An All-Organic D-A System for Visible-Light-Driven Overall Water Splitting

Article abstract of DOI:10.1002/smll.202003914

Direct water splitting over photocatalysts is a prospective strategy to convert solar energy into hydrogen energy. Nevertheless, because of the undesirable electron accumulation at the surface, the overall water-splitting efficiency is seriously restricted by the poor charge separation/transfer ability. Here, an all-organic donor–acceptor (D-A) system through crafting carbon rings units-conjugated tubular graphitic carbon nitride (C-TCN) is proposed. Through a range of characterizations and theoretical calculations, the incorporation of carbon rings units via continuous π-conjugated bond builds a D-A system, which can drive intramolecular charge transfer to realize highly efficient charge separation. More importantly, the tubular structure and the incorporated carbon rings units cause a significant downshift of the valence band, of which the potential is beneficial to the activation for O₂ evolution. When serving as photocatalyst for overall water splitting, C-TCN displays considerable performance with H₂ and O₂ production rates of 204.6 and 100.8 μmol g^?1 h^?1, respectively. The corresponding external quantum efficiency reaches 2.6% at 405 nm, and still remains 1.7% at 420 nm. This work demonstrates that the all-organic D-A system conceptualized from organic solar cell can offer promotional effect for overall water splitting by addressing the charge accumulation problem rooted in the hydrogen evolution reaction.

Full text of DOI:10.1002/smll.202003914

FULL PAPER
www.small-journal.com
An All-Organic D-A System for Visible-Light-Driven
Overall Water Splitting
Zhao Mo, Jun Di, Pengcheng Yan, Chade Lv, Xingwang Zhu, Daobin Liu, Yanhua Song,
Chuntai Liu, Qing Yu, Huaming Li, Yucheng Lei, Hui Xu,* and Qingyu Yan*
1. Introduction
Direct water splitting over photocatalysts is a prospective strategy to
convert solar energy into hydrogen energy. Nevertheless, because of the
undesirable electron accumulation at the surface, the overall water-splitting
efficiency is seriously restricted by the poor charge separation/transfer
ability. Here, an all-organic donor–acceptor (D-A) system through crafting
carbon rings units-conjugated tubular graphitic carbon nitride (C-TCN) is
proposed. Through a range of characterizations and theoretical calcula-
tions, the incorporation of carbon rings units via continuous π-conjugated
bond builds a D-A system, which can drive intramolecular charge transfer
to realize highly efficient charge separation. More importantly, the tubular
structure and the incorporated carbon rings units cause a significant
downshift of the valence band, of which the potential is beneficial to the
activation for O2 evolution. When serving as photocatalyst for overall water
splitting, C-TCN displays considerable performance with H2 and O2 produc-
tion rates of 204.6 and 100.8 µmol g−1 h−1, respectively. The corresponding
external quantum efficiency reaches 2.6% at 405 nm, and still remains 1.7%
at 420 nm. This work demonstrates that the all-organic D-A system concep-
tualized from organic solar cell can offer promotional effect for overall water
splitting by addressing the charge accumulation problem rooted in the
hydrogen evolution reaction.
Hydrogen production by means of photo-
catalytic water splitting represents
a
green way to convert solar energy into the
renewable and clean hydrogen energy.[1–5]
For this reason, photocatalytic water split-
ting holds great promise in addressing the
long-standing environmental and energy
issues, which has attracted extensive
attention of researchers from different
scientific and engineering domains.[5–9]
Despite the intriguing prospect of photo-
catalytic water splitting, recent photocata-
lytic systems are still far from meeting
the requirements of industrial standard
for water splitting reaction, especially the
challenges associated with the low solar-
to-hydrogen (STH) conversion efficiency
(the targeted STH conversion efficiency
for commercial application threshold is
10%).[10–12] The unfavorable STH con-
version efficiency can be explained
from three different aspects, which are
photoabsorption performance, charge
separation/transfer efficiency, and sur-
face reactions.[13] Therefore, in order to
enhance the STH conversion efficiency, the selection of suitable
photocatalytic materials and the design of efficient photocata-
lytic systems appear to be especially important.[14]
Z. Mo, P. Yan, X. Zhu, Prof. Q. Yu, Prof. H. Li, Prof. Y. Lei, Prof. H. Xu
School of Materials Science and Engineering
Institute for Energy Research
Jiangsu University
Among various photocatalytic materials, graphitic carbon
nitride (g-C3N4) has taken tremendous interest in photocatalytic
overall water splitting because of its suitable electronic structure
(Eg: 2.7 eV, EVB: +1.6 V, and ECB: −1.1 V, which are the prereq-
uisite for photocatalytic overall water splitting).[15–17] However,
bulk g-C3N4 (Bulk CN) is also facing the abovementioned obsta-
cles.[18] From this perspective, numerous strategies, including
nanostructuring, cocatalyst loading, elemental doping, het-
erojunction construction, and so on, have been developed to
enhance the photocatalytic activity of Bulk CN through opti-
mizing one or more of the above three aspects.[16,19–21] Despite
tremendous effort devoted to optimizing the three aspects men-
tioned above, the photocatalytic enhancement is insufficient to
reach parity with those metal-based photocatalysts, let alone
satisfying the requirements of industrial standard.[2,16] Hence,
further insight into how material design affect the performance
of g-C3N4 is essential to provide basic guideline for achieving
enhancement in photocatalytic activity. Considering that the
Zhenjiang 212013, P. R. China
E-mail: xh@ujs.edu.cn
Z. Mo, Dr. J. Di, Dr. C. Lv, Dr. D. Liu, Prof Q. Yan
School of Materials Science and Engineering
Nanyang Technological University
Singapore 639798, Singapore
E-mail: Alexyan@ntu.edu.sg
Dr. Y. Song
School of Environmental and Chemical Engineering
Jiangsu University of Science and Technology
Zhenjiang 212003, P. R. China
Prof. C. Liu
Key Laboratory of Materials Processing and Mold
Ministry of Education
Zhengzhou University
Zhengzhou 450002, P. R. China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/smll.202003914.
DOI: 10.1002/smll.202003914
2003914 (1 of 10)
© 2020 Wiley-VCH GmbH
Small 2020, 2003914

www.advancedsciencenews.com
www.small-journal.com
Figure 1. a) Schematic representation, b) SEM image, c) TEM image, and d) HRTEM image of C-TCN (The circular area represents carbon rings units
and the rectangular area is g-C3N4); Chemical element mapping images of C-TCN: e) C and f) N.
high adjustability and compatibility of g-C3N4 due to its intrinsic
polymeric feature and tunable chemical composition, rational
molecular design is an efficient strategy that can be used to
optimize the photocatalytic performance of g-C3N4.[18,22–24]
Molecular design strategy can deliberately introduce some
appointed elements, functional groups, or molecules into the
basal plane of g-C3N4, which intrinsically optimizes the band
structure and textual property of g-C3N4, thus achieving out-
standing photocatalytic efficiency.[25,26] Even so, the conven-
tional molecular design strategy still suffers from unfavorable
charge separation/transfer efficiency, thereby limiting the
photocatalytic performance and stability.
In the past few years, the feasibility of donor–acceptor
(D-A) system has been demonstrated in organic solar cells,
which can effectively overcome the limit of charge diffu-
sion length of polymer and promote the charge separation
via intramolecular charge transfer.[27–30] Inspired by the high
performance of D-A system in organic solar cells in addition
to the high malleability of g-C3N4, we therefore believe that
the D-A system also can duplicate its success in the photo-
catalysis field. With this in mind, molecular design strategy
is utilized to build a g-C3N4-based D-A system by precisely
embedding suitable small organic molecules into the basal
plane of g-C3N4. Recently, Wang and co-workers developed an
efficient D-A system by cocondensation of urea with oxamide
in molten salt, where the corresponding apparent quantum
efficiency reached up to 57% at λ = 420 nm.[24] Wang and co-
workers also built another D-A system by preparing a sele-
nium and cyanamide-functionalized heptazine-based melon
to accelerate the charge transfer, which further proved the
feasibility of D-A system.[30] In particular, carbon rings units
can be seamlessly grafted into the basal plane of g-C3N4 via
continuous π-conjugated bond due to their similar aromatic
structures.[31–35] Theoretically, the electron-withdrawing prop-
erty of carbon rings units is stronger than that of g-C3N4, so
carbon rings units and the basal plane of g-C3N4 can serve
as the electron acceptor and donor, respectively. Once the
g-C3N4-based D-A polymer is photoexcited, the excited electron
is rapidly extracted from the basal plane of g-C3N4 (the elec-
tron donor) to carbon rings units (the electron acceptor),
which can reduce the potential for charge accumulation by
accelerating the charge separation, resulting in the optimiza-
tion of photocatalytic performance and even the realization of
overall water splitting.
As a proof-of-concept, we constructed an all-organic D-A
system by grafting carbon rings units on the basal plane of
g-C3N4. In addition, melamine and glucose served as the pre-
cursor of g-C3N4 units and carbon rings units, meanwhile
supramolecular self-assembly strategy was employed for the
assembly of 1D tubular structure. On one hand, carbon rings-
conjugated tubular g-C3N4 (C-TCN) has stronger oxidation
ability than the Bulk CN, which is more likely to enable the O2
evolution reaction that is inaccessible by the Bulk CN. On the
other hand, the incorporation of carbon rings units can also
extend the light absorption and promotes the charge separation
due to the intramolecular charge transfer, as verified by density
functional theory (DFT) calculations. Consequently, the C-TCN
exhibits excellent and stable photocatalytic overall water split-
ting activity.
2. Results and Discussions
2.1. Synthesis and Structure of C-TCN
As illustrated in Figure 1a, we report a template-free preas-
sembly strategy to prepare C-TCN. In the first step, melamine
and glucose are used as the precursors for the synthesis of
supramolecular intermediate via molecular self-assembly. It is
postulated that the self-assembly process starts with the reac-
tion between melamine and H2O to produce cyanuric acid at
200 °C for 12 h before the cyanuric acid assembles with the
unreacted melamine into the supramolecular intermediate
through hydrogen bonding.[36] Thanks to the abundant hydroxyl
groups of glucose, such unique feature ensures that glucose
is perfectly combined into the supramolecular intermediate
Small 2020, 2003914
2003914 (2 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
through hydrogen bonding. The supramolecular assembly
process has a significant effect on the synthesis of final photo-
catalyst. On the one hand, the supramolecular intermediate is
employed as a self-template to control the formation of TCN. In
the second step, the as-prepared supramolecular intermediate
is further polymerized under a N2 atmosphere at 550 °C for 4 h.
Due to the similar aromatic structures, carbon rings units can
be in situ incorporated into the conjugated network of TCN via
continuous π-conjugated bond, and named C-TCN.[31–33,35]
The significance of supramolecular assembly process can be
further illustrated from scanning electron microscope (SEM)
and transmission electron microscopy (TEM) images. As illus-
trated in Figure S1a (Supporting Information), the melamine is
random-shaped particles with a wide size distribution. In con-
trast, the supramolecular intermediate exhibits uniform hexag-
onal rod morphology after hydrothermal reaction (Figure S1b,
Supporting Information). The X-ray diffraction (XRD) patterns
(Figure S3a, Supporting Information) and Fourier transform
infrared (FTIR) spectra (Figure S3b, Supporting Information)
also confirm that the melamine has turned into the supramolec-
ular intermediate via molecular self-assembly.[36,37] As shown in
Figure S3a (Supporting Information), the characteristic diffrac-
tion peaks of supramolecular intermediate accord well with the
structure in previous report, the strongest peak at 28.1° assigns
to the graphite-like stacking of sheet-like structure, and the
other peaks at about 11.8°, 17.7°, and 21.9° are attributed to
the in-planar packing in (100), (110), and (200) planes, respec-
tively.[38,39] The above result is further substantiated by FTIR
spectra (Figure S3b, Supporting Information), the peaks located
at 763, 1442, and 1528 cm−1 are attributed to the breathing mode
of triazine units, the stretching of CN-heterocycles, and the
stretching mode of the triazine units, respectively. The peaks
at 1727 and 1778 cm−1 are assigned to the stretching modes of
the carbonyl group, and the bands observed at 3200–3400 cm−1
are attributed to the stretching vibration band of the amino
group.[40] In order to further investigate the important role of
the supramolecular intermediate in determining the mor-
phology of the final product, SEM and TEM analyses on the
C-TCN were carried out (Figure 1b,c and Figure S1c,d, Sup-
porting Information). The Bulk CN derived from melamine
is composed of macrosized particles with irregular blocky
structure (Figure S1c, Supporting Information). In contrast, as
shown in Figure 1b,c and Figure S1d (Supporting Information),
the C-TCN exhibits uniform hexagonal tubular morphology
(diameter: 1–3 µm, length: 10–30 µm), demonstrating that the
C-TCN retains the original 1D structure of the supramolecular
intermediate well. Additionally, as can be seen from the magni-
fied TEM image of the C-TCN in Figure S2a (Supporting Infor-
mation), the C-TCN also possesses plentiful nanoscale pores
on the tubular structure, which might result in larger specific
surface area than the Bulk CN. Such a presumption is veri-
fied by the N2 adsorption–desorption experiments (Figure S2c,
Supporting Information), in which the specific surface area of
the C-TCN (116.8 m2 g−1) is clearly larger than that of the Bulk
CN (8.6 m2 g−1). The high-resolution TEM (HRTEM) image
in Figure 1d exhibits that the C-TCN consists of amorphous
and crystalline regions. The amorphous region belongs to the
disordered triazine units of the g-C3N4, which is the same as
the HRTEM image of the Bulk CN (Figure S2b, Supporting
Information). On the other hand, the crystalline region that fea-
tures a lattice space of 0.35 nm can be assigned to the (002)
plane of the carbon rings.[32] The energy-dispersive X-ray (EDX)
elemental mapping images of the C-TCN exhibit the homoge-
neous distribution of C and N elements over the entire nano-
tube (Figure 1e,f).
XRD, FTIR, synchrotron-based X-ray absorption near-edge
structure (XANES) spectroscopy, Raman, solid-state 13C NMR
spectra, and X-ray photoelectron spectroscopy (XPS) analyses
were used to characterize the chemical composition of the
C-TCN. The XRD patterns are shown in Figure 2a, the C-TCN
exhibits two typical diffraction peaks, which is consistent with
that of the Bulk CN, indicating the C-TCN retains the crystal
structure of g-C3N4. More specifically, the (100) diffraction peak
at around 13.1° originates from the in-plane repeated units of
tri-s-triazine,[41] while the (002) diffraction peak at 27.3° is attrib-
uted to the interlayer stacking of g-C3N4.[42] Compared with
the Bulk CN, the intensities of (002) peak in TCN and C-TCN
become weaker, which is ascribed to the evident size-dependent
properties of tubular structure.[36] The intensity of the (002)
peak in C-TCN tends to be weaker than that of TCN, indicating
the introduction of glucose in calcination process may cause
weaker crystallinity.[31,43] Furthermore, the structural informa-
tion about the C-TCN is further characterized by FTIR analysis
shown in Figure S4a (Supporting Information). No significant
difference is observed between the FTIR absorption peaks of
the C-TCN and Bulk CN, and all the peaks can be assigned to
the basic crystal structure of g-C3N4.[44] Relevant structural char-
acteristics of C-TCN are further performed by C and N K-edge
XANES measurement (Figure 2b,c). In the C K-edge XANES
spectra (Figure 2b), both samples display a typical struc-
 

ture of g-C3N4, which can be indexed to σ*(C N C/C N),
[31,45]
 


π*(C N C), and π*(C C/C C).
Interestingly, C-TCN


exhibits higher peak intensity of π*(C C/C C) than that of
Bulk CN and TCN, which may be derived from the introduction
of carbon rings units. In consideration of the higher electron-
withdrawing property of carbon rings than g-C3N4, the electron
migration from the basal plane of g-C3N4 to carbon rings units
 
could be expected in C-TCN. The peak intensity of π*(C N C)
increases a lot after the incorporation of carbon rings units into
the conjugated network of TCN, indicating a stronger interfa-
cial interaction between carbon rings units and the basal plane
of g-C3N4.[46] Such stronger interfacial interaction can create
a conductive path across the layers to accelerate the electron
transport in C-TCN, which brings efficient charge separation/
transfer and then promotes the photocatalytic performance.[47]
In addition, the N K-edge XANES spectra (Figure 2c), for Bulk
CN, TCN, and C-TCN show four typical features at 399.7, 401.4,
 
402.5, and >405 eV, which can be assigned to π*(C N C),
[45]


 

π*(N 3C), π*(N 3C bridging), and σ*(C N C/C N). The
 
enhancement of π*(C N C) feature for C-TCN suggests a
strengthen interfacial interaction between carbon rings units
and the basal plane of g-C3N4, which is consistent with the
result of C K-edge XANES spectra. This newly formed carbon
rings units can be further verified by the solid-state 13C NMR
spectra (Figure 2d). For Bulk CN, TCN, and C-TCN, two typ-
ical peaks at the chemical shifts of 162.5 and 154.4 ppm are
observed, which are assigned to the basic crystal structure of
g-C3N4.[48] Different from Bulk CN and TCN, two new peaks
2003914 (3 of 10)
© 2020 Wiley-VCH GmbH
Small 2020, 2003914

www.advancedsciencenews.com
www.small-journal.com
Figure 2. a) XRD patterns, b) C K-edge XANES spectra, c) N K-edge XANES spectra, and d) solid-state 13C MAS NMR spectra of C-TCN.

assigned to the incorporated carbon rings units are clearly
found at the chemical shifts of 128.1 and 118.7 ppm, which con-
firms the successful incorporation of carbon rings units into
the basal plane of g-C3N4.[49] Raman spectra in Figure S4b (Sup-
porting Information) are carried out to further prove the pres-
ence of carbon rings units in the C-TCN. Typically, Bulk CN and
TCN show no characteristic absorption bands. In comparison,
two characteristic peaks corresponding to characteristic fea-
tures of crystalline graphite are observed at 1350 and 1574 cm−1
in the C-TCN, confirming that crystalline graphite unit indeed
exists in the C-TCN.[32,50] Furthermore, XPS was conducted to
examine the presence of crystalline graphite unit in the basal
plane of the g-C3N4. XPS survey spectra (Figure S4c, Supporting
Information) and elemental composition (Table S1, Supporting
Information) indicate that C-TCN has the higher C content than
Bulk CN. In order to further determine the origin of increased
C content, the high-resolution C 1s spectra in Figure S4d
(Supporting Information) are deconvoluted into three areas
for π-excitations of the C N conjugated structure.[52] The above
analysis results evidence the successful incorporation of carbon
rings units into the conjugated network of TCN via continuous
π-conjugated bond during pyrolysis.
The incorporation of carbon rings units into the basal plane
of g-C3N4 can dramatically influence the optical property, which
is measured by UV–vis absorption spectra in Figure 3a. Spe-
cifically, an obvious blueshift is observed in the absorption
edge of the TCN with respect to the Bulk CN, which is usually
ascribed to the strong quantum confinement effect arises from
the ultrathin and porous structure of the TCN.[53] Interestingly,
for C-TCN, a redshift emerges in the absorption edge with
regard to the TCN, due to the efficient incorporation of carbon
rings units. Meanwhile, the incorporation of carbon rings units
in the basal plane of g-C3N4 causes the absorption profile fur-
ther stretches into the visible range, making the full range of
visible light usable for photocatalysis. Correspondingly, the
bandgaps of Bulk CN, TCN, and C-TCN are calculated to be
2.78, 2.86, and 2.81 eV by Kubelka–Munk function (Figure 3b).
The valence band (VB) positions of the Bulk CN and C-TCN are
measured by VB XPS spectra (Figure 3c), which are about 1.74,
2.01, and 2.05 V, respectively. Obviously, the C-TCN possesses
the strongest oxidation ability due to the deepest VB level,
which is beneficial for O2 evolution reaction. The outstanding
oxidation ability of the C-TCN is verified by the electron spin
resonance (ESR) shown in Figure S5 (Supporting Informa-
tion). As expected, for Bulk CN, only ESR signals of DMPO-
 
at 284.8, 286.3, and 288.2 eV, which are assigned to N C N,
C NHx, and C C/C C, respectively.[51] Note that C-TCN shows





higher content of C C/C C bonds than Bulk CN (Table S2,
2


Supporting Information). The sp hybridization of C C/C C
bonds is ascribed to crystalline graphite, demonstrating the
formation of carbon rings units in the C-TCN. Meanwhile,
the high-resolution N 1s spectrum (Figure S4e, Supporting
Information) can be deconvoluted into four areas, including
2
 
398.7 eV for sp -hybridized nitrogen (C N C), 400.1 eV for
−

 
the tertiary N bonding to carbon atoms N (C)3 or H N (C)2,
401.1 eV for terminal amino functions (C N H), and 404.4 eV
•O2 but no ESR signal of DMPO-•OH are observed because
the VB potential of the Bulk CN is higher than the potential of
 
Small 2020, 2003914
2003914 (4 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 3. a) UV–vis absorption spectra, b) the estimated bandgaps, c) VB XPS spectra, and d) band alignment of C-TCN.
OH−/•OH (1.99 V).[54] Interestingly, the characteristic signals of
DMPO-•OH are detected in the C-TCN, and the VB potential
of the C-TCN is sufficiently positive for the oxidation of OH− to
photocatalytic performance of C-TCN is significantly improved
in comparison with only using Pt (3%) as the cocatalyst, and
the average H2 and O2 evolution rates are estimated to be 10.23
and 5.04 µmol h−1, respectively (Figure 4b). To our knowledge,
the photocatalytic activity is superior to most of the previous
reported g-C3N4-based photocatalysts (Table S3, Supporting
Information). Moreover, no obvious decay is observed in a 25 h
reaction, which further implies the remarkable photostability
of the C-TCN. The control experiment of photocatalytic overall
water splitting is further performed, and no gases are detected
when using no cocatalytst or CoOx (1%) as the cocatalytst
(Figure S6a, Supporting Information). The influence of the
weight ratio between carbon rings units and TCN was further
evaluated in Figure 4c. After the incorporation of carbon rings
units, the photocatalytic activity is dramatically improved, and
the optimized amount of carbon rings units in our system is
when the addition of glucose reaches 0.01 g. In order to further
prove that the morphological effect endowed by the 1D tubular
structure and D-A system play critical roles in the enhance-
ment of photocatalytic activity, the control experiments were
conducted. Figure S6b (Supporting Information) shows that
no gases are detected for the Bulk CN or TCN photocatalyst,
while C(0.02)/Bulk CN, C(0.02)/TCN, and C-TCN are photo-
catalytically active toward overall water splitting, indicating that
the introduction of carbon rings units makes the above process
more favorable. Furthermore, the C-TCN shows the highest
overall water splitting performance among the three photocata-
lysts, followed by the C(0.02)/TCN and C(0.02)/Bulk CN, which
also implies that the in-plane incorporation of carbon rings
units and 1D tubular structure have a real impact on the water
splitting performance. Furthermore, the external quantum
•
generate OH. Therefore, the above analysis indicates that the
C-TCN has better oxidation ability superior to Bulk CN, infer-
ring that O2 evolution reaction is more likely to happen in the
C-TCN. According to the combined analysis of bandgaps and
VB edges, the CB potentials of the Bulk CN, TCN, and C-TCN
are calculated to be −1.04, −0.85, and −0.76 V, respectively.
Although the CB potential of the C-TCN reveals a positive shift
with regard to the Bulk CN, it is still more negative than the
reduction potential of H+/H2 (−0.41 V vs NHE, pH = 7). The
corresponding band alignments of the Bulk CN and C-TCN are
schematically depicted in Figure 3d.
2.2. Photocatalytic Overall Water Splitting
In view of the excellent photoelectric properties, the C-TCN is
expected to be a promising photocatalyst. First, the photocata-
lytic performance of the C-TCN was evaluated by examining the
efficiency of photocatalytic overall water splitting using Pt (3%)
as the cocatalyst at λ > 400 nm irradiation in pure water, the
photocatalytic water splitting equipment is an on-line photo-
catalytic analysis system (Lbsolar-IIIAG, PerfectLight, Beijing).
As revealed in Figure 4a, the C-TCN exhibits excellent photo-
catalytic activity (2.69 µmol h−1 H2 and 1.30 µmol h−1 O2) with a
H2/O2 ratio of 2:1. Meanwhile, no noticeable attenuation of the
photocatalytic performance is observed in a 25 h reaction, dem-
onstrating that the C-TCN has a promising photostability. As
expected, when CoOx (1%) is further deposited on C-TCN, the
2003914 (5 of 10)
© 2020 Wiley-VCH GmbH
Small 2020, 2003914

www.advancedsciencenews.com
www.small-journal.com
Figure 4. The overall water splitting performance of C-TCN under visible light irradiation: a) using Pt (3%) as a cocatalytst, and b) using Pt (3%) and
CoOx (1%) as the cocatalytsts; c) The overall water splitting performance of C-TCN with different compositions; d) Wavelength dependence of EQE for
photocatalytic overall water splitting using Pt (3%) and CoOx (1%) as the cocatalytsts.
efficiency (EQE) of the C-TCN for photocatalytic overall water
splitting using Pt (3%) and CoOx (1%) as the cocatalytsts was
further measured (Figure 4d). The trend of EQE curve fits
nicely with the UV–vis absorption spectra, suggesting that the
light absorption behavior significantly affects the photocatalytic
performance of the C-TCN. Specifically, C-TCN exhibits a quite
high EQE up to 2.6% at 405 nm, and even the EQE values can
also reach 1.7%, 0.8%, and 0.2% at longer visible light wave-
length of 420, 435, and 450 nm, respectively. To confirm the
source of the generated O2, the H218O isotopic labeling experi-
ment is performed in a ratio H216O/H218O of 4:1. As shown in
Figure S6c (Supporting Information), the peaks at m/z = 32,
m/z = 34, and m/z = 36 in the mass spectra can be assigned
to 16-16O2 (81.7%), 16-18O2 (16.3%), and 18-18O2 (2.0%), respec-
tively. The two types of O2 species (16-18O2 and 18-18O2) further
affirm that the oxygen source of O2 is indeed derived from H2O
caused by photocatalytic overall water splitting.[55,56]
The photocatalytic performance of the C-TCN was further
evaluated by H2 evolution half-reaction. First, the photocata-
lytic H2 evolution was further evaluated without any cocata-
lysts with just the presence of triethanolamine (TEOA) that
acts as the hole sacrificial reagent at λ > 400 nm. As shown
in Figure S7a (Supporting Information), the Bulk CN displays
a negligible H2 evolution rate of 0.01 µmol h−1. Thanks to the
morphological effect conferred by the 1D tubular structure, the
TCN exhibits much higher H2 evolution rate of 1.66 µmol h−1.
After grafting carbon rings units into the basal plane of
g-C3N4, the H2 evolution rate of the C-TCN can be enhanced
significantly. The H2 evolution rate of the C-TCN can even
reach 5.65 µmol h−1, which is 3.4 times as much as that of the
value of the TCN. The improved water splitting performance
stems from the electron-withdrawing effect imposed by the
carbon rings units, which swiftly extract electrons from the
basal plane of the g-C3N4 to participate in H2 evolution reac-
tion after being photoexcited. In order to further illustrate
the role of carbon rings units, 3 wt% Pt is selected as the
cocatalyst. As revealed in Figure S7b (Supporting Informa-
tion), the TCN displays superior photocatalytic performance
of 105.18 µmol h−1, which is ≈11.35 times higher than that of
the Bulk CN (9.27 µmol h−1). The improved water splitting
performance mainly attributes to the 1D tubular structure,
large surface area, and excellent charge separation/transfer
efficiency. After the incorporation of carbon rings units, the
photocatalytic activity of the C-TCN is further improved. The
H2 evolution rate increases up to 221.80 µmol h−1, which is
about 23.93 and 2.11 times as much as that of the Bulk CN and
TCN, respectively. This phenomenon implies that the implan-
tation of carbon rings units into the basal plane of g-C3N4 can
facilitate the photocatalytic H2 evolution. When the addition
of glucose is higher than 0.02 g, the photocatalytic H2 evolu-
tion rate declines. The possible reason is that the presence of
excess carbon rings units reduces the visible light utilization of
the sample and disrupts the crystal structure of g-C3N4. More-
over, the photocatalytic stability test of the C-TCN is shown in
Figure S7c,d (Supporting Information) and no obvious decay
is observed after five cycles in 25 h, which implies the remark-
able photostability of C-TCN. In addition, the EQE values of
C-TCN for H2 evolution half-reaction at 420, 435, 450, and
550 nm are calculated to be 9.3%, 4.8%, 3.2%, and 0%, respec-
tively (Figure S8a, Supporting Information).
Small 2020, 2003914
2003914 (6 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 5. a) Room-temperature PL spectra, b) time-resolved PL spectra, c) Mott–Schottky plots, d) transient photocurrent response, e) EIS Nyquist
plots, and f) the calculated free-energy diagram for hydrogen evolution reaction of C-TCN.
2.3. Possible Photocatalytic Mechanism
Bulk CN, TCN, and C-TCN are 2.38 × 1031, 2.13 × 1031, and
5.96 × 1031 cm−3, respectively. The higher ND value signifies the
enhanced transfer efficiency of photogenerated carriers, which
further confirm that carbon rings units can effectively promote
charge transfer because carbon rings units can effectively avoid
electron accumulation via withdrawing photogenerate electron
from the basal plane of g-C3N4. A similar trend is observed in
photocurrent and EIS measurements, which contributes to
more in-depth understanding on the charge transfer process.
As shown in Figure 5d, all the samples exhibit stable photo-
electrochemical responses under intermittent light irradiation.
The photocurrent intensities of the TCN and C-TCN are all
higher than that of the Bulk CN, and C-TCN has the highest
photocurrent intensity, which further demonstrates that C-TCN
possesses the best charge separation/transfer efficiency. From
the EIS measurement shown in Figure 5e, the arc radius of
the C-TCN is smallest among all the samples, indicating its
lowest charge transfer resistance and further confirming that
the improved charge transport ability conferred by the in-plane
incorporated carbon rings units.[60,61] The optoelectronic perfor-
mance of the C-TCN is further investigated through DFT calcu-
lations (Figure S9, Supporting Information), which is fitting to
the experimental results. It is found that C-TCN exhibits better
refractive index and photoconductivity, which means C-TCN
possesses the better optoelectronic performance and thus
improving the photocatalytic performance.[62] Therefore, com-
bining the characterization and DFT calculations of photoelec-
tricity property shown above, it is concluded that the improved
water splitting performance arises from the charge separation/
transfer capability of the C-TCN after the carbon rings units are
incorporated.
Apart from the bandgap structure, the incorporation of carbon
rings units in the basal plane of the g-C3N4 also promotes the
charge separation/transfer efficiency, which is mainly driven by
the intramolecular charge transfer in a D-A system. The charge
carrier behavior can be measured by photoluminescence (PL)
spectra, time-resolved PL spectra, Mott–Schottky plots, photo-
current measurement, and electrochemical impedance spec-
troscopy (EIS).[57] As shown in Figure 5a, the emission peak
of the TCN exhibits a slight blueshift with respect to the Bulk
CN, and a gradual redshift is observed in the C-TCN with the
incorporation of carbon rings units, which is in accordance
with the result of UV–vis absorption spectra. Furthermore, the
PL intensities of the TCN and C-TCN are lower than that of
the Bulk CN, confirming that the recombination rate of the
photogenerated charge carriers can be efficiently suppressed
by the 1D tubular structure and the introduction of carbon
rings units.[58] The time-resolved PL spectra are also conducted
in Figure 5b. The C-TCN shows longer lifetime (7.84 ns) com-
pared with Bulk CN (3.29 ns) and TCN (5.07 ns), suggesting
the improved charge transfer efficiency. To further verify the
improved charge transfer efficiency, the Mott–Schottky plots are
measured, as shown in Figure 5c. The carrier density (ND) is
obtained by the Mott–Schottky equation
−1
ND = 2e−1ε−1ε0−1 d C−2 / dV
(1)
(
)
0
where e0 is the elemental charge value (e0 = 1.60 × 10−19 C), ε
is the relative permittivity of CN (ε = 5.5), and ε0 is the per-
mittivity of vacuum (ε0 = 8.85 × 10−14).[59] The ND values of the
2003914 (7 of 10)
© 2020 Wiley-VCH GmbH
Small 2020, 2003914

www.advancedsciencenews.com
www.small-journal.com
Figure 6. Differential charge density map of C-TCN: a) the Bader charge between the basal plane of g-C3N4 and carbon rings units, b) the Bader charge
of two adjacent tubes (Δq is the Bader charge change of the TCN, the yellow and blue regions indicate electron accumulation and depletion, respec-
tively); c) Schematic illustration of the proposed mechanism for photocatalytic overall water splitting about the all-organic D-A system.
Furthermore, Gibbs free energy change of H* adsorption
(ΔGH*) is considered to be an important descriptor to evaluate the
H2 evolution performance, which is calculated by DFT calcula-
tions.[63–65] As shown in Figure 5f and Table S4 (Supporting Infor-
mation), for Bulk CN and TCN, the calculated ΔGH* values are
−0.41 and −0.45 eV. Dramatically, after grafting carbon rings units
into the basal plane of TCN, the ΔGH* value of C-TCN is reduced
to −0.25 eV, which is very close to that of Pt (−0.09 eV), indicating
that the H2 evolution reaction activity of C-TCN is greatly pro-
moted. In order to explain the existence of intramolecular charge
transfer between the basal plane of g-C3N4 and carbon rings
units, the Bader charge of C-TCN is conducted by DFT calcula-
tions. For clarity, Figure 6a shows the Bader charge between the
basal plane of g-C3N4 and carbon rings units in the structure of
C-TCN, the negative value of the calculated Bader charge change
means that the electrons can transfer from the basal plane of
g-C3N4 to the carbon rings units, signifying the existence of intra-
molecular charge transfer between them. On the other hand, the
Bader charge of two adjacent tubes is close to zero (Figure 6b),
indicating that there is almost no charge transfer.
Based on above analysis and discussion, a possible photocat-
alytic mechanism to elucidate the overall water splitting process
on the C-TCN is proposed in Figure 6c. First, a D-A system is
formed via the effective incorporation of carbon rings units in
the basal plane of the g-C3N4, which is the major driving force for
intramolecular charge transfer. Meanwhile, the energy bands of
the C-TCN match well with the requirements of photocatalytic
overall water splitting. Once the C-TCN is exposed to the light,
the photogenerated electron–hole pairs in the basal plane of the
g-C3N4 are separated. Due to the strong electron-withdrawing
property, carbon rings units can serve as the electron acceptor
to allow a smooth transfer of the excited electrons in the CB
from the basal plane of g-C3N4 to the carbon rings units. Such
an intramolecular charge transfer can effectively suppress the
recombination of the photogenerated electron–hole pairs. After-
ward, the electrons that are being transferred to the carbon ring
reduce H2O to H2 while the photogenerated holes react with
H2O to produce O2. In this process, carbon rings units that
act as the electron acceptor are responsible for inhibiting rapid
recombination of the photogenerated electron–hole pairs, thus
optimizing the water splitting performance.
3. Conclusions
In summary, we have successfully built an all-organic D-A
system to optimize the photocatalytic overall water splitting
activity through the implantation of carbon rings units into the
basal plane of g-C3N4 via continuous π-conjugated bond. Owing
to the effective intramolecular charge transfer of the D-A
system, the C-TCN exhibits enhanced charge separation capa-
bility, which contributes to its remarkable photocatalytic per-
formance (EQE = 1.7% at 420 nm). Moreover, the VB position
can be effectively downshifted due to the tubular structure and
the incorporation of carbon rings units, which can bypass the
limit of water oxidation reaction. Therefore, the all-organic D-A
system can efficiently execute water splitting reaction without
any sacrificial reagents. This work successfully extends the con-
cept of D-A system usually employed in organic solar cells to
photocatalysis field.
Small 2020, 2003914
2003914 (8 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
[13] S. Cao, T. Chan, Y. Lu, X. Shi, B. Fu, Z. Wu, H. Li, K. Liu, S. Alzuabi,
P. Cheng, M. Liu, T. Li, X. Chen, L. Piao, Nano Energy 2020, 67,
104287.
[14] J. Di, C. Chen, S. Z. Yang, S. Chen, M. Duan, J. Xiong, C. Zhu,
R. Long, W. Hao, Z. Chi, H. Chen, Y. X. Weng, J. Xia, L. Song, S. Li,
H. Li, Z. Liu, Nat. Commun. 2019, 10, 2840.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[15] P. Kumar, E. Vahidzadeh, U. K. Thakur, P. Kar, K. M. Alam,
A. Goswami, N. Mahdi, K. Cui, G. M. Bernard, V. K. Michaelis,
K. Shankar, J. Am. Chem. Soc. 2019, 141, 5415.
[16] G. Zhang, Z. A. Lan, X. Wang, Chem. Sci. 2017, 8, 5261.
[17] X. Chen, R. Shi, Q. Chen, Z. Zhang, W. Jiang, Y. Zhu, T. Zhang,
Nano Energy 2019, 59, 644.
Acknowledgements
This study was supported by National Natural Science Foundation
of China (21776118, 21878134, 51902138), Natural Science Foundation
of Jiangsu Province (BK20190981, BK20190835), Jiangsu Fund for
Distinguished Young Scientists (BK20190045), China Postdoctoral
Science Foundation (2019M661765), High-tech Research Key laboratory
of Zhenjiang (SS2018002), and A Project Funded by the Priority Academic
Program Development of Jiangsu Higher Education Institutions, the
high-performance computing platform of Jiangsu University.
[18] Z. Teng, N. Yang, H. Lv, S. Wang, M. Hu, C. Wang, D. Wang,
G. Wang, Chem 2019, 5, 664.
[19] J. Qi, W. Zhang, R. Cao, Adv. Energy Mater. 2018, 8, 1701620.
[20] X. W. Zhu, J. M. Yang, X. J. She, Y. H. Song, J. C. Qian, Y. Wang,
H. Xu, H. M. Li, Q. Y. Yan, J. Mater. Chem. A 2019, 7, 5209.
[21] T. Takata, K. Domen, ACS Energy Lett. 2019, 4, 542.
[22] H. Ou, C. Tang, X. Chen, M. Zhou, X. Wang, ACS Catal. 2019, 9,
2949.
Conflict of Interest
The authors declare no conflict of interest.
[23] Y. Zhang, L. Wu, X. Zhao, Y. Zhao, H. Tan, X. Zhao, Y. Ma, Z. Zhao,
S. Song, Y. Wang, Y. Li, Adv. Energy Mater. 2018, 8, 1801139.
[24] G. Zhang, G. Li, Z. A. Lan, L. Lin, A. Savateev, T. Heil, S. Zafeiratos,
X. Wang, M. Antonietti, Angew. Chem., Int. Ed. 2017, 56, 13445.
[25] J. Fu, J. Yu, C. Jiang, B. Cheng, Adv. Energy Mater. 2018, 8, 1701503.
[26] L. Chen, J. Song, Adv. Funct. Mater. 2017, 27, 1702695.
[27] P. Roy, A. Jha, V. B. Yasarapudi, T. Ram, B. Puttaraju, S. Patil,
J. Dasgupta, Nat. Commun. 2017, 8, 1716.
Author Contributions
Z.M. and J.D. contributed equally to this work. Z.M., J.D., H.X., and
Q.Y. designed the experiments; Z.M., P.Y., C.L., D.L., and X.Z. carried out
the experiments; Y.S., C.L., Q.Y., H.L., and Y.L. analyzed the experimental
results; Z.M. wrote the manuscript.
[28] J. Lee, S. M. Lee, S. Chen, T. Kumari, S. H. Kang, Y. Cho, C. Yang,
Adv. Mater. 2019, 31, 1804762.
[29] D. Dang, D. Yu, E. Wang, Adv. Mater. 2019, 31, 1807019.
[30] H. Ou, X. Chen, L. Lin, Y. Fang, X. Wang, Angew. Chem., Int. Ed.
2018, 57, 8729.
Keywords
[31] W. Che, W. Cheng, T. Yao, F. Tang, W. Liu, H. Su, Y. Huang, Q. Liu,
J. Liu, F. Hu, Z. Pan, Z. Sun, S. Wei, J. Am. Chem. Soc. 2017, 139,
3021.
[32] W. Niu, K. Marcus, L. Zhou, Z. Li, L. Shi, K. Liang, Y. Yang, ACS
Catal. 2018, 8, 1926.
[33] A. Du, S. Sanvito, Z. Li, D. Wang, Y. Jiao, T. Liao, Q. Sun, Y. H. Ng,
Z. Zhu, R. Amal, S. C. Smith, J. Am. Chem. Soc. 2012, 134, 4393.
[34] Y. Gong, J. Lin, X. Wang, G. Shi, S. Lei, Z. Lin, X. Zou, G. Ye,
R. Vajtai, B. I. Yakobson, H. Terrones, M. Terrones, B. K. Tay, J. Lou,
S. T. Pantelides, Z. Liu, W. Zhou, P. M. Ajayan, Nat. Mater. 2014, 13,
1135.
[35] C. Huang, C. Chen, M. Zhang, L. Lin, X. Ye, S. Lin, M. Antonietti,
X. Wang, Nat. Commun. 2015, 6, 7698.
[36] Z. Mo, H. Xu, Z. Chen, X. She, Y. Song, J. Wu, P. Yan, L. Xu, Y. Lei,
S. Yuan, H. Li, Appl. Catal., B 2018, 225, 154.
[37] Y. Xiao, G. Tian, W. Li, Y. Xie, B. Jiang, C. Tian, D. Zhao, H. Fu,
J. Am. Chem. Soc. 2019, 141, 2508.
all-organic donor–acceptor system, charge separation, g-C3N4, overall
water splitting, photocatalysis
Received: June 28, 2020
Revised: September 25, 2020
Published online:
[1] C. F. Fu, X. Wu, J. Yang, Adv. Mater. 2018, 30, 1802106.
[2] S. Chen, T. Takata, K. Domen, Nat. Rev. Mater. 2017, 2, 17050.
[3] Q. Wang, K. Domen, Chem. Rev. 2020, 120, 919.
[4] J. Tournet, Y. Lee, S. K. Karuturi, H. H. Tan, C. Jagadish, ACS Energy
Lett. 2020, 5, 611.
[5] W. Liu, L. Cao, W. Cheng, Y. Cao, X. Liu, W. Zhang, X. Mou, L. Jin,
X. Zheng, W. Che, Q. Liu, T. Yao, S. Wei, Angew. Chem., Int. Ed.
2017, 56, 9312.
[6] S. Cao, H. Li, T. Tong, H.-C. Chen, A. Yu, J. Yu, H. M. Chen, Adv.
Funct. Mater. 2018, 28, 1802169.
[38] M. Shalom, S. Inal, C. Fettkenhauer, D. Neher, M. Antonietti, J. Am.
Chem. Soc. 2013, 135, 7118.
[7] J. Di, J. Xia, M. F. Chisholm, J. Zhong, C. Chen, X. Cao, F. Dong,
Z. Chi, H. Chen, Y. X. Weng, J. Xiong, S. Z. Yang, H. Li, Z. Liu,
S. Dai, Adv. Mater. 2019, 31, 1807576.
[8] J. Yi, W. El-Alami, Y. Song, H. Li, P. M. Ajayan, H. Xu, Chem. Eng. J.
2020, 382, 122812.
[9] Y. Wan, L. Wang, H. Xu, X. Wu, J. Yang, J. Am. Chem. Soc. 2020, 142, 4508.
[10] L. Wang, Y. Zhang, L. Chen, H. Xu, Y. Xiong, Adv. Mater. 2018, 30,
1801955.
[11] M. Xiao, Z. Wang, M. Lyu, B. Luo, S. Wang, G. Liu, H. M. Cheng,
L. Wang, Adv. Mater. 2018, 31, 1801369.
[39] S. Guo, Z. Deng, M. Li, B. Jiang, C. Tian, Q. Pan, H. Fu, Angew.
Chem., Int. Ed. 2016, 55, 1830.
[40] P. Meng, A. Brock, Y. Xu, C. Han, S. Chen, C. Yan, J. McMurtrie,
J. Xu, J. Am. Chem. Soc. 2020, 142, 479.
[41] C. Lv, Y. Qian, C. Yan, Y. Ding, Y. Liu, G. Chen, G. Yu, Angew. Chem.,
Int. Ed. 2018, 57, 10246.
[42] S. Wang, L. Chen, X. Zhao, J. Zhang, Z. Ao, W. Liu, H. Wu, L. Shi,
Y. Yin, X. Xu, C. Zhao, X. Duan, S. Wang, H. Sun, Appl. Catal., B
2020, 278, 119312.
[43] Y. Yu, W. Yan, X. Wang, P. Li, W. Gao, H. Zou, S. Wu, K. Ding, Adv.
Mater. 2018, 30, 1705060.
[44] H. Xu, X. She, T. Fei, Y. Song, D. Liu, H. Li, X. Yang, J. Yang, H. Li,
L. Song, P. M. Ajayan, J. Wu, ACS Nano 2019, 13, 11294.
[12] Y. Wang, A. Vogel, M. Sachs, R. S. Sprick, L. Wilbraham,
S. J. A. Moniz, R. Godin, M. A. Zwijnenburg, J. R. Durrant,
A. I. Cooper, J. Tang, Nat. Energy 2019, 4, 746.
2003914 (9 of 10)
© 2020 Wiley-VCH GmbH
Small 2020, 2003914

www.advancedsciencenews.com
www.small-journal.com
[45] Z. Mo, X. Zhu, Z. Jiang, Y. Song, D. Liu, H. Li, X. Yang, Y. She, Y. Lei,
S. Yuan, H. Li, L. Song, Q. Yan, H. Xu, Appl. Catal., B 2019, 256,
117854.
[46] K. Feng, J. Zhong, B. Zhao, H. Zhang, L. Xu, X. Sun, S. Lee, Angew.
Chem., Int. Ed. 2016, 55, 11950.
[47] D. Zhao, C. Dong, B. Wang, C. Chen, Y. Huang, Z. Diao, S. Li,
L. Guo, S. Shen, Adv. Mater. 2019, 31, 1903545.
[48] P. Xia, M. Antonietti, B. Zhu, T. Heil, J. Yu, S. Cao, Adv. Funct. Mater.
2019, 29, 1900093.
[55] B. Qiu, L. Cai, N. Zhang, X. Tao, Y. Chai, Adv. Sci. 2020, 7, 1903568.
[56] H. A. Vignolo-González, S. Laha, A. Jiménez-Solano, T. Oshima,
V. Duppel, P. Schützendübe, B. V. Lotsch, Matter 2020, 3, 464.
[57] Y. Yang, G. Zeng, D. Huang, C. Zhang, D. He, C. Zhou, W. Wang,
W. Xiong, B. Song, H. Yi, S. Ye, X. Ren, Small 2020, 16, 2001634.
[58] X. W. Zhu, S. Q. Huang, Q. Yu, Y. B. She, J. M. Yang, G. L. Zhou,
Q. D. Li, X. J. She, J. J. Deng, H. M. Li, H. Xu, Appl. Catal., B 2020,
269, 118760.
[59] S. Sun, G. Shen, J. Jiang, W. Mi, X. Liu, L. Pan, X. Zhang, J. J. Zou,
Adv. Energy Mater. 2019, 9, 1901505.
[60] J. Xi, H. Xia, X. Ning, Z. Zhang, J. Liu, Z. Mu, S. Zhang, P. Du,
X. Lu, Small 2019, 15, 1902744.
[49] J. Zhang, G. Zhang, X. Chen, S. Lin, L. Mohlmann, G. Dolega,
G. Lipner, M. Antonietti, S. Blechert, X. Wang, Angew. Chem., Int.
Ed. 2012, 51, 3183.
[50] Z. Chen, T. T. Fan, X. Yu, Q. L. Wu, Q. H. Zhu, L. Z. Zhang, J. H. Li,
W. P. Fang, X. D. Yi, J. Mater. Chem. A 2018, 6, 15310.
[51] Z. Wei, M. L. Liu, Z. J. Zhang, W. Q. Yao, H. W. Tan, Y. F. Zhu, Energy
Environ. Sci. 2018, 11, 2581.
[61] J. Di, C. Zhu, M. Ji, M. Duan, R. Long, C. Yan, K. Gu, J. Xiong,
Y. She, J. Xia, H. Li, Z. Liu, Angew. Chem., Int. Ed. 2018, 57, 14847.
[62] C. F. Fu, R. Zhang, Q. Luo, X. Li, J. Yang, J. Comput. Chem. 2019, 40,
980.
[52] Y. Wang, P. Du, H. Pan, L. Fu, Y. Zhang, J. Chen, Y. Du, N. Tang,
G. Liu, Adv. Mater. 2019, 31, 1807540.
[53] W. Xing, W. Tu, Z. Han, Y. Hu, Q. Meng, G. Chen, ACS Energy Lett.
2018, 3, 514.
[54] X. She, J. Wu, H. Xu, J. Zhong, Y. Wang, Y. Song, K. Nie, Y. Liu,
Y. Yang, M. Rodrigues, R. Vajtai, J. Lou, D. Du, H. Li, P. Ajayan, Adv.
Energy Mater. 2017, 7, 1700025.
[63] Y. Wang, Y. Ma, X. B. Li, L. Gao, X. Y. Gao, X. Z. Wei, L. P. Zhang,
C. H. Tung, L. Qiao, L. Z. Wu, J. Am. Chem. Soc. 2020, 142, 4680.
[64] X. Zhang, H. Liang, H. Li, Y. Xia, X. Zhu, L. Peng, W. Zhang,
L. Liu, T. Zhao, C. Wang, Z. Zhao, C. T. Hung, M. M. Zagho,
A. A. Elzatahry, W. Li, D. Zhao, Angew. Chem., Int. Ed. 2019, 59, 2.
[65] Z. Luo, Y. Ouyang, H. Zhang, M. Xiao, J. Ge, Z. Jiang, J. Wang,
D. Tang, X. Cao, C. Liu, W. Xing, Nat. Commun. 2018, 9, 2120.
Small 2020, 2003914
2003914 (10 of 10)
© 2020 Wiley-VCH GmbH