Research Article

www.afm-journal.de

Single-Atom Co-Decorated MoS2 Nanosheets Assembled

on Metal Nitride Nanorod Arrays as an Efficient Bifunctional
Electrocatalyst for pH-Universal Water Splitting
Thi Luu Luyen Doan, Dinh Chuong Nguyen, Sampath Prabhakaran, Do Hwan Kim,
Duy Thanh Tran, Nam Hoon Kim,* and Joong Hee Lee*

carbon dioxide emissions.[1,2] Given its

The commercialization of electrochemical water splitting technology requires simplicity, cost-effectiveness, high energy
electrocatalysts that are cost-effective, highly efficient, and stable. Herein, an conversion efficiency, and environmental
advanced bifunctional electrocatalyst based on single-atom Co-decorated MoS2 compatibility, electrochemical water split-
nanosheets grown on 3D titanium nitride (TiN) nanorod arrays (CoSAs-MoS2/ ting has attracted considerable attention
and has emerged as the most effective
TiN NRs) has been developed for overall water splitting in pH-universal elec-
method of high-purity H2 gas production
trolytes. When applied as a self-standing cathodic electrode, the CoSAs-MoS2/ in recent years.[3–5] The developed water
TiN NRs requires overpotentials of 187.5, 131.9, and 203.4 mV to reach a HER splitting device with high energy conver-
current density of 10 mA cm–2 in acidic, alkaline, and neutral conditions, respec- sion and good stability is, however, bot-
tively, which are superior to the most previously reported non-noble metal HER tlenecked by the high activation energies
of both the cathodic and anodic electrodes
electrocatalysts at the same current density. The CoSAs-MoS2/TiN NRs anodic
due to the sluggish kinetics of the two
electrode also shows low OER overpotentials of 454.9, 340.6, and 508.0 mV, electrochemical reactions, the hydrogen
respectively, at a current density of 10 mA cm–2 in acidic, alkaline, and neutral evolution reaction (HER) and oxygen evo-
mediums, markedly outperforming current OER catalysts reported elsewhere. lution reaction (OER), respectively, which
More importantly, an electrolyzer delivered from the cathodic and anodic hinders the commercialization of water
CoSAs-MoS2/TiN NRs electrodes exhibits an extraordinary overall water split- splitting devices.[6] Generally, noble metal-
based electrocatalysts, such as Pt/C and
ting performance with good stability and durability in pH-universal conditions.
RuO2/IrO2, show good HER and OER
activities, respectively, but the scarcity,
high cost, and insufficient stability of
1. Introduction these noble metal catalysts greatly limit their large-scale appli-
cations.[7–9] Accordingly, cost-effective, highly active, and stable
Hydrogen gas (H2), a promising renewable energy carrier, is a alternative bifunctional electrocatalysts are quite valuable to the
good substitute for carbon-based fuel sources in future energy effort to improve the overall performance and reduce the cost
systems owing to its high gravimetric energy density and zero of water splitting. Moreover, it has been demonstrated that the
HER reaction rate in an acidic environment is much faster than
Dr. T. L. L. Doan, Dr. D. C. Nguyen, S. Prabhakaran, Prof. D. T. Tran, that in an alkaline environment, while the OER efficiency in
Prof. N. H. Kim, Prof. J. H. Lee alkaline conditions is better than that in acidic conditions;[10,11]
Department of Nano Convergence Engineering thus, the overall water splitting process is generally conducted
Jeonbuk National University
Jeonju, Jeonbuk 54896, Republic of Korea in two different pH conditions. To eliminate the extra energy
E-mail: [email protected]; [email protected] that is required to maintain the pH difference between HER
Prof. D. H. Kim and OER during water splitting, the development of highly effi-
Division of Science Education cient electrocatalysts with noble-metal-like activity suitable for a
Graduate School of Department of Energy wide pH range is necessary, though the challenging pursuit.[6,12]
Storage/Conversion Engineering
Jeonbuk National University
Recently, 2D molybdenum disulfide (MoS2) has earned a
Jeonju, Jeonbuk 54896, Republic of Korea great deal of attention in the field of catalysis. Both computa-
Prof. J. H. Lee tional and experimental studies have shown it to be a typical
Carbon Composite Research Center HER electrocatalyst.[13–15]. However, the catalytic activity of
Department of Polymer-Nano Science and Technology MoS2 is far lower than that of benchmark catalysts due to its
Jeonbuk National University low intrinsic conductivity and poor active site exposure.[16] It is
Jeonju, Jeonbuk 54896, Republic of Korea
thus of paramount importance to optimize and activate MoS2-
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adfm.202100233.
based catalysts for simultaneous HER and OER if MoS2 is to
be exploited for overall water splitting. Among various strate-
DOI: 10.1002/adfm.202100233 gies, hybridization of metal sulfides (MSs) with earth-abundant

Adv. Funct. Mater. 2021, 2100233 2100233 (1 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

transition metals has recently emerged as an effective approach active sites for reactant molecules, and optimize the adsorp-
to improving the catalytic behavior of MSs. According to pre- tion/desorption energies of intermediate species during water
vious studies, the electron density distribution of MSs can electrolysis. Furthermore, the CoSAs-MoS2/TiN NRs catalyst
be adjusted through electronic interactions between metal possesses a large specific surface area and a typical mesoporous
and host MSs that involve electron transfer from the metal to feature, ultrathin MoS2 NSs with abundant edge site exposure,
the MSs and vice versa, thereby optimizing the intermediate and highly conductive TiN NRs as the mainstay and effec-
adsorption and desorption possibilities of the host MSs.[17–19] tive electron transfer pathway. The synergistic effects of these
For example, Feng and coworkers reported Cu nano­ dots- advantageous factors yield an efficient bifunctional electrocata-
decorated Ni3S2 nanotubes supported on carbon fibers as an lyst that evidences effective electrocatalytic behaviors toward
efficient electrocatalyst for the HER; due to electronic inter- both the HER and OER in pH-universal environments.
action between Cu and Ni3S2, the two became positively and
negatively charged, respectively, which encourages H adsorp-
tion and desorption and leads to a significant enhancement of 2. Results and Discussion
HER performance.[18] Xing and coworkers demonstrated a 3D
MoS2@Ni core/shell nanosheets (NSs) structure as a novel CoSAs-MoS2/TiN NRs was developed as highly efficient, cost-
bifunctional electrocatalyst for water splitting; its excellent cata- effective, and pH-universal electrocatalyst for overall water split-
lytic performance was mainly due to synergistic electrocatalytic ting by a versatile and simple strategy, as illustrated in Scheme 1.
effects between the MoS2-based core and Ni-based shell.[20] Figure S1a–c, Supporting Information provide SEM images of
Considering that the integration of transition metals in MSs TiO2 NRs, which show that the CC surface was uniformly coated
has been proven to be an efficient approach to improving the by aligned TiO2 NRs with an average diameter of 150 nm. The
catalytic activity of MSs, we believe that a transition metal-MoS2 energy-dispersive X-ray spectroscopy (EDAX) pattern of TiO2
hybridization strategy can further improve the catalytic perfor- NRs (Figure S1d, Supporting Information) includes the Ti, O,
mance for the HER and OER processes. and C signals. TiN NRs were converted from TiO2 NRs by high-
Cobalt (Co) has been confirmed as promising non-noble temperature annealing treatment under an NH3 atmosphere;
metal candidate for fabrication of bifunctional electrocatalysts SEM images of the TiN NRs are shown in Figure S2a,b, Sup-
for overall water splitting due to its natural merits of high abun- porting Information. It is evident that the surface of the TiN NRs
dance, low cost, and specific chemical/physical properties.[12,21] is rougher and more porous than that of the pristine TiO2 NRs,
These excellent advantages make Co a competitive promoter of which will promote the growth of secondary active materials on
MoS2. In order to meet the requirements of cost effectiveness the surface owing to the high degree of void chamber exposure.
and enhanced catalytic activity, construction of Co single sites Moreover, the roughness and porosity of the TiN NRs will also
has been accomplished, which showed some promising prelim- be good for physical adhesion of TiN and secondary materials,
inary results.[22] Kuznetsov and coworkers reported the utiliza- thus benefiting long-term stability. The EDAX spectrum of the
tion of Co single atoms (CoSAs) as a promoter of 2D Mo2C; this TiN NRs (Figure S2c, Supporting Information) reveals Ti, N, and
Mo2C/Co catalyst exhibited extraordinary activity and good sta- C signals, confirming the complete conversion of TiO2 into the
bility toward the HER.[23] In another study, the Dilpazir group TiN phase. To further demonstrate the structural characteristics
demonstrated CoSAs-promoting N-doped CNTs as bifunctional of TiN NRs, Transmission electron microscopy (TEM) image of
catalysts for the OER and oxygen reduction reaction; the OER an individual TiN nanorod is shown in Figure S2d, Supporting
performance of the obtained CoSAs/NCNTs was even higher Information, which exhibits a 1D nanostructure with a diameter
than that of the state-of-the-art IrO2 catalyst.[24] The advan- of ≈150 nm and noteworthy porosity. An HR-TEM image (inset
tages of the single-atom strategy, which include maximizing in Figure S2d, Supporting Information) shows a lattice fringe of
the exposed reactive metal atoms and enhancing the atom uti- 0.251 nm, which coincides with the (111) crystal plane coming
lization efficiency, are the main reasons for the excellent per- from TiN, corroborating the formation of TiN NRs.[29] MoS2 NSs
formance of the abovementioned catalysts.[25–27] Furthermore, were vertically and densely deposited on the surface of TiN NRs
the unique size quantum effects of single atom catalysts also via a hydrothermal method to form the MoS2/TiN NRs. The
support their excellent activity for different electrochemical SEM images in Figure S3a–c, Supporting Information and the
reactions, including the HER and OER.[28] Despite the great EDAX pattern in Figure S3d, Supporting Information evidence
advantages, hybridization of CoSAs and 2D layered MoS2 to the successful fabrication of MoS2/TiN NRs. Then, CoSAs were
design efficient non-noble metal catalysts for both the HER and atomically distributed on the surface of MoS2/TiN NRs by chem-
OER has not yet been reported. ical bath deposition. After heating and acid etching, the CoSAs-
In this context, single-atom-dispersed Co site-promoted MoS2/TiN NRs were successfully prepared. SEM images of the
ultrathin MoS2 NSs vertically grown on 3D TiN nanorod arrays CoSAs-MoS2/TiN NRs at low magnifications are provided in
(NRs) (denoted CoSAs-MoS2/TiN NRs) as a highly active Figure 1a,b, which clearly show the nanorod array structure of
bifunctional electrocatalyst for overall water splitting in uni- the CoSAs-MoS2/TiN NRs. The average diameter of CoSAs-
versal pH conditions has been developed by a versatile synthetic MoS2/TiN NRs is 300 nm, which is 150 nm larger than that of
approach. The atomic dispersion of Co can accomplish max- the TiN NRs, evidencing the formation of a shell layer with a
imum atomic exposure and remarkably enhance the density of thickness of ≈75 nm. Magnified image in Figure 1c shows that
active Co-containing sites. Importantly, the strong interactions the surface of TiN NRs is fully covered by ultrathin CoSAs-MoS2
between CoSAs and layered MoS2 modulate the electron density NSs without defects or fractures, forming a unique hierarchical
distributions of both Co and MoS2, providing more catalytically core/shell architecture. Noticeably, the morphology of

Adv. Funct. Mater. 2021, 2100233 2100233 (2 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Scheme 1. Schematic illustration of the fabrication of the CoSAs-MoS2/TiN hybrid.

CoSAs-MoS2/TiN NRs is unchanged as compared to that of the Figure 1c) confirms the existence of Co, Mo, S, Ti, and N ele-
MoS2/TiN NRs, indicating that dispersion of CoSAs had no ments in the CoSAs-MoS2/TiN NRs. For comparison, pure
influence on the material structure. The EDAX pattern (inset in CoSAs-MoS2 NSs were also synthesized by the same procedure

Figure 1. a–c) FE-SEM images of the CoSAs-MoS2/TiN NRs (Inset in (c): EDAX pattern of the CoSAs-MoS2/TiN NRs). d,e) TEM images of the CoSAs-
MoS2/TiN NRs. f) HR-TEM image of the outer CoSAs-MoS2 nanosheet (Inset in (f): SAED pattern of the outer CoSAs-MoS2 nanosheet). g,h) Atomic-
resolution TEM images of the outer CoSAs-MoS2 nanosheet (CoSAs are highlighted by yellow circles). i) STEM image and j–n) EDS mapping of the
CoSAs-MoS2/TiN NRs.

Adv. Funct. Mater. 2021, 2100233 2100233 (3 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

(see Supporting Information) and their morphology was investi- species in hexagonal 2H-MoS2. Moreover, Raman spectra
gated by SEM. Figure S4, Supporting Information shows that (Figure S5c,f,i, Supporting Information) reveal two character-
the CoSAs-MoS2 NSs were directly deposited on the CC surface istic peaks of 2H-MoS2, including E 2g 1
at ≈380 eV and A1g at
in the form of stacked particles. Such agglomeration can cause ≈410 eV, which match well with the literatures of 2H-MoS2 and
poor exposure of active edge sites, as well as inefficient utiliza- atomic-resolution TEM observation, further demonstrating 2H
tion of reactive active sites, leading to induced inferior catalytic crystal phase of MoS2 in our developed materials.[35,39] Scanning
activity in comparison with that of the hierarchical core/shell TEM (STEM) and energy dispersive spectroscopy (EDS) map-
CoSAs-MoS2/TiN NRs. TEM images of the CoSAs-MoS2/TiN ping images taken of the shell layer (Figure S6, Supporting
NRs at different magnifications are provided in Figure 1d,e to Information) exhibit no local aggregation of Co atoms on the
further demonstrate their typical core/shell structure con- entire MoS2 NSs surface, which provide evidence for the forma-
structed by self-assembly of ultrathin CoSAs-MoS2 NSs on the tion of single-atom Co sites on the surface of MoS2. Further-
surface of an inner TiN nanorod. In this structure, the diameter more, to verify the homogeneous distribution of elements in
of the core part is ≈150 nm and the thickness of the shell layer the catalyst structure, STEM and EDS mapping analyses were
is ≈75 nm, which are consistent with SEM observations. The conducted for individual CoSAs-MoS2/TiN nanorods. The
spatially open structure of the CoSAs-MoS2 NSs will be particu- center of the nanorod is rich in Ti and N elements (Figure 1j,k),
larly beneficial to electrode/electrolyte interactions, as well as whereas the outer area of the selected region contains Co, Mo,
diffusion and penetration of electrolyte/ions. TiN NRs not only and S elements (Figure 1l–n), demonstrating the successful for-
act as a mainstay for firm attachment of CoSAs-MoS2 NSs, but mation of a unique core/shell structure with a TiN nanorod as
also serve as an efficient charge transport channel due to their the core and CoSAs-decorated MoS2 NSs as the shell. Sporadic
high electrical conductivity, as demonstrated by previous dispersion of Co element over the shell layer (Figure 1n) further
reports.[30,31] Based on the abovementioned advantages, the as- substantiates the formation of CoSAs.
synthesized CoSAs-MoS2/TiN NRs are expected to effectively The crystalline phases of the samples were identified via
catalyze the HER and OER and boost the performance of the X-ray photoelectron spectroscopy (XRD) analysis. Figure 2a
overall water splitting process. An edge view of the shell layer compares the XRD patterns of the TiN NRs, MoS2/TiN NRs,
(Figure 1f) reveals that it contains several NSs with an interlayer and CoSAs- MoS2/TiN NRs. The XRD pattern of the TiN NRs
distance of 0.65 nm, which is assigned to the (002) plane of shows diffraction peaks located at 36.8°, 42.8°, 62.1°, 74.5°, and
MoS2.[32] This finding proves the presence of the MoS2 phase in 78.4° corresponding to the (111), (200), (220), (311), and (222)
the shell part. A selected area electron diffraction (SAED) pat- crystal planes of the osbornite TiN phase (PDF#87-0630).[40,41]
tern was also recorded at the shell region; the SAED spectrum In the XRD pattern of MoS2/TiN NRs, in addition to the char-
(inset in Figure 1f) identifies polycrystalline diffraction rings cor- acteristic peaks of TiN, other diffraction peaks appear at 14.4°,
responding to the (110) and (100) facets of MoS2, reconfirming 32.7°, 39.5°, 49.7°, 58.3°, and 70.1°, which match well with the
the presence of the MoS2 phase.[33] Notably, the SEM, TEM, and (002), (100), (103), (105), (110), and (108) crystal planes of the
even high-resolution TEM images demonstrate the absence of hexagonal MoS2 phase (PDF#37-1492),[42] indicating successful
particles on the surface of MoS2 NSs, suggesting a single-atomic formation of MoS2 NSs on the TiN NRs surface. The XRD pat-
dispersion of the Co component. In the atomic-resolution TEM tern of CoSAs-MoS2/TiN NRs shows only the diffraction peaks
images of the shell layer of the CoSAs-MoS2/TiN NRs of TiN and MoS2; no diffraction peaks related to a Co compo-
(Figure 1g,h), a large number of dark spots (yellow circles) nent can be observed, corroborating the existence of CoSAs in
corresponding to CoSAs with ultra-small sizes of ≈0.15 nm is the structure.
observed, clearly demonstrating the uniform dispersion of The chemical compositions and valence states of the Co, Mo,
CoSAs over the MoS2 NSs. The atomic-resolution TEM images S, Ti, and N elements in the CoSAs-MoS2/TiN NRs, MoS2/TiN
also show an additional 0.274 nm lattice fringe that is coinci- NRs, and TiN NRs were studied in detail by X-ray photoelectron
dent with the (100) plane of MoS2, further indicating the pres- spectroscopy (XPS) characterization. The XPS survey spectrum
ence of the MoS2 phase in the hierarchical structure.[34] of the CoSAs-MoS2/TiN NRs (Figure 2b) reveals the coexist-
Moreover, the atomic-resolution TEM image of CoSAs-MoS2/ ence of Co2p (779.6 eV), Ti2p (459.3 eV), N1s (400.6 eV), Mo3d
TiN NRs (Figure S5a, Supporting Information) indicates that (229.5 eV), and S2p (162.3 eV) orbitals, belonging to CoSAs,
atomic arrangement along the basal plane of MoS2 nanosheets MoS2, and TiN. The appearance of the O1s orbital may be due
is similar in appearance to 2H-MoS2.[35,36] The same results are to oxidized surface upon exposure to air. Figure 2c compares
found for MoS2/TiN NRs and CoSAs-MoS2 NSs (Figure S5d,g, the high-resolution Ti2p spectra of the CoSAs-MoS2/TiN NRs,
Supporting Information, respectively), suggesting that their MoS2/TiN NRs, and TiN NRs. The Ti2p spectrum of the pris-
crystal phases are also hexagonal 2H-MoS2. Based on these tine TiN NRs shows two primary peaks of Ti2p3/2 and Ti2p1/2
TEM images, distance between two adjacent Mo atoms is found located at binding energies of ≈458.7 and ≈464.4 eV, respec-
to be ≈3.3 Å for CoSAs-MoS2/TiN NRs, MoS2/TiN NRs, and tively. The broad Ti2p3/2 peak can be split into three minor
CoSAs-MoS2 NSs, consistent with previous reports related to peaks of TiN (455.8 eV) corresponding to the TiN state,[41]
2H-MoS2 materials.[37,38] Regarding corresponding intensity TiNO (457.2 eV) and TiO (458.6) that is attributable to
profiles obtained along the lines revealed in the atomic-resolution surface oxidation.[32] After deposition of MoS2 NSs on the TiN
TEM images (Figure S5b,e,h, Supporting Information), all NRs surface, the Ti2p3/2 peak of the MoS2/TiN NRs shows a
intensity profiles of CoSAs-MoS2/TiN NRs, MoS2/TiN NRs, and significant decrease in the intensity of the core-level TiO peak;
CoSAs-MoS2 NSs materials show the peaks originated from Mo this is due to protection from surface oxidation derived from
and S, which correspond with the coordination of Mo and S the MoS2 shell layer. Though the number of TiO bonds is

Adv. Funct. Mater. 2021, 2100233 2100233 (4 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 2. a) XRD patterns and b) XPS survey spectra of the as-prepared samples ((i): TiN NRs, (ii): MoS2/TiN NRs, and (iii): CoSAs-MoS2/TiN NRs).
c–g) High-resolution XPS spectra of Ti2p, N1s, Mo3d, S2p, and Co2p, respectively, for the CoSAs-MoS2/TiN NRs, MoS2/TiN NRs, and TiN NRs. h) N2
adsorption-desorption curves and i) pore size distributions of the CoSAs-MoS2/TiN NRs, MoS2/TiN NRs, and TiN NRs.

limited, the presence of this bond is beneficial for the chemical Mo3d3/2 peaks shift slightly to lower binding energies of 226.2,
connectivity of the TiN core and MoS2 shell, as it boosts charge 228.8, and 232.4 eV, respectively, as seen in the Mo3d spectrum
transfer ability and promotes chemical coupling between TiN of the CoSAs-MoS2/TiN NRs, indicating an increase in electron
and MoS2.[32,43] The high-resolution Ti2p spectrum of CoSAs- densities around the Mo and S species in MoS2.[45] This change
MoS2/TiN NRs is similar in appearance to that of pure MoS2/ can be ascribed to the strong electronic interactions between
TiN NRs, indicating no influence of dispersed CoSAs on the CoSAs and MoS2 in the CoSAs-MoS2/TiN NRs.[47,48] Likewise,
electronic properties of the TiN core. Figure 2d shows a com- a comparison between the S2p XPS spectra of MoS2/TiN NRs
parison of N1s XPS peaks between the CoSAs-MoS2/TiN NRs, and CoSAs-MoS2/TiN NRs (Figure 2f) reveals that the S2p3/2
MoS2/TiN NRs, and TiN NRs. The N1s region of the pristine and S2p1/2 peaks are negatively shifted after the introduction of
TiN NRs is deconvoluted into two primary peaks of TiN CoSAs, reconfirming the presence of good electronic contact
(396.1 eV) and TiNO (397.3 eV), consistent with that of Ti2p. between the CoSAs and the host MoS2. The existence of Co in
Noticeably, a new peak at 394.6 eV corresponding to Mo3p is this structure is further demonstrated by the Co2p spectrum
observed in the high-resolution XPS spectra of N1s for MoS2/ (Figure 2g), which is deconvoluted into two peaks of Co2p1/2
TiN NRs and CoSAs-MoS2/TiN NRs, which is due to the peak and Co2p3/2 at 794.5 and 779.8 eV, respectively.[12,49] It is worth
overlap of Mo3p and N1s.[44,45] The high-resolution Mo3d XPS noting that the binding energy of the Co2p3/2 peak is more pos-
spectrum of MoS2/TiN NRs (Figure 2e) can be fitted with four itive than that of the metallic Co peak (778.1–778.8 eV) reported
core-level peaks at 226.2, 228.7, 232.4, and 236.2 eV, which elsewhere,[50] proving a lower electron density around Co spe-
match well with MoS bonding, Mo3d5/2, Mo3d3/2, and Mo6+ cies. This result is also evidence in support of electronic cou-
originating from oxidized surface, evidencing the presence of pling between CoSAs and MoS2. Due to the higher electronega-
the MoS2 phase in the obtained structure.[46,47] After dispersion tivity of Mo and S as compared to Co, electrons are transferred
of CoSAs over the MoS2 surface, the MoS, Mo3d5/2, and from CoSAs to MoS2, leading to a decrease in positive charge

Adv. Funct. Mater. 2021, 2100233 2100233 (5 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

on Mo and an increase in negative charge on the S of MoS2 The electrocatalytic activities of the CoSAs-MoS2/TiN NRs
in the neighborhood of CoSAs, as well as an enhanced positive for the HER and OER were investigated using a typical three-
charge on the Co.[18,45] Such variation in the charges of Mo, S, electrode system in pH-universal conditions, including 0.5 m
and Co not only benefits the adsorption of intermediate species H2SO4, 1.0 m KOH, and 1.0 m PBS. First, the HER perfor-
on active sites of the catalyst, but also may promote the desorp- mance of the CoSAs-MoS2/TiN NRs was assessed in acidic
tion of adsorbed intermediates from the catalyst surface to form electrolyte. Reference samples, such as the MoS2/TiN NRs, TiN
gaseous species, thus improving the catalytic activities of the NRs, CoSAs-MoS2 NSs, and commercial Pt/C, were also exam-
CoSAs-MoS2/TiN NRs for the HER and OER.[18,45,51] Therefore, ined for comparison purposes. The iR-corrected polarization
the XRD and XPS results clearly demonstrate successful for- curves of the different samples upon HER tests in 0.5 m H2SO4
mation of TiN, MoS2, and CoSAs in the CoSAs-MoS2/TiN NRs (Figure 3a) show that the CoSAs-MoS2/TiN NRs have better
structure, together with strong electronic interactions between activity in terms of overpotential (Eo) and catalytic activity than
CoSAs and MoS2 that properly adjust electronic structure. their counterparts. Figure S8a, Supporting Information shows a
The formation of core/shell nanostructures is also an comparison of Eo at a current density (J) of 10 and 50 mA cm–2
important reason to highly enhance the catalytic activity of the between the CoSAs-MoS2/TiN NRs and the reference sam-
CoSA-MoS2/TiN NRs as compared to the CoSA-MoS2 NSs. We ples. The commercial Pt/C catalyst shows good HER catalytic
conducted XPS analysis for CoSA-MoS2/TiN NRs and CoSA- activity, with Eo of 51.2 and 134.3 mV at J of 10 and 50 mA cm–2,
MoS2 NSs and the results are provided in Figure S7a,b, Sup- respectively. The CoSAs-MoS2/TiN NRs only require Eo of 187.5
porting Information. As seen, after formation of core/shell and 229.3 mV to generate a J of 10 and 50 mA cm–2, respec-
structure, high resolution XPS spectra of Mo3d and S2p for tively. Both pristine TiN NRs (485.0 and 614.7 mV) and pure
CoSA-MoS2/TiN NRs have positive binding energy shift in CoSAs-MoS2 NSs (302.5 and 402.9 mV) show inferior perfor-
comparison with the CoSA-MoS2 NSs, demonstrating the for- mance to the CoSAs-MoS2/TiN NRs, suggesting that the core/
mation of electronic synergistic effect between the inner TiN shell architecture is one of the main contributory factors in the
and the outer CoSA-MoS2.[52,53] Such good electronic inter- high catalytic performance of the CoSAs-MoS2/TiN NRs. In
action results in not only regulated electronic structure for addition, the far lower HER activity of MoS2/TiN NRs (277.9
each part toward improving catalytic behavior, but also gener- and 357.4 mV) than CoSAs-MoS2/TiN NRs also demonstrates
ated an effective pathway for electron transportation between the critical role of CoSAs in the HER enhancement. To the best
core and shell part. The XPS analyses also show the relative of our knowledge, the Eo at J of 10 mA cm–2 in an acidic condi-
similarity of atomic percentages of Mo, S, and Co elements tion of the CoSAs-MoS2/TiN NRs is lower than those of other
in these materials (Figure S7c, Supporting Information), sug- HER electrocatalysts reported recently (Figure S8b, Supporting
gesting that the better activity of CoSA-MoS2/TiN than that of Information), proving the superior activity and competitiveness
CoSA-MoS2 is main due to the formation of its unique core/ of the proposed catalyst in practicability. Tafel plots (Figure 3b)
shell structure. show that the CoSAs-MoS2/TiN NRs has a smaller Tafel slope
Specific surface area and porosity are well-known as crucial (53.5 mV dec–1) than the MoS2/TiN NRs (95.8 mV dec–1), TiN
factors that significantly impact catalytic performance of cata- NRs (180.6 mV dec–1), and CoSAs-MoS2 NSs (96.3 mV dec–1),
lysts.[54] To evaluate the specific surface area and porosity of the implying that the CoSAs-MoS2/TiN NRs has more favorable
obtained materials, N2 adsorption–desorption isotherms were and faster reaction kinetics upon catalyzing the HER. The small
conducted; the results are presented in Figure 2h. As seen in Tafel slope of the CoSAs-MoS2/TiN NRs also suggests that the
the figure, the CoSAs-MoS2/TiN NRs has the biggest Brunauer− HER process using CoSAs-MoS2/TiN NRs follows the Volmer–
Emmett−Teller (BET) surface area of the three samples, Heyrovsky mechanism, in which the rate-determining step of
53.2 m2 g–1, compared to 51.9 m2 g–1 for the MoS2/TiN NRs, and the full reaction is the Heyrovsky step:[56,57]
41.7 m2 g–1 for the TiN NRs. Notably, the CoSAs-MoS2/TiN NRs
and MoS2/TiN NRs have comparable surface areas, which is H+ + e − + Cat. → Cat. − Hads ( Volmer step ) (1)
due to the similarities of their morphologies, as demonstrated
by SEM analysis. In contrast, the far higher specific surface H+ + e − + Cat. − Hads → H2 + Cat. (Heyrovsky step )
area of CoSAs-MoS2/TiN NRs than TiN NRs demonstrates that (2)
the core/shell architecture majorly contributes to surface area To evaluate electrocatalytic stability under acidic conditions,
improvement. The pore size distributions of these samples are chronoamperometric measurement was performed for the
provided in Figure 2i, which clearly shows that the average pore CoSAs-MoS2/TiN NRs in 0.5 m H2SO4. The chronoampero-
diameters of the CoSAs-MoS2/TiN NRs and MoS2/TiN NRs are metric curve in Figure 3c shows no significant change in the
about 1.9 and 2.0 nm, respectively, due to the connectedness of current density driven by an operating potential of −0.229 V
NSs in the shell; this result suggest their meso-porosity. The (versus RHE) for 45 h. The current retention of the CoSAs-
pore diameter of TiN NRs is around 0.5–1 nm, consistent with MoS2/TiN NRs is 94.8% after a 45 h test, demonstrating the high
the SEM and TEM analyses, proving its highly mesoporous stability and durability of the CoSAs-MoS2/TiN NRs for acidic
structure. These results imply that the CoSAs-MoS2/TiN NRs HER. A comparison of polarization curves recorded before
possesses large specific surface area with mesoporous charac- and after the chronoamperometric test (inset in Figure 3c)
teristics, expected to provide more active sites and void spaces shows that the polarization curve after 45 h is not significantly
to promote mass transfer, as well as boost reagent diffusion and changed relative to the initial curve, suggesting good mainte-
penetration during catalytic processes, all of which contribute to nance of catalytic activity of the CoSAs-MoS2/TiN NRs under
improving OER and HER performance.[55] prolonged operating conditions. Thus, the high electrocatalytic

Adv. Funct. Mater. 2021, 2100233 2100233 (6 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 3. HER activity of the as-prepared samples in 0.5 m H2SO4: a) iR-corrected polarization curves and b) the corresponding Tafel plots of different
samples; c) chronoamperometric curve of the CoSAs-MoS2/TiN NRs at an applied potential of −0.229 V (versus RHE) (Inset in (c): Polarization curves
recorded before and after the chronoamperometric test). HER activity of the as-prepared samples in 1.0 m KOH: d) iR-corrected polarization curves and
e) the corresponding Tafel plots of different samples; f) Chronoamperometric curve of the CoSAs-MoS2/TiN NRs at an applied potential of −0.233 V
(versus RHE) (Inset in (f): Polarization curves recorded before and after the chronoamperometric test). HER activity of the as-prepared samples in 1.0 m
PBS: g) iR-corrected polarization curves and h) the corresponding Tafel plots of different samples; i) Chronoamperometric curve of the CoSAs-MoS2/
TiN NRs at an applied potential of −0.384 V (versus RHE) (Inset in (i): Polarization curves recorded before and after the chronoamperometric test).
((i): TiN NRs, (ii): CoSAs-MoS2 NSs, (iii) MoS2/TiN NRs, (iv) CoSAs-MoS2/TiN NRs, and (v) Pt/C).

performance and good stability for the HER under acidic condi- CoSAs-MoS2 NSs (242.5 and 383.0 mV); this suggests that the
tions make the catalyst as a promising candidate to solve the CoSAs-MoS2/TiN NRs has the best HER performance among
challenges of the acidic HER in terms of the activity, stability, all of the samples tested (Figure S9a, Supporting Information).
and cost demands. The HER performance of the CoSAs-MoS2/ Furthermore, the HER performance of CoSAs-MoS2/TiN NRs
TiN NRs was also studied in alkaline conditions, along with its is well ahead of the recently reported performance of alkaline
counterparts and the commercial Pt/C catalyst for comparison HER electrocatalysts (Figure S9b, Supporting Information),
purposes. The iR-corrected polarization curves (Figure 3d) further demonstrating the high catalytic activity of the CoSAs-
reveal the superior activity of the CoSAs-MoS2/TiN NRs for the MoS2/TiN NRs for alkaline HER. To further understand the
HER over the MoS2/TiN NRs, TiN NRs, and CoSAs-MoS2 NSs, HER kinetics, we evaluated the Tafel slope of the as-prepared
possibly originating from the synergistic effects of its unique samples. A Tafel slope of 56.9 mV dec–1 is found for the CoSAs-
advantageous factors, which include single atom effects, strong MoS2/TiN NRs (Figure 3e), which is smaller than that of the
electronic interactions between CoSAs and MoS2, the merits MoS2/TiN NRs (65.5 mV dec–1), TiN NRs (181.2 mV dec–1), and
of the core/shell structure, large specific surface area, and CoSAs-MoS2 NSs (81.5 mV dec–1), indicating that the kinetics
meso-porosity. The Eo at J of 10 and 50 mA cm–2 are 131.9 and of the HER reaction are more favorable when using the CoSAs-
232.8 mV, respectively, for the CoSAs-MoS2/TiN NRs, which are MoS2/TiN NRs. The above results reconfirm the important
much lower than the corresponding values for the MoS2/TiN role of the CoSAs and hierarchical architecture in improving
NRs (210.8 and 361.7 mV), TiN NRs (398.7 and 652.2 mV), and the catalytic activity and kinetics of the CoSAs-MoS2/TiN NRs

Adv. Funct. Mater. 2021, 2100233 2100233 (7 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 4. OER activity of the as-prepared samples in 0.5 m H2SO4: a) iR-corrected polarization curves and b) the corresponding Tafel plots of different
samples; c) Chronoamperometric curve of the CoSAs-MoS2/TiN NRs at an applied potential of 1.85 V (versus RHE) (Inset in (c): Polarization curves
recorded before and after the chronoamperometric test). OER activity of the as-prepared samples in 1.0 m KOH: d) iR-corrected polarization curves
and e) the corresponding Tafel plots of different samples; f) Chronoamperometric curve of the CoSAs-MoS2/TiN NRs at an applied potential of 1.65 V
(versus RHE) (Inset in (f): Polarization curves recorded before and after the chronoamperometric test). OER activity of the as-prepared samples in 1.0
m PBS: g) iR-corrected polarization curves and h) the corresponding Tafel plots of different samples; i) Chronoamperometric curve of the CoSAs-MoS2/
TiN NRs at an applied potential of 2.11 V (versus RHE) (Inset in (i): Polarization curves recorded before and after the chronoamperometric test). ((i):
TiN NRs, (ii): CoSAs-MoS2 NSs, (iii): MoS2/TiN NRs, (iv): CoSAs-MoS2/TiN NRs, and (v): RuO2).

toward the HER in 1.0 m KOH. The stability of the CoSAs- stable. In addition, the coexistence of Co2p, Mo3d, S2p, Ti2p,
MoS2/TiN NRs for the HER in 1.0 m KOH was evaluated by and N1s orbitals in the XPS survey spectrum of the post-HER
chronoamperometric testing at −0.233 V (versus RHE) for CoSAs-MoS2/TiN NRs (Figure S10d, Supporting Information)
45 h. The current density derived from that operating poten- suggests that the chemical composition of the CoSAs-MoS2/
tial remains stable throughout the chronoamperometric test TiN NRs is fundamentally intact after long-term operation. The
(Figure 3f). After the 45 h stability test, the current retention is high-resolution XPS spectra of Co2p, Mo3d, S2p, Ti2p, and N1s
90.0% for the CoSAs-MoS2/TiN NRs. On the other hand, there for the post-HER CoSAs-MoS2/TiN NRs (Figure S10e–i, Sup-
is no significant change in the initial polarization curve and porting Information) are also unchanged from the pre-HER
the curve after 45 h (inset in Figure 3f). These results are evi- state. The CoSAs-MoS2/TiN NRs catalyst thus possesses high
dence of the good stability of the CoSAs-MoS2/TiN NRs under catalytic activity and robust stability for the HER in alkaline
alkaline HER conditions. The structural features and chemical conditions, indicating that it is a promising substitute for noble
composition of the CoSAs-MoS2/TiN NRs after 45 h HER oper- metal catalysts in high-pH water reduction. In neutral condi-
ation were investigated by SEM, TEM, and XPS. The SEM and tions (1.0 m PBS), the CoSAs-MoS2/TiN NRs also needs a lower
TEM images (Figure S10a–c, Supporting Information) show Eo of 203.4 and 383.9 mV to reach J of 10 and 50 mA cm–2
no obvious changes in morphology of the post-HER CoSAs- (Figure 3g and Figure S11a, Supporting Information), with a
MoS2/TiN NRs compared with the pristine morphology, indi- smaller Tafel slope of 82.7 mV dec–1 (Figure 3h) compared to
cating that the CoSAs-MoS2/TiN NRs are highly structurally its counterparts, such as MoS2/TiN NRs (323.5 and 527.3 mV,

Adv. Funct. Mater. 2021, 2100233 2100233 (8 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

137.6 mV dec–1), TiN NRs (417.3 and 826.8 mV, 196.3 mV dec–1), CoSAs-MoS2 NSs to gain the same J, proving the exceptional
and CoSAs-MoS2 NSs (372.6 and 656.9 mV, 139.0 mV dec–1), catalytic performance toward OER of the CoSAs-MoS2/TiN NRs
implying that the new catalyst has higher activity and faster in alkaline electrolyte. The smaller Tafel slope of 81.2 mV dec–1
HER kinetics under neutral conditions. The Eo of CoSAs-MoS2/ for the CoSAs-MoS2/TiN NRs also demonstrates the enhance-
TiN NRs at J of 10 mA cm–2 is comparable to or even lower ment of OER reaction kinetics, and therefore better OER
than of other recently reported neutral HER electrocatalysts catalytic performance, in comparison with MoS2/TiN NRs
(Figure S11b, Supporting Information), corroborating the desir- (95.1 mV dec–1), TiN NRs (138.5 mV dec–1), CoSAs-MoS2 NSs
able electrocatalytic activity of the CoSAs-MoS2/TiN NRs in a (114.4 mV dec–1), and RuO2 (73.5 mV dec–1), as presented in
neutral environment. In addition to good activity for HER, the Figure 4e. Note that the OER performance of the CoSAs-MoS2/
CoSAs-MoS2/TiN NRs can operate continuously over 45 h in 1.0 m TiN NRs at a J of 10 mA cm–2 is also superior to that of other
PBS with only a slight decrease of its performance. A current reported non-noble metal-based OER catalysts (Figure S13b,
retention of 92.2% after long-term catalysis revealed good sta- Supporting Information). Figure 4f displays the chronoampero-
bility of the CoSAs-MoS2/TiN NRs for the HER in a neutral-pH metric curve of CoSAs-MoS2/TiN NRs toward the OER with an
electrolyte. operating voltage of 1.65 V (versus RHE). The corresponding J
The catalytic activities of the CoSAs-MoS2/TiN NRs for the of 50 mA cm–2 seems to remain constant over the 45 h stability
OER were also studied in a wide pH range. In 0.5 m H2SO4 test. The current retention is 95.2% after chronoamperometry.
electrolyte, the iR-corrected polarization curves (Figure 4a) In addition, the comparison of polarization curves before and
show that the OER performance of CoSAs-MoS2/TiN NRs is after 45 h catalysis shows a negligible performance degradation.
superior to that of all of its counterparts. The CoSAs-MoS2/ These results indicate the excellent stability of CoSAs-MoS2/TiN
TiN NRs can reach J of 10 and 50 mA cm–2 at Eo of 454.9 and NRs under alkaline OER conditions. SEM and TEM images of the
623.8 mV, which are much lower than those of the MoS2/TiN CoSAs-MoS2/TiN NRs taken after the stability test (Figure S14a–c,
NRs (578.1 and 904.6 mV), TiN NRs (838.0 and 1216.0 mV), Supporting Information) show that the morphology of
and CoSAs-MoS2 NSs (620.3 and 955.7 mV) (Figure S12a, CoSAs-MoS2/TiN NRs is almost unaffected by the continuous
Supporting Information). Furthermore, the CoSAs-MoS2/TiN catalysis process, demonstrating good structural stability. The
NRs present better catalytic activity in comparison with other XPS survey spectrum of the post-OER CoSAs-MoS2/TiN NRs
recently reported acidic OER catalysts (Figure S12b, Supporting (Figure S14d, Supporting Information) also reveals the coex-
Information), further demonstrating its extraordinary activity istence of Co, Mo, S, Ti, and N elements. There is no obvious
toward the OER in acidic electrolytes. In addition, an investi- change in Co2p spectrum compared to the original spectrum
gation of the OER reaction kinetics for the developed samples (Figure S14e, Supporting Information), indicating that single-
with the RuO2 catalyst was conducted based on the Tafel plot atom Co sites are stable under alkaline OER conditions. The
method. As expected, the Tafel slope of the CoSAs-MoS2/TiN enhanced intensity of the Mo6+ peak (Figure S14f, Supporting
NRs (165.5 mV dec–1) is lower than those of all other samples: Information) suggests the partial conversion of MoS2 into
200.2 mV dec–1 for MoS2/TiN NRs, 438.5 mV dec–1 for TiN MoO3 owing to in situ oxidation of MoS2 during the OER pro-
NRs, 264.3 mV dec–1 for CoSAs-MoS2 NSs (Figure 4b). These cess, which is consistent with a previous report.[58] The OER
findings suggest that the OER kinetics using the CoSAs-MoS2/ performance of CoSAs-MoS2/TiN NRs is retained for at least
TiN NRs are faster than when other catalysts are used. Further- 45 h of continuous catalysis despite partial phase conversions,
more, the stability of CoSAs-MoS2/TiN NRs for the OER in an reconfirming its extraordinary stability for the alkaline OER. In
acidic electrolyte was also probed through chronoamperometric the neutral condition, in comparison with the set of reference
measurement for 45 h. At an operating voltage of 1.85 V (versus samples, the CoSAs-MoS2/TiN NRs presents enhanced OER
RHE), a stable J of 50 mA cm–2 was observed for the entirety activity and kinetics (Figure 4g,h, and Figure S15a, Supporting
of the stability test, without remarkable changes (Figure 4c). Information), corresponding to an Eo of 508.0 and 882.3 mV at
The current retention of 96.6% after continuous catalysis J of 10 and 50 mA cm–2 and a Tafel slope of 183.1 mV dec–1.
under acidic OER conditions provides favorable evidence of In comparison to recently reported OER electrocatalysts in neu-
good stability for the CoSAs-MoS2/TiN NRs. Additionally, the tral electrolyte, the CoSAs-MoS2/TiN NRs also exhibits better
polarization curve recorded after the chronoamperometric electrocatalytic activity in terms of an Eo at J of 10 mA cm–2
measurement is similar to the initial curve, providing additional (Figure S15b, Supporting Information). After the 45 h chrono-
evidence of the robust stability and durability of the CoSAs- amperometric test, the current retention is 96.1% (Figure 4i),
MoS2/TiN NRs (inset in Figure 4c). We then assessed the OER demonstrating almost no degradation in catalytic activity. Also,
activity of the CoSAs-MoS2/TiN NRs in an alkaline environ- the polarization curve after long-term running shows almost
ment (1.0 m KOH). The iR-corrected polarization curves of the no change from the initial curve (inset in Figure 4i), providing
CoSAs-MoS2/TiN NRs and references, including MoS2/TiN additional evidence of the superior stability of the CoSAs-MoS2/
NRs, TiN NRs, CoSAs-MoS2 NSs, and RuO2, are summarized TiN NRs for the OER in a neutral electrolyte.
in Figure 4d. Compared with the selected reference samples, Therefore, the developed CoSAs-MoS2/TiN NRs catalyst has
the CoSAs-MoS2/TiN NRs shows great enhancement of OER excellent activities in the HER and OER reactions under a wide
activity. In detail, the CoSAs-MoS2/TiN NRs can afford a J of 10 range of pH condition. To assess the possibility of utilizing the
and 50 mA cm–2 with an Eo of 340.6 and 423.7 mV (Figure S13a, CoSAs-MoS2/TiN NRs in practical applications, CoSAs-MoS2/
Supporting Information), compared with Eo of 345.7 and TiN NRs was employed as the working electrodes at both the
499.0 mV for RuO2, 472.7 and 640.0 mV for MoS2/TiN NRs, cathode and anode in a pH-universal overall water splitting
610.7 and 813.5 mV for TiN NRs, and 525.1 and 698.1 mV for process and the performance of the obtained electrolyzer was

Adv. Funct. Mater. 2021, 2100233 2100233 (9 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 5. Overall water-splitting measurements with a symmetrical electrode system using CoSAs-MoS2/TiN NRs as both cathode and anode in
a) 0.5 m H2SO4, b) 1.0 m KOH, and c) in 1.0 m PBS, in comparison with the Pt/C (−) // RuO2 (+) couple. A comparison of operating voltages at J
values of 10 and 50 mA cm–2 between CoSAs-MoS2/TiN NRs (−,+) and Pt/C (−) // RuO2 (+) in d) 0.5 m H2SO4, e) 1.0 m KOH, and f) 1.0 m PBS.
Chronoamperometric curves of the CoSAs-MoS2/TiN NRs (−,+) in g) 0.5 m H2SO4, h) 1.0 m KOH, and i) 1.0 m PBS (Inset in (g–i): Polarization curves
recorded before and after the chronoamperometric test).

evaluated by linear sweep voltammetry (LSV) measurement. densities after a 30 h stability test in acidic, alkaline, and neu-
Commercial Pt/C and RuO2 catalysts were also applied as the tral electrolytes, respectively. Furthermore, there are essentially
cathode and anode, respectively, for comparison. The CoSAs- no obvious changes in the polarization curve after 30 h (inset
MoS2/TiN NRs electrodes yields water-splitting J values of in Figure 5g–i), proving the excellent stability of CoSAs-MoS2/
10 and 50 mA cm–2 at low operating voltages of 1.70 and 2.00 V; TiN NRs catalyst during water splitting under a wide range of
1.65 and 2.02 V; and 1.66 and 2.26 V in acidic, alkaline, and neu- pH conditions. Consequently, these results are also favorable
tral electrolytes, respectively (Figure 5a–f), slightly higher than evidence of the promise of the CoSAs-MoS2/TiN NRs for long-
those of the commercial Pt/C(−)//RuO2(+) couple (1.60 and term practical applications.
1.91 V; 1.58 and 2.06 V; 1.59 and 1.93 V in acidic, alkaline, and Next, the FE for H2 and O2 evolution in an alkaline electro-
neutral electrolytes, respectively) and are even superior to lyte were evaluated for the CoSAs-MoS2/TiN NRs by comparing
those of other recently reported bifunctional electrocatalysts the measured amount of produced gas to the theoretical amount
(Tables S1–S3, Supporting Information). Regarding the stability of produced gas, according to Equation (8).[59] The amount of
of the CoSAs-MoS2/TiN NRs for pH-universal overall water produced gas was measured with a lab-made alkaline electro-
splitting, we conducted chronoamperometric measurements lyzer with the CoSAs-MoS2/TiN NRs as bifunctional electro-
of the CoSAs-MoS2/TiN NRs (+,−) couple in different pH con- catalyst by chronoamperometric measurement at 2.19 V (versus
ditions. The chronoamperometric curves in Figure 5g–i show RHE) (Figure 6a). Dense H2 and O2 bubbles are observed on
smooth and stable features. In addition, the CoSAs-MoS2/TiN the cathode and anode surfaces, respectively, during chrono-
NRs retains 96.9%, 98.5%, and 93.14% of the original current amperometric measurement (Figure 6a), further confirming the

Adv. Funct. Mater. 2021, 2100233 2100233 (10 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 6. a) Images of a lab-made alkaline electrolyzer assembled with CoSAs-MoS2/TiN NRs as both the cathode and anode, along with sur-
face detail images of the CoSAs-MoS2/TiN NRs electrodes during chronoamperometric measurement. b) The measured and theoretical amounts of
produced gas against time of the alkaline electrolyzer when using CoSAs-MoS2/TiN NRs. c–e) Nyquist plots of the CoSAs-MoS2/TiN NRs, MoS2/TiN
NRs, TiN NRs, and CoSAs-MoS2 NSs in 0.5 m H2SO4, 1.0 m KOH, and 1.0 m PBS, respectively ((i): TiN NRs, (ii): CoSAs-MoS2 NSs, (iii): MoS2/TiN NRs,
and (iv): CoSAs-MoS2/TiN NRs). f) CV curves at 10 mV s–1 and g) current density variation (∆J = Janode − Jcathode, at 0.97 V (versus RHE)) as a function
of scan rate of (i): CC, (ii): TiN NRs, (iii): CoSAs-MoS2 NSs, (iv): MoS2/TiN NRs, and (v): CoSAs-MoS2/TiN NRs in 1.0 m KOH.

high performance of the developed CoSAs-MoS2/TiN NRs-based electron conductivity, resulting in a significant boost in HER
alkaline electrolyzer. Figure 6b displays a good agreement of and OER kinetics upon CoSAs-MoS2/TiN NRs utilization.[47,60]
measured and theoretical amounts of produced gas. We meas- In addition, ECSA values of the as-synthesized samples were
ured very high FE values of 99.5% and 99.1% for the HER and assessed by determining Cdl from cyclic voltammetry (CV)
OER, respectively, with a ≈2:1 ratio of O2 and H2. Moreover, the curves at different scan rates in alkaline electrolyte (Figure S16,
Rct and electrochemical active surface area (ECSA) were assessed Supporting Information). The CV curve at 10 mV s–1
to provide a basic understanding of the reason for the excellent of CoSAs-MoS2/TiN NRs exhibits much greater current and
catalytic behavior of the CoSAs-MoS2/TiN NRs, along with its area than those of the MoS2/TiN NRs, TiN NRs, CoSAs-MoS2
counterparts for comparison. The Nyquist plots in Figure 6c–e NSs, and CC at the same scan rate (Figure 6f), suggesting that
show that the semicircle radius of the CoSAs-MoS2/TiN NRs is the CoSAs-MoS2/TiN NRs results in a higher Cdl. Indeed, the
much smaller than those of the MoS2/TiN NRs, TiN NRs, and Cdl of CoSAs-MoS2/TiN NRs is 56.2 mF cm–2, which is much
CoSAs-MoS2 NSs at all pH levels, indicating that the CoSAs- higher than those of MoS2/TiN NRs (43.9 mF cm–2), TiN NRs
MoS2/TiN NRs has the smallest Rct. The lower Rct suggests (10.9 mF cm–2), CoSAs-MoS2 NSs (12.4 mF cm–2), and CC
more rapid electron transport kinetics and higher intrinsic (10.9 mF cm–2) (Figure 6g). The CoSAs-MoS2/TiN NRs thus

Adv. Funct. Mater. 2021, 2100233 2100233 (11 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

Figure 7. DFT calculations of the MoS2 and CoSAs-MoS2: a) The corresponding structure configurations for the two steps of HER in alkaline condi-
tion on Co active sites of CoSAs-MoS2; b) The corresponding structure configurations for the four steps of OER in alkaline condition on Mo active
sites of CoSAs-MoS2 (cyan, purple, yellow balls represent molybdenum, cobalt, sulfur atoms, respectively); c) The DOS calculation of the MoS2 and
CoSAs-MoS2; d) The Gibbs free-energy diagram for the two steps of HER in alkaline conditions on different sites of the MoS2 and CoSAs-MoS2 ((1):
H-Co(CoSAs-MoS2), (2): H-Mo(CoSAs-MoS2), (3): H-S(CoSAs-MoS2), (4): H-Mo(MoS2), and (5): H-S(MoS2)); e,f) The Gibbs free-energy diagram for
the four steps of OER in alkaline condition on Mo sites of the MoS2 and CoSAs-MoS2, respectively.

possesses the largest ECSA, indicating more electrocatalytic active conditions on Co active sites and the four steps of OER in alka-
sites and, thus, markedly higher HER and OER performance. line conditions on Mo active sites of CoSAs-MoS2 are also pro-
Density functional theory (DFT) calculations were also per- vided in Figure 7a,b. The density of states (DOS) calculation
formed for the developed CoSAs-MoS2/TiN NRs to achieve provided in Figure 7c clearly shows the electronic states around
insight into the synergistic effects generating this desirable the Fermi level of both MoS2 and CoSAs-MoS2. Importantly,
electrocatalytic performance. Figure S17a,b, Supporting Infor- after dispersion of CoSAs over the MoS2 surface, the band
mation show top and side views of the geometric structures structure of the CoSAs-MoS2 is superior to that of pure MoS2
of the MoS2 shell layer of MoS2/TiN NRs and the CoSAs-MoS2 in the Fermi level region, suggesting remarkably improved
shell layer of CoSAs-MoS2/TiN NRs, respectively, with their electrical conductivity as well as increased electron mobility in
primary structural parameters. Additionally, the corresponding the CoSAs-MoS2.[61,62] These modifications are highly beneficial
structure configurations for the two steps of HER in alkaline for electron transfer to active sites on the catalyst surface, thus

Adv. Funct. Mater. 2021, 2100233 2100233 (12 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

significantly enhanced electrocatalytic activity. In addition, the evidence supporting the essential role of chemical coupling
hydrogen adsorption free energy (∆GHads) of individual sites of between CoSAs and host MoS2 in the enhanced electrocatalytic
the MoS2 and CoSAs-MoS2, including Mo, S, and Co sites, was performance toward the HER and OER.
also calculated to determine their hydrogen adsorption behavior
(Figure 7d).[63] The ∆GHads is well-known as a primary indicator
of the HER activity of a catalyst; the ∆GHads value of a good 3. Conclusions
catalyst for the HER should be close to 0 eV.[64] For the MoS2,
the Mo and S sites show largely positive ∆GHads of 0.775 and In this study, we successfully synthesized a novel CoSAs-MoS2/
0.737 eV, respectively, suggesting weak adsorption of Hads on TiN NRs catalyst showing excellent catalytic performance and
the surface of pure MoS2, which has a negative influence on the stability toward simultaneous HER and OER in a wide pH
HER catalysis process.[65] In contrast, the hydrogen adsorption range. The superior performance is derived from the syner-
behavior of CoSAs-MoS2 is better than that of pure MoS2, as gistic effects of advantageous structural and electronic factors.
shown by the near-zero ∆GHads of Mo sites (0.603 eV) and S site A strong CoSAs-MoS2 interaction results in regulation of elec-
(0.651 eV). The synergistic effect between CoSAs and MoS2 is tronic structures toward an optimized intermediate adsorption/
considered a major contributing factor in the optimized ∆GHads. desorption ability, while the unique 3D core/shell architecture
Furthermore, the ∆GHads of Co sites is 0.224 eV, which is very coupled with the mesoporous structure provides large surface
close to the ideal value of 0 eV, suggesting that the CoSAs also area and high number of active sites, thus enlarging the contact
serve as extra active sites for the HER in the CoSAs-MoS2 struc- surface between reactants and active sites and facilitating reac-
ture—besides, the Co sites is better active sites than that of the tant penetration/diffusion and the release of gaseous species.
Mo and S sites for catalyzing the HER. Additionally, we used a Additionally, both outer CoSAs-MoS2 NSs and inner TiN NRs
DFT calculation to clarify the fundamental mechanism of the display as effective conductive channels for charge transport
alkaline OER process. The OER can be described according to toward dramatically enhanced reaction kinetics. The CoSAs-
the following reactions:[66] MoS2/TiN NRs demonstrate good HER efficiency, with a small
Eo of 187.5, 131.9, and 203.4 mV at a J of 10 mA cm–2 under
Cat. + OH− → Cat. − OHads + e − ( ∆GOHads ) (3) 0.5 m H2SO4, 1.0 m KOH, and 1.0 m PBS, respectively. The Eo
values for the OER are 454.9 mV in 0.5 m H2SO4, 340.6 mV in
Cat. − OHads + OH− → Cat. − Oads + H2O ( ∆GOads ) (4) 1.0 m KOH, and 508.0 mV in 1.0 m PBS, making it a promising
candidate for commercialized pH-universal water oxidation.
In particular, the developed catalyst shows remarkable stability
Cat. − Oads + OH− → Cat. − OOHads + e − ( ∆GOOHads ) (5)
under pH-universal HER and OER conditions with high cur-
rent response after 45 h of continuous work. By employing the
Cat. − OOHads + OH− → Cat. + O2 + H2O ( ∆GO2 ) (6) CoSAs-MoS2/TiN NRs as both the cathodic and anodic elec-
trodes for water splitting in 0.5 m H2SO4, 1.0 m KOH, and 1.0 m
where OHads, Oads, OOHads are intermediate species adsorbed PBS, operating voltages of merely 1.70, 1.65, and 1.66 V, respec-
on the active sites of catalyst and ∆GOHads, ∆GOads, ∆GOOHads, tively, are required to generate a J of 10 mA cm–2, together with
and ∆GO2 are Gibbs free energy changes in each step. In gen- superior long-term stability for over 30 h in pH-universal water
eral, difference of ∆GOHads and ∆GOads (adsorption energy dif- splitting conditions.
ference, ΔG = |ΔGOHads − ΔGOads| (eV)) is recognized as an
indicator of the OER activity of a catalyst; a good catalyst for
the OER possesses a low adsorbed energy difference value.[67,68] 4. Experimental Section
The calculated Gibbs free-energy change, adsorption energy dif-
ference, and theoretical OER overpotential values of the MoS2 Chemical Reagents: Carbon cloth (CC), titanium(IV) chloride
(TiCl4, 99.9%), hydrochloric acid (HCl, 37%), titanium n-butoxide
and CoSAs-MoS2 are provided in Table S4. The Gibbs free-
(Ti(OC4H9)n, 97%), thiourea (NH2CSNH2, ≥ 99.0%), Nafion (5 wt.%),
energy diagram for the OER in an alkaline condition on the sodium molybdate dihydrate (Na2MoO4.2H2O, ≥ 99.0%), cobalt(II)
Mo center of the MoS2 and CoSAs-MoS2 was established and nitrate hexahydrate (Co(NO3)2.6H2O, ≥ 98%), potassium hydroxide
given in Figure 7e,f. For pure MoS2, the high adsorption energy (KOH, ≥ 85%), ethanol (CH3CH2OH, ≥ 99.8%), platinum on carbon
difference value (1.33 eV) and high ∆GOHads (1.66 eV), which (Pt/C, 20 wt.%), sodium sulfate (Na2SO4, ≥ 99.0%), phosphate buffered
suggests too weak adsorption of OHads on the surface of pure saline (PBS, pH 7.4), acetone (CH3COCH3, ≥ 99.9%), sulfuric acid
MoS2, confirm the poor catalytic activity of MoS2 for the OER; (H2SO4, 99.9%), triblock copolymer PEOn-PPOm-PEOn (poly(ethylene
oxide)-block-poly(propylene oxide)-block-poly(ethylene oxide)), and
this is reconfirmed by a high theoretical OER overpotential of ruthenium oxide (RuO2, 99.9%) were purchased from Sigma Co. (USA).
1.62 eV. In contrast, the adsorption energy difference value of Synthesis of CoSAs-MoS2/TiN NRs: A CC substrate (5 × 2 cm) was
CoSAs-MoS2 reduces to 1.27 eV, implying facilitated reaction pretreated with HCl (3 m), acetone, and deionized (DI) water for 5 h
kinetics and better OER catalytic activity of the CoSAs-MoS2 each under magnetic stirring before being used in the experimental
than that of pure MoS2.[69] Also, the synergistic effects between process. First, TiO2 NRs were directly grown on CC by a seed-assisted
CoSAs and MoS2 endows the CoSAs-MoS2 to have optimum hydrothermal approach. Briefly, to form TiO2 seeds on CC surface, the
cleaned CC was dipped in a mixture of 0.43 mL of TiCl4 and 20 mL of DI
Gibbs free-energies, thereby achieving a much lower overpo-
water for 10 min, followed by blow-drying and heating at 350 °C in air for
tential of 0.86 eV than that of pure MoS2, implying the essen- 30 min. In the next step, 20 mL of HCl, 0.6 mL of Ti(OC4H9)n, and 20 mL
tial role of CoSAs in OER reaction. Therefore, in addition to of DI water were mixed together and then stirred for 48 h to obtain a
the experimental results, DFT calculations provide compelling transparent solution. The as-formed solution and the TiO2 seed-assisted

Adv. Funct. Mater. 2021, 2100233 2100233 (13 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

CC were hydrothermally reacted in a 50 mL Teflon-lined stainless-steel the as-prepared electrolyzer was investigated through LSV measurement
autoclave at 160 °C for 5 h. After cooling to room temperature, the TiO2 at 10 mV s–1 in all pH conditions. In addition, chronoamperometric
NRs were collected, washed with ethanol and DI water three times each, measurements were taken to examine the stability and durability of the
and then dried at 60 °C overnight. catalysts under overall water splitting conditions. Faradaic efficiency (FE)
To prepare TiN NRs, the TiO2 NRs were annealed at 900 °C at a was determined according to the following equation:[6]
heating rate of 1 °C min–1 in an Ar/NH3 atmosphere (100:50 sccm) for 2 h.
Ultrathin MoS2 NSs were vertically grown on the TiN NRs surface nmeasured
FE = × 100 (8)
through a hydrothermal method. Typically, 0.6 g of (NH2)2CS, 0.3 g of J · t/nF
Na2MoO4 · 2H2O, and 0.27 g of triblock copolymer PEOn-PPOm-PEOn where nmeasured is the measured amount of O2 or H2 gas (mol), J is the
were dissolved in 40 mL of DI water with stirring for 30 min. The solution constant oxidation/reduction current density (mA cm–2), t is electrolysis
was poured into a 50 mL Teflon-lined stainless-steel autoclave. After the time (s), n is the electron transfer number, and F is the Faraday constant
TiN NRs sample was dipped in that solution, the autoclave was sealed (96 485 Cmol–1). A water drainage method was used to measure the
and maintained at 200 °C for 6 h. Upon completion of the reaction, the amount of evolved H2 or O2 gas.
MoS2 NSs grown on the TiN NRs sample (denoted MoS2/TiN NRs) was
rinsed and dried at 60 °C for 12 h.
CoSAs were dispersed on the MoS2/TiN NRs surface as follows.
1.74 g of Co(NO3)2 . 6H2O was dissolved in 50 mL of DI water with Supporting Information
stirring for 30 min. The MoS2/TiN NRs were immersed in that solution
and maintained at 70 °C for 18 h. After washing with ethanol and DI Supporting Information is available from the Wiley Online Library or
water several times and drying at 60 °C overnight, the obtained sample from the author.
was annealed at 400 °C at a heating rate of 2 °C min–1 for 2 h under an Ar
atmosphere (100 sccm). Finally, acid etching was conducted by soaking
the as-annealed sample in 50 mL of 1 m HCl solution for 1 h, followed
by rinsing with ethanol and DI water, and then the CoSAs-MoS2/TiN Acknowledgements
NRs samples were collected for the next experiments. For comparison, This research was supported by the Basic Science Research Program
individual CoSAs-MoS2 NSs were fabricated by the above method (2019R1A2C1004983) and the Regional Leading Research Center Program
(Supporting information). (2019R1A5A8080326) through the National Research Foundation funded
Material Characterizations: Field emission scanning electron by the Ministry of Science and ICT of Republic of Korea.
microscopy (FE-SEM, Carl Zeiss AG-SUPRA 40), TEM (Hitachi-H-7650),
and atomic-resolution TEM with spherical aberration (Cs) corrector
(CS-FE-TEM, JEM-ARM200F) were performed to characterize the
morphology and structure of the samples. The XRD (Rigaku-D/ Conflict of Interest
Max 2500 V/PC) patterns were recorded from 10° to 90° using Cu Kα
radiation (λ = 1.54 Å) to determine the phase compositions of the The authors declare no conflict of interest.
samples. XPS (Thermo Fisher Scientific, Theta Probe) was performed
to further investigate the elemental compositions of the samples. N2
adsorption-desorption isotherms were recorded on an ASAP 2020 Plus
system (Micromeritics Instrument Corp., USA) to evaluate the BET
Data Availability Statement
specific surface area and pore size of the samples. Research data are not shared.
Electrochemical Measurement: All electrochemical measurements
were carried out on a CH electrochemical workstation (CH660D-CH
Instruments) at room temperature. A basic three-electrode cell system
was assembled with a graphite rod counter electrode, an Ag/AgCl Keywords
reference electrode, and a piece of sample (1 × 1 cm) as the working
bifunctional catalyst, core/shell structure, pH-universal electrolytes,
electrode. All measured potentials were calibrated to the reversible
single-atom Co, water splitting
hydrogen electrode (RHE) scale according to the Nernst equation:
Received: January 11, 2021
E RHE = E Ag/AgCl + 0.059 × pH + 0.198 V (7) Revised: April 6, 2021
Before electrochemical testing, electrolyte solutions including 0.5 m Published online:
H2SO4, 1.0 m KOH, and 1.0 m PBS were deaerated by N2 gas for 30 min.
iR compensations were applied to all polarization data. The polarization
curves were recorded by LSV at 10 mV s–1. Double layer capacitance
(Cdl) was determined in an alkaline environment by running CV at [1] D. McDowall, B. J. Greeves, R. Clowes, K. McAulay,
10, 15, 20, 25, and 30 mV s–1 in a potential range of −0.1 to −0.01 V A. M. Fuentes-Caparrós, L. Thomson, N. Khunti, N. Cowieson,
(versus Ag/AgCl). Electrochemical impedance spectroscopy (EIS) was
M. C. Nolan, M. Wallace, A. I. Cooper, E. R. Draper, A. J. Cowan,
scanned from 0.01 Hz to 100 kHz at open circuit potential at all pH
D. J. Adams, Adv. Energy Mater. 2020, 10, 2002469.
values to measure the charge-transfer resistance (Rct) of the samples.
[2] K. T. Møller, T. R. Jensen, E. Akiba, H. wen Li, Prog. Nat. Sci. 2017,
Chronoamperometric measurement was conducted at current responses
of 50 mA cm–2 to assess the stability and durability of the obtained 27, 34.
samples. Also, multistep chronoamperometric measurement was used [3] Y. Zheng, Y. Jiao, M. Jaroniec, S. Z. Qiao, Angew. Chem. 2015, 54, 52.
to evaluate the stability of the samples at various operating potentials. [4] Y. Shi, B. Zhang, Chem. Soc. Rev. 2016, 45, 1529.
A lab-made water splitting device was constructed using CoSAs-MoS2/ [5] T. L. Luyen Doan, D. T. Tran, D. C. Nguyen, H. Tuan Le, N. H. Kim,
TiN NRs as both cathode and anode in a symmetrical two-electrode J. H. Lee, Appl. Catal. B 2020, 261, 118268.
configuration, in which the cathode chamber was separated from the [6] L. Wang, X. Duan, X. Liu, J. Gu, R. Si, Y. Qiu, Y. Qiu, D. Shi, F. Chen,
anode chamber by a Nafion membrane. Another symmetrical two- X. Sun, J. Lin, J. Sun, Adv. Energy Mater. 2020, 10, 1903137.
electrode system was also prepared for comparison purposes using Pt/C [7] Z. Yin, Y. Sun, Y. Jiang, F. Yan, C. Zhu, Y. Chen, ACS Appl. Mater.
on CC as the cathode and RuO2 on CC as the anode. The performance of Interfaces 2019, 11, 27751.

Adv. Funct. Mater. 2021, 2100233 2100233 (14 of 15) © 2021 Wiley-VCH GmbH
www.advancedsciencenews.com www.afm-journal.de

[8] H. Bin Wu, B. Y. Xia, L. Yu, X. Y. Yu, X. W. Lou, Nat. Commun. 2015, [39] S. Wang, D. Zhang, B. Li, C. Zhang, Z. Du, H. Yin, X. Bi, S. Yang,
6, 6512. Adv. Energy Mater. 2018, 8, 1801345.
[9] T. L. L. Doan, D. T. Tran, D. C. Nguyen, D. H. Kim, N. H. Kim, [40] A. M. Beyene, J. H. Yun, S. A. Ahad, B. Moorthy, D. K. Kim, Appl.
J. H. Lee, Adv. Funct. Mater. 2020, 2007822. Surf. Sci. 2019, 495, 143544.
[10] E. H. Yu, X. Wang, U. Krewer, L. Li, K. Scott, Energy Environ. Sci. [41] Q. Zhang, Y. Wang, Y. Wang, A. M. Al-Enizi, A. A. Elzatahry,
2012, 5, 5668. G. Zheng, J. Mater. Chem. A 2016, 4, 5713.
[11] Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, M. Jaroniec, [42] L. David, R. Bhandavat, G. Singh, ACS Nano 2014, 8, 1759.
S. Z. Qiao, J. Am. Chem. Soc. 2016, 138, 16174. [43] Z. Cui, C. Zu, W. Zhou, A. Manthiram, J. B. Goodenough, Adv.
[12] S. Yuan, Z. Pu, H. Zhou, J. Yu, I. S. Amiinu, J. Zhu, Q. Liang, J. Yang, Mater. 2016, 28, 6926.
D. He, Z. Hu, G. Van Tendeloo, S. Mu, Nano Energy 2019, 59, 472. [44] S. Cwik, D. Mitoraj, O. Mendoza Reyes, D. Rogalla, D. Peeters,
[13] X. Meng, L. Yu, C. Ma, B. Nan, R. Si, Y. Tu, J. Deng, D. Deng, J. Kim, H. M. Schütz, C. Bock, R. Beranek, A. Devi, Adv. Mater. Inter-
X. Bao, Nano Energy 2019, 61, 611. faces 2018, 5, 1800140.
[14] S. Deng, M. Luo, C. Ai, Y. Zhang, B. Liu, L. Huang, Z. Jiang, [45] Z. Shi, K. Nie, Z. J. Shao, B. Gao, H. Lin, H. Zhang, B. Liu, Y. Wang,
Q. Zhang, L. Gu, S. Lin, X. Wang, L. Yu, J. Wen, J. Wang, G. Pan, Y. Zhang, X. Sun, X. M. Cao, P. Hu, Q. Gao, Y. Tang, Energy Environ.
X. Xia, J. Tu, Angew. Chem., Int. Ed. 2019, 58, 16289. Sci. 2017, 10, 1262.
[15] D. C. Nguyen, D. T. Tran, T. L. L. Doan, D. H. Kim, N. H. Kim, [46] J. Lin, P. Wang, H. Wang, C. Li, X. Si, J. Qi, J. Cao, Z. Zhong, W. Fei,
J. H. Lee, Adv. Energy Mater. 2020, 10, 1903289. J. Feng, Adv. Sci. 2019, 6, 1900246
[16] Q. Xiong, Y. Wang, P. F. Liu, L. R. Zheng, G. Wang, H. G. Yang, [47] Y. Liu, S. Jiang, S. Li, L. Zhou, Z. Li, J. Li, M. Shao, Appl. Catal. B
P. K. Wong, H. Zhang, H. Zhao, Adv. Mater. 2018, 30, 1801450. 2019, 247, 107.
[17] P. Wang, X. Zhang, J. Zhang, S. Wan, S. Guo, G. Lu, J. Yao, [48] L. An, Y. Li, M. Luo, J. Yin, Y. Q. Zhao, C. Xu, F. Cheng, Y. Yang,
X. Huang, Nat. Commun. 2017, 8, 14580. P. Xi, S. Guo, Adv. Funct. Mater. 2017, 27, 1703779.
[18] J. X. Feng, J. Q. Wu, Y. X. Tong, G. R. Li, J. Am. Chem. Soc. 2018, [49] J. Di, C. Chen, S. Z. Yang, S. Chen, M. Duan, J. Xiong, C. Zhu,
140, 610. R. Long, W. Hao, Z. Chi, H. Chen, Y. X. Weng, J. Xia, L. Song, S. Li,
[19] D. Y. Wang, M. Gong, H. L. Chou, C. J. Pan, H. A. Chen, Y. Wu, H. Li, Z. Liu, Nat. Commun. 2019, 10, 2840.
M. C. Lin, M. Guan, J. Yang, C. W. Chen, Y. L. Wang, B. J. Hwang, [50] M. Zhang, Q. Dai, H. Zheng, M. Chen, L. Dai, Adv. Mater. 2018, 30,
C. C. Chen, H. Dai, J. Am. Chem. Soc. 2015, 137, 1587. 1705431.
[20] Z. Xing, X. Yang, A. M. Asiri, X. Sun, ACS Appl. Mater. Interfaces [51] J. Durst, A. Siebel, C. Simon, F. Hasché, J. Herranz, H. A. Gasteiger,
2016, 8, 14521. Energy Environ. Sci. 2014, 7, 2255.
[21] X. Han, X. Ling, Y. Wang, T. Ma, C. Zhong, W. Hu, Y. Deng, Angew. [52] Y. Guo, J. Tang, Z. Wang, Y. M. Kang, Y. Bando, Y. Yamauchi, Nano
Chem., Int. Ed. 2019, 131, 5413. Energy 2018, 47, 494.
[22] W. Liu, L. Cao, W. Cheng, Y. Cao, X. Liu, W. Zhang, X. Mou, L. Jin, [53] M. Shao, F. Ning, M. Wei, D. G. Evans, X. Duan, Adv. Funct. Mater.
X. Zheng, W. Che, Q. Liu, T. Yao, S. Wei, Angew. Chem., Int. Ed. 2014, 24, 580.
2017, 129, 9440. [54] J. Zhang, R. Cui, C. Gao, L. Bian, Y. Pu, X. Zhu, X. Li, W. Huang,
[23] D. A. Kuznetsov, Z. Chen, P. V. Kumar, A. Tsoukalou, A. Kierzkowska, Small 2019, 15, 1904688.
P. M. Abdala, O. V. Safonova, A. Fedorov, C. R. Müller, J. Am. Chem. [55] P. Yan, M. Huang, B. Wang, Z. Wan, M. Qian, H. Yan, T. T. Isimjan,
Soc. 2019, 141, 17809. J. Tian, X. Yang, J. Energy Chem. 2020, 47, 299.
[24] S. Dilpazir, H. He, Z. Li, M. Wang, P. Lu, R. Liu, Z. Xie, D. Gao, [56] M. Zhang, J. Chen, H. Li, P. Cai, Y. Li, Z. Wen, Nano Energy 2019,
G. Zhang, ACS Appl. Mater. Interfaces 2018, 1, 3283. 61, 576.
[25] C. Zhu, S. Fu, Q. Shi, D. Du, Y. Lin, Angew. Chem., Int. Ed. 2017, 56, [57] S. Drouet, J. Creus, V. Collière, C. Amiens, J. García-Antón, X. Sala,
13944. K. Philippot, Chem. Commun. 2017, 53, 11713.
[26] P. Liu, Y. Zhao, R. Qin, S. Mo, G. Chen, L. Gu, D. M. Chevrier, [58] J. Bai, T. Meng, D. Guo, S. Wang, B. Mao, M. Cao, ACS Appl. Mater.
P. Zhang, Q. Guo, D. Zang, B. Wu, G. Fu, N. Zheng, Science 2016, Interfaces 2018, 10, 1678.
352, 797. [59] C. Tang, R. Zhang, W. Lu, L. He, X. Jiang, A. M. Asiri, X. Sun, Adv.
[27] D. Ji, L. Fan, L. Li, S. Peng, D. Yu, J. Song, S. Ramakrishna, S. Guo, Mater. 2017, 29, 1602441.
Adv. Mater. 2019, 31, 1808267. [60] Y. Guo, J. Tang, H. Qian, Z. Wang, Y. Yamauchi, Chem. Mater. 2017,
[28] H. Zhang, G. Liu, L. Shi, J. Ye, Adv. Energy Mater. 2018, 8, 1701343. 29, 5566.
[29] X. Peng, A. M. Qasim, W. Jin, L. Wang, L. Hu, Y. Miao, W. Li, Y. Li, [61] T. Ling, T. Zhang, B. Ge, L. Han, L. Zheng, F. Lin, Z. Xu, W. Bin Hu,
Z. Liu, K. Huo, K. yin Wong, P. K. Chu, Nano Energy 2018, 53, 66. X. W. Du, K. Davey, S. Z. Qiao, Adv. Mater. 2019, 31, 1807771.
[30] W. S. Williams, Prog. Solid. State Chem. 1971, 6, 57. [62] Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, A. Du,
[31] N. C. Saha, H. G. Tompkins, J. Appl. Phys. 1992, 72, 3072. M. Jaroniec, S. Z. Qiao, Nat. Commun. 2014, 5, 3783.
[32] M. Yu, S. Zhao, H. Feng, L. Hu, X. Zhang, Y. Zeng, Y. Tong, X. Lu, [63] K. Sillar, A. Hofmann, J. Sauer, J. Am. Chem. Soc. 2009, 131, 4143.
ACS Energy Lett. 2017, 2, 1862. [64] J. K. Nørskov, T. Bligaard, A. Logadottir, J. R. Kitchin, J. G. Chen,
[33] M. H. Hsieh, G. A. Li, W. C. Chang, H. Y. Tuan, J. Mater. Chem. A S. Pandelov, U. Stimming, J. Electrochem. Soc. 2005, 152, J23.
2017, 5, 4114. [65] J. Greeley, T. F. Jaramillo, J. Bonde, I. Chorkendorff, J. K. Nørskov,
[34] D. Wang, Q. Li, C. Han, Z. Xing, X. Yang, Appl. Catal. B 2019, 249, 91. Nat. Mater. 2006, 5, 909.
[35] B. Shang, X. Cui, L. Jiao, K. Qi, Y. Wang, J. Fan, Y. Yue, H. Wang, [66] N. T. Suen, S. F. Hung, Q. Quan, N. Zhang, Y. J. Xu, H. M. Chen,
Q. Bao, X. Fan, S. Wei, W. Song, Z. Cheng, S. Guo, W. Zheng, Nano Chem. Soc. Rev. 2017, 46, 337.
Lett. 2019, 19, 2758. [67] H. Yan, Y. Xie, A. Wu, Z. Cai, L. Wang, C. Tian, X. Zhang, H. Fu, Adv.
[36] D. He, H. Ooka, Y. Kim, Y. Li, F. Jin, S. H. Kim, R. Nakamura, Proc. Mater. 2019, 31, 1901174.
Natl. Acad. Sci., U. S. A. 2020, 117, 31631. [68] P. Li, X. Duan, Y. Kuang, Y. Li, G. Zhang, W. Liu, X. Sun, Adv. Energy
[37] C. Ataca, H. Şahin, E. Aktuörk, S. Ciraci, J. Phys. Chem. C 2011, 115, 3934. Mater. 2018, 8, 1703341.
[38] Y. Fang, J. Pan, J. He, R. Luo, D. Wang, X. Che, K. Bu, W. Zhao, P. Liu, [69] Y. Gu, S. Chen, J. Ren, Y. A. Jia, C. Chen, S. Komarneni, D. Yang,
G. Mu, H. Zhang, T. Lin, F. Huang, Angew. Chem. 2018, 130, 1246. X. Yao, ACS Nano 2018, 12, 245.

Adv. Funct. Mater. 2021, 2100233 2100233 (15 of 15) © 2021 Wiley-VCH GmbH