Nano Research 2017, 10(9): 3103–3112

DOI 10.1007/s12274-017-1527-1

Au/Ni12P5 core/shell single-crystal nanoparticles as

oxygen evolution reaction catalyst
Yingying Xu1,§, Sibin Duan1,§, Haoyi Li2, Ming Yang3, Shijie Wang3, Xun Wang2, and Rongming Wang1 ()

1
Beijing Key Laboratory for Magneto-Photoelectrical Composite and Interface Science, School of Mathematics and Physics, University
of Science and Technology Beijing, Beijing 100083, China
2
Department of Chemistry, Tsinghua University, Beijing 100084, China
3
Institute of Materials Research and Engineering, A*STAR, 2 Fusionopolis Way, Singapore 138634, Singapore
§
Yingying Xu and Sibin Duan contributed equally to this work.

Received: 25 November 2016 ABSTRACT

Revised: 10 February 2017 We have demonstrated the improved performance of oxygen evolution reactions
Accepted: 11 February 2017 (OER) using Au/nickel phosphide (Ni12P5) core/shell nanoparticles (NPs) under
basic conditions. NPs with a Ni12P5 shell and a Au core, both of which have
© Tsinghua University Press well-defined crystal structures, have been prepared using solution-based synthetic
and Springer-Verlag Berlin routes. Compared with pure Ni12P5 NPs and Au-Ni12P5 oligomer-like NPs, the
Heidelberg 2017 core/shell crystalline structure with Au shows an improved OER activity. It
affords a current density of 10 mA/cm2 at a small overpotential of 0.34 V, in 1 M
KEYWORDS KOH aqueous solution at room temperature. This enhanced OER activity may
oxygen evolution reaction, relate to the strong structural and effective electronic coupling between the
nickel phosphide, single-crystal core and the shell.
core/shell nanoparticles,
interfacial coupling

1 Introduction effective, highly active catalytic electrodes for oxygen

evolution reactions (OERs) has become a critical topic
In the pursuit of sustainable energy, the hydrogen cycle in water splitting. In order to develop highly active
has attracted considerable attention as hydrogen is a catalysts based on abundant elements, some oxides
clean, non-polluting, and CO2-free energy carrier for of first-row transition metals (Co, Mn, Ni, Fe, etc.)
energy storage options. However, hydrogen production have been extensively studied as electrode materials
processes such as direct solar [1] and electricity-driven [3–6], in contrast to the well-known noble metal oxide
[2] water splitting (H2O → H2 + O2) are always catalysts (such as RuO2 and IrO2 [7, 8]). These ceramic
associated with substantial energy losses. Most of the oxides with the spinel (such as Co3O4 [3, 4] and NiCo2O4
overpotential giving rise to the energy loss is related [5]) or perovskite (mainly LaNiO3, LaCoO3 [6], etc.)
to the oxygen evolution process. The design of cost- structures exhibit efficient electrochemical activity

Address correspondence to [email protected]

3104 Nano Res. 2017, 10(9): 3103–3112

for OER, as well as high electronic conductivity and mismatch, or interfacial interactions between the core
stability in alkaline solutions. and the shell [24]. With substantial lattice constant
Other non-metal elements like B, N, P, or S have mismatches at the core/shell interfaces, homogenous
also been explored to improve the activities of metal strain distribution and strong coupling effects can be
electrocatalysts in recent works [2, 9–13]. In order to induced in single-crystal hybrid metal/semiconductor
further lower the overpotential in water splitting, other core/shell heterostructures [25, 26]. It is believed that
kinds of transition metal compounds with considerable these strong coupling interactions could induce core/
conductivity and stability have been successfully shell synergetic effects and enhance the catalytic
used as catalysts in OER, including metal sulfides (Ni3S2) performance.
[9], nitrides (CoN) [10], and phosphates (Co-Pi) [2, 13]. In this study, we report that NPs consisting of a
Transition metal phosphides (Ni2P, Co2P, Fe2P, etc.) single crystalline nickel phosphide (Ni12P5) shell and
are well known as a group of stable catalysts with a Au core display efficient OER catalytic ability. The
promising high-activity in fuel hydroprocessing synthesized Au/Ni12P5 core/shell NPs can afford a
technology [14, 15] and organic dye degradation [16, 17]. current density of 10 mA/cm2 at a small overpotential
Recently, they have also been shown to have pro- of 0.34 V in 1 M KOH aqueous solution, which is
mising electrochemical catalytic performance in water much improved compared with that of Au-Ni12P5
splitting, especially in hydrogen evolution reactions oligomer-like and pure Ni12P5 NPs.
(HER) [18–22]. In addition, the physical properties of
the metal phosphides are desirable for application as
2 Experimental
electrochemical catalytic electrodes. It is reported that
metal-rich phosphides are good conductors of heat The Au/Ni12P5 core/shell NPs are synthesized via a
and electricity (resistivity < 200 μΩ·cm) similar to metals, solution-based route [27], using the in situ transition
with hardness and strength close to those of ceramics. method based on our previous work [26]. In a typical
These phosphides also display high thermal and synthesis, Ni(acac)2 (0.5 g) is dissolved in 10 mL
chemical stability [15]. oleylamine (OAm), and then heated up to 100 °C
One way to achieve a large reaction surface and for 60 min, forming a blue-green solution. After that,
tunable activity in catalytic design is to develop a freshly prepared solution of HAuCl4·4H2O (0.05 g
nanostructured catalysts. By adjusting the components in 5 mL of toluene) was added dropwise with
and their interface, heterostructured catalysts can stirring and maintained at 100 °C for 120 min. Then,
provide the desired flexibility. Due to their synergetic triphenylphosphine (TPP) (1.0 g) was added and the
effects, the coupling interaction between the catalyst reaction stirred for 60 min. The solution was then
and the substrate can lead to high OER activity. It is heated to 270 °C for 180 min and then cooled to room
reported that the OER turnover frequency (TOF) for temperature. Finally, the prepared Au/Ni12P5 core/shell
‫׽‬0.4 monolayer (ML) of cobalt oxide deposited on NPs were collected by centrifugation and washed
Au is nearly three times higher than that for bulk several times with acetone, then resolved in toluene.
Ir [23], due to the strain or structural deformation in Au-Ni12P5 oligomer-like NPs were synthesized using
the nanostructure, which modifies the reactivity of the traditional seeded growth method for comparison.
the nanostructure surface and the energy levels of The details of the preparation procedure can be found
the bonding electrons [24]. The catalytic activity and in S1 (in the Electronic Supplementary Material (ESM)).
selectivity can be further improved by tuning the stress The crystal structures, morphologies, and chemical
in the nanostructures. compositions of the prepared Au/Ni12P5 NPs were
Compared with other systems such as thin films, studied using X-ray diffraction (XRD, X'Pert Pro MPD
the core/shell nanoparticle (NP) structure is one of system, Cu-Kα) and transmission electron microscopy
the most effective ways to induce stress. Stress in the (TEM, JEOL 2100FS with 200 kV field emission gun).
core/shell NPs may come from the finite size effect, The energy dispersive X-ray spectroscope (EDS) results
defects, surface reconstruction, interfacial lattice are conducted with a field emission scanning electron

| www.editorialmanager.com/nare/default.asp
Nano Res. 2017, 10(9): 3103–3112 3105

microscope (Zeiss Supra 55) to identify the elemental 3 Results and discussion
composition of the sample. X-ray photoelectron
spectroscopy (XPS) measurements were performed 3.1 Morphology and structure
using a VG ESCA LAB 220i-XL XPS system with a
The prepared NPs were first identified as core/shell
monochromatic Al Kα source. All XPS spectra were
nanostructures by TEM characterization. From the TEM
aligned using the C 1s core level peak (284.6 eV).
image in Fig. 1(a), the prepared samples are nearly
The electrocatalytic activities of the synthesized
spherical core/shell particles, with a core diameter of
samples in alkaline solutions were characterized
~5 nm and a shell thickness of ~5.5 nm. High-resolution
with electrochemical methods by a PARSTAT 4000
transmission electron microscopy (HR-TEM) was used
Potentiostats-Electrochemistry Workstation. The OER
measurements were carried out in a three-electrode to investigate the crystal structure of a single particle
electrochemical cell containing 1 M KOH aqueous in detail (Fig. 1(b)). The lattice spacing measured in
solution as the electrolyte with Pt foil as the counter the shell is 0.186 and 0.217 nm, corresponding to the
electrode, and a saturated calomel electrode (SCE) as tetragonal Ni12P5 (312) and (040) planes. The crystal
the reference electrode. To prepare the working electrode, lattices in the shells are consecutive, indicating the
the electroactive samples were mixed with CABOT single crystalline nature of the semiconductor shells,
VXC72 carbon black and Nafion. A loading mount which is beneficial for ionic diffusion and electron
of 25 μg of sample was prepared on a glassy carbon transport [26].
electrode (GCE) of 5 mm in diameter (~0.13 mg/cm2). The EDS results for the NPs are shown in Fig. 1(c).
The polarization data tests were conducted at a scan The atom ratio Ni:P:Au is calculated to be 68:30:2.
rate of 5 mV/s between 0 and 0.8 V vs. SCE (about The Ni/P atom ratio is 2.27, confirming the nickel
1.1–1.9 V vs. reversible hydrogen electrode (RHE)). phosphide phase is more likely to be Ni12P5 instead of
Electrochemical impedance spectroscopy (EIS) exper- Ni2P. The ratio is less than the stoichiometric ratio
iments were carried out from 100 kHz to 0.1 Hz at for Ni12P5 (= 2.4). This is most likely due to some TPP
1.57 V (vs. RHE). residue coated on the NP surfaces. The weight ratio of

Figure 1 Structural characterizations of Au/Ni12P5 core/shell NPs. (a) The morphology of Au/Ni12P5 core/shell NPs revealed by TEM
characterization. (b) HR-TEM image of an individual Au/Ni12P5 core/shell NP and the corresponding fast Fourier transformation (FFT)
pattern (inset), confirming the single crystalline nature of the Ni12P5 shell. Scale bar: 5 nm. (c) The EDS spectrum of the synthesized
Au/Ni12P5 NPs. The atom ratio of Ni:P:Au is confirmed to be 68:30:2.

www.theNanoResearch.com∣www.Springer.com/journal/12274 | Nano Research

3106 Nano Res. 2017, 10(9): 3103–3112

Au:Ni12P5 is calculated to be 7.3:92.7, which is consistent 0.8646 nm and c = 0.5070 nm), it could be inferred that
with the results of the XRD Rietveld analysis. the tetragonal crystal structure is a little extracted
Further evidence for the Au/Ni12P5 heterostructure (~1.1%) in the c direction and contracted (~0.3%) in
was obtained by XRD. As shown in Fig. 2(a), all the a and b directions.
diffraction peaks can be indexed as the body-centered The chemical states of Ni and P in Au/Ni12P5
tetragonal (bct) phase of Ni12P5 (JCPDS No. 74-1381) core/shell NPs were investigated by XPS, as shown in
and the face-centered cubic (fcc) phase of Au (JCPDS Figs. 2(b)–2(d). In nickel phosphides, electrons transfer
No. 89-3697), respectively. The grain sizes are estimated from Ni to P. In Fig. 2(b), the Ni 2p1/2 binding energy
to be ‫׽‬6 nm for the Au using the (002) peak and around 853.3 eV is higher than the reported values
‫׽‬15 nm for the Ni12P5 using the (121) peak by Scherrer’s for Ni metal (852.5–852.9 eV) and lower than those
equation, indicating the integrity of the crystal structure reported for Ni in NiO (853.5–854.1 eV) [28], indicating
in the particles. that the Ni cations in Ni12P5 possess a partial positive
The phase composition and lattice parameters of charge between 0 and 2 [29]. In the P 2p peak in Fig. 2(c),
the samples were evaluated by Rietveld analysis on the P binding energy shifts to 129.5 eV for 2p3/2 and
the data of the XRD measurement. Estimation with 130.5 eV for 2p1/2, slightly lower than the values for the
Rietveld refinement implies that Ni12P5 makes up neutral atom (about 130.4 eV for black phosphorus
92 wt.% (~97 vol.%) of the whole Au/Ni12P5 particle, and 130.6 eV for red phosphorus [30]). The weak
with refined lattice parameters of tetragonal a = shoulder in the range of 131–133 eV indicates a slight
0.8622 nm and c = 0.5129 nm. The calculated plane amount of high oxidation state P, which may come
distance and the volume percentage of the Ni12P5 from residual TPP. Figure 2(d) shows the binding
shell have a good agreement with the values directly energy peaks at 84.1 and 87.8 eV specific to Au 4f7/2
observed from the TEM image. Referring to the lattice and Au 4f5/2, respectively. The characteristic features
parameter data in the literature (JCPDS 89-3697, a = of these peaks show the presence of metallic gold in

Figure 2 (a) The XRD pattern of the synthesized Au/Ni12P5 NPs. : Ni12P5; : Au. The simulation curve (red line) is calculated using
Rietveld analysis based on the observed data (black line) of the measurement. (b)–(d) XPS spectra of Au/Ni12P5 core/shell NPs: (b) Ni 2p,
(c) P 2p, and (d) Au 4f regions. Black line: observed data; red and blue lines: curve fitting; yellow line: background.

| www.editorialmanager.com/nare/default.asp
Nano Res. 2017, 10(9): 3103–3112 3107

the NPs. 3.2 Electrochemical properties of Au-modified

The Au-Ni12P5 oligomer-like and pure Ni12P5 NPs Ni12P5 NPs
were also characterized using TEM. The morphology
of the pure Ni12P5 NPs, Au-Ni12P5 oligomer-like and The OER activities of the nickel phosphide NPs with
Au/Ni12P5 core/shell NPs are shown in Fig. 3. All three different nanostructures were investigated by electro-
varieties of NP were of similar sizes (diameters of chemical methods in alkaline solutions. All three nickel
around 16 nm). Obviously, the differences in catalytic phosphide NPs behave as effective electrocatalysts
properties mainly come from the composition and towards OER, as illustrated in Fig. 4, and summarized
structural differences due to the inclusion of Au. in Table 1. From the polarization curves in Fig. 4(a),

Figure 3 TEM characterization and structure schematic illustration of (a) pure Ni12P5 NPs; (b) Au-Ni12P5 oligomer-like NPs; and (c)
Au/Ni12P5 core/shell NPs. Scale bar: 10 nm.

Figure 4 (a) Polarization curves, (b) Tafel plots and (d) Nyquist plots of reported Au/Ni12P5 core/shell NPs, Au-Ni12P5 oligomer-like
NPs and pure Ni12P5 NPs loaded on a GCE in 1.0 M KOH solution. The loading amount of the catalysts on the GCE are 0.13 mg/cm2.
The Tafel plots have been corrected for the iR drop of the solution. Tafel slope are calculated to be 42–49 mV/dec by linear fitting. Inset
of (d): the corresponding equivalent circuit diagram, consisting of Rs, Rt, and CPE. (c) Chronoamperometric response at a constant
potential of 1.57 V for Au/Ni12P5 core/shell NPs.

www.theNanoResearch.com∣www.Springer.com/journal/12274 | Nano Research

3108 Nano Res. 2017, 10(9): 3103–3112

Table 1 The OER catalytic parameters of the Au/Ni12P5 core/shell in which a pair of Ni2+/Ni3+ redox peaks can be
NPs, Au-Ni12P5 oligomer-like NPs and pure Ni12P5 NPs observed (see S2 and Fig. S1 in the ESM), indicating
Au-Ni12P5 Au/Ni12P5 good reversibility after the oxidation reaction. An
Ni12P5
oligomer-like core/shell
NPs oxidation peak is located at ~1.38 V, indicating the
NPs NPs
oxidation of Ni2+ to Ni3+ in Ni12P5, and the initial
Overpotential at j =
0.38 0.36 0.34 absorption of –O or –OH to Ni atoms at the catalyst
10 mA/cm2 (V)
Tafel slope (mV/dec) 43 42 49 surface before the oxygen evolution, which is common
in the OER process (as depicted in previous studies
Electrolyte resistance
7.4 6.7 5.3 [33, 34]).
Rs (Ω)
Charge-transfer The chronoamperometric response at a constant
46.6 28.8 11.4
resistance Rt (Ω) potential of 1.57 V for the Au/Ni12P5 core/shell NPs
CPEa: Y0 (S·sn) 0.0026 0.0021 0.016 was recorded, as shown in Fig. 4(d). The OER activity
a
CPE : n 0.70 0.72 0.73 is reserved with an anodic current attenuation of
a
The impedance of a CPE ZQ = (jω) −n
·Y0−1, where ω is the angular 22% after 104 s. The NPs demonstrate considerable
frequency in the circuit. durability in alkaline electrolytes, however there is still
room for better performance with improved adhesion
the inclusion of the Au core in the nickel phosphide of the powder to the electrode.
shell has lowered the overpotential to 0.34 V at a These results demonstrate that the different Ni12P5
current density of 10 mA/cm2, which is comparable based NPs, i.e. Au/Ni12P5 core/shell, Au-Ni12P5 oligomer-
to the value for oxide catalysts such as Co3O4 (~0.31 V) like and pure Ni12P5 NPs, are all efficient catalysts
[31]. This is obviously an improvement over Au-Ni12P5 for OER, of which the Au/Ni12P5 core/shell NPs are
oligomer-like (0.36 V) and pure Ni12P5 NPs (0.38 V). the most effective. To identify the exact electrical con-
At room temperature with the same loading, the ductivity of the three electrodes and investigate
Au/Ni12P5 core/shell NPs exhibited an area-specific the structure-property relationships, electrochemical
activity of 10 mA/cm2 at 1.57 V (vs. RHE), which is impedance spectra were first measured at room
242% and 95% greater than that of pure Ni12P5 temperature over a frequency range from 100 kHz to
NPs (2.92 mA/cm2) and Au-Ni12P5 oligomer-like NPs 0.1 Hz under open-circuit conditions. As the semi-
(5.14 mA/cm2), respectively. It is obvious that the circular Nyquist plots show in Fig. 4(d), the system was
performance of Ni12P5 NP catalysts is greatly enhanced typically modeled with an equivalent circuit model
with the inclusion of Au, especially in the core/shell consisting of an electrolyte resistance (Rs), a charge-
structure. transfer resistance (Rt) and a constant-phase element
As shown in Fig. 4(b), all our NPs displayed a small (CPE) (shown in Fig. 4(d) inset) and simulated by
Tafel slope, which is about 49, 42, and 43 mV/dec for Zview software. The Faradaic resistance (charge-transfer
Au/Ni12P5 core/shell, Au-Ni12P5 oligomer-like and pure resistance) Rt of the samples, which intuitively
Ni12P5 NPs, respectively. The relatively small Tafel corresponds to the circular diameter in Nyquist plots,
slopes indicate the superior reaction kinetics and good represents the resistance from the mass transfer during
catalytic activities of these nickel phosphide NPs. For electrocatalytic reactions, which is an important
example, some well-known OER catalysts have Tafel reference for the reaction efficiency of the electrode
slopes of 40 mV/dec (pure RuO2), 60 mV/dec (pure [21, 35]. Rt was calculated to be 46.6 Ω for pure Ni12P5
IrO2) [7] and 70 mV/dec (Co3O4 hybrid system) [4]. The NPs, 28.8 Ω for Au-Ni12P5 oligomer-like NPs and
Tafel slopes of our samples are comparable or even 11.4 Ω for the Au/Ni12P5 core/shell NPs. Obviously,
lower than some known OER catalysts. core/shell NPs have the lowest charge-transfer resistance
The oxidation peak at the voltage of ~1.39 V can be among the three samples, which can be attributed to
associated with the Ni2+/Ni3+ couple, which is common the enhanced electric conductivity and short charge
in some Ni contained systems [32]. To confirm this, diffusion length due to the synergistic effect. This is
we measured the cyclic voltammogram (CV) curve, in good agreement with the fact that the core/shell

| www.editorialmanager.com/nare/default.asp
Nano Res. 2017, 10(9): 3103–3112 3109

crystals display the highest catalytic activity. the other hand, lattice-engineered interfacial effects
The capacitance-like Y0 of CPE reflects the double between Au and Ni12P5 also affect the electronic and
layer capacitance (Cd) at the surface of the NPs. The catalytic properties of nanoparticles. As Fig. 1(b) shows,
double layer capacitances often reflect the position of the lattice spacing of the Au (111) facets was measured
absorbed molecules, which is related to the effective to be 0.217 nm, which is 7.9% less than its standard
area of the surface. The value of Y0 for the core/shell value (0.235 nm), indicating that the Au (111) facets
crystal is much larger than the Au-Ni12P5 oligomer-like are suppressed by the outer Ni12P5 shells. The atomic
and pure Ni12P5 NPs, as shown in Table 1. By simplifying positions drift away from the normal lattice points,
the model to R(CR) using Cd instead of Y0, we can creating strain or other structural changes. Such changes
compare the double layer capacitances Cd of the three affect the electronic and catalytic properties of the
structures by simulation. In fact, Cd can be deduced by nanoparticles. Moreover, the dislocations and lattice
strains can be clearly observed. With this tensile stress
Cd = 1/ωBRt (1) exerted by the Au core support, the shell overlayer
undergoes structural distortion which may result in
directly, where ωB is the angular frequency at the
tuning the d-band center and the adsorbate binding
top point of each semicircle. The double layer capa-
energy in the surface [36].
citance of 0.97 mF for pure Ni12P5 NPs and 1.1 mF for In addition, the lattice matching between Au and
oligomer-like NPs are very close, while the Au/Ni12P5 Ni12P5 could also be verified by the lattice continuity
core/shell NPs show a much larger capacitance of between Au and Ni12P5. The orientation relationships
11.0 mF. The Au cores could multiply the double between the fcc Au core and the bct Ni12P5 shell were
layer capacitance of the Ni12P5 NPs by 10. This result determined as Ni12P5(312)//Au(210) and Ni12P5(040)//
discloses the distinct microstructural differences bet- Au(111), as illustrated in Fig. 1(b). The lattice matching
ween the core/shell and oligomer-like NPs. For our between metal core and semiconductor shell serve
NPs with similar size and phase, the much larger as efficient electron channels during the reaction,
double layer capacitance of the Au/Ni12P5 core/shell correspondingly enhancing its catalytic performance.
NPs implies a higher concentration of electroactive Finally, the electrical coupling effect from Au will
positions at the surface, which reflect the difference increase the proportion of high oxidation state metal
of the electrical structure in the shell. atoms in the surface, which are essential intermediates
in OER. For our NP catalyst, when the metallic Au
3.3 Synergetic effect in the single crystalline
and nickel phosphide shell are in contact, the free
core/shell structure
electrons will transfer between them due to the work
Heterostructures have attracted great interest in function difference. As a highly electronegative transition
catalytic enhancement due to the potential synergetic metal, Au can act as an electron sink to facilitate
effect. The single crystal nature of the Ni12P5 shells the oxidation [23]. Compared to the oligomer-like
and lattice modulation between Au and Ni12P5 play an structure, the Au core in the core/shell structure can
important role in their electrocatalytic performance. affect the whole surface of the NPs, which may be
more helpful for OER.
It is known that the single crystalline semiconductor
shell benefits the study of semiconductor physics at 3.4 Catalytic mechanism of metal phosphides as
the nanoscale and promotes research on technological OER electrocatalysts
devices with optimum physicochemical properties
[25, 26]. Crystalline defects in the semiconductor shell Ni12P5 has a body-centered tetragonal structure with
weaken nanomaterial properties and degrade device lattice constant of a = 0.86 nm and c = 0.51 nm, as shown
performance. As shown in Fig. 1(b) and many similar in Fig. 5(a). The Ni–P bond lengths vary between
HR-TEM images, the Ni12P5 forms a single crystal, 0.22–0.26 nm. The P atoms have two equivalent sites in
which favors ionic diffusion and electron transport, Ni12P5: One is located at the vertex and body-center
and is accordingly beneficial for its OER activity. On positions (2a Wyckoff position) with an 8-Ni-atom

www.theNanoResearch.com∣www.Springer.com/journal/12274 | Nano Research

3110 Nano Res. 2017, 10(9): 3103–3112

promising electrocatalyst. In contrast, the bonding

character in the NiO surface is predominantly ionic,
with a much larger variation of electrostatic potential,
as shown in Fig. 5(c). This leads to strong binding
of OH* groups at the surface. For NiO, it has been
calculated that the binding energy is too strong (at
low coverage) to reach the theoretical best catalyst
activities [33, 41]. Thus, we can expect improved OER
activity with Ni12P5 compared with that with NiO.

4 Conclusion
Based on detailed structural and electrochemical
study, we find improved OER catalytic activity using
Au/Ni12P5 NPs compared to previously reported
Figure 5 (a) The structure schematic of the Ni12P5 lattice. electrocatalysts, which is ascribed to their weak
Calculated total electron density of the (b) Ni12P5 (001) surface hydride accepting ability at the surface, and strong
and (c) NiO (001) surface, mapped by the electrostatic potential. interfacial coupling between the crystalline metallic
The isosurface value is 0.2 e/Å3. Electrostatic potential is color cores and the thin semiconductor shells. Our results
coded as in the scale bar: The blue corresponds to negatively
suggest that these NPs could find use as promising
charged regions, while the red represents positively charged
electrocatalysts for water splitting applications.
regions. (Color of atoms: navy, Ni; pink, P; red, O.)

cuboid; the other is located at the 8h Wyckoff position Acknowledgements

with a 10-Ni-atom tetrakaidekahedron. For binary
nickel phosphides, it has been reported that the electronic This work was supported by the National Natural
polarizability of phosphorus atoms will increase with Science Foundation of China (Nos. 11674023, 51371015,
coordination to the neighboring Ni atoms [28]. The 51331002, and 51501004), the Beijing Municipality
high average P–Ni coordination number (8 and 10) in Natural Science Foundation (No. 2142018), and the
Ni12P5 results in the metallic behavior, which might Fundamental Research Funds for the Central
also explain the improved OER efficiency. Universities (No. FRF-BR-15-023A). We would like to
The covalent bonding nature in Ni12P5 was calculated thank Prof. Feng Yuanping at Centre for Advanced
2D Materials and Graphene Research of NUS for
using first-principles calculations using Vienna Ab
discussions on computing results. We also thank
initio Simulation Package (VASP) [37, 38]. As supported
the National Supercomputing Centre Singapore for
by the Bader charge analysis [39, 40], there are about
providing the computing resource.
0.16 electrons transferred from each surface Ni atom to
neighboring P atoms, suggesting a dominant covalent Electronic Supplementary Material: Supplementary
bond character. It is also notable that the variation of material (the details of the sample preparation and the
electrostatic potential in Ni12P5 surface is within 1 V, calculation methods, the CV characteristic curve and
as shown in Fig. 5(b). In principle, regions with positive extra TEM images) is available in the online version
electrostatic potential may become hydride acceptors, of this article at http://dx.doi.org/10.1007/s12274-017-
which are usually the sites of Ni atoms. This weak 1527-1.
electrostatic potential distribution and the covalent
bonding nature of Ni-P might explain the relatively
References
weaker adsorption sites at the Ni12P5 surface, and
therefore the improved OER activity. It is also noted [1] Gray, H. B. Powering the planet with solar fuel. Nat. Chem.
that the NiO compound has been proposed as a 2009, 1, 7.

| www.editorialmanager.com/nare/default.asp
Nano Res. 2017, 10(9): 3103–3112 3111

[2] Kanan, M. W.; Nocera, D. G. In situ formation of an [15] Oyama, S. T. Novel catalysts for advanced hydroprocessing:
oxygen-evolving catalyst in neutral water containing Transition metal phosphides. J. Catal. 2003, 216, 343–352.
phosphate and Co2+. Science 2008, 321, 1072–1075. [16] Mi, K.; Ni, Y. H.; Hong, J. M. Solvent-controlled syntheses
[3] Singh, R. N.; Mishra, D.; Anindita; Sinha, A. S. K.; Singh, A. of Ni12P5 and Ni2P nanocrystals and photocatalytic property
Novel electrocatalysts for generating oxygen from alkaline comparison. J. Phys. Chem. Solids 2011, 72, 1452–1456.
water electrolysis. Electrochem. Commun. 2007, 9, 1369–1373. [17] Ni, Y. H.; Liao, K. M.; Li, J. In situ template route
[4] Ma, T. Y.; Dai, S.; Jaroniec, M.; Qiao, S. Z. Metal-organic for synthesis of porous Ni12P5 superstructures and their
framework derived hybrid Co3O4-carbon porous nanowire applications in environmental treatments. CrystEngComm
arrays as reversible oxygen evolution electrodes. J. Am. 2010, 12, 1568–1575.
Chem. Soc. 2014, 136, 13925–13931. [18] Liu, P.; Rodriguez, J. A. Catalysts for hydrogen evolution
[5] Cui, B.; Lin, H.; Li, J. B.; Li, X.; Yang, J.; Tao, J. Core-ring from the [NiFe] hydrogenase to the Ni2P(001) surface: The
structured NiCo2O4 nanoplatelets: Synthesis, characterization, importance of ensemble effect. J. Am. Chem. Soc. 2005,
and electrocatalytic applications. Adv. Funct. Mater. 2008, 127, 14871–14878.
18, 1440–1447. [19] Popczun, E. J.; McKone, J. R.; Read, C. G.; Biacchi, A. J.;
[6] Suntivich, J.; May, K. J.; Gasteiger, H. A.; Goodenough, Wiltrout, A. M.; Lewis, N. S.; Schaak, R. E. Nanostructured
J. B.; Shao-Horn, Y. A perovskite oxide optimized for
nickel phosphide as an electrocatalyst for the hydrogen
oxygen evolution catalysis from molecular orbital principles.
evolution reaction. J. Am. Chem. Soc. 2013, 135, 9267–9270.
Science 2011, 334, 1383–1385.
[20] Paseka, I. Hydrogen evolution reaction on Ni-P alloys: The
[7] Mattos-Costa, F. I.; de Lima-Neto, P.; Machado, S. A. S.;
internal stress and the activities of electrodes. Electrochim.
Avaca, L. A. Characterisation of surfaces modified by sol-gel
Acta 2008, 53, 4537–4543.
derived RuxIr1–xO2 coatings for oxygen evolution in acid
[21] Huang, Z. P.; Chen, Z. B.; Chen, Z. Z.; Lv, C. C.; Meng, H.;
medium. Electrochim. Acta 1998, 44, 1515–1523.
Zhang, C. Ni12P5 nanoparticles as an efficient catalyst for
[8] Over, H. Surface chemistry of ruthenium dioxide in
hydrogen generation via electrolysis and photoelectrolysis.
heterogeneous catalysis and electrocatalysis: From fundamental
ACS Nano 2014, 8, 8121–8129.
to applied research. Chem. Rev. 2012, 112, 3356–3426.
[22] Zhang, G.; Wang, G. C.; Liu, Y.; Liu, H. J.; Qu, J. H.; Li, J. H.
[9] Zhou, W. J.; Wu, X. J.; Cao, X. H.; Huang, X.; Tan, C. L.;
Highly active and stable catalysts of phytic acid-derivative
Tian, J.; Liu, H.; Wang, J. Y.; Zhang, H. Ni3S2 nanorods/Ni
transition metal phosphides for full water splitting. J. Am.
foam composite electrode with low overpotential for
Chem. Soc. 2016, 138, 14686–14693.
electrocatalytic oxygen evolution. Energy Environ. Sci. 2013,
6, 2921–2924. [23] Yeo, B. S.; Bell, A. T. Enhanced activity of gold-supported
[10] Zhang, Y. Q.; Ouyang, B.; Xu, J.; Jia, G. C.; Chen, S.; cobalt oxide for the electrochemical evolution of oxygen.
Rawat, R. S.; Fan, H. J. Rapid synthesis of cobalt nitride J. Am. Chem. Soc. 2011, 133, 5587–5593.
nanowires: Highly efficient and low-cost catalysts for oxygen [24] Wu, J. B.; Li, P. P.; Pan, Y. T.; Warren, S.; Yin, X.; Yang, H.
evolution. Angew. Chem., Int. Ed. 2016, 55, 8670–8674. Surface lattice-engineered bimetallic nanoparticles and their
[11] Cao, X. Y.; Wang, H. N.; Lu, S. F.; Chen, S.; Xiang, Y. catalytic properties. Chem. Soc. Rev. 2012, 41, 8066–8074.
An Ni–P/C electro-catalyst with improved activity for [25] Zhang, J. T.; Tang, Y.; Lee, K.; Ouyang, M. Nonepitaxial
the carbohydrazide oxidation reaction. RSC Adv. 2016, 6, growth of hybrid core–shell nanostructures with large lattice
91956–91959. mismatches. Science 2010, 327, 1634–1638.
[12] Wang, H. N.; Cao, D.; Xiang, Y.; Liang, D. W.; Lu, S. F. [26] Duan, S. B.; Wang, R. M. Au/Ni12P5 core/shell nanocrystals
Novel Pd-decorated amorphous Ni–B/C catalysts with from bimetallic heterostructures: In situ synthesis, evolution
enhanced oxygen reduction reaction activities in alkaline and supercapacitor properties. NPG Asia Mater. 2014, 6,
media. RSC Adv. 2014, 4, 51126–51132. e122.
[13] Surendranath, Y.; Kanan, M. W.; Nocera, D. G. Mechanistic [27] Carenco, S.; Portehault, D.; Boissière, C.; Mézailles, N.;
studies of the oxygen evolution reaction by a cobalt- Sanchez, C. 25th anniversary article: Exploring nanoscaled
phosphate catalyst at neutral pH. J. Am. Chem. Soc. 2010, matter from speciation to phase diagrams: Metal phosphide
132, 16501–16509. nanoparticles as a case of study. Adv. Mater. 2014, 26,
[14] Song, H.; Dai, M.; Song, H. L.; Wan, X.; Xu, X. W.; Jin, Z. 371–389.
S. A solution-phase synthesis of supported Ni2P catalysts [28] Franke, R. X-ray absorption and photoelectron spectroscopy
with high activity for hydrodesulfurization of dibenzothiophene. investigation of binary nickel phosphides. Spectrochim.
J. Mol. Catal. A-Chem. 2014, 385, 149–159. Acta A-Mol. Biomol. Spectrosc. 1997, 53, 933–941.

www.theNanoResearch.com∣www.Springer.com/journal/12274 | Nano Research

3112 Nano Res. 2017, 10(9): 3103–3112

[29] Wu, S. K.; Lai, P. C.; Lin, Y. C. Atmospheric hydro- on IrO2-based DSA® type electrodes: Kinetics analysis
deoxygenation of guaiacol over nickel phosphide catalysts: of Tafel lines and EIS. Int. J. Hydrogen Energy 2004, 29,
Effect of phosphorus composition. Catal. Lett. 2014, 144, 791–797.
878–889. [36] Hammer, B.; Nørskov, J. K. Theoretical surface science and
[30] Franke, R.; Chassé, T.; Streubel, P.; Meisel, A. Auger catalysis—Calculations and concepts. Adv. Catal. 2000, 45,
parameters and relaxation energies of phosphorus in solid 71–129.
compounds. J. Electron Spectrosc. Relat. Phenom. 1991, 56, [37] Kresse, G.; Furthmüller, J. Efficiency of ab-initio total energy
381–388. calculations for metals and semiconductors using a plane-
[31] Liang, Y. Y.; Li, Y. G.; Wang, H. L.; Zhou, J. G.; Wang, J.; wave basis set. Comput. Mater. Sci. 1996, 6, 15–50.
Regier, T.; Dai, H. J. Co3O4 nanocrystals on graphene as [38] Kresse, G.; Furthmüller, J. Efficient iterative schemes for
a synergistic catalyst for oxygen reduction reaction. Nat. ab initio total-energy calculations using a plane-wave basis
Mater. 2011, 10, 780–786. set. Phys. Rev. B 1996, 54, 11169–11186.
[32] Su, X. R.; Xu, Y. Y.; Liu, J. L.; Wang, R. M. Controlled [39] Yu, M.; Trinkle, D. R. Accurate and efficient algorithm for
synthesis of Ni0.25Co0.75(OH)2 nanoplates and their electro- Bader charge integration. J. Chem. Phys. 2011, 134, 064111.
chemical properties. CrystEngComm 2015, 17, 4859–4864. [40] Tang, W.; Sanville, E.; Henkelman, G. A grid-based Bader
[33] Rossmeisl, J.; Qu, Z. W.; Zhu, H.; Kroes, G. J.; Nørskov, J. K. analysis algorithm without lattice bias. J. Phys.: Condens.
Electrolysis of water on oxide surfaces. J. Electroanal. Matter 2009, 21, 084204.
Chem. 2007, 607, 83–89. [41] Man, I. C.; Su, H. Y.; Calle-Vallejo, F.; Hansen, H. A.;
[34] Rossmeisl, J.; Logadottir, A.; Nørskov, J. K. Electrolysis of Martínez, J. I.; Inoglu, N. G.; Kitchin, J.; Jaramillo, T. F.;
water on (oxidized) metal surfaces. Chem. Phys. 2005, 319, Nørskov, J. K.; Rossmeisl, J. Universality in oxygen evolution
178–184. electrocatalysis on oxide surfaces. ChemCatChem 2011, 3,
[35] Hu, J. M.; Zhang, J. Q.; Cao, C. N. Oxygen evolution reaction 1159–1165.

| www.editorialmanager.com/nare/default.asp