iwase, S. Yoshino, T. Takayama, Y. H. Ng, R. Amal, A. Kudo, J.
Am. Chem. Soc. 2016, 138, 10260.
[27] H. Li, W. Tu, Y. Zhou, Z. Zou, Adv. Sci. 2016, 3, 1500389.
[28] P. Zhou, J. Yu, M. Jaroniec, Adv. Mater. 2014, 26, 4920.
[29] K. Maeda, ACS Catal. 2013, 3, 1486.
[30] S. Shen, S. A. Lindley, X. Chen, J. Z. Zhang, Energy Environ. Sci.
2016, 9, 2744.
[49] X. Hu, J. C. Yu, J. Gong, Q. Li, G. Li, Adv. Mater. 2007, 19, 2324.
[50] H. Rao, L. C. Schmidt, J. Bonin, M. Robert, Nature 2017, 548, 74.
[51] X. G. Liu, G. J. Dong, S. P. Li, G. X. Lu, Y. P. Bi, J. Am. Chem. Soc.
2016, 138, 2917.
[52] J. Low, B. Dai, T. Tong, C. Jiang, J. Yu, Adv. Mater. 2019, 31, 1802981.
©
Small 2020, 2002140
2002140 (8 of 8)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim