Electrochimica Acta 355 (2020) 136754

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

Electroplated synthesis of semi-rigid MoS2 –rGO–Cu as efficient

self-supporting electrode for hydrogen evolution reaction
Junying Huang a, Mengting Chen a, Tao Tang a, Weipeng Liu a,b, Yingju Liu a,b,∗
a
Key Laboratory for Biobased Materials and Energy of Ministry of Education, College of Materials and Energy, South China Agricultural University,
Guangzhou 510642, China
b
Guangdong Laboratory of Lingnan Modern Agriculture, Guangzhou 510642, China

a r t i c l e i n f o a b s t r a c t

Article history: The self-supporting catalytic electrode is useful in the industry, since it can reduce the complex pre-
Received 29 April 2020 treatment of catalytic materials and avoid the use of binders. In this work, a ternary self-supporting
Revised 4 July 2020
electrode including Cu mesh-supported graphene and molybdenum disulfide (MoS2 ) was prepared by
Accepted 9 July 2020
electroplating zinc and hydrothermal reaction. Due to the excellent conductivity of copper and the hydro-
Available online 21 July 2020
gen evolution catalytic ability of MoS2 , this self-supporting electrode showed a low Tafel slope as 54 mV
Keywords: dec−1 and a current density of 400 mA cm−2 at an overpotential of -300 mV in hydrogen evolution reac-
Self-supporting electrode tion (HER). In addition, due to the metal flexibility of the Cu mesh substrate, the MoS2 –rGO–Cu ternary
Hydrogen evolution reaction self-supporting electrode can be bent or cut into any desired shape. Even shaped as an elongated one,
Molybdenum disulfide curved "S", this self-supporting electrode still kept itself from any deformation when hydrogen bubbles
Graphene oxide were rapidly generated. Therefore, a strategy for green, simple, and low-cost self-supporting electrode
was provided, which may increase the large-scale operation of the catalytic electrode in HER.
© 2020 Elsevier Ltd. All rights reserved.

1. Introduction carbon electrode or ITO. A binder such as Nafion or polytetraflu-

oroethylene is also required, but the binder can cover the active
The long-term use of fossil energy has caused the depletion of sites of the material and create an open circuit between catalytic
fossil energy and the global warming. Hydrogen, as a renewable units [9,10]. Thus, the catalytic performance and electron transport
and carbon-free energy source, hasn’t been popularized due to the capacity are influenced. Moreover, when the powder materials are
preparation problems [1,2]. At this stage, the best catalytic materi- dropped on the conductive surface, the accumulation and aggre-
als for hydrogen evolution are precious metals such as Pt, Ir and Ru gation of the material are inevitable [11]. As the area of the con-
[3]. However, its high cost and unsatisfactory stability make them ductive surface increases, the aggregation of powder materials be-
difficult to popularize. As a possible substitute for noble metal cat- comes more serious. Especially, when the current density increases
alysts, MoS2 has attracted attention due to its easy processing and with the voltage, hydrogen bubbles are also produced between the
abundant reserves. Many researchers have been devoted to regulat- catalytic material and the conductive surface, which can limit the
ing MoS2 morphologies, such as the preparation of nanoparticles, current density of the catalytic material and reduce the hydrogen
nano-ribbons, and ultra-thin nanosheets to increase active sites [4– production rate [12].
6]. In addition, MoS2 -based composites such as carbon nanotubes In contrast, the self-supporting catalytic electrode is more ben-
and graphene have been used to enhance the electron transport eficial to industrial application, since it reduces the complex pre-
capacity [7,8]. treatment of catalytic materials and avoids the use of binders
When investigating the properties of MoS2 , the macroscopic [13,14]. However, most self-supporting catalytic materials are usu-
state of catalytic materials is usually powder, which may hinder ally prepared from foamed nickel or foamed copper [15,16]. It is
the application in the hydrogen production. The powder usually difficult to bend them in a desired shape, thus the flexibility is in-
needs to be modified on the conductive interface such as glassy sufficient. In contrast, graphene sheets-based self-supporting elec-
trodes have unparalleled flexibility and excellent bending prop-
erties [17,18]. However, the poor rigidity of graphene sheets can
∗
Corresponding author at: Key Laboratory for Biobased Materials and Energy of cause the moving of graphene sheets in the solution by hydrogen
Ministry of Education, College of Materials and Energy, South China Agricultural bubbles [19].
University, Guangzhou 510642, China.
E-mail address: liuyingju@hotmail.com (Y. Liu).

https://doi.org/10.1016/j.electacta.2020.136754
0013-4686/© 2020 Elsevier Ltd. All rights reserved.
2 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754

In this work, graphene and MoS2 were coated on the surface and maintained at 200 °C for 24 h. After soaking in water, ethanol
of the Cu mesh by double electroplating of zinc and hydrother- and water for every 20 min, the MoS2 –rGO–Cu product was ob-
mal reaction. As the common metal material in electric and elec- tained. As a control group to investigate the effect of second galva-
tronic components, a Cu mesh was used as a basic conductive nization, the product obtained under the same conditions without
substrate. After covering with graphene, the Cu mesh provided a the second galvanization or the immersion of the molybdenum salt
complete and continuous electron transport channel for the frag- solution, was labeled as MoS2 –rGO–Cu/WSZ. The WSZ in MoS2 –
mented graphene. Thus, the electron transport capacity was greatly rGO–Cu/WSZ was used to mark the self-supporting electrode with-
enhanced by the Cu network. It also provided a rigid skeleton for out the second-electroplating zinc.
the graphene sheet, which improved the mechanical strength of
the catalytic electrode. Additionally, the graphene not only pro- 2.3. Preparation of MoS2 -rGO sheet
vided a platform for the subsequent MoS2 , but also formed a film
for the Cu mesh, which can protect the Cu mesh from high temper- The MoS2 –rGO sheet was prepared as a reference. The 3 × 2
ature corrosion and oxidative corrosion. The zinc nanosheets pro- cm2 zinc flake was immersed in ethanol and sonicated for 20 min.
duced by the double electro-galvanization provided sufficient re- After one side was glued with transparent adhesive, the zinc flake
action sites for the loading of the graphene layer and the forma- was immersed in a 0.5 g/L GO solution at 4 °C for 20 min and fur-
tion of the molybdenum layer. The final self-supported electrode ther in 8.3 g/L ammonium molybdate solution at room temper-
showed excellent electron conductivity to realize the current den- ature for 3 h. After the zinc flake was lyophilized at −20 °C for
sity of 400 mA cm−2 at −300 mV and the Tafel Slope as low as 5 h, the graphene sheet was peeled off and dried. Additionally,
54 mV dec−1 . ammonium molybdate tetrahydrate (0.05 mmol) and l-cysteine
(2.8 mmol) were successively dissolved in 60 mL of deionized wa-
2. Experimental ter and stirred for 10 min, and then 450 μL of concentrated HCl
was added and stirred for another 30 min. After the solution was
2.1. Chemicals and instruments transferred to a 100 mL PTFE reactor, the above graphene sheet was
fixed in a PTFE frame, placed in a stainless steel autoclave, and re-
Graphite flakes (99.8%, Alfa Aesar), ammonium molybdate acted at 200 °C for 24 h. The MoS2 –rGO product was soaked in wa-
tetrahydrate (Aladdin) and l-cysteine (Aladdin) were directly used ter, ethanol and water for each 20 min.
without any purification. The Cu mesh (99.9%, Meidi family) was
cut into 3 × 0.5 cm2 , soaked inhydrochloric acid (3.6%) for 30 s, and 2.4. Electrochemical measurement
washed with deionized water. Then, it was immersed in ethanol
for 10 min ultra-sonication, and the ethanol solution was changed Before the test, all the samples were cut to the size of 1 × 0.5
every 2.5 min. The clean copper mesh was electroplated by using cm2 and fixed with titanium wire electrode. After the clamping
a constant voltage current device (wanptek-KPS603D) as a power electrode was tightened, a part of the self-supporting electrode
source and the Pt plate (1 cm2 ) as an anode. The surface change was exposed outside with the area of 0.5 × 0.5 cm2 . All tests were
of the self-supporting electrode was observed by scanning elec- carried out in a 0.5 M H2 SO4 solution with uninterrupted flow of
tron microscopy (SEM, Hitachi S-4800). X-ray powder diffractome- N2 (99.999%). The potential in the work was regulated by a re-
ter (XRD, Rigaku-Ultima IV) and energy dispersive spectroscopy versible hydrogen potential (RHE). The polarization curve was mea-
(EDS, Hitachi) were used to detect the type and distribution of sured between 0.1 V and −0.4 V at a scan rate of 1 mV/s. The in-
elements of the material. X-ray photoelectron spectroscopy (XPS, ternal resistance of the three-electrode system was automatically
Thermo fisher) was used to further investigate the elemental va- compensated by the IM6ex electrochemical workstation. The elec-
lence and chemical bonds of the sample. The macroscopic proper- trochemical impedance spectroscopy (EIS) was measured from 100
ties of the material at different preparation stages and the hydro- KHz to 0.1 Hz at a potential of −0.25 V. The stability of MoS2 –
gen evolution of the self-supporting electrode were recorded by a rGO–Cu can be judged by comparing the polarization curves be-
digital camera (Olympus, SZ-14). Electrochemical performance was tween 0.2 V and −0.2 V before and after 20 0 0 cycles of cyclic
measured by an electrochemical workstation (Zahner-IM6ex) and a voltammetry (CV) at a scan rate of 50 mV/s.
standard three-electrode system from Hg|Hg2 SO4 electrode (stan-
dard electrode, filled in saturated K2 SO4 ), carbon rod (counter elec- 3. Results and discussion
trode) and the self-supporting electrode (working electrode).
3.1. Digital characterization of the preparation steps for the
2.2. Preparation of MoS2 -rGO–Cu MoS2 -rGO–Cu

Firstly, graphene oxide (GO) was prepared by an improved The MoS2 –rGO–Cu was prepared with five steps as Scheme 1.
Hummer method. The Cu mesh was electroplated in 125 g/L ZnSO4 The color of the material changed from reddish brown (A), going
solution for 1 min at a constant voltage of 4.0 V, and the galva-
nized Cu mesh (Zn-Cu) was washed with ethanol. Then, it was im-
mersed in a 0.5 g/L GO solution for 10 min at 4 °C. The graphene-
loaded copper mesh (rGO–Cu) was further electroplated in the
above ZnSO4 solution for another 20 s at the same voltage. The
secondary galvanized Cu Mesh (Zn–rGO–Cu) was immersed in an
ammonium molybdate solution (8.3 g/L) for 3 h at room tempera-
ture (Mo-Zn–rGO–Cu). In addition, ammonium molybdate tetrahy-
drate (0.025 mmol) and l-cysteine (1.4 mmol) were successively
dissolved in 30 mL of deionized water and stirred for 10 min, then
75 μL of concentrated HCl was added and stirred for another
30 min. After the solution was transferred to a 50 mL polytetraflu-
oroethylene (PTFE) reactor, the Mo-Zn–rGO–Cu was fixed in a PTFE Scheme 1. Scheme for the synthesis of self-supporting electrode. (A) Cu mesh, (B)
frame. Then, the frame was placed in a stainless steel autoclave, Zn-Cu, (C) rGO–Cu, (D) Zn–rGO–Cu, (E) Mo-Zn–rGO–Cu, (F) MoS2 –rGO–Cu.
 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754 3

Fig. 1. SEM images at different magnification of (A-C) rGO–Cu and (D-F) Zn–rGO–Cu.

off-white (B), bright black (C), gray-black (D), silver-gray (E), to the in the GO solution (Fig. 1) suggested that a large amount of GO
final matte black (F). The five color changes in B-F corresponded was reduced due to the replacement reaction of zinc nanosheets.
to the first galvanization, dipping in GO solution, the second galva- rGO sheets were uniformly attached to the Cu mesh, forming a
nization, dipping in ammonium molybdate solution, and the final macroscopic integrated graphene layer on Cu mesh. In order to
hydrothermal reaction. The graphene layer was successfully loaded study the adhesion of the graphene layer on the Cu mesh, the
on the surface of the copper mesh after the first galvanization outer graphene of rGO–Cu was slightly destroyed and observed by
and immersion of the GO solution, while the second electroplat- SEM. In Fig. 1A, the graphene sheets formed a complete plane. By
ing of zinc and immersion in ammonium molybdate solution made observing the cracks in the graphene layer (Fig. 1B), a large cylin-
the molybdenum layer on the graphene surface to provide bind- drical structure under the outer layer of graphene can be found,
ing sites for MoS2 loading during the hydrothermal reaction. As a which can be identified as part of the copper mesh by comparison
control experiment, the amount of hydrochloric acid was doubled with Fig. S3. Therefore, under the graphene layer, pure and con-
under the same process (MoS2 –rGO–Cu-HD), or the hydrothermal tinuous copper wires were staggered to form an excellent electron
reaction was directly performed on the copper mesh, which can transport network. In Fig. 1C, under the flat graphene layer (outer
make the structure of the materials collapse (Fig. S1 and S2). graphene), there was also a layer of graphene (inner graphene),
Scheme 1A was the initial Cu mesh. After the first 1-min elec- which wrapped around the Cu mesh to further enhance the cor-
trodeposition of zinc nano-flakes, the reddish-brown Cu mesh was rosion resistance. The zinc nanosheets on the surface of the cop-
uniformly transformed into grayish white (Scheme 1B). After wash- per mesh in the previous step can firmly fix the graphene sheet.
ing with ethanol and drying, the Zn-Cu was immersed in a 0.5 g/L Without graphene, the copper mesh may dissolve during the hy-
GO solution at 4 °C for 10 min, and then washed with deionized drothermal reaction (Fig. S2).
water. Thus, the graphene was uniformly covered on the Cu mesh In order to provide growth sites for the subsequent hydrother-
(Scheme 1C). Then, the rGO–Cu was immersed in zinc sulfate so- mal reaction of MoS2 , a second electrodeposition was carried out
lution for a second 20 s electroplating, which can provide attach- on rGO–Cu as a pre-reaction. The color of Zn–rGO–Cu changed
ment sites for MoS2 . Compared with rGO–Cu, the surface color from black to gray-black after the electrodeposition of zinc. In
of Zn–rGO–Cu changed from black with metallic luster to gray- Fig. 1D, the smooth outer layer of graphene became very rough
ish black (Scheme 1D). Successively, the Zn–rGO–Cu was immersed after the second electrodeposition. At a larger magnification in
in an ammonium molybdate solution for 3 h (Scheme 1E). Finally, Fig. 1E, the surface of the graphene layer was densely covered with
such modified Cu mesh was treated with 24 h of hydrothermal zinc nanosheets. Zinc nanosheets, as in Fig. 1F, were tilted up to
reaction at 200 °C, and the final product was obtained. Although reveal vertical edges.
the product in Scheme 1F was very similar to the black color in In Fig. 2, the surface of Mo-Zn–rGO–Cu became bright silver af-
Scheme 1D, the rGO–Cu had a metal-like reflection on the surface, ter soaking in the molybdenum salt solution. Compared with the
but the MoS2 –rGO–Cu was a deep and matte black. The control previous Zn–rGO–Cu, the metal reflection on the Mo-Zn–rGO–Cu
group, the dosage of HCl was doubled to obtain the product of surface was obvious. In the SEM image of Fig. 2A, after immersing
MoS2 –rGO–Cu-HD. It was easily broken and the Cu mesh wrapped in the molybdenum salt solution, the Cu mesh was significantly
by inner graphene was dissolved in Fig. S1. In addition, if the Cu smooth compared with the Zn–rGO–Cu. From the further enlarged
mesh wasn’t soaked in GO solution and directly subjected to the SEM image (Fig. 2B), zinc nanosheets were wrapped by the sub-
hydrothermal reaction, the structure was collapsed (Fig. S2). sequent molybdenum to form a relatively flat metal plane. There
were some cracks on the surface of molybdenum layer, since some
3.2. Morphological characterization of the MoS2 -rGO–Cu zinc nanocrystals were stuck up to prevent the molybdenum layer
from fully closing during the immersion in molybdenum salt solu-
In order to understand the change of zinc electrodeposition on tion. Further enlargement of the crack position, from Fig. 2C, the
the Cu mesh, the SEM was used to observe the Cu mesh before and edge contour of the zinc nanosheet can be seen on the surface of
after galvanization. After zinc was deposited on the copper mesh, the flat molybdenum layer. In the cracks, what can be seen was
the color turned gray. A more intuitive distinction can be observed not the bare copper mesh, but the outer layer of flat graphene.
from the SEM. The Cu mesh before electrodeposition (Fig. S3A) This indicated that cracks formed by the combination of molybde-
showed a relatively smooth surface without obvious undulations. num cannot damage the integrity of the outer layer of graphene,
After electrodeposition, the surface of the galvanized Cu mesh (Fig. thus the integrity of electric channel was maintained and the cor-
S3B) showed a flannel-like surface, and the zinc nanosheets were rosion resistance of the inner Cu mesh was not affected. In order to
evenly adhered to the surface of the Cu mesh. The further soaking further understand the change in the sample after soaking in the
4 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754

Fig. 2. (A–C) SEM image at different magnification of Mo-Zn–rGO-Cu.

Fig. 3. (A–D) SEM images at different magnification of MoS2 –rGO–Cu, and the EDS mapping of (E) Cu, (F) Mo and (G) S.

molybdenum salt solution, a relatively flat position (Fig. S4A) was In Fig. 4A, the same peaks of six samples were clearly seen at
selected on the Cu mesh for elemental testing. In Fig. S4B-D for the 43.5°, 50.6° and 74.2°, which represented the crystal faces of cop-
EDS of Mo, Cu, and Zn, such elements were uniformly distributed per (111), (200) and (220) [20,21]. As the experiment progressed,
on the surface of the material. At the same time, the Cu mesh was different materials were loaded on the surface of the Cu mesh,
acted as an electron transport channel under the graphene layer to and the XRD spectrum showed regular changes. The characteris-
enhance the conductivity of the surface. tic peaks of zinc, graphene and MoS2 appeared successively. At
After the hydrothermal reaction, as in Fig. 3, the MoS2 –rGO–Cu the first and second electrodeposition of zinc, the same charac-
showed only a deep black color. The effect of hydrothermal reac- teristic peak of Zn appeared in the XRD spectrum. From Fig. 4B,
tion on the copper mesh was also observed by SEM. In Fig. 3A, the the XRD magnified the view of Zn–rGO–Cu, showing that the two
surface of the self-supporting electrode became very rough with characteristic peaks represented the (002) and (100) crystal faces
cracks and some granular materials. By increasing the magnifica- of Zn [22]. After immersing Zn-Cu in GO solution, the character-
tion, the material under the cracks in Fig. 3B was not a Cu wire, istic peak of Zn on rGO–Cu was weakened (Fig. 4A). In addition,
but a flat outer graphene layer on the surface of the Cu mesh. the peaks of rGO and GO appeared simultaneously in the spec-
Therefore, the hydrothermal reaction cannot destroy the outer trum [23,24]. This indicated that the reduction adsorption of Zn
graphene or affect the structural stability of the self-supporting for GO wasn’t completely reduced the oxidized groups of GO, so
electrode. At the larger magnification of Fig. 3C, it was loaded with rGO–Cu still had a certain hydrophilicity. After the second galva-
countless flower-like MoS2 and active sites were exposed. From nizing and soaking in molybdenum salt solution (Fig. 4C), the peak
the SEM image, there was a flat graphene layer under the large of Zn substantially disappeared. Three new characteristic peaks of
amount of MoS2 , which implied that the material on the surface of 12.8°, 25.7°, and 39.0° corresponded to the (020), (040), and (060)
the self-supporting electrode was a mixture of graphene and MoS2 . crystal faces of MoO3 [25]. From the characteristic peak of ZnO in
The formation of MoS2 on the surface of the self-supporting elec- the Zn 2p spectrum (Fig. S5), the formation of the molybdenum
trode was prepared by a two-step course. First, the second electro- oxide layer can be inferred [26]. When the electrode undergone
plating zinc was formed on the surface of the graphene layer. Due the second electroplating zinc and immersed in the ammonium
to the reduction ability of zinc, ammonium molybdate was trans- molybdate solution, zinc reduced the ammonium radical to am-
formed into a molybdenum layer bonded with the graphene sur- monia [27]. Zn2+ forms a Mo-O-Zn bond with molybdate, making
face. Second, by using such molybdenum layer as the binding sites, the formation of the molybdenum oxide layer on the surface of
the free molybdenum source can transform to MoS2 with the aid the graphene [28]. Finally, after the hydrothermal reaction, as in
of l-cysteine during the hydrothermal reaction. Fig. 4D, the peak of GO almost disappeared. The most obvious peak
As for the EDS of self-supporting electrode (Fig. 3D), the el- of the self-supporting electrode, in addition to the previous peak
ement types of the products were monitored. From Fig. 3E-G, of Cu and MoO3 , was corresponded to the (100), (103) and (110)
the surface was uniformly distributed with elements such as Cu, planes of MoS2 [29]. This proved that MoS2 was successfully at-
Mo, and S. In comparison with Fig. S4B-D, MoS2 was successfully tached to the surface of the self-supporting electrode, while MoO3
bonded to the surface of the self-supporting electrode. The copper may be the reaction sites of MoS2 during hydrothermal reaction, as
element was from the copper mesh on the self-supporting elec- well as the adhesive between MoS2 and self-supporting electrode
trode substrate. The existence of copper mesh and its synergistic [30].
effect with MoS2 can improve the conductivity of the material and In order to better understand the influence of hydrothermal
optimize the electrocatalytic performance of the electrode [19]. reaction on the formation of self-supporting electrode, Mo-Zn–
 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754 5

Fig. 4. (A) XRD images of Cu, Zn-Cu, rGO–Cu, Zn–rGO–Cu, Mo-Zn–rGO–Cu and MoS2 –rGO–Cu, and (B-D) enlarged images of Zn–rGO–Cu, Mo-Zn–rGO–Cu and MoS2 –rGO–Cu.

rGO–Cu and MoS2 –rGO–Cu were tested by XPS. In the Cu 2p MoO3 was reduced as MoS2 . From the Mo 3d spectra of MoS2 –
spectrum of Fig. 5A, before the hydrothermal reaction, the Mo- rGO–Cu, the original peak area of Mo6+ was greatly reduced. In ad-
Zn–rGO–Cu had double peaks as Cu 2p1/2 (∼952.56 eV) and Cu dition, double peaks of 3d3/2 (∼ 233.8 eV) and 3d5/2 (∼230.3 eV) of
2p3/2 (∼ 932.7 eV) of Cu element [40], while the CuO 2p3/2 Mo5+ can be found, suggesting the incomplete reduction of Mo6+
(∼934.6 eV), CuO 2p1/2 (∼954 eV) and CuO 2p3/2 , sat (∼942.87 eV) [36]. The 3d3/2 (∼ 232.2 eV) and 3d5/2 (∼ 229.1 eV) of Mo4+ , as
peaks also appeared [31]. After the hydrothermal reaction, the Cu well as the 2p1/2 (∼ 163.05 eV) and 2p3/2 (∼ 161.9 eV) of S 2p spec-
2p spectra of MoS2 –rGO–Cu only had the peaks of Cu, and the tra (Fig. 5E), proved the successful synthesis of MoS2 [37]. The va-
CuO peak was reduced. As for the C 1 s spectra of Fig. 5B, the lence state change between the above elements indicated that the
most characteristic peak of Mo-Zn–rGO–Cu and MoS2 –rGO–Cu was graphene layer and the MoO3 layer of Mo-Zn–rGO–Cu had a re-
C–C/C=C (∼284.8 eV), which was due to the graphene [32]. It was duction reaction during the hydrothermal process. The Mo6+ on
further found that Mo-Zn–rGO–Cu also contained peaks of C–O the electrode surface and the solution together can be served as a
(∼285.85 eV) and C=O (∼287.5 eV) [33]. However, after hydrother- molybdenum source, so the flower-like MoS2 was stably bonded to
mal reaction, the C=O group on MoS2 –rGO–Cu disappeared. In ad- the surface of the self-supporting electrode. The unreduced parts
dition, the peak area of C–C/C=C on MoS2 –rGO–Cu was larger than of the MoO3 layer can act as a binder between MoS2 and the outer
that on Mo-Zn–rGO–Cu, where the peak area of C–O was smaller layer of graphene.
than that on Mo-Zn–rGO–Cu. Therefore, the proportion of C–O be-
came smaller in MoS2 –rGO–Cu, further indicating that the self- 3.3. Electrochemical performance of the MoS2 -rGO–Cu towards HER
supporting material underwent a reduction reaction during the hy-
drothermal reaction. In Fig. 6A, the current densities of the pure Cu mesh and
In the O 1 s spectra (Fig. 5C), both Mo-Zn–rGO–Cu and MoS2 – rGO–Cu were very low, thus the catalytic hydrogen evolution per-
rGO–Cu contained C–O (532.5 eV) and C=O (∼531.7 eV) peaks. In formance was poor. After graphene and MoS2 were supported
addition to the peak of the oxidizing groups on the surface of the on the surface, the current density can be greatly improved. Un-
graphene, the peak of Mo-Zn–rGO–Cu at ∼530.7 eV represented der the overpotential of −300 mV, the current density of the
CuO and MoO3 [34]., [35] The two peak positions of CuO and MoS2 –rGO sheet was 21 mA cmgeo −2 , while the current density of
MoO3 were basically the same, so their characteristic peaks were the MoS2 –rGO–Cu/WSZ was 57 mA cmgeo −2 . The MoS2 –rGO sheet
overlapped. However, the peak of MoS2 –rGO–Cu at the position had a low current density since the graphene sheets had many
of ∼ 530.7 eV only contained MoO3 . CuO was reduced in MoS2 – transverse faults and poor longitudinal electron transport capac-
rGO–Cu, which can be confirmed by the Cu 2p spectra that the ity [38,39]. However, the MoS2 –rGO–Cu/WSZ was inferior to the
MoS2 –rGO–Cu had only the peak of Cu. Turning attention to the MoS2 –rGO–Cu because of its weak bonding ability with MoS2 . In
change of molybdenum, in the Mo 3d spectrum of Mo-Zn–rGO–Cu addition to the Pt electrode, MoS2 –rGO–Cu had the best perfor-
in Fig. 5D, there were only 3d3/2 (∼ 235.8 eV) and 3d5/2 (∼233.1 eV) mance. At the overpotential of −300 mV, the current density of
of MoO3 . It indicated that the peak of ∼530.7 eV of Mo-Zn–rGO–Cu MoS2 –rGO–Cu reached 400 mA cmgeo −2 . This current density was
in O 1 s spectrum not only represented copper oxide, but also rep- also relatively high in self-supporting electrode due to excellent
resented molybdenum trioxide. After the hydrothermal reaction, electron transport performance (Table S1).
6 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754

Fig. 5. XPS deconvolution of Cu 2p (A), C 1 s (B), O 1 s (C), Mo 3d (D) and S 2p (E) for Mo-Zn–rGO–Cu and MoS2 –rGO–Cu.

To further understand the electrochemical performance of Cu mesh had the highest Tafel slope, indicating its catalytic abil-
MoS2 –rGO–Cu, the polarization curves of each sample at low cur- ity was the worst. When the Cu mesh was covered with graphene,
rent density were observed. Fig. 6B showed that the current den- the slope of Tafel was slightly improved, since the graphene on the
sity of pure copper mesh and rGO–Cu was less than 10 mA cmgeo −2 surface of the Cu mesh formed a flat surface and the catalytic re-
even at 400 mV. As for the polarization curves of other three action area increased. However, both graphene and Cu mesh can-
samples, at the current density of 10 mA cmgeo −2 , the overpo- not significantly promote the catalytic performance of hydrogen
tential had an obvious sequence as MoS2 –rGO sheet (236 mV) > evolution. When the surface of graphene was further loaded with
MoS2 –rGO–Cu/WSZ (210 mV) > MoS2 –rGO–Cu (200 mV). The cur- MoS2 , the Tafel slope of the composite material decreased due to
rent density at 0 mV was not 0 mA cmgeo −2 , suggesting that the the excellent HER performance of flower-like MoS2 . At the same
electrical double layer on the surface was slightly discharged [40]. time, the catalytic performance of MoS2 –rGO–Cu/WSZ was better
The electrical double layer current density (at 0 mV) of MoS2 – than MoS2 –rGO sheet due to the electron transport channel con-
rGO sheet and MoS2 –rGO–Cu were 0.37 mA cmgeo −2 , and the on- structed by copper mesh. From the highlight of the Tafel slope of
set potential of MoS2 –rGO sheet (126 mV) was smaller than MoS2 – each sample, the influence of the surface structure on the catalytic
rGO–Cu (152 mV). However, the overpotential of the MoS2 –rGO HER can be further investigated. In Fig. S6, as the increase of cur-
sheet at a current density of 10 mA cmgeo −2 was larger than that rent density, the Tafel slope first appeared as a plateau area and
of MoS2 –rGO–Cu, which was due to the dispersion of MoS2 on the then increased. The hydrogen bubbles created on the electrode sur-
surface of graphene. In the preparation of hydrothermal reaction, face had no time to disassemble and began to interfere with the
the amount of HCl for the synthesis of the MoS2 –rGO sheet was reaction between the electrolyte and the catalytic electrode, re-
three times that of MoS2 –rGO–Cu. HCl can prevent the formation sulting in an increase in the Tafel slope. The rougher the surface,
of large crystal nuclei of MoS2 and enhance its dispersion, while the greater the upper limit of the Tafel slope. In Fig. S6, a clear
the dispersed MoS2 nanosheets can provide more active sites. At sequence of Tafel upper limit can be found as MoS2 –rGO sheet
the same time, since MoS2 –rGO sheets lacked electron transport (39.8 mA cmgeo −2 ) > MoS2 –rGO–Cu (34.7 mA cmgeo −2 ) > MoS2 –
channel, the current density increased at a low speed, and the fi- rGO–Cu/WSZ (13.5 mA cmgeo −2 ) > Pt (5.89 mA cmgeo −2 ) > rGO–Cu
nal current was smaller than that of MoS2 –rGO–Cu. The electrical (2.14 mA cmgeo −2 ) > Cu mesh (0.64 mA cmgeo −2 ). MoS2 –rGO sheets
double layer current density of MoS2 –rGO–Cu/WSZ was the largest, had the highest Tafel upper limit, since the MoS2 on the surface
about 1 mA cmgeo −2 , but the increase rate of subsequent current was mostly dispersed, which can remove the generated hydrogen
density was lower and the overpotential was larger due to its lack bubbles quickly to increase the upper limit. In order to strike a
of sufficient catalytic active sites. balance between protecting the copper mesh and dispersing MoS2 ,
In Fig. 6C, the Tafel slope of the Pt electrode was 31 mV dec−1 , the amount of HCl should be reduced in MoS2 –rGO–Cu. The Tafel
which was the same as the values in earlier reports [41], indi- upper limit of MoS2 –rGO–Cu/WSZ was lower than that of MoS2 –
cating that the data of the electrochemical test was reliable. In rGO–Cu. The lack of second galvanization for MoS2 –rGO–Cu/WSZ
Fig. 6C, a significant order of the Tafel slope between the six sam- can decrease binding sites for MoS2 loading, thus a smooth surface
ples can be discovered as MoS2 –rGO–Cu (54 mV dec−1 ) < MoS2 – resulted in a smaller Tafel upper limit.
rGO–Cu/WSZ (72 mV dec−1 ) < MoS2 –rGO sheet (83 mV dec−1 ) < In order to further understand the electron transport proper-
rGO–Cu (97 mV dec−1 ) < Cu Mesh (101 mV dec−1 ). The smaller ties of the materials, the electrochemical impedance spectroscopy
Tafel slope stands for the easier HER and the faster reaction rate. (EIS) of each sample was tested by a three-electrode system. The
 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754 7

Fig 6. Polarization curves at (A) high current density and (B) low current density and (C) Tafel slopes, (D) Nyquist plots and (E) its partial enlarged zone at the overpotential
of −250 mV of Cu mesh, rGO–Cu, MoS2 –rGO sheet, MoS2 –rGO–Cu/WSZ, MoS2 –rGO–Cu, Pt, and (F) polarization curves of MoS2 –rGO–Cu before (black) and after (red) 20 0 0
cycles of CV at 50 mV/s.

equivalent circuit of the test system was in the inset of Fig. 6D. As
can be seen from the Nyquist plots in Fig. 6D, the charge trans-
fer impedance (Rct ) of the copper mesh and rGO–Cu were signif-
icantly greater than that of other three samples, suggesting that
the presence or absence of MoS2 had a significant effect on mate- Fig. 7. Digital photos of “s, c, a, u”-shaped MoS2 –rGO–Cu electrode in actual work.
rial properties. From EIS, the most significant influence of MoS2 on
the sample was the charge transfer impedance in the electrolysis
of hydrogen production. The Cu mesh and rGO–Cu without MoS2 the hydrothermal reaction, the growth of molybdenum layer and
had much higher Rct than the samples with MoS2 . The impedance MoS2 on the surface of the graphene further strengthened the sta-
of rGO–Cu was slightly smaller than that of copper mesh, because bility. In addition, there was no void between the catalytic host
graphene provided a larger area to contact with the electrolyte. and the conductive support due to the integrated process of MoS2 –
From the enlarged EIS image of Fig. 6E, the MoS2 –rGO–Cu had the rGO–Cu, which allowed the resulting hydrogen bubbles to derive
smallest impedance. The impedance of MoS2 –rGO sheet was larger from the electrode surfaces rapidly.
than that of MoS2 –rGO–Cu/WSZ, suggesting copper mesh can ef- In Video S1, the surface of four electrodes shaped with "s, c,
fectively improve the conductivity of the material. a, u" was full of fine bubbles which were constantly generated
As in Fig. 6F, the polarization curve after running the CV of and discharged. In Fig. 7 of the digital photos, the electrode ma-
20 0 0 cycles was basically matched with the original curve. The re- terial can be cut into the letter pattern, which did not show any
sults showed that the HER performance of MoS2 –rGO–Cu was rel- deformation or breakage during the hydrogen evolution reaction.
atively stable. As in Fig. S7, the surface of the self-supporting elec- Therefore, as a semi-rigid self-supporting electrically conductive
trode was not seriously detached or damaged after 20 0 0 cycles. material, MoS2 –rGO–Cu can be cut or bent into any desired shape.
Zinc nanosheets on the surface of the graphene reinforced the cov- The self-supporting hydrogen evolution catalytic materials such as
erage of graphene as a binder. By the displacement reaction and graphene paper may be pushed away from the electrolyte due to
8 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754

the slow exhaust rate of hydrogen bubbles and the lack of rigidity [4] X. Qian, J. Ding, J. Zhang, Y. Zhang, Y. Wang, E. Kan, X. Wang, J. Zhu, Ultrathin
of the materials. In contrast, MoS2 –rGO–Cu can maintain relative molybdenum disulfide/carbon nitride nanosheets with abundant active sites
for enhanced hydrogen evolution, Nanoscale 10 (2018) 1766–1773, doi:10.1039/
mechanical strength in the catalytic process to avoid the deforma- c7nr07213a.
tion. [5] Y. Ouyang, Q. Li, L. Shi, C. Ling, J. Wang, Molybdenum sulfide clusters immobi-
lized on defective graphene: a stable catalyst for the hydrogen evolution reac-
tion, J. Mater. Chem. A 6 (2018) 2289–2294, doi:10.1039/c7ta09828f.
4. Conclusions [6] Y. Liu, Y. Zhu, X. Fan, S. Wang, Y. Li, F. Zhang, G. Zhang, W. Peng, (0D/3D) MoS2
on porous graphene as catalysts for enhanced electrochemical hydrogen evo-
lution, Carbon N Y 121 (2017) 163–169, doi:10.1016/j.carbon.2017.05.092.
In summary, the semi-rigid self-supporting electrode including [7] X. He, F. Yin, B. Chen, G. Li, H. Yin, “Bulk” 1T/2H-MoS2 with tunable phases and
the Cu mesh skeleton, graphene layer and flower-like MoS2 was residual S, N co-doped carbon as a highly active and durable catalyst for hy-
successfully prepared by the combination of double electroplat- drogen evolution, ACS Appl. Energy 2 (2019) 2022–2033, doi:10.1016/10.1021/
acsaem.8b02109.
ing and hydrothermal reaction. The Cu mesh provided an excel-
[8] F. Wang, S. Song, J. Li, J. Pan, Wang X, H. Zhang, S,N co-doped carbon
lent electron transport channel and a template for the electrode. nanotubes decorated with ultrathin molybdenum disulfide nanosheets with
Graphene not only protected the mesh from corrosion, but also highly electrochemical performance, Nanoscale 9 (2017) 6346–6352, doi:10.
1039/c7nr01665d.
provided the attachment sites for MoS2 . Thanks to the catalytic
[9] H. Sun, Z. Yan, F. Liu, W. Xu, F. Cheng, J. Chen, Self-supported transition-
ability of MoS2 and the electron transport capability of the Cu metal-based electrocatalysts for hydrogen and oxygen evolution, Adv. Mater.
mesh, the self-supporting electrode had a Tafel slope as low as 32 (2019) 1806326, doi:10.1002/adma.201806326.
54 mV decade−1 and reached 400 mA cmgeo −2 at overpotential of [10] A. Ali, D. Akyuz, M.A. Asghar, A. Koca, B. Keskin, Free-standing carbon nan-
otubes as non-metal electrocatalyst for oxygen evolution reaction in water
−300 mV. The self-supporting electrode obtained by the comple- splitting, Int. J. Hydrogen Energy 43 (2018) 1123–1128, doi:10.1016/j.ijhydene.
mentary combination of graphene and copper mesh can be folded 2017.11.060.
or fabricated into any desired shape, while the original shape can [11] L.G. Bach, M.L.N. Thi, N.T. Son, Q.B. Bui, H. T.Nhac-Vu, P.H. Ai-Le, Mesoporous
gold nanoparticles supported cobalt nanorods as a free-standing electrochemi-
be remained when a large amount of hydrogen bubbles were gen- cal sensor for sensitive hydrogen peroxide detection„ J. Electroanal. Chem. 848
erated by the electrolysis. Our research has greatly reduced the op- (2019) 113359, doi:10.1016/j.jelechem.2019.113359.
eration difficulty and preparation cost of self-supporting electrode, [12] X. Wang, X. Gan, T. Hu, K. Fujisawa, Y. Lei, Z. Lin, B. Xu, Z. Huang, F. Kang,
M. Terrones, R. Lv, Noble-metal-free hybrid membranes for highly efficient hy-
which can promote the practical application of non-precious met- drogen evolution, Adv. Mater. 29 (2017) 1603617, doi:10.1002/adma.201603617.
als in hydrogen evolution reaction. [13] R. Dong, M. Pfeffermann, H. Liang, Z. Zheng, X. Zhu, J. Zhang, X. Feng,
Large-area, free-standing, two-dimensional supramolecular polymer single-
layer sheets for highly efficient electrocatalytic hydrogen evolution, Angew.
Declaration of Competing Interest Chem. Int. Edit. 54 (2015) 12058–12063, doi:10.1002/anie.201506048.
[14] J. Wang, H. Zhu, J. Chen, B. Zhang, M. Zhang, L. Wang, M. Du, Small and well-
dispersed Cu nanoparticles on carbon nanofibers: self-supported electrode ma-
The authors declare that they have no known competing finan-
terials for efficient hydrogen evolution reaction, Int. J. Hydrogen Energy 41
cial interests or personal relationships that could have appeared to (2016) 18044–18049, doi:10.1016/j.ijhydene.2016.08.058.
influence the work reported in this paper. [15] J. Zhou, F. Qi, Y. Chen, Z. Wang, B. Zheng, X. Wang, CVD-grown three-
dimensional sulfur-doped graphene as a binder-free electrocatalytic electrode
for highly effective and stable hydrogen evolution reaction, J. Mater. Sci. 53
CRediT authorship contribution statement (2018) 7767–7777, doi:10.1007/s10853- 018- 2118- 6.
[16] K. Kim, A. Tiwari, G. Hyun, T.G. Novak, S. Jeon, Improving electrochemical ac-
tive area of MoS2 via attached on 3D-ordered structures for hydrogen evolu-
Junying Huang: Methodology, Investigation, Formal analysis, tion reaction, Int. J. Hydrogen Energy 44 (2019) 28143–28150, doi:10.1016/j.
Writing - original draft. Mengting Chen: Investigation, Methodol- ijhydene.2019.09.071.
ogy. Tao Tang: Investigation, Methodology. Weipeng Liu: Visual- [17] J. Gao, Z. Cheng, C. Shao, Y. Zhao, Z. Zhang, L. Qu, A 2D free-standing film-
inspired electrocatalyst for highly efficient hydrogen production, J. Mater.
ization. Yingju Liu: Conceptualization, Writing - original draft, Su- Chem. A 5 (2017) 12027–12033, doi:10.1039/c7ta03228e.
pervision. [18] D. Liu, W. Xu, Q. Liu, Q. He, Y.A. Haleem, C. Wang, T. Xiang, C. Zou, W. Chu,
J. Zhong, Z. Niu, L. Song, Unsaturated-sulfur-rich MoS2 nanosheets decorated
on free-standing SWCNT film: synthesis, characterization and electrocatalytic
Acknowledgements application, Nano Res. 9 (2016) 2079–2087, doi:10.1007/s12274- 016- 1098- 6.
[19] X. Zhang, M. Zhang, Y. Tian, J. You, C. Yang, J. Su, Y. Li, Y. Gao, H. Gu, In
situ synthesis of MoS2 /graphene nanosheets as free-standing and flexible elec-
This work was supported by the National Natural Science Foun- trode paper for high-efficiency hydrogen evolution reaction, RSC Adv. 8 (2018)
dation of China (21874048, 21705051), the Scientific Foundation 10698–10705, doi:10.1039/c8ra01226a.
of Guangdong Province (2017A030313077), National Key Research [20] D. Goranova, E. Lefterova, R. Rashkov, Electrocatalytic activity of Ni-Mo-Cu and
Ni-Co-Cu alloys for hydrogen evolution reaction in alkaline medium, Int. J. Hy-
and Development Program of China (SQ2017YFC160089), the Pro- drogen Energy 42 (2017) 28777–28785, doi:10.1016/j.ijhydene.2017.10.002.
gram for the Top Young Innovative Talents of Guangdong Province [21] J. Niu, Z. Tang, H. Huang, J. Pei, H. Zhang, G. Yuan, W. Ding, Research on a Zn-
(2016TQ03N305), and the Foundation for High-level Talents in Cu alloy as a biodegradable material for potential vascular stents application,
Mat. Sci. Eng. C-Mater. 69 (2016) 407–413, doi:10.1016/j.msec.2016.06.082.
South China Agricultural University.
[22] H. Saleem, M. Haneef, H.Y. Abbasi, Synthesis route of reduced graphene oxide
via thermal reduction of chemically exfoliated graphene oxide, Mater. Chem.
Phys. 204 (2018) 1–7, doi:10.1016/j.matchemphys.2017.10.020.
Supplementary materials
[23] Y. Si, Y. Xia, S. Shang, X. Xiong, X. Zeng, J. Zhou, Y. Li, Enhanced visible light
driven photocatalytic behavior of BiFeO3 /reduced graphene oxide composites,
Supplementary material associated with this article can be Nanomaterials 8 (2018) 526, doi:10.3390/nano8070526.
found, in the online version, at doi:10.1016/j.electacta.2020.136754. [24] J. Han, X. Ji, X. Ren, G. Cui, L. Li, F. Xie, H. Wang, B. Li, X. Sun, MoO3 nanosheets
for efficient electrocatalytic N-2 fixation to NH3 , J, Mater. Chem. A 6 (2018)
12974–12977, doi:10.1039/c8ta03974g.
References [25] F. Li, L. Zhang, J. Li, X. Lin, X. Li, Y. Fang, J. Huang, W. Li, M. Tian, J. Jin, R. Li,
Synthesis of Cu-MoS2 /rGO hybrid as non-noble metal electrocatalysts for the
[1] D. Parra, L. Valverde, F.J. Pino, M.K. Patel, A review on the role, cost and value hydrogen evolution reaction, J. Power Sources 292 (2015) 15–22, doi:10.1016/j.
of hydrogen energy systems for deep decarbonisation, Renew. Sust. Energ. Rev. jpowsour.2015.04.173.
101 (2019) 279–294, doi:10.1016/j.rser.2018.11.010. [26] I.G. Morozov, O.V. Belousova, D. Ortega, M.K. Mafina, M.V. Kuznetcov, opti-
[2] M.H. Suliman, A. Adam, M.N. Siddiqui, Z.H. Yamani, M. Qamar, Facile synthesis cal Structural, XPS and magnetic properties of Zn particles capped by ZnO
of ultrathin interconnected carbon nanosheets as a robust support for small nanoparticles, J. Alloy Compd. 633 (2015) 237–245, doi:10.1016/j.jallcom.2015.
and uniformly-dispersed iron phosphide for the hydrogen evolution reaction, 01.285.
Carbon N Y 144 (2019) 764–771, doi:10.1016/j.carbon.2018.12.106. [27] N.V. Nikolenko, I.V. Kozhevnikov, A.O. Kostyniuk, H. Bayahia, Y.V. Kalashnykov,
[3] R. Mulla, C.W. Dunnill, Powering the hydrogen economy from waste heat: Preparation of iron molybdate catalysts for methanol to formaldehyde oxida-
a review of heat-to-hydrogen concepts, ChemSusChem 12 (2019) 3882–3895, tion based on ammonium molybdoferrate(II) precursor, J. Saudi. Chem. Soc. 22
doi:10.1002/cssc.201901426. (2018) 372–379, doi:10.1016/j.jscs.2016.04.002.
 J. Huang, M. Chen and T. Tang et al. / Electrochimica Acta 355 (2020) 136754 9

[28] D. Levin, S.L. Soled, J.Y. Ying, Chimie douce synthesis of a layered ammonium [35] B.V.R. Chowdari, K.L. Tan, W.T. Chia, R. Gopalakrishnan, X-ray photoelectron
zinc molybdate, Chem. Mater. 8 (1996) 836–843, doi:10.1021/cm950427i. spectroscopic studies of molybdenum phosphate glassy system, J. Non-cryst
[29] H. Huang, J. Huang, W. Liu, Y. Fang, Y. Liu, Ultradispersed and single-layered Solids 119 (1990) 95–102, doi:10.1016/0022- 3093(90)90245- H.
MoS2 nanoflakes strongly coupled with graphene: an optimized structure with [36] S. Podila, S.F. Zaman, H. Driss, Y.A. Alhamed, A.A. Al-Zahrani, L.A. Petrov, Hy-
high kinetics for the hydrogen evolution reaction, ACS Appl. Mater. Interfaces drogen production by ammonia decomposition using high surface area Mo2 N
9 (2017) 39380–39390, doi:10.1021/acsami.7b12038. and Co3 Mo3 N catalysts, Catal. Sci. Technol. 6 (2016) 1496–1506, doi:10.1039/
[30] C. Zhu, H. Wang, H. Li, B. Cai, W. Lv, C. Cai, C. Wang, L. Yan, Q. Liu, L. Ma, c5cy00871a.
Selective hydrodeoxygenation of 5-hydroxymethylfurfural to 2,5-dimethylfuran [37] D. Vikraman, A.A. Arbab, S. Hussain, N.K. Shrestha, S.H. Jeong, J. Jung, S.A. Patil,
over alloyed Cu-Ni encapsulated in biochar catalysts, ACS Sustain. Chem. Eng. H.S. Kim, Design of WSe2 /MoS2 heterostructures as the counter electrode to
7 (2019) 19556–19569, doi:10.1021/acssuschemeng.9b04645. replace Pt for dye-sensitized solar cell, ACS Sustain. Chem. Eng. 7 (2019)
[31] F. Parmigiani, G. Pacchioni, F. Illas, P.S. Bagus, Studies of the CuO bond in cupric 13195–13205, doi:10.1021/acssuschemeng.9b02430.
oxide by X-ray photoelectron spectroscopy and ab initio electronic structure [38] S. Chinkanjanarot, J.M. Tomasi, J.A. King, G.M. Odegard, Thermal conductivity of
models, J. Electron. Spectrosc. 59 (1992) 255–269, doi:10.1016/0368-2048(92) graphene nanoplatelet/cycloaliphatic epoxy composites: multiscale modelling,
87005-7. Carbon N Y 140 (2018) 653–663, doi:10.1016/j.carbon.2018.09.024.
[32] J. Huang, M. Chen, X. Zhang, W. Liu, Y. Liu, P-doped 3D graphene network sup- [39] G. Calogero, N.R. Papior, B. Kretz, A. Garcia-Lekue, T. Frederiksen, M. Brandbyge,
porting uniformly vertical MoS2 nanosheets for enhanced hydrogen evolution Electron transport in nanoporous graphene: probing the Talbot effect, Nano
reaction, Int. J Hydrogen. Energ. 45 (2020) 4043–4053, doi:10.1016/j.ijhydene. Lett. 19 (2019) 576–581, doi:10.1021/acs.nanolett.8b04616.
2019.12.014. [40] W. Chen, Z. Liu, J. Hou, Y. Zhou, X. Lou, Y. Li, Enhancing performance of mi-
[33] J. Huang, M. Chen, X. Li, X. Zhang, L. Lin, W. Liu, Y. Liu, A facile layer- crobial fuel cells by using novel double-layer-capacitor-materials modified an-
by-layer fabrication of three dimensional MoS2 –rGO-CNTs with high perfor- odes, Int. J. Hydrogen Energy 43 (2018) 1816–1823, doi:10.1016/j.ijhydene.2017.
mance for hydrogen evolution reaction, Electrochim. Acta 300 (2019) 235–241, 11.034.
doi:10.1016/j.electacta.2019.01.107. [41] H. Huang, W. Huang, Z. Yang, J. Huang, J. Lin, W. Liu, Y. Liu, Strongly coupled
[34] V.I. Nefedov, M.N. Firsov, I.S. Shaplygin, Electronic structures of MRhO2 , MoS2 nanoflake-carbon nanotube nanocomposite as an excellent electrocata-
MRh2 O4 , RhMO4 and Rh2 MO6 on the basis of X-ray spectroscopy and ESCA lyst for hydrogen evolution reaction, J. Mater. Chem. A 5 (2017) 1558–1566,
data, J. Electron. Spectrosc. 26 (1982) 65–78, doi:10.1016/0368-2048(82) doi:10.1039/c6ta09612c.
87006-0.