Introduction

Nowadays, the increasing use of fossil fuels has not only brought a shortage of fossil fuels as a non-renewable resource but also exposed people to several serious problems, such as greenhouse gas emissions and environmental degradation [1, 2]. Therefore, alternative and clean energy sources are being urgently sought. As a clean energy source with high energy density and non-polluting combustion, hydrogen energy has attracted increasing attention and studies [3,4,5]. Compared with other mainstream steam methane reforming and coal gasification for hydrogen production [6, 7], electrocatalytic water splitting is more favored as it requires only eco-friendly electricity and sufficient water supply. In addition, the oxygen evolution reaction (OER) and hydrogen evolution reaction (HER) occur separately at the anode and cathode, ensuring the cleanliness of the whole process and the products [8, 9]. However, due to the superpotential of HER at the cathode consumes more electrical energy, many scholars focus on the preparation of various functional cathodes to decrease the superpotential. As a well-known fact, platinum group metals and their alloys or metal oxides are excellent catalysts for HER, but their widespread applications for industries are limited by low element abundance and high extraction costs [10, 11]. Therefore, it is highly urgent to develop high-activity and continuously durable non-precious metal electrocatalysts for HER.

Up to now, various studies for HER have been conducted on this aim [12]. For instance, transition metal and their alloys [13, 14], transition metal phosphides [15], carbides [16], nitrides [17], and sulfides [18] have been studied as effective electrocatalysts by different modification means. Especially, benefiting from the wide distribution and high abundance of nickel (Ni) and cobalt (Co) elements on the geological surface, Ni-based and Co-based electrocatalysts show excellent catalytic performance and stability in alkaline solutions, which have been extensively investigated in recent years and have been expected to replace precious metals to achieve efficient HER [19,20,21]. And the electronic structure and d-spectrum generation centers of transition metal-based catalysts can be modulated by non-metallic intercalation, which provides a realizable and effective tactic to enhance the activity of electrocatalysts [22]. In addition, metal oxides with spinel structures are considered as promising electrocatalysts for water splitting because of their good physicochemical properties and electrocatalytic activity [23]. On the other hand, excellent electrocatalysts would be combined with high-conductive substrates for sustained water splitting. With the merits of high conductivity, superb specific surface area, and favorable HER behavior, three-dimensional porous nickel foam (NF) is widely chosen as the conductive substrate. For instance, Wang et al. [24] successfully manufactured self-supported electrocatalysts of NF with P-modified Co3O4 nanowire arrays which were adjusted with different P doping ratios. And the work revealed that P-doping can change the electronic structure of catalysts, showing excellent performance in overall water splitting. Cheng et al. [25] fabricated P-NiCo2O4/NF electrocatalyst via a facile hydrothermal and a subsequent phosphatizing treatment, which demonstrated that P incorporation helped to improve HER kinetics and enhance HER activity.

Considering the production of hydrogen in industrial water splitting, the migration of cations from the solution during HER will move to the cathode. Part of them reacts with the rich hydroxyl groups on the cathode surface to generate precipitates that wrap the active sites on the porous substrate, making the electrocatalyst prone to toxicity and ineffectiveness. The three-dimensional porous NF matrix undoubtedly accelerates the above reaction, which is contrary to our original intention. In addition, ordinary NF exists drawbacks such as easy deformation under pressure and brittle after annealing, so it is only suitable for limited mild modification means [26]. Therefore, a self-supporting matrix electrocatalyst conforming to the characteristics of large pore size and rigidity is urgently needed for the HER process in actual water splitting to achieve a win–win situation of hydrogen energy supply and electrode consumption reduction in an alkaline medium. Thus, we chose NM as a substrate to face HER performed in complex electrolytes.

Inspired by the above ideas and combined with the practical needs of water splitting, we designed a Ni-Co-P nanosheet electrocatalyst grown at macroporous NM for HER in an alkaline medium by hydrothermal method and phosphorylation. With promising HER electrocatalytic performance in 1 M KOH electrolyte, NCP/NM was able to achieve 10 mA cm−2 with only 126 mV overpotential. Continuous constant voltage electrolysis and multi-step constant current electrolysis confirmed the promising durability and applicability of the electrode. The related physicochemical analysis and characterization showed little change in the micromorphological structure and physical phase after long-term working, corroborating the favorable structural stability of the electrolysis. NCP/NM prepared by the above strategy provided an inexpensive, high electrocatalytic performance, and durable alternative cathode for hydrogen production in an alkaline medium.

Experimental

Materials

Nