Construction of CoS2/Zn0.5Cd0.5S S-Scheme Heterojunction for Enhancing H2 Evolution Activity Under Visible Light

Article abstract of DOI:10.1002/chem.202102811

In the field of photocatalysis, building a heterojunction is an effective way to promote electron transfer and enhance the reducibility of electrons. Herein, the S-scheme heterojunction photocatalyst (CoS₂/Zn_0.5Cd_0.5S) of CoS₂ nanospheres modified Zn_0.5Cd_0.5S solid solution was synthesized and studied. The H₂ evolution rate of the composite catalyst reached 25.15 mmol g^?1 h^?1, which was 3.26 times that of single Zn_0.5Cd_0.5S, whereas pure CoS₂ showed almost no hydrogen production activity. Moreover, CoS₂/Zn_0.5Cd_0.5S had excellent stability and the hydrogen production rate after six cycles of experiments only dropped by 6.19 %. In addition, photoluminescence spectroscopy and photoelectrochemical experiments had effectively proved that the photogenerated carrier transfer rate of CoS₂/Zn_0.5Cd_0.5S was better than CoS₂ or Zn_0.5Cd_0.5S single catalyst. In this study, the synthesized CoS₂ and Zn_0.5Cd_0.5S were both n-type semiconductors. After close contact, they followed an S-scheme heterojunction electron transfer mechanism, which not only promoted the separation of their respective holes and electrons, but also retained a stronger reduction potential, thus promoting the reduction of H⁺ protons in photocatalytic experiments. In short, this work provided a new basis for the construction of S-scheme heterojunction in addition to being used for photocatalytic hydrogen production.

Full text of DOI:10.1002/chem.202102811

Accepted Article
Accepted Article
Title: Construction of CoS2/Zn0.5Cd0.5S S-scheme heterojunction for
enhancing H2 evolution activity under visible light
Authors: Lijun Ma, Jing Xu, Sheng Zhao, Lingjiao Li, and Ye Liu
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: Chem. Eur. J. 10.1002/chem.202102811
Link to VoR: https://doi.org/10.1002/chem.202102811
01/2020

10.1002/chem.202102811
Chemistry - A European Journal
Construction of CoS2/Zn0.5Cd0.5S S-scheme heterojunction for enhancing H2
evolution activity under visible light
Lijun Ma[a], Jing Xu*[a,b,c], Sheng Zhao[a], Lingjiao Li[a], Ye Liu[a]
[a] L. J. Ma, Prof. J. Xu, S. Zhao, L. J. Li, Y. Liu
School of Chemistry and Chemical Engineering, North Minzu University, Yinchuan 750021,
PR China.
[b] Prof. J. Xu
Key Laboratory of Chemical Engineering and Technology (North Minzu University), State
Ethnic Affairs Commission, Yinchuan 750021, PR China.
[c] Prof. J. Xu
Ningxia Key Laboratory of Solar Chemical Conversion Technology, North Minzu University,
Yinchuan 750021, PR China.
Corresponding author (Prof. J. Xu) email: wgyxj2000@163.com
Abstract
In the field of photocatalysis, building a heterojunction is an effective way to promote electron
transfer and enhance the reducibility of electrons. Herein, the S-scheme heterojunction
photocatalyst (CoS2/Zn0.5Cd0.5S) of CoS2 nanospheres modified Zn0.5Cd0.5S solid solution was
synthesized and studied. The H2 evolution rate of composite catalyst reached 25.15 mmol·g-1·h-1,
which was 3.26 times that of single Zn0.5Cd0.5S, while pure CoS2 showed almost no hydrogen
production activity. Moreover, CoS2/Zn0.5Cd0.5S had excellent stability, and the hydrogen
production rate after six cycles of experiments only dropped by 6.19%. In addition,
photoluminescence spectroscopy and photoelectrochemical experiments had effectively proved
that the photogenerated carrier transfer rate of CoS2/Zn0.5Cd0.5S was better than CoS2 or
Zn0.5Cd0.5S single catalyst. In this study, the synthesized CoS2 and Zn0.5Cd0.5S were both n-type
semiconductors. After close contact, they followed an S-scheme heterojunction electron transfer
mechanism, which not only promoted the separation of their respective holes and electrons, but
also retained a stronger reduction potential, thus promoting the reduction of H+ protons in
photocatalytic experiments. In short, this work provided a new basis for the construction of
S-scheme heterojunction in addition to being used for photocatalytic hydrogen production.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
Keywords: Photocatalysis; CoS2; Zn0.5Cd0.5S; S-scheme heterojunction
Introduction
Hydrogen energy is a storable clean energy and an energy carrier with extremely low
environmental pollution.[1,2] Studies have found that the use of solar energy to convert water into
hydrogen is the most promising technology for obtaining hydrogen energy.[3,4] Therefore, the
construction of visible light-induced semiconductor catalysts is of great significance for improving
global environmental degradation and energy shortages.[5,6] In recent years, many excellent
semiconductors have been widely used in the field of photocatalytic H2 evolution,[7-10] such as
metal oxides, phosphides, metal sulfides, etc. However, the poor stability of semiconductor
catalysts, the low utilization of sunlight, and the easy recombination of electrons after being
excited by light, which are still problems that need to be solved urgently in this field.[11-13]
So far, Pt-based catalyst is a catalyst material with strong hydrogen evolution reaction (HER)
activity, but its scarcity and high price limit its development in the field of hydrogen
production.[14,15] Therefore, researchers have been working to find low-cost materials with high
photocatalytic activity to replace precious metal catalysts. Surprisingly, in recent years, transition
metal compounds (for example, WS2,[10] MoS2,[16] Ni2P,[17] etc.) have been frequently reported due
to their excellent H2 evolution activity in photocatalytic reactions. Among them, cobalt pyrite
(CoS2) has metal-like conductivity and exhibits excellent HER activity. Because of the abundant
content of Co element on the earth, CoS2 is expected to become a low-cost hydrogen evolution
catalyst to replace precious metals.[18,19] However, reports on CoS2 are more inclined to
electrocatalytic hydrogen production, and the research on photocatalytic hydrogen production is
not prominent.
As we all know, CdS is one of the representative catalysts for photocatalytic H2 production
reaction,[20] but the holes generated after CdS are excited by light will oxidize S-2 in CdS, that is,
CdS has a serious photo-corrosion phenomenon, which leads to poor stability in HER.[21] Based on
CdS, the solid solution (ZnxCd1-xS) of ZnS and CdS formed by doping Zn atoms is widely used for
photocatalytic hydrogen production.[22] Although ZnxCd1-xS has the stability of ZnS and a band
structure similar to CdS, there are still some non-negligible shortcomings, such as the serious
agglomeration phenomenon and low carrier separation efficiency of ZnxCd1-xS.[23] In order to
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
improve photocatalytic activity, ZnxCd1-xS and other semiconductors (for example, NiCo2S4,[22]
α-Ni(OH)2,[24] CoFe2O4,[25] and MoS2,[26]) can be used to construct heterojunction.
Based on the above discussion, a new CoS2/Zn0.5Cd0.5S composite was prepared in this paper
and used for hydrogen evolution reaction with 10 vol% lactic acid solution as sacrifice agent.
Experiments showed that CoS2/Zn0.5Cd0.5S had very excellent HER activity, and its H2 evolution
rate reached 25.15 mmol·g-1·h-1, which was much higher than other materials containing
ZnxCd1-xS. This was not only attributable to the successful coupling of CoS2 and Zn0.5Cd0.5S to
improve the serious agglomeration of Zn0.5Cd0.5S, but also to the close contact between CoS2 and
Zn0.5Cd0.5S, formed an S-scheme heterojunction. The heterojunction retained a strong hydrogen
reduction potential. The study of CoS2/Zn0.5Cd0.5S photocatalyst will provide a new strategy for
the use of CoS2 in the field of hydrogen production and the construction of S-scheme
heterojunction.
Experimental section
Materials
NaOH (sodium hydroxide, 96.0%), Cd(Ac)2·2H2O (cadmium acetate dihydrate, 99.0%),
Zn(Ac)2·2H2O (Zinc acetate dihydrate, 99.0%), Na2S2O3·5H2O (sodium thiosulfate, 99.5%),
C2H5NS (thioacetamide, 99.0%), Co(NO3)2·6H2O [Cobalt nitrate hexahydrate, 99.9%], absolute
ethanol (AE, 99.7%) and L-lactic acid (85%).
Synthesis of CoS2
Added Na2S2O3·5H2O and Co(NO3)2·6H2O with a molar ratio of 2:1 into the beaker, as well as
30 mL of AE and 40 mL of deionized water (DW). Then the mixed solution was continuously
stirred for 1 h. After the solution was uniformly mixed, it was transferred to a 100 mL
polytetrafluoroethylene (PTFE) reactor and maintained at 180°C for 12 h. Finally, it was washed
several times with DW and AE, and then the prepared CoS2 was dried for 8 h at 60 ℃.
Synthesis of Zn0.5Cd0.5S
Weighed 12.5 mmol C2H5NS, 5 mmol Cd(Ac)2·2H2O and 5 mmol Zn(Ac)2·2H2O, added them
to a beaker containing 75 mL DW, stirred for 1 h, added 0.8 g NaOH during stirring, and then
transferred to a PTFE reactor of 100 mL, heated for 24 h at 180 °C. After cooling, the products
were washed and dried as before. The synthesis method of other ZnxCd1-xS pure substances was
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
the same as above, only the molar ratio of Zn(Ac)2·2H2O and Cd(Ac)2·2H2O was changed. The
quality of the required drugs was shown in Table S1 (Supporting Information).
Synthesis of CoS2/Zn0.5Cd0.5S (CZCS-y, y=1,2,3,4,5)
Taking CZCS-3 as an example, added 40 mg of CoS2 powder to a beaker containing 30 mL of
AE and 0.1 g of Zn0.5Cd0.5S. After ultrasonic treatment for 20 min, it was heated and stirred in a
water bath at 80 °C until it became viscous, then dried in a thermostat for 12 h at 60°C, and finally
ground to obtain CZCS-3 powder. Scheme 1 was a preparation process of CZCS-3 composite
catalyst. The same method was used to prepare a composite catalyst with a mass ratio of
ZnxCd1-xS (x=0.3, 0.4, 0.6, 0.7) and CoS2 of 1:0.4. The required mass and catalyst abbreviation
were shown in Table S2 and Table S3 (Supporting Information).
Scheme 1: Preparation process of CZCS-3 composite catalyst.
Characterization
Characterization methods and instruments were recorded in detail in the Supporting
Information.
Photocatalytic hydrogen production experiments
Added 12 mg of photocatalyst powder and 30 mL of 10 vol% L-lactic acid aqueous solution
into a 65 mL quartz bottle equipped with a sealing rubber sheet, and then ultrasonicated for 5 min
and stirred for 15 min. Filled the flat-bottomed quartz bottle with N2, kept it for 15 min. Then put
the flat-bottomed quartz reactor into a multichannel photocatalytic reaction system, and conducted
a hydrogen evolution experiment for 5 h under simulated visible light (5W LED lamp, λ≥420 nm),
and the amount of hydrogen precipitated was detected by a gas chromatograph (SP-2100, 13X
column, TCD, N2 carrier). In addition, the apparent quantum efficiency (AQE) of the CZCS-3
sample was studied. The hydrogen production data required for the CZCS-3 at different
wavelengths was measured by using a 300 W Xe lamp as the light source and matching
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
monochromatic light cut-off filters of different wavelengths. The photon flux of the incident light
at different wavelengths was detected by the strong light optical power meter (CEL-NP2000).
Formula (1) was the calculation formula of AQE.[27]
ꢀꢁꢂ(%) = 2×ꢃℎꢄ ꢅꢆꢇꢈꢄꢉ ꢊꢋ ꢄꢌꢊꢍꢌꢄꢎ ℎꢏꢎꢉꢊꢐꢄꢅ ꢇꢊꢍꢄꢑꢆꢍꢄꢒ ×100%
(1)
ꢃℎꢄ ꢅꢆꢇꢈꢄꢉ ꢊꢋ ꢓꢅꢑꢓꢎꢄꢅꢃ ꢔℎꢊꢃꢊꢅ
Electrochemical measurement
In this part, an electrochemical workstation (VersaSTA4-400) was used to measure
photoelectrochemical characteristics such as linear sweep voltammetry (LSV) and photocurrent
response (I-T) curve. The electrodes and electrolyte solutions used were detailed in the Supporting
Information. The photocurrent response curve test also required a 300 W Xe lamp (λ≥420 nm) to
provide the light source.
Results and discussion
The phase structure of the prepared samples were identified and confirmed by X-ray
diffractometer (XRD). It could be found from the XRD pattern of Figure 1a that compared with
the standard cards of cubic phase CdS (PDF#65-2887) and cubic phase ZnS (PDF#65-1691), the
position of the diffraction peak of Zn0.5Cd0.5S was continuously shifted to a certain extent. The
characteristic peaks corresponded to the (311), (220), (200) and (111) crystal planes of the
Zn0.5Cd0.5S sample were between the main diffraction peaks of cubic CdS and ZnS.[28] That was,
the diffraction peak of Zn0.5Cd0.5S shifted to a small-angle compared to ZnS as a whole, and
shifted to a large-angle compared to CdS as a whole. This could be explained as the synthesis of
Zn0.5Cd0.5S solid solution instead of simple mixture of CdS and ZnS, this was consistent with
previous reports.[21,29] Comparing the XRD patterns of CZCS-3 and Zn0.5Cd0.5S, it could be seen
that CZCS-3 contained the characteristic peak of Zn0.5Cd0.5S without change, indicating that the
lattice structure of Zn0.5Cd0.5S did not change after coupling with CoS2. In Figure 1b, the
characteristic peaks of the CoS2 sample corresponded well to the diffraction peaks of the CoS2
standard card (PDF#89-1492), indicating that a single CoS2 catalyst had also been successfully
synthesized. And the characteristic peaks of the CZCS-3 sample at 32.3°, 36.2°, 39.7°, 54.9° and
60.2° corresponded to CoS2.[30] Except for the diffraction peaks of CoS2 and Zn0.5Cd0.5S, the
CZCS-3 sample did not contain other impurity peaks, indicating that the synthesized composite
sample had high purity. (Deposition Number(s) 1646235 (for CoS2), 1605240 (for CdS), 1666388
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
(for ZnS) contain(s) the supplementary crystallographic data for this paper. These data are
provided free of charge by the joint Cambridge Crystallographic Data Centre and
Fachinformationszentrum Karlsruhe Access Structures service.)
Figure 1. XRD pattern: (a) Comparison between Zn0.5Cd0.5S and CZCS-3, (b) comparison between CoS2 and
CZCS-3.
The morphology of the catalyst is an important characterization for studying the catalytic
performance of the sample, and it is also a side verification of the successful synthesis of the
catalyst. Figures 2a-c were the scanning electron microscope (SEM) images of Zn0.5Cd0.5S, CoS2
and CZCS-3 samples respectively. It could be seen that Zn0.5Cd0.5S exhibited uneven smaller
particles and had serious agglomeration. The particle size distribution diagram of Zn0.5Cd0.5S
(Figure S1, Supporting Information) was obtained from Figure 2a by SEM particle size analysis
method. It could be seen that most of the particles of Zn0.5Cd0.5S were distributed in 45-75 nm,
mainly concentrated in 45 nm, and the average particle size was 59.9 nm. However, CoS2 was a
stacked structure of spheres with uneven sizes of about 0.2-2 μm. As could be seen from the SEM
of CZCS-3 (Figure 2c), a large number of Zn0.5Cd0.5S particles uniformly loaded on the surface of
CoS2, which not only retained the spherical accumulation structure of CoS2, but also broke the
agglomeration state of Zn0.5Cd0.5S particles. This structure can provide more active sites to
promote the photocatalytic H2 production capacity under visible light. Figure 2d was a
transmission electron microscope (TEM) image of the CZCS-3 sample. The irregular spherical
shape with a light-colored outline was CoS2. It could be seen more clearly that a large amount of
Zn0.5Cd0.5S was covered on the surface of CoS2, and the size was basically the same as estimated
in SEM. The high-resolution TEM (HRTEM) of the CZCS-3 sample was magnified (Figure 2e),
and the clear lattice fringes with different orientations could be seen. The measured interplanar
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
spacing d of 0.247 nm and 0.324 nm corresponded to the (210) facet of CoS2 and the (111) facet of
Zn0.5Cd0.5S, respectively.[31,32] Figures 2f-g were the energy dispersive X-ray (EDX) spectrum and
element mapping of the CZCS-3 sample. It could be seen that CZCS-3 was composed of Cd, Zn,
Co, and S elements, and the distribution of each element was relatively uniform.
Figure 2. SEM images of (a) Zn0.5Cd0.5S, (b) CoS2 and (c) CZCS-3. (d) TEM, (e) HR-TEM, and (f) EDX energy
spectrum of CZCS-3. (g-k) EDS mapping images of Cd, Zn, Co and S in CZCS-3.
In order to further study the valence states of the elements contained in the CZCS-3 composite
catalyst, CZCS-3 was scanned by X-ray photoelectron spectroscopy (XPS). Figure 3a was the full
spectrum of the CZCS-3 sample, Cd, Co, S and Zn elements were all detected, which were
consistent with the previous EDX spectrum and element mapping results. In Figure 3b, the
characteristic peak of Zn 2p3/2 was located at 1022.1 eV, and the characteristic peak of Zn 2p1/2
was founded at 1045.1 eV. For Cd 3d (Figure 3c), similar to Zn 2p, two characteristic peaks
belonged to Cd 3d3/2 (411.8 eV) and Cd 3d5/2 (405.1 eV) could be observed. This showed that both
Zn and Cd elements in CZCS-3 existed in the +2 valence state.[33,34] From the XPS spectrum of S
2p (Figure 3d), it could be seen that the two peaks at the binding energy of 162.8 eV and 161.6 eV
corresponded to S 2p1/2 and S 2p3/2, respectively, indicated the presence of S2- in CZCS-3.[24]
Figure 3e showed the peak area of Co 2p. The two peaks at 781.9 eV and 798.1 eV corresponded
to Co 2p3/2 and Co 2p1/2 (the two peaks at 803.5 eV and 786.5 eV were satellite peaks), this was
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
consistent with our previous report.[30] Figure S2 (Supporting Information) was the comparison of
the XPS spectra of CoS2, Zn0.5Cd0.5S and CZCS-3. It was worth noting that compared with pure
Zn0.5Cd0.5S, the binding energy of Zn 2p in CZCS-3 showed a slight positive shift in the absence
of light illumination. This phenomenon was further confirmed in the XPS spectra of Cd 3d and S
2p, which means that Zn0.5Cd0.5S would lose electrons after being in close contact with CoS2.[35]
Compared with CoS2, the XPS spectra of Co 2p and S 2p in CZCS-3 both shifted to the direction
of lower binding energy. The obvious shift of the fine spectrum of each element indicated that
there was a strong interaction between Zn0.5Cd0.5S and CoS2, forming an effective heterojunction
structure.[22] These results also indicated that before CZCS-3 was illuminated, electrons transferred
from Zn0.5Cd0.5S to CoS2, resulting in an internal electric field at the interface of the two
materials.[36] Except for the shift of the fine spectrum of the element, the peak shape of each fine
spectrum had almost no change, which indicated that the construction of the heterojunction had no
effect on the chemical valence of the element.
Figure 3. XPS spectra: (a) Full spectrum of CZCS-3; (b) Zn 2p; (c) Cd 3d; (d) S 2p; (e) Co 2p.
Ultraviolet visible diffuse reflectance spectroscopy (UV-Vis DRS) can be used to study the
photoresponse performance of the photocatalyst. As shown in Figure 4a, the CoS2 sample showed
strong light absorption between 250-800 nm, while Zn0.5Cd0.5S had an inherent light absorption
edge around 500 nm. Compared with Zn0.5Cd0.5S, the CZCS-3 sample had an upward tail
absorption phenomenon in the region > 500 nm, which could indicate that a mixed interface
structure was formed between Zn0.5Cd0.5S and CoS2.[37] It could also be explained that, compared
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
with a single sample of Zn0.5Cd0.5S, the addition of CoS2 improved the light response performance
of CZCS-3 in the 500-800 nm region. Figures 4b-d were the sample band gap (Eg) value
determined according to Tauc formula: αhv = A(hv - Eg)n.[22] The Eg values of CoS2, Zn0.5Cd0.5S
and CZCS-3 were 2.47 eV, 2.35 eV and 1.87 eV, respectively, and the Eg values of the two pure
substances were consistent with previous reports.[31,33] Among them, the Eg value of CZCS-3 was
the smallest, indicating that in the CZCS-3 sample, the mixed interface structure formed between
Zn0.5Cd0.5S and CoS2 made CZCS-3 easier to be excited by light to generate electrons, which was
beneficial to the photocatalytic H2 evolution reaction.
Figure 4. (a) UV-vis DRS of CoS2, Zn0.5Cd0.5S and CZCS-3; Band gap plots for (b) CoS2, (c) Zn0.5Cd0.5S, and (d)
CZCS-3.
Figure 5a was the N2 adsorption-desorption isotherm (77K) of CoS2, Zn0.5Cd0.5S and CZCS-3. It
could be seen that these catalysts were all type IV curves with H3 type hysteresis loop, it had
obvious mesoporous characteristics. And from the pore diameter distribution of the three catalysts
in Figure 5b and the average pore size in Table 1, the pore diameter of the sample was distributed
between 0-50 nm. According to the definition of IUPAC, the pore size structure in this range was
divided into micropores and mesopores.[38] In Table 1, the SBET (specific surface area) of CoS2,
Zn0.5Cd0.5S and CZCS-3 were 18.57, 7.78 and 14.14 m2·g-1, respectively. The SBET of CZCS-3 was
between the two simple substances, the density difference between Zn0.5Cd0.5S and CoS2 and their
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
SBET per unit mass were the main reasons for the SBET change of the final composite material.[39] It
further showed that there was a mixed interface structure between Zn0.5Cd0.5S and CoS2, and the
good synergy could help improve the performance of photocatalytic H2 evolution reaction.
Figure 5. (a) N2 adsorption-desorption curves of CoS2, Zn0.5Cd0.5S and CZCS-3 and (b) pore size distribution
curves.
Table 1. The physical arameters of CoS2, Zn0.5Cd0.5S and CZCS-3.
Catalyst
SBET(m2·g-1)
Pore volume(cm3·g-1)
Average pore size (nm)
Zn0.5Cd0.5S
CoS2
18.57
7.78
0.18
0.02
0.08
35.60
13.03
23.93
CZCS-3
14.14
In order to prove the superiority of the CoS2/Zn0.5Cd0.5S photocatalyst, under the condition of
L-lactic acid solution (10 vol%) as the sacrificial agent, the H2 evolution rate of the prepared
photocatalyst was measured and compared. It could be found from Figures 6a-b and Figure S3a
(Supporting Information) that pure CoS2 basically did not exhibit hydrogen evolution activity in
lactic acid solution, and the H2 evolution amount was negligible. The H2 evolution amount of a
single Zn0.5Cd0.5S in the fifth hour was 463.51 μmol. However, among the composite catalysts,
CZCS-3 had the highest hydrogen production, reaching 1508.91 μmol (the hydrogen production
rate had reached 25.15 mmol·g-1·h-1), which was about 3.26 times the H2 evolution of pure
Zn0.5Cd0.5S. On the one hand, it showed that after the addition of CoS2, the CoS2/Zn0.5Cd0.5S
heterojunction was formed, and the good interface contact and synergy between CoS2 and
Zn0.5Cd0.5S were beneficial to photocatalytic H2 evolution reaction. On the other hand, it showed
that the mass ratio of CoS2 and Zn0.5Cd0.5S during the preparation of the composite catalyst had an
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
important influence on the activity of HER. This may be because when the mass of CoS2 in the
CoS2/Zn0.5Cd0.5S photocatalyst was small, the agglomeration phenomenon of Zn0.5Cd0.5S was still
serious. A large amount of Zn0.5Cd0.5S covered the surface of CoS2, which masked the active sites
of CoS2 or blocked the absorption of light by CoS2, thereby reducing the electron concentration of
CoS2, and the photocatalytic activity also decreases.
Figure 6c and Figure S3b (Supporting Information) showed the effect of the ratio of Cd to Zn in
ZnxCd1-xS on the composite photocatalyst on the basis of keeping the mass ratio of ZnxCd1-xS to
CoS2 at the optimal ratio (ZnxCd1-xS: CoS2=1: 0.4). Obviously, the ratio of Zn to Cd had a great
influence on the H2 evolution reaction activity of the composite photocatalyst, which could be
attributed to the different energy band structure and photosensitive strength of ZnxCd1-xS materials
with different mass ratios of Zn to Cd,[22] this was also reflected in the H2 evolution of pure
ZnxCd1-xS with different mass ratios of Zn to Cd (Figure S4, Supporting Information).
Figure 6d was the comparative experimental data obtained by changing the amount of CZCS-3
photocatalyst in the H2 production experiment. It could be seen that when 12 mg of CZCS-3
sample was added to 30 mL of sacrificial agent, the photocatalytic H2 evolution rate was the
highest, indicating that in the photocatalytic HER, it was not that the more photocatalyst was
added, the better the hydrogen production rate. The AQE of the CZCS-3 sample under different
wavelengths of monochromatic light was shown in Figure 6e, a high AQE of 6.74% was obtained
at 420 nm. Figure 6f studied the hydrogen production activity of the CZCS-3 sample under
long-term illumination (20 h). The hydrogen production rate of the CZCS-3 sample dropped
slightly after the tenth hour. This might be because the reaction time was too long, the pressure in
the hydrogen production bottle increased, which inhibited the progress of the hydrogen evolution
reaction or a small amount of gas leaked. However, with the prolongation of the illumination time,
the hydrogen production still showed an increasing trend, which showed that the CoS2/Zn0.5Cd0.5S
heterojunction has excellent light stability.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
Figure 6. Photocatalytic H2 evolution comparison of (a) CoS2, Zn0.5Cd0.5S, CZCS-3, (b) CZCS-Y (Y=1, 2, 3, 4, 5),
(c) CZ3CS-3, CZ4CS-3, CZCS-3, CZ6CS-3, and CZ7CS-3, and (d) by changing the quality of CZCS-3 catalyst; (e)
AQE of H2 evolution for CZCS-3 under different wavelength irradiation; (f) Hydrogen production test of CZCS-3
for 20 h.
In order to further confirm that the CoS2/Zn0.5Cd0.5S photocatalyst had excellent cycle stability,
the cycle hydrogen production of CZCS-3 and two single catalysts were tested. Surprisingly, the
cycle hydrogen production rate of the CZCS-3 sample within 30 h did not change significantly
(Figure 7a). Compared with the first round, the H2 production of the sixth round of stability
experiment only decreased by 6.19%, while the hydrogen production of Zn0.5Cd0.5S had a
significant downward trend, indicating that there was a photo-corrosion phenomenon similar to
CdS. Compared with Zn0.5Cd0.5S, the stability of the CZCS-3 sample was enhanced. This was
attributed to the construction of the heterojunction structure between CoS2 and Zn0.5Cd0.5S solid
solution. Figure 7b compared the XRD patterns of the CZCS-3 sample before and after the
hydrogen evolution reaction. Obviously, the XRD diffraction peaks of the CZCS-3 catalyst after
the HER were basically the same as those before the reaction, indicating that the prepared CZCS-3
catalyst did not undergo a phase change after participating in the catalytic reaction. The SEM of
the CZCS-3 catalyst after H2 production (Figure S5, Supporting Information) and before H2
production also did not change significantly, which can further illustrate that the constructed
CoS2/Zn0.5Cd0.5S heterojunction photocatalyst had a high degree of stability.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
Figure 7. (a) Cycle test of CoS2, Zn0.5Cd0.5S and CZCS-3 photocatalyst; (b) XRD patterns before and after
CZCS-3 reaction.
As shown in Figure 8a, in order to study the electron-hole recombination rate of the prepared
sample, the Photoluminescence (PL) spectrum of the samples were measured under the condition
of an excitation wavelength of 350 nm. Photoluminescence was hardly observed with pure CoS2,
while Zn0.5Cd0.5S and CZCS-3 had similar fluorescence emission peaks at the excitation
wavelength of 469 nm.[23] Moreover, the fluorescence quenching effect of CZCS-3 was better than
that of Zn0.5Cd0.5S alone. On the one hand, the added CoS2 had no photoluminescence effect at this
excitation wavelength.[22] On the other hand, it was because the electrons in CZCS-3 could be
transferred quickly, the electron-hole recombination was inhibited. Figure 8b was the
time-resolved fluorescence (TRPL) decay curve of the sample obtained by triple exponential
decay fitting, the average life (Ʈav) calculated through fitting was shown in Table 2.[40] The
average lifetime of pure Zn0.5Cd0.5S was 0.37 ns, while the average lifetime of the CZCS-3 sample
was reduced to 0.07 ns. This indicated that the formation of CoS2/Zn0.5Cd0.5S heterojunction
effectively inhibited the recombination of photogenerated electrons and holes in Zn0.5Cd0.5S, and
made the transfer of photogenerated electrons easier.[35] Therefore, CZCS-3 showed better
photocatalytic H2 evolution performance than Zn0.5Cd0.5S.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
Figure 8. (a) Steady-state fluorescence of Zn0.5Cd0.5S, CZCS-3, and CoS2; (b) Time-resolved fluorescence (TRPL)
decay curves of Zn0.5Cd0.5S and CZCS-3.
Table 2. Fluorescence lifetime parameters of Zn0.5Cd0.5S and CZCS-3 samples.
χ2
Systems
ZnCdS
Lifetime <Ʈ>(ns)
Pre-exponential
factors A/%
Average
lifetime,<Ʈ >(ns)
av
Ʈ1=2.51
Ʈ2=11.80
Ʈ3=0.18
A1=41.37
A2=12.37
A3=46.26
0.37
0.07
1.79
Ʈ1=2.17
Ʈ2=9.29
Ʈ3=0.05
A1=24.73
A2=9.73
A3=65.54
CZCS-3
1.34
In addition to PL spectra, photoelectrochemical experiments were used to further study the
transfer and separation efficiency of generated photoelectrons in the catalyst. Figure 9a recorded
the I-T curves of CoS2, Zn0.5Cd0.5S, and CZCS-3 samples. All three samples had obvious
photocurrent response intensity under four switching cycles of illumination. Generally, the
stronger the photocurrent response intensity, the higher the electron transfer rate in the catalyst,
and the longer the electron lifetime.[41] The comparison showed that the CZCS-3 composite
catalyst exhibits the strongest photocurrent response. This indicated that the coupling effect
between CoS2 and Zn0.5Cd0.5S accelerated the migration of electrons and prolonged the life of
electrons, so that the photocatalytic activity of the CZCS-3 complex was better than that of the
two pure substances.
Figure 9b showed the LSV polarization curve of the sample. Generally, the photocatalytic
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
hydrogen evolution reaction activity is highly dependent on the over-potential of the reaction.[42]
Obviously, under the same voltage, the overpotential of the sample was CoS2, Zn0.5Cd0.5S, and
CZCS-3 in descending order. It further showed that the CZCS-3 sample had good photocatalytic
activity and was a promising material for photocatalytic hydrogen production.
Electrochemical impedance spectroscopy (EIS) is an effective way to further study the electron
migration and recombination in materials. Theoretically, the greater the radius of the arc of the
Nyquist curve, the greater the impedance during electron transfer.[12] In Figure 9c, The arc radius
of CZCS-3 material was the smallest, which effectively verified the increase in the charge transfer
rate, which was consistent with the H2 evolution rate of the CZCS-3 material.
Figure 9. (a) I-T curves, (b) LSV polarization curves and (c) EIS spectra of Zn0.5Cd0.5S, CZCS-3, CoS2.
In order to determine the flat band potential (Efb) and semiconductor type of the CoS2 and
Zn0.5Cd0.5S two single catalysts, the Mott-Schottky curve of the materials were measured in the
photoelectrochemical experiment. The results were shown in Figures 10a and 10c. The slope of
the Mott-Schottky curve was positive, indicating that the measured substance is an n-type
semiconductor.[43,44] Both CoS2 and Zn0.5Cd0.5S were n-type semiconductors, the same as the
previous report.[22,31] One of the conditions for the formation of S-type heterojunction was the
coupling of two n-type semiconductors.[45] In addition, the Efb of CoS2 and Zn0.5Cd0.5S were -0.45
eV and -0.68 eV (vs. saturated calomel electrode (SCE)), respectively. Generally, the relationship
between the conduction band potential (ECB) of an n-type semiconductor and the Efb is: ECB = Efb -
0.10 V,[25] and there is also a certain conversion relationship between the SCE electrode and the
standard hydrogen electrode (NHE), namely: ENHE = ESCE + 0.241 V.[46] Through these two
relations, the ECB values of pure CoS2 and Zn0.5Cd0.5S relative to NHE could be estimated, which
were -0.31 eV and -0.54 eV, respectively. Then, combined with the Eg values obtained from the
previous analysis, the valence band potential (EVB) of pure CoS2 and Zn0.5Cd0.5S relative to NHE
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
were 2.16 eV and 1.81 eV, respectively. The detailed band structure was shown in Figures 10b and
10d.
Figure 10. (a) (c) Mott-Schottky curves of CoS2 and Zn0.5Cd0.5S; (b) (d) Energy band structure diagram of CoS2
and Zn0.5Cd0.5S.
Based on the above characterization analysis and experimental results, the photocatalytic
hydrogen production mechanism of CZCS-3 composite material was inferred. Figure 11a was a
schematic diagram of the relative energy levels before the two pure substances were in contact.
Compared to the valence band (VB) and conduction band (CB) positions of CoS2 and Zn0.5Cd0.5S,
it was not difficult to see that the CB position of Zn0.5Cd0.5S was more negative (with stronger
reducibility), and the VB position of CoS2 was more positive (with stronger oxidizing property). If
the electron transfer followed the type-II heterojunction mode, the overall redox capability of the
heterojunction was weakened. In addition, the latest report claimed that the existence of
electrostatic repulsion made it difficult to realize the electron transfer mode of type-II
heterojunction.[47] Here, we speculated that the CZCS-3 composite material was an S-scheme
heterojunction photocatalyst. As we all know, the Fermi energy of n-type semiconductors is close
to the conduction band.[48,49] It was not difficult to know that compared with CoS2, Zn0.5Cd0.5S had
a larger Fermi energy and a smaller work function. Therefore, as shown in Figure 11b, when
Zn0.5Cd0.5S was in contact with CoS2, the electrons of Zn0.5Cd0.5S were spontaneously dispersed
on CoS2 until the Fermi level was aligned to the same level, and the energy band of Zn0.5Cd0.5S
bended upwards and the energy band of CoS2 bended downwards. At the same time, an internal
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
electric field from Zn0.5Cd0.5S to CoS2 was formed.[47,50] This was consistent with the results of
XPS characterization. When irradiated by visible light (Figure 11c), CoS2 and Zn0.5Cd0.5S were
excited by light to generate electrons (e-) and transition from their respective VB to CB, leaving
oxidizing holes (h+) on the VB. The e- on the CB of CoS2 and the h+ on the VB of Zn0.5Cd0.5S were
affected by the three factors of Coulomb attraction, internal electric field and energy band bending,
and recombined at the contact interface, that was, a stepped (S-scheme) transfer path was
formed.[51,52] Thus, the electrons in Zn0.5Cd0.5S CB, which were relatively strong reduction, were
retained for the reduction of hydrogen. In the meantime, the h+ in the VB of CoS2 and the L-lactic
acid sacrificial agent underwent an oxidation reaction and were consumed, which further inhibited
the recombination of the electrons and holes, and the photocatalytic activity was also enhanced.
Figure 11. Proposed mechanism of H2 production: (a) CoS2 and Zn0.5Cd0.5S before contact, (b) CoS2 and
Zn0.5Cd0.5S after contact, (c) S-scheme electron transfer mechanism after visible light irradiation.
Conclusion
From the results of XRD, XPS and TEM, it was found that CoS2 nanospheres and Zn0.5Cd0.5S
solid solution were successfully supported together, and a new photocatalytic composite material
was prepared. Among them, when the mass ratio of Zn0.5Cd0.5S to CoS2 was 1:0.4 (CZCS-3), the
composite material had the most excellent hydrogen production performance, and the H2
production rate in the fifth hour could reach 25.15 mmol·g-1·h-1. Moreover, CoS2/Zn0.5Cd0.5S had
excellent stability, and the hydrogen production rate of the sixth repeated experiment was only
6.19% lower than that of the first. From the PL spectroscopy and electrochemical characterization
analysis, it was known that the good contact between CoS2 and Zn0.5Cd0.5S made the electron
transfer of CZCS-3 easier, which benefited from the construction of the CoS2/Zn0.5Cd0.5S
S-scheme heterojunction. This S-scheme electron transfer mechanism formed between two n-type
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
semiconductors retained a stronger redox potential, thereby effectively promoting the
photocatalytic hydrogen production activity.
Conflicts of interest
The authors declare that they have no competing interests.
Acknowledgments
This work was financially supported by the Fundamental Research Funds for the Central
Universities of North University for Nationalities (2021KJCX02). This work also was supported
by the Natural Science Foundation of Ningxia Province (2021AAC03180) and the Xixia District
Science and Technology Plan Project (XXKJ1901).
References
[1] H. Chang, H. Wu, Y. T. Yang, L. D. Xie, W. J. Fan, M. Zou, G. Q. Ma, Z. Q. Jiang, Y. Zhang,
Int. J. Hydrogen Energy 2020, 45, 30571-30582.
[2] Y. W. Jiang, M. Z. Li, Y. Mi, L. Y. Guo, W. J. Fang, X. H. Zeng, T. Zhou, Y. S. Liu, Nano
Energy 2021, 85, 105949.
[3] H. Y. Tian, X. Liu, Z. Q. Liang, P. Y. Qiu, X. Qian, H. Z. Cui, J. Tian, J. Colloid Interface Sci.
2019, 557, 700-708.
[4] X. L. Hu, S. C. Lu, J. Tian, N. Wei, X. J. Song, X. Z. Wang, H. Z. Cui, Appl. Catal. B Environ.
2019, 241, 329-337.
[5] W. Zhao, T. T. She, J. Y. Zhang, G. X. Wang, S. J. Zhang, W. Wei, G. Yang, L. L. Zhang, D.
H. Xia, Z. P. Cheng, H. B. Huang, J. Mater. Sci. Technol. 2021, 85, 18-29.
[6] L. Tian, S. X. Min, F. Wang, Appl. Catal. B Environ. 2019, 259, 118029.
[7] M. B. Tahir, M. Sagir, N. Abas, Int. J. Hydrogen Energy 2019, 44, 24690-24697.
[8] X. Liao, J. J. Chen, M. M. Wang, Z. X. Liu, L. J. Ding, Y. Li, J. Alloys Compd. 2016, 658,
642-648.
[9] S. Zhao, J. Xu, Z. T. Li, Z. Y. Liu, Y. R. Li, J. Colloid Interface Sci. 2019, 555, 689-701.
[10] Y. Q. Jing, X. Y. Mu, C. W. Xie, H. L. Liu, R. J. Yan, H. T. Dai, C. L. Liu, X. D. Zhang, Int.
J. Hydrogen Energy 2019, 44, 809-818.
[11] X. Luo, Y. M. Ke, L. Yu, Y. Wang, K. P. Homewood, X. X. Chen, Y. Gao, Appl. Surf. Sci.
2020, 515, 145970.
[12] L. J. Li, J. Xu, S. Zhao, M. Mao, X. H. Li, Int. J. Hydrogen Energy 2021, 46, 1934-1944.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
[13] R. C. Shen, Y. N. Ding, S. B. Li, P. Zhang, Q. J. Xiang, Y. H. Ng, X. Li, Chin. J. Catal. 2021,
42, 25-36.
[14] S. Wan, Y. P. Liu, G. D. Li, X. T. Li, D. J. Wang, X. X. Zou, Catal. Sci. Technol. 2016, 6,
4545-4553.
[15] L. Tian, F. Wang, Z. G. Zhang, S. X. Min, Int. J. Hydrogen Energy 2020, 45, 19017-19026.
[16] L. L. Zhang, H. W. Zhang, C. K. Jiang, J. Yuan, X. Y. Huang, P. Liu, W. H. Feng, Appl.
Catal. B Environ. 2019, 259, 118073.
[17] S. Q. Peng, Y. Yang, J. N. Tan, C. Gan, Y. X. Li, Appl. Surf. Sci. 2018, 447, 822-828.
[18] T. T. Li, K. L. Niu, M. H. Yang, N. K. Shrestha, Z. D. Gao, Y. Y. Song, J. Power Sources
2017, 356, 89-96.
[19] J. Y. Zhang, Y. C. Liu, B. R. Xia, C. Q. Sun, Y. G. Liu, P. T. Liu, D. Q. Gao, Electrochim.
Acta 2018, 259, 955-961.
[20] B. Girginer, G. Galli, E. Chiellini, N. Bicak, Int. J. Hydrogen Energy 2009, 34, 1176-1184.
