Localized NiS2 Quantum Dots on g-C3N4 Nanosheets for Efficient Photocatalytic Hydrogen Production from Water

Article abstract of DOI:10.1002/cctc.201801510

Developing high-efficiency yet low-cost photocatalyst for solar hydrogen production by avoiding the use of noble metals has received adequate interest but remains a great challenge to date. This work reports a seed-mediated hydrothermal approach for the synthesis of NiS₂ quantum dots (QDs) anchored two-dimensional graphitic carbon nitride (g-C₃N₄) nanosheets. This hybrid shows superior performance toward photocatalytic H₂ evolution from water. The highest H₂ evolution rate reaches 4.841 μmol h^?1, with an apparent quantum efficiency of 2 % at 425 nm, which is even much higher than that of Pt-modified g-C₃N₄ photocatalyst (2.865 μmol h^?1). Moreover, the composite presents good stability without notable activity decay after several cycled tests. It is found that NiS₂ QDs are essential for this improvement. These small nanoclusters not only benefit rapid and vectorial diffusion of photogenerated electrons from g-C₃N₄ to NiS₂, but also promote H₂ evolution by decreasing the thermodynamic overpotential for proton reduction. This work thus marks an important step toward designing good-performance and low-cost photocatalytic materials for solar H₂ conversion.

Full text of DOI:10.1002/cctc.201801510

Accepted Article
Title: Localized NiS2 Quantum Dots on g-C3N4 Nanosheets for
Efficient Photocatalytic Hydrogen Production from Water
Authors: Fei Xue, Maochang Liu, Cheng Cheng, Junkai Deng, and
Jinwen Shi
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201801510
Link to VoR: http://dx.doi.org/10.1002/cctc.201801510
A Journal of
www.chemcatchem.org

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
Localized NiS2 Quantum Dots on g-C3N4 Nanosheets for Efficient
Photocatalytic Hydrogen Production from Water
Fei Xue,[a] Maochang Liu,*[a][b] Cheng Cheng,[a] Junkai Deng,[c] and Jinwen Shi[a]
Abstract: Developing high-efficiency yet low-cost photocatalyst for
solar hydrogen production by avoiding the use of noble metals has
received adequate interest but remains a great challenge to date. This
formed between the cocatalyst and g-C3N4 not only facilitates the
rapid separation of photogenerated charge carriers but also
provides numerous reactive sites to promote H2 generation. It is
worth being reminded that the noble metals such as Pt, Au, Ag,
etc., are underlying cocatalyst candidates for enhancing the
photocatalytic activity of g-C3N4.[5,17-20] Despite their extraordinary
activity, the natural scarcity and preciousness of these noble
metals largely limit their practicability for this technology. As a
result, it is extremely attractive to explore alternative noble-metal-
free cocatalysts that are sufficiently active. To date, tremendous
earth-abundant transition metal compounds have been
developed as cocatalysts including Ni, NiSx (x=1 or 2), MoS2, WS2,
Ni12P5, Ni2P, Ni(OH)2, etc., for boosting the photocatalytic H2-
evolution activity of g-C3N4.[10,11,21-29]
Very recently, transition metal sulfides have received
intensive attention as active species for photocatalytic H2
evolution.[10,21-23,30] Nonetheless, the metal sulfides decorated g-
C3N4 are usually obtained via a solution-deposition strategy under
mechanical stirring.[10,27,28] The insufficient contact between g-
C3N4 and the cocatalyst taken by this method will lead to a
weakened interfacial interaction. As a result, the stability of the
composite as well as the associated charge transfer behavior has
been significantly restricted. In comparison with the mechanical
deposition method, hydrothermal approach has been
demonstrated to be versatile to fabricate high-quality metal sulfide
nanoparticles with finely tuned size and morphology.[14,21,31] It is
thus reasonable to expect integration of well-defined metal
sulfides onto g-C3N4 surface with enhanced physicochemical
property via hydrothermal synthesis.
Herein, we report a facile two-step approach, involving
thermal polymerization and subsequent seed-mediated mild
hydrothermal treatment, for embedding uniform NiS2 quantum
dots (QDs) onto the surface of g-C3N4 nanosheet. The composite
shows excellent and enhanced photocatalytic performance
toward H2 generation in an aqueous solution containing
triethanolamine (TEOA) under visible irradiation. This
amelioration is mainly attributed to the promoted charge
separation efficiency at the solid interface as well as the
accelerated proton reduction rate localized on NiS2 QDs. It is
expected that this work could inspire ongoing interest by
extending the use of noble-metal-free cocatalyst to other highly
efficient photocatalytic systems.
work reports
a seed-mediated hydrothermal approach for the
synthesis of NiS2 quantum dots (QDs) anchored two-dimensional
graphitic carbon nitride (g-C3N4) nanosheets. This hybrid shows
superior performance toward photocatalytic H2 evolution from water.
The highest H2 evolution rate reaches 4.841 µmol h-1, with an
apparent quantum efficiency of 2% at 425 nm, which is even much
higher than that of Pt-modified g-C3N4 photocatalyst (2.865 µmol h-1).
Moreover, the composite presents good stability without notable
activity decay after several cycled tests. It is found that NiS2 QDs are
essential for this improvement. These small nanoclusters not only
benefit rapid and vectorial diffusion of photogenerated electrons from
g-C3N4 to NiS2, but also promote H2 evolution by decreasing the
thermodynamic overpotential for proton reduction. This work thus
marks an important step toward designing good-performance and
low-cost photocatalytic materials for solar H2 conversion.
Introduction
Semiconductor-based photocatalytic H2 evolution from water
splitting has been deemed to be one of the most promising
technologies to mitigate the energy and environmental crises.[1-4]
Up to now, the reasonable design of sustainable and efficient
visible-light-response photocatalysts is still a challenging mission.
As an economical polymer semiconductor, graphitic carbon
nitride (g-C3N4) has been identified as one of the most arresting
photocatalytic materials on account of its nontoxicity, splendid
endurance, appropriate band structure, convenient availability,
and so forth.[5-9] However, high recombination rate of
photoinduced charges as well as the limited capacity for H2
evolution inevitably hinders its large-scale application for solar
water reduction.[10-12] The last decade has witnessed significant
progress by combining g-C3N4 with an effective cocatalyst to
overcome these shortages.[13-16] This kind of heterojunction
[a]
F. Xue, Prof. M. Liu, C. Cheng, Prof. J. Shi
International Research Center for Renewable Energy & State Key
Laboratory of Multiphase Flow in Power Engineering,
Xi'an Jiaotong University
Xi'an, Shaanxi 710049, P. R. China
E-mail: maochangliu@mail.xjtu.edu.cnF.
Prof. M. Liu
Suzhou Academy of Xi’an Jiaotong University,
Xi’an Jiaotong University
Suzhou, Jiangsu 215123, P. R. China
Prof. J. Deng
State Key Laboratory for Mechanical Behavior of Materials,
Xi'an Jiaotong University
[b]
[c]
Results and Discussion
As characterized by TEM, the bulk g-C3N4 after the thermal
polymerization can be easily exfoliated into nanosheet by the
ultrasonic treatment (Figure S1). While the bulk g-C3N4 showed a
clear three-dimensional stacked structure, the nanosheet was
found with uniform two-dimensional layered structure. The
thicknesses of the bulk g-C3N4 and nanosheet were about 9.0 and
Xi'an, Shaanxi 710049, P. R. China
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
3.6 nm, respectively, as determined by the atomic force
microscope (AFM) (Figure S2). The ultra-thin feature of the
nanosheets makes them at short-range entanglement, leading to
the formation of regular crimp wave (Figure S1b, as marked by
arrows). Moreover, the approximately twofold activity toward
photocatalytic hydrogen production of the g-C3N4 nanosheet to
that of the bulk counterpart provided a direct proof to certify the
superiority of the exposed surface (Figure S3). This activity
together with the subsequent hydrothermal treatment enables
strong interaction between g-C3N4 and NiS2 via seed-mediated
hetero-nucleation. Despite the amounts of Ni precursor were
employed, the precise NiS2 weight percentages in final
synthesized g-C3N4-NiS2 composites were further determined by
ICP-MS measurement, as depicted in Figure S4. Figure 1, a-c
represents the typical ordinary, bright-field, and large-
magnification TEM images of the hybrid that contains 25.3 wt%
NiS2. Clearly, while the nanosheet structures were largely
maintained, many monodispersed QDs with a mean size (in
diameter) of 2.62 nm (see Figure S5 for statistic results) were
found to be anchored onto the surface of g-C3N4 nanosheets. As
per the synthetic procedure, it is reasonable to speculate these
QDs are NiS2. To determine the nanostructure at atomic level, the
composite was then investigated by high resolution TEM (HRTEM,
see Figure 1d). Different from g-C3N4 nanosheet, the QDs were
found with obvious lattice fringes. The interplanar spacings of 2.00
Å and 2.30 Å could be indexed to the characteristic NiS2 {220}
and {211} diffractions, respectively. Moreover, the intimate
interfacial contact between the QDs and g-C3N4 surface was also
clearly demonstrated. These NiS2 QDs are expected to get the
utmost out of the photogenerated electrons from g-C3N4, thus
fundamentally facilitating the H2-evolution kinetics. Note that
when the content of NiS2 exceeds a critical value, monolithic bulk
structure instead of QDs will emerge (Figure S6). This structural
transformation stems from limited capacity of entangled g-C3N4 to
prevent the fusion and further crystallization of NiS2 particles in
the case of plentiful NiS2 monomer. The hybrid was further
investigated by the high angle annular dark field-scanning TEM
(HAADF-STEM) and corresponding EDX elemental mapping
analyses as shown in Figure 1, e-i. The uniform distribution
densities of C and N certified the nanosheet structure of g-C3N4.
For S and Ni, notable deviation in elemental scope was observed.
While the same distribution of S and Ni explicitly illustrates the
formation of NiS2, it also indicates quantum sized NiS2
nanoclusters localized on the surface of g-C3N4 nanosheets. In
addition to those well-resolved larger nanoparticles, many sub-
nanometer, or even atomic sized NiS2 structures were also
noticed. In fact, these sub-structures are hardly to be resolved by
traditional TEM method, but could still be characterized by STEM.
These results provide direct evidence for the successful
fabrication of the binary g-C3N4/NiS2 composite.
Figure 1. (a) Typical low-magnification, (b) bright-field, and (c) high-
magnification TEM images of g-C3N4/25.3%-NiS2 nanodots composite
photocatalyst; (d) HRTEM images of g-C3N4/25.3%-NiS2 nanodots composite
photocatalyst; (e)-(i) STEM-EDX elemental mappings of g-C3N4/25.3%-NiS2
nanodots composite photocatalyst. The scale bars in (e)-(i) are 20 nm.
respectively. Our result is consistent with the previously reported
g-C3N4 synthesized via similar approach.[10-12,18,19,25,26] Generally,
the position of the diffraction peak of g-C3N4 kept unchanged after
incorporation with NiS2, indicating the prominent property of the
nanosheet to maintain the graphitic-packing layer structure.
Further examination reveals that the diffraction intensity of g-C3N4
gradually decreases with the increment of NiS2 content. In
principle, it should be the marked inhibition effect of NiS2 QDs on
the two-dimensional structure that induces this diffraction
reduction.[24] Moreover, hydrothermal treatment in the presence
of NH3·H2O could induce more defect on the surface g-C3N4,
which also contributes to the reduction of the peak intensity. In
contrast, the diffraction peaks, pertaining to the characteristic
diffraction pattern of standard cubic NiS2 (JCPDS 01-089-
3059),[21,32] become more evident when the content of NiS2
increased to be more than 12.2 wt%. No NiS2 diffraction peaks
could be discerned below this content. These characteristic
diffraction peaks of both g-C3N4 and NiS2 in XRD profile
demonstrate the successful fabrication of g-C3N4/NiS2 composites.
To acquire insight into the chemical state and composition of
g-C3N4/NiS2 samples, XPS analysis was then employed (see
Figure S7 for the survey spectra). Figure 3 shows the high-
resolution XPS spectra of C 1s, N 1s, Ni 2p, and S 2p of g-
C3N4/NiS2 composite sample. For simplicity, we took the hybrid
containing 25.3 wt% NiS2 as an example and pristine g-C3N4 as a
comparison. The identical profiles of C 1s and N 1s spectra in g-
C3N4/NiS2 composite and pristine g-C3N4 again provide a direct
proof of the slight impact taken by hydrothermal treatment (see
We further examined the crystal structure of the as-prepared
g-C3N4/NiS2 hybrids by powder XRD. As depicted in Figure 2,
despite the poor crystallinity of g-C3N4, two distinct peaks
centered at 13.2o and 27.5o were observed, which could be
indexed to the (100) and (002) diffraction. Structurally, the two
diffraction peaks stem from the in-plane ordering of tri-s-triazine
motifs and periodic interlayer stacking of aromatic segments,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
Figure 2. XRD patterns of g-C3N4/NiS2 nanodots composites with different NiS2
amounts.
Figure 3. High-resolution XPS core-level spectra of (a) C 1s, (b) N 1s, (c) Ni 2p,
and (d)
photocatalyst, respectively.
S 2p for pristine g-C3N4 and g-C3N4/25.3%-NiS2 composite
Figure 3, a and b). Three peaks centered at 284.8, 286.2, and
288.3 eV are obtained for the C 1s spectrum after deconvolution.
They are typical adventitious C species from the graphitic carbon,
sp2 hybridized C connected to -NH2, and sp2 C bonded to N in
aromatic ring (N-C=N), respectively.[10,11,19,25] High-resolution N 1s
scan can be fit into one dominant peak and three tiny ones with
the binding energies of 398.5, 399.7, 401.2, and 404.2 eV,
respectively. These signals derive from the sp2 N atoms in triazine
units, bridging N in N-(C)3, N in the heterocycles and cyano
groups (C-N-H), and -excitation in the polymeric g-C3N4
structures, respectively.[10,11,19,25] For Ni 2p spectrum (Figure 3c),
vibration at of tri-s-triazine units,[19,26,35] indicating a more ordered
in-plane structural packing motif of g-C3N4. Significantly, the FTIR
scan without showing notable signals that represent S-N or S-C
groups, suggests that no sulfide is doped into the structure of g-
C3N4. Our FTIR results, on one hand, verify the well-maintained
g-C3N4 nanosheet structure; on the other hand, demonstrate the
intrinsic QD behavior of NiS2.
two typical peaks at 855.9 eV (Ni 2p3/2) and 874.1 eV (Ni 2p1/2
)
accompanied with two obvious satellite peaks are presented,
certified the Ni(II) state.[32-34] This chemical state was further
validated by the high-resolution S 2p scan, in which two typical
(S2)2- peaks located at 162.7 and 163.9 eV were observed (see
Figure 3d).[32-34] Moreover, another conspicuous peak observed at
the range of 166-172 eV can be assigned to the oxidized metal
sulfide species,[33,34] which is in accordance with the Ni 2p satellite
signal shown in Figure 3c.
In addition to XPS, g-C3N4, NiS2, and g-C3N4/NiS2
composites were also investigated by FTIR. As shown in Figure
4, the adsorption band of NiS2 has been largely covered by that
of g-C3N4. In this case, we did not observe any adsorption bands
that represented to NiS2 in the hybrid even 56.5 wt% NiS2 was
introduced. It also leads to the same FTIR spectra of pure g-C3N4
and typical g-C3N4/NiS2. Taken C3N4/NiS2 shown in Figure 4a as
an example, the broad bands at around 3100-3400 cm-1 can be
assigned to the stretching vibrational modes of the N-H, which are
generated by the uncondensed terminal amino and -OH
groups.[10,11,19,24-26] The peaks ranging from 1200 to 1700 cm-1 are
typically a set of stretching vibration modes of aromatic heptazine-
derived repeating units, including the typical sp2 C=N stretching
modes and out-of-plane bending of the sp3 C-N bonds.[10,19,24,26]
Moreover, the characteristic absorption peak centered at a
smaller wavenumber, i.e., 810 cm-1, should be the breathing
Figure 4. (a) FTIR spectra of NiS2, pristine g-C3N4, and g-C3N4/25.3%-NiS2
composite. (b) FTIR spectra of g-C3N4/NiS2 nanodots composites with different
NiS2 contents.
The textural properties such as surface area and porous
structure of pristine g-C3N4 and g-C3N4/25.3%-NiS2 composites
were also characterized by nitrogen adsorption-desorption
isotherms. As clearly depicted in Figure 5, both isotherms, having
typical type-IV profiles with the distinctive H3 hysteresis loops,[36]
imply the predominant mesoporous character of the two
photocatalysts. Significantly, the lower N2 uptake for g-
C3N4/25.3%-NiS2 compared with that of pristine g-C3N4 indicates
weakened textural behavior after the hydrothermal treatment
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
which yet yields the narrowed pore size distribution (mean/peak
pore diameter: 14.94/2.78 and 4.26/3.36 nm for pristine g-C3N4
and g-C3N4/25.3%-NiS2, respectively, see the inset of Figure 5).
The corresponding textural parameters of these two
photocatalysts are summarized in Table S1. BET specific surface
areas of pristine g-C3N4 and g-C3N4/25.3%-NiS2 were calculated
as 157.46 and 62.01 m2 g-1, giving corresponding pore volumes
of 0.61 cm3 g-1 and 0.10 cm3 g-1, respectively. It indicates the
composite. We also benchmarked the mass activities against the
extensively investigated g-C3N4/Pt photocatalysts.[5,21,25,39] In this
comparison, Pt was photo-deposited onto the surface of g-C3N4.
As shown in Figure 6a, the improved photocatalytic activity by
using NiS2 is comparable or even higher than that by using Pt,
indicating NiS2 as a potential alternative for the replacing scarce
Pt. More importantly, we also systematically compared the
performance with that of many reported nickel-based cocatalysts
formation of partial NiS2 quantum dots in the mesopores of g-C3N4, modified g-C3N4 (Table S2). The superiority of NiS2 clusters is
resulting in partially filled mesoporous structures.[10,37] Although
the reduction in surface area and pore volume are generally
unfavorable for photocatalysis, we still expect improved
photocatalytic activity over the heterojunction.
thus clearly demonstrated. Meanwhile, the much lower activity of
mechanical mixture g-C3N4 and 25.3 wt% NiS2 than that of g-
C3N4/25.3%-NiS2 clearly demonstrated the advantage of close
contact induced by hydrothermal treatment (see Figure 6a,
column D and I). Such heterostructure was expected to be a
promising and effective noble-metal-free candidate photocatalyst
for H2 evolution. We also evaluated the photostability of the
composite photocatalysts by taking g-C3N4/30.7%-NiS2 as an
example. As shown in Figure 6b, continuous H2 evolution without
notable deactivation was achieved during a consecutive 15-h
photocatalytic reaction. Figure S8a showed the typical H2-
evolution kinetics of different samples under visible-light
irradiation. All samples present approximate linear H2-evolution
profiles, except the mechanical mixing g-C3N4/25.3%-NiS2, in the
initial 5-h reaction, confirming their relatively good photostability.
The generally decreased photocatalytic activity of the mechanical
mixed sample indicated the necessity of tight contact of NiS2
nanodot and g-C3N4 for maintaining the stability. Figure S8b
depicts the wavelength-dependent AQY of g-C3N4/25.3%-NiS2
composite with respect to the corresponding UV-vis spectra. It
shows a typical bandgap-transition-dependent proton reduction
thus H2-evolution behavior.[14] The AQY of g-C3N4/25.3%-NiS2
composite is about 2% at 425 nm.
Figure 5. N2 adsorption-desorption isotherms and the corresponding pore size
distribution curves (inset) of pristine g-C3N4 and g-C3N4/25.3%-NiS2
photocatalysts.
The as-prepared nanostructures were then evaluated as
photocatalyst for visible-light-driven H2 production from water.
Blank experiments showed that no H2 evolution was detected
without either illumination or photocatalyst, indicating the
photocatalytic nature of the reaction. As shown in Figure 6a, the
photocatalytic activity of g-C3N4/NiS2 composites was significantly
improved compared to that of pure g-C3N4. The result indicates
the surface area and pore volume are not decisive factors of the
photocatalytic property. Moreover, a volcano-type trend for the
mass activity as a function of loading amount of NiS2 was obtained
(Figure 6a). The highest H2 production rate of 4.841 µmol h-1 was
gained when the proportion of NiS2 reached 25.3 wt%. The rapid
reduction of H2-generation activity beyond this optimized loading
amount can be explained by a shielding effect of NiS2 species.
The overcrowding of NiS2 led to a reduction of both visible-light
absorption and the number of surface-active sites.[10-12,38] Note
that simply hydrothermal treatment of the g-C3N4 without NiS2
also could enhance the photocatalytic activity (see Figure 6a,
Figure 6. (a) The initial rate of hydrogen evolution rate over (A) pristine g-C3N4,
(B) g-C3N4/56.5%-NiS2, (C) g-C3N4/39.2%-NiS2, (D) g-C3N4/25.3%-NiS2, (E) g-
C3N4/12.2%-NiS2, (F) g-C3N4/5.6%-NiS2, (G) g-C3N4/2.5%-NiS2, (H) g-
C3N4/0.0%-NiS2, (I) mechanical mixture of g-C3N4 and 25.3 wt%-NiS2, (J) g-
C3N4/1%-Pt, and (K) g-C3N4/3%-Pt photocatalysts. (b) Time courses of recycling
tests toward photocatalytic H2 evolution over g-C3N4/25.3%-NiS2 composite
photocatalyst. Error bars in (a) are standard error values of three tests (n=3).
To reveal the photocatalytic mechanism that is responsible
for this enhancement, the dynamics of photoinduced charge
transfer was investigated by a collective use of PL, transient
photocurrent measurement, EIS, and polarization curve. PL
measurement is a common method to assess separation
efficiency of photogenerated carriers.[10,11,40] Figure 7a shows the
room-temperature PL spectra of g-C3N4 and g-C3N4/25.3%-NiS2
column
A and H for comparison). It indicates that both
hydrothermal treatment and integration of NiS2 could contribute to
the enhancement of the photocatalytic activity for g-C3N4/NiS2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
photocatalysts that were excited at 370 nm. Pure g-C3N4 exhibited
a strong intrinsic emission peak centered at ~420 nm, indicating
the severe recombination of the photogenerated charge carriers
in the g-C3N4. Although similar trend of the spectrum was
observed over g-C3N4/25.3%-NiS2, it showed slight blue shift and
significantly decreased PL intensity. In principle, the fate of
photogenerated charges determines the quenching efficiency of
photoluminescence. Here, NiS2 cocatalysts could act as charge
carrier trapping center to extract electrons from g-C3N4 to the
demonstrate that the intensive electron migration could positively
influence the reaction kinetics of water reduction.[43]
interface of NiS2. As
a result, the recombination of the
photogenerated electrons-holes pairs can be enormously
restrained, which is favorable for photocatalytic reaction. To
further understand the interfacial charge separation efficiency,
transient photocurrent measurement was accordingly performed.
Figure 7b shows the corresponding results that were obtained
under intermittent visible light illumination and at a positive bias
voltage of 0.6 V. While both sample exhibit rapid and completely
reversible photocurrent responses, g-C3N4/25.3%-NiS2 produces
an elevated photocurrent density, further demonstrating the
efficaciously promoted mobility and lifetime of the charge carriers.
Of particular note is that the decline of transient photocurrent
derives from the detachment of film from electrode (properly
because of the deficient preparation method).
Figure 7. (a) Photoluminescence spectra of g-C3N4 and g-C3N4/25.3%-NiS2
samples at room temperature with an excitation wavelength of 370 nm. (b)
Transient photocurrent responses (I-t curves) of g-C3N4 and g-C3N4/25.3%-NiS2
samples under visible light irradiation (> 420 nm) at a potential of 0.6 V (vs.
Ag/AgCl), and (c) the corresponding Nyquist plots of electrochemical
impedance spectroscopy in the dark obtained at a frequency range from 1 Hz
to 10 kHz. (d) Polarization curves in the dark of pristine g-C3N4, pristine NiS2,
and g-C3N4/25.3%-NiS2 composite. Note an aqueous solution containing 0.5 M
Na2SO4 was employed as the electrolyte.
EIS is a powerful electrochemical technique commonly
adopted to illustrate the electron-transfer efficiency in the three-
electrode systems.[10,11,41] The Nyquist impedance plots for pure
g-C3N4 and g-C3N4/25.3%-NiS2 samples are thus monitored as
shown in Figure 7c. An electrical equivalent-circuit model was
applied to simulate the plots (inset of Figure 7c), where Rs, CPE,
and Rct represent the Ohmic resistance of the electrolyte solution,
constant-phase element for the electrode-electrolyte interface,
and the charge-transfer resistance of contacting interface,
respectively.[41] Our result indicates the semicircular Nyquist plot
for both g-C3N4 and g-C3N4/25.3%-NiS2. Nonetheless, g-
C3N4/25.3%-NiS2 features prominently smaller arc radius. This
notion manifests a lower charge-transfer impedance and thus
more smooth charge immigration toward the intimated interface
between g-C3N4 and NiS2. The larger semicircle radius for pristine
g-C3N4 is imputed to the inferior electrical conductivity that hinders
the electron migration from g-C3N4 to the back-contact
electrode.[42] It also indicates that NiS2 quantum dots in the hybrid
could bridge g-C3N4 and back-contact electrode, thus significantly
accelerating the charge separation and transport. We also
measured the polarization curve for H2 evolution over g-
C3N4/25.3%-NiS2 photocatalyst, together with g-C3N4 and pure
NiS2 photocatalysts for comparison (see Figure 7d). The tests
were performed in dark with 0.5 M Na2SO4 act as the electrolyte,
and scanning potential ranging from -0.6 to -1.8 V. The lower
overpotential of NiS2 forebodes the feasibility of being an
alternative cocatalyst for photocatalytic H2 evolution. On the
contrary, pristine g-C3N4 produced a rather low cathodic current
even the potential was set to be -1.6 V, implying the enormous
energy barrier for water reduction. Significantly, an abrupt onset
for cathodic current at -1.3 V can be achieved after the integration
of NiS2 and g-C3N4. The more pronounced cathodic current
density and lower overpotential for g-C3N4/25.3%-NiS2
In addition, the optical properties of all the samples were
measured using UV-vis absorption spectroscopy. As depicted in
Figure S9, compared with pure g-C3N4, an apparent red shift of
absorption band is observed for all the hydrothermal treated
samples. The similar band edges indicate that the hydrothermal
treatment plays a major role in narrowing the band gap. Moreover,
all binary composites exhibit prominently enhanced absorption
intensities with the increase of NiS2 content, especially in the long-
wavelength region. This extended light absorption, in terms of
both intensity and band edge, is thus attributed to the synergistic
effect between the NiS2 and g-C3N4, as well as the mild
hydrothermal treatment. The bandgap energy (Eg) of can be
estimated by using the Kubelka-Munk method based upon the
tangent lines of Tauc plots, as shown in Figure 8a, where F(R) is
the diffuse absorption coefficient and hν is the photon energy.[31,40]
The obtained values of the bandgap for g-C3N4, g-C3N4/0.0%-NiS2,
and g-C3N4/25.3%-NiS2 are calculated to be 2.94, 2.63 and 2.62
eV, respectively. The result is consistent with the blue-shift of
fluorescence spectrum (Figure 7a). In the meantime, the results
again demonstrated the dominant role of hydrothermal treatment
on narrowing the band gap. We next sought to determine the
conduction band (CB) edge by Mott-Schottky (MS) plots over g-
C3N4, g-C3N4/0.0%-NiS2 and g-C3N4/30.7%-NiS2 photocatalysts
(Figure 8b). Apparently, all three samples feature in typical n-type
as demonstrated by the positive tangent slopes of straight
lines.[10,27] The flat band potential values of g-C3N4, g-C3N4/0.0%-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
NiS2 and g-C3N4/25.3%-NiS2 were calculated to be -1.30,-1.41
and -1.42 V (vs. Ag/AgCl), respectively. The CB edges could be
accordingly obtained by negatively shift the flat band potential by
-0.2 V.[44] The similar CB positions of g-C3N4/0.0%-NiS2 and g-
C3N4/25.3%-NiS2 also certify that the band level changes stem
from the hydrothermal procedure rather than NiS2 cluster
anchoring. Taken bandgap energies into consideration, the
valence band (VB) positions for pristine g-C3N4 and g-C3N4/0.0%-
NiS2 are estimated to be 1.44 and 1.02 V (vs Ag/AgCl),
respectively.
is above its VB edge. The mild hydrothermal treatment on g-C3N4
can elevate both the CB and VB levels, which not only enhances
the reduction ability of adsorbed H+ ions, but also promotes water
oxidation. Moreover, the less negative overpotential of NiS2
(−1.00 V vs Ag/AgCl, pH=7) allows it as a receptor to trap the
photogenerated electrons from g-C3N4. On the basis of the above
analysis, the underlying mechanism for the photocatalytic
reaction on g-C3N4/NiS2 is proposed as shown in Scheme 1b.
Upon the irradiation of visible light, g-C3N4 absorbs photons via
bandgap excitation to generate electrons and holes in the CB and
VB, respectively. The photogenerated electrons in CB of g-C3N4
can easily transfer to the embedded NiS2 quantum dots for rapid
water reduction. On the other hand, the holes will diffuse within
the VB of g-C3N4 and be captured by adsorbed TEOA molecules.
More importantly, the quantum size of NiS2 shall be also favorable
for efficient electron transfer. These advantages that are induced
by NiS2 quantum dots can synergistically enhance the
photocatalytic activity toward hydrogen evolution.
Conclusions
Figure 8. (a) Tauc plots of the UV-vis spectra (the relationship between
(F(R)hν)1/2 and photon energy, herein, the (F(R)hν)1/2 donates as Kubelka-Munk
function) for pristine g-C3N4 and g-C3N4/25.3%-NiS2 photocatalyst, and (b) the
corresponding Mott-Schottky (MS) plots of two different photocatalysts in the
form of film electrodes. MS plots were obtained at a frequency of 1 kHz in an
aqueous solution of Na2SO4
In summary, a series of g-C3N4/NiS2 hybrid photocatalysts have
been fabricated successfully. The success relies on a mild
hydrothermal procedure for decoration of NiS2 nanodots on the
surface of g-C3N4. These hybrids showed notable enhancement
of solar photocatalytic hydrogen production. The highest activity
was achieved over g-C3N4/25.3%-NiS2 photocatalyst, with a H2
production rate of 4.841 µmol h-1 and an AQY of 2% at 425 nm.
The efficiency is much higher than that of pristine g-C3N4, or even
higher than that of Pt/g-C3N4. This improved photocatalytic
property can be explained by the synergy taken by NiS2 quantum
dots for expanded visible-light harvest, improved separation and
transportation efficiency of charge carriers, and reduced energy
barrier for water reduction. Moreover, the composite is also
relatively stable as demonstrated by cycling tests. This work
provides new experimental and theoretical insights into the
photocatalytic process by using an efficient but low-cost g-
C3N4/NiS2 hybrid photocatalyst.
Consequently, the electronic band structures of these two
different semiconductors can be illustrated in Scheme 1a.
Specifically, the reduction potential of H+/H2 is positioned below
the CB edge of g-C3N4, while the oxidation potential of H2O to O2
Experimental Section
Chemicals and materials
Urea, triethanolamine (TEOA), thiourea, nickel (II) chloride hexahydrate
(NiCl2·6H2O), sodium citrate dehydrate (C6H5Na3O7), ammonium
hydroxide (NH3·H2O), and absolute ethanol were all purchased from
Sinopharm Chemical Reagent Co., Ltd. All reagents were used as
received without further purification. Deionized (DI) water with a resistivity
of 18.2 MΩ·cm was used for preparing all samples.
Preparation of pristine g-C3N4
Scheme 1. (a) and (b) schematic diagrams showing the possible mechanism
involved in the process of photocatalytic H2 production over pristine g-C3N4 and
g-C3N4/25.3%-NiS2 photocatalysts under visible-light irradiation.
The pristine g-C3N4 was synthesized with a direct polycondensation
method by thermal treatment of urea under ambient pressure without the
use of any additives. Typically, 10 g of urea was put into a covered alumina
crucible and calcined in a muffle furnace at 600 oC for 4 h in static air at a
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
ramp rate of 5 oC min-1 and cooled naturally to room temperature. The
product was subsequently grinded into powders thoroughly with a mortar
and pestle for further processing and investigation.
(DuPont) was added to the above mixture to form a homogeneous
suspension. After ultrasonic treatment for 4 h, 1 mL of dispersion solution
was then spread onto a 2×3 fluorine-doped tin oxide (FTO) glass plate.
Then the working electrodes were obtained after drying them under
infrared lamp. For the preparation of the working electrodes for EIS,
polarization curve, and MS measurements, 1 mg of sample and 10 μL of
10% Nafion (DuPont) were firstly dispersed in 1.0 mL of water/ethanol
(volume ratio: 1/1) solution. The mixture was then under ultrasonic
Preparation of g-C3N4/NiS2 composite photocatalyst
The g-C3N4/NiS2 composite was synthesized using a hydrothermal method.
In a typical synthesis (see Scheme S1 for detailed procedure), 40 mg of g-
C3N4 was dispersed into 50 mL of deionized water and then the mixture
was place in ultrasonic bath for 1 h to form homogeneously dispersing
suspension. Then certain amounts of NiCl2·6H2O (2.975 mg , or 5.95 mg,
or 11.9 mg, or 23.8 mg, or 35.7 mg, or 47.6 mg), C6H5Na3O7 (3.675 mg ,
or 7.35 mg, or 14.7 mg, or 29.4 mg, or 58.8 mg, or 117.6 mg), and thiourea
(2.85 mg , or 5.7 mg, or 11.4 mg, or 22.8 mg, or 45.6 mg, or 91.2 mg) were
added rapidly in sequence under violent stirring. After dissolution of these
precursors, the pH of the suspension was adjusted to 12 by adding
NH3·H 2O. The obtained suspension was stirred continuously for another
10 min. After that, the mixture was transferred to a 100 mL Teflon lined
autoclave and heated at 120 oC for 24 h. After cooling to room temperature,
the resultant product was separated by centrifugation and washed with
deionized water and ethanol for several times to remove soluble impurities
thoroughly. Finally, it was dried at 80 oC in a vacuum oven to obtain g-
C3N4/NiS2 composite photocatalyst. Pure NiS2 was prepared using the
similar process except without adding g-C3N4.
treatment for
1 h in order to get the homogeneous suspension.
Subsequently, 10 μL of this dispersion was drop-casted onto the top of a
glassy carbon electrode (GCE) at a diameter of 3 mm. Finally, the GCE
was naturally dried under ambient temperature to get the working
electrode. Time-dependent transient photocurrent responses (I-t curve)
were measured at 0.4 V (vs. Ag/AgCl). For EIS measurement, a small
amplitude (10 mV) alternate current (AC) signal with a frequency ranged
from 1 Hz to 100 kHz at open circuit potential was provided. The same
electrochemical system was also used to determine the flat-band potential
of the working electrodes by the MS method. The measurements were
performed in darkness by scanning the electric potential from -0.9 to -1.5
V. The impedance-potential characteristics were recorded at a frequency
of 1 kHz. The polarization curves were conducted in the same system,
while the bias sweep range was changed from -0.6 to -1.8 V (vs. Ag/AgCl,
10 mV/s).
Instrumentations
Photocatalytic tests
The powder X-ray diffraction (XRD) patterns were recorded on
a
PANalytical X’pert MPD Pro diffractometer using Ni-filtered Cu Kα
irradiation (wavelength: 1.5406 Å) conducting at 40 kV and 40 mA. The
Fourier transform infrared spectrophotometer (FTIR) spectra were
collected on a Bruker Vertex 70 FTIR spectrophotometer by using the KBr
pellet technique. The crystallite morphologic micrograph and
microstructures were characterized by a transmission electron microscope
(TEM) of FEI Tecnai G2 F30 S-Twin transmission electron microscope
attached with an OXFORD MAX-80 energy-dispersive X-ray detector
(EDX). X-ray photoelectron spectroscopy (XPS) was measured with a
Kratos spectrometer (AXIS Ultra DLD) equipped with a standard and
monochromatic Al Kα line as the excitation source (hν = 1486.6 eV). The
C1s peak at 284.8 eV of adventitious carbon was used for calibration.
Optical properties were determined by a Hitachi double-beam U-4100 UV-
vis-near-IR spectrophotometer equipped with an integrating sphere in
which BaSO4 acted as the background. Nitrogen adsorption-desorption
isotherms were conducted at 77 K using an Accelerated Surface Area and
Porosimetry Analyzer (ASAP 2020, Micromeritics). The Brunauer-Emmett-
Teller (BET) approach was employed to determine the specific surface
area. Steady-state photoluminescence (PL) properties were characterized
Photocatalytic reactions of hydrogen production were conducted in a 50
mL side-irradiation Pyrex gas-tight circulation system with a flat optical
entry window at ambient temperature using a 300 W Xe lamp equipped
with a UV cut off filter (420 nm). An aluminum alloy shell was employed
outside the Pyrex cell to reflect and gather the light. 5 mg of obtained
photocatalyst powder was suspended under sonication in a solution
containing 36 mL of water and 4 mL of triethanolamine (electron donor).
The suspension was evacuated for about 20 min with N2 gas to remove
oxygen prior to the irradiation experiments. The evolved gases were
analyzed by gas chromatography (Bruker GC-450) equipped with a
thermal conductivity detector (TCD) and high-purity argon as the carrier
gas. The apparent quantum yield (AQY) for H2 evolution was measured
using band-pass filters with various wavelengths and calculated according
to the following equation:
AQY (%) = Ne/Np = 2NH2/Np
where, Ne represents for the number of reacted electrons. NH2 is the
number of generated H2 molecules. Np stands for the number of incident
photons.
at room temperature using
a time-resolved PTIQM-4 fluorescence
spectrophotometer at an excitation wavelength of 370 nm. The
concentrations of Ni ions were determined using a Perkin Elmer inductively
coupled plasma mass spectrometer (ICP-MS, NexION 300Q). The
thickness of bulk and thin layer g-C3N4 were measured by the atomic force
microscopy (AFM, Bruker INNOVA).
(Photo)electrochemical Measurements
Transient photocurrent, the electrochemical impedance spectra (EIS), the
polarization curve and Mott-Schottky (MS) plot analyses were carried out
on a CHI600D electrochemical analyzer (Chenhua Instruments Co.,
Shanghai) in a standard three-electrode configuration with the sample,
Ag/AgCl (in saturated KCl), and Pt plate as the working electrode, the
counter electrode, and reference electrode, respectively. A sodium sulfate
aqueous solution (0.5 M) served as the electrolyte in all the
(photo)electrochemical measurements. For photoelectrochemical
measurements, a 300 W Xe lamp equipped with a simulated solar light
filter (AM 1.5) was used as the light source. The working electrodes for
transient photocurrent test were fabricated using a following method.
Specifically, 5 mg of catalyst was dispersed in 1.5 mL of water/ethanol
(volume ratio: 1/1) solution, and then 20 µL of 10% Nafion solution
Acknowledgements
This work was supported by the National Nature Science
Foundation of China (No. 51502240), the Natural Science
Foundation of Jiangsu Province (No. BK20150378), China
Postdoctoral Science Foundation (No. 2014M560769), and the
China Fundamental Research Funds for the Central Universities.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
[23] Y. Hou, Y. Zhu, Y. Xu, X. Wang, Appl. Catal. B: Environ. 2014, 156, 122-
127.
The authors also appreciate the help of Mr. Miao Wang and Mr.
Kejian Lu for technical assistance.
[24] L. Kong, Y. Dong, P. Jiang, G. Wang, H. Zhang, N. L. Zhao, J. Mater.
Chem. A 2016, 4, 9998-10007.
Keywords: g-C3N4 • NiS2 • photocatalysis • quantum dot •
[25] J. Yu, S. Wang, B. Cheng, Z. Lin, F. N. Huang, Catal. Sci. Technol. 2013,
3, 1782-1789.
hydrogen evolution
[26] D. Zeng, W-J. Ong, H. Zheng, M. Wu, Y. Chen, D-L. Peng, M-Y. Han, J.
Mater. Chem. A 2017, 5, 16171-16178.
[1] X. Chen, S. Shen, L. Guo, S. S. Mao, Chem. Rev. 2010, 110, 6503-6570.
[27] Y. Lu, D. Chu, M. Zhu, Y. Du, P. Yang, Phys. Chem. Chem. Phys. 2015,
17, 17355-17361.
[2]
S. J. A Moniz, S. A. Shevlin, D. J. Martin, Z-X. Guo, J. Tang, Energy
Environ. Sci. 2015, 8, 731-759.
[28] J. Yuan, J. Wen, Y. Zhong, X. Li, Y. Fang, S. Zhang, W. Liu, J. Mater.
Chem. A 2015, 3, 18244-18255.
[3] Y. Tachibana, L. Vayssieres, J. R. Durrant, Nat. Photon. 2012, 6, 511-518.
[4] S. Ye, R. Wang, M. Z. Wu, Y. P. Yuan, Appl. Surf. Sci. 2015, 358, 15-27.
[29] J. Xu, Y. Li, S. Peng, Int. J. Hydrogen Energy 2015, 40, 353-362.
[30] L. Jiang, L. Wang, G. Xu, L. Gu, Y. Yuan, Sustainable Energy Fuels 2018,
2, 430-435.
[5]
X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K.
Domen, M. Antonietti, Nat. Mater. 2009, 8, 76-80.
J. Fu, J. Yu, C. Jiang, B. Cheng, Adv. Energy Mater. 2017, 1701503.
W-J. Ong, L-L. Tan, Y. H. Ng, S-T. Yong, S-P. Chai, Chem. Rev. 2016,
116, 7159-7329.
[6]
[7]
[31] F. Xue, W. Fu, M. Liu, X. Wang, B. Wang, L. Guo, Int. J. Hydrogen Energy
2016, 41, 20455-20464.
[32] P. Luo, H. Zhang, L. Liu, Y. Zhang, J. Deng, C. Xu, N. Hu, Y. Wang, ACS
Appl. Mater. Interfaces 2017, 9, 2500-2508.
[8]
[9]
Y. Li, S. Ouyang, H. Xu, X. Wang, Y. Bi, Y. Zhang, J. Ye, J. Am. Chem.
Soc. 2016, 138, 13289-13297.
[33]
P. Thangasamy, V. Maruthapandian, V. Saraswathy, M. Sathish, Catal.
Sci. Technol. 2017, 7, 3591-3597.
Y. Guo, J. Li, Y. Yuan, L. Li, M. Zhang, C. Zhou, Z. Lin, Angew. Chem.
Int. Ed. 2016, 55, 14693-14697.
[34] J. Yu, G. Cheng, W. Luo, J. Mater. Chem. A 2017, 5, 15838-15844.
[10] J. Wen, J. Xie, H. Zhang, A. Zhang, Y. Liu, X. Chen, X. Li, ACS Appl.
Mater. Interfaces 2017, 9, 14031-14042.
[35]
Z. Mao, J. Chen, Y. Yang, D. Wang, L. Bie, B. D. Fahlman, ACS Appl.
Mater. Interfaces 2017, 9, 12427-12435.
[11] H. Zhao, S. Sun, P. Jiang, Z. J. Xu, Chem. Eng. J. 2017, 315, 296-303.
[12] C. Li, Y. Du, D. Wang, S. Yin, W. Tu, Z. Chen, M. Kraft, G. Chen, R. Xu,
Adv. Funct. Mater. 2017, 27, 1604328.
[36]
M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F. Rodriguez-
Reinoso, J. Rouquerol, K. S. W. Sing, Pure Appl. Chem. 2015, 87,1051-
1069.
[13] S. Cao, J. Low, J. Yu, M. Jaroniec, Adv. Mater. 2015, 27, 2150-2176.
[14] M. Liu, Y. Chen, J. Su, J. Shi, X. Wang, L. Guo, Nat. Energy 2016, 1,
16151.
[37] X. Li, H. Li, S. Huo, Z. Li, Kinet. Catal. 2010, 51, 754-761.
[38] Z. Sun, B. Lv, J. Li, M. Xiao, X. Wang, P. Du, J. Mater. Chem. A 2016, 4,
1598-1602.
[15] J. Ran, J. Zhang, J. Yu, M. Jaroniec, S. Z. Qiao, Chem. Soc. Rev. 2014,
43, 7787-7812.
[39] G. Zhang, Z-A. Lan, L. Lin, S. Lin, X. Wang, Chem. Sci. 2016, 7, 3062-
3066.
[16] P. Ye, Xi. Liu, J. Iocozzia, Y. Yuan, L. Gu, G. Xu, Z. Lin, J. Mater. Chem.
A 2017, 5, 8493-8498.
[40] X. Wang, M. Liu, Y. Chen, W. Fu, B. Wang, L. Guo, Nanoscale 2016, 8,
15970-15977.
[17] Y. Di, X. Wang, A. Thomas, M. Antonietti, ChemCatChem. 2010, 2, 834-
838.
[41] Z. Qin, Y. Chen, Z. Huang, J. Su, L. Guo, J. Mater. Chem. A 2017, 5,
19025-19035.
[18] X. Li, W. Bi, L. Zhang, S. Tao, W. Chu, Q. Zhang, Y. Luo, C. Wu, Y. Xie,
Adv. Mater. 2016, 28, 2427-2431.
[42] X-X. Li, G-F. Chen, K. Xiao, N. Li, T-Y. Ma, Z-Q. Liu, Electrochim. Acta
2017, 255, 153-159.
[19] J. Qin, J. Huo, P. Zhang, J. Zeng, T. Wang, H. Zeng, Nanoscale, 2016, 8,
2249-2259.
[43] Y-P. Zhu, T-Z. Ren, Z-Y. Yuan, ACS Appl. Mater. Interfaces 2015, 7,
16850-16856.
[20] J. Xu, Y. Li, S. Peng, G. Lu, S. Li, Phys. Chem. Chem. Phys. 2013,15,
7657-7665.
[44] X. Li, J. Yu, J. Low, Y. Fang, J. Xiao, X. Chen, J. Mater. Chem. A 2015,
3, 2485-2534.
[21] L. Yin, Y. Yuan, S. Cao, Z. Zhang, C. Xue, RSC Adv. 2014, 4, 6127-3132.
[22] Y. Hou, A. B. Laursen, J. Zhang, G. Zhang, Y. Zhu, X. Wang, S. Dahl, I.
Chorkendorff, Angew. Chem. Int. Ed. 2013, 52, 3621-3625.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801510
ChemCatChem
FULL PAPER
FULL PAPER
NiS2 quantum dot successfully
anchored on the surface of g-C3N4
promotes both vectorial charge
diffusion and rapid proton reduction
during photocatalytic water splitting.
Fei Xue, Maochang Liu*, Cheng Cheng
Page No. – Page No.
Localized NiS2 Quantum Dots on g-
C3N4 Nanosheets for Efficient
Photocatalytic Hydrogen Production
from Water
This article is protected by copyright. All rights reserved.