[21] D. Shu, H. B. Wang, Y. Wang, Y. Li, X. Liu, X. Chen, X. N. Peng, X. N. Wang, P. Ruterana,
H. Wang, Int. J. Hydrogen Energy 2017, 42, 20888-20894.
[22] S. Zhao, J. Xu, M. Mao, L. J. Li, X. H. Li, Appl. Surf. Sci. 2020, 528, 147016.
[23] H. M. Gong, X. J. Zhang, G. R. Wang, Y. Liu, Y. B. Li, Z. L. Jin, Mol. Catal. 2020, 485,
110832.
[24] L. J. Zhang, G. R. Wang, Z. L. Jin, New J. Chem. 2019, 43, 6411-6421.
[25] Z. W. Shao, T. T. Zeng, Y. N. He, D. F. Zhang, X. P. Pu, Chem. Eng. J. 2019, 359, 485-495.
[26] W. X. Liu, L. L. Lu, Q. Li, B. Y. Wu, R. Z. Zhang, W. Shi, P. Cheng, Chem. Eur. J. 2020, 26,
12206-12211.
[27] X. Q. Hao, Z. W. Cui, J. Zhou, Y. C. Wang, Y. Hu, Y. Wang, Z. G. Zou, Nano Energy, 2018,
52, 105-116.
[28] L. M. Song, D. liu, S. J. Zhang, J. F. Wei, Int. J. Hydrogen Energy 2019, 44, 16327-16335.
[29] P. F. Wang, S. H. Zhan, H. T. Wang, Y. G. Xia, Q. L. Hou, Q. X. Zhou, Q. X. Zhou, Y. Li, R.
R. Kumar, Appl. Catal. B Environ. 2018, 230, 210-219.
[30] L. J. Ma, J. Xu, L. J. Li, M. Mao, S. Zhao, New J. Chem. 2020, 44, 18326-18336.
[31] L. j. Li, J. Xu, Appl. Surf. Sci. 2020, 505, 143937.
[32] F. Wang, Y. H. Su, S. X. Min, Y. N. Li, Y. G. Lei, J. H. Hou, J. Solid State Chem. 2018, 260,
23-30.
[33] L. J. Li, J. Xu, X. H. Li, Z. Y. Liu, J. Mater. Sci.: Mater. Electron. 2020, 31, 10657-10668.
[34] X. X. Zhao, J. R. Feng, J. Liu, W. Shi, G. M. Yang, G. C. Wang, P. Cheng, Angew. Chem. Int.
Ed. 2018, 57, 9790−9794.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
[35] H. Z. Deng, X. G. Fei, Y. Yang, J. J. Fan, J. G. Yu, B. Cheng, L. Y. Zhang, Chem. Eng. J.
2021, 409, 127377.
[36] J. Wang, G. H. Wang, B. Cheng, J. G. Yu, J. J. Fan, Chin. J. Catal. 2021, 42, 56-68.
[37] W. K. Wei, Q. F. Tian, H. S. Sun, P. Liu, Y. Zheng, M. Z. Fan, J. D. Zhuang, Appl. Catal. B
Environ. 2020, 260, 118153.
[38] Z. L. Jin, Y. Liu, X. Q. Hao, J. Colloid Interface Sci. 2020, 567, 357-368.
[39] L. J. Zhang, X. Q. Hao, J. K. Li, Y. P. Wang, Z. L. Jin, Chin. J. Catal. 2020, 41, 82-94.
[40] K. Das, S. N. Sharma, M. Kumar, S. K. De, J. Phys. Chem. C 2009, 113, 14783-1479.
[41] X. F. Feng, Z. X. Yu, Y. X. Sun, M. Y. Shan, R. Long, X. H. Li, Sep. Purif. Technol. 2021, 266,
118606.
[42] X. Q. Hao, Y. Hu, Z. W. Cui, J. Zhou, Y. Wang, Z. G. Zou, Appl. Catal. B Environ. 2019, 244,
694-703.
[43] Q. N. Wu, Q. J. Bu, S. Li, Y. H. Lin,X. X. Zou, D. J. Wang, T. F. Xie, J. Alloys Compd. 2019,
803, 1105-1111.
[44] S. L. Bai, J. Y. Han, Y. Y. Zhao, H. M. Chu, S. Q. Wei, J. H. Sun, L. X. Sun, R. X. Luo, D. Q.
Li, A. F. Chen, Renewable Energy 2020, 148, 380-387.
[45] J. K. Li, M. Li, Z. L. Jin, J. Colloid Interface Sci. 2021, 592, 237-248.
[46] Y. B. Li, Z. L. Jin, T. S. Zhao, Chem. Eng. J. 2020, 382, 123051.
[47] Q. L. Xu, L. Y. Zhang, B. Cheng, J. J. Fan, J. G. Yu, Chem, 2020, 6, 1543-1559.
[48] Y. Xu, M. A.A. Schoonen, Am. Mineral. 2000, 85, 543-556.
[49] K. L. Zhang, M. Zhou, C. L. Yu, K. Yang, X. X. Li, W. X. Dai, J. Guan, Q. Shu, W. Y. Huang,
Dyes Pigm. 2020, 180, 108525.
[50] A.Y. Meng, B. Cheng, H. Y. Tan, J. J. Fan, C. L. Su, J. G. Yu, Appl. Catal. B Environ. 2021,
289, 120039.
[51] Y. P. Wang, X. Q. Hao, L. J. Zhang, Z. L. Jin, T. S. Zhao, Catal. Sci. Technol. 2021, 11,
943-955.
[52] T. Yan, H. Liu, Z. L. Jin, Energy Fuels 2021, 35, 856-867.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102811
Chemistry - A European Journal
Table of Contents entry
In this study, Zn0.5Cd0.5S and CoS2 formed an S-type heterojunction, the mixed interface
structure between them provided a new hydrogen-producing active center for each. In addition,
the S-type heterojunction electron transfer mechanism not only promoted the separation of
electrons (e-) and holes (h+) between Zn0.5Cd0.5S and CoS2, but also retained electrons with strong
reducibility, thereby promoting the photocatalytic hydrogen evolution reaction. (In the figure: CB
is the conduction band, VB is the valence band, hv is the photon energy).
This article is protected by copyright. All rights reserved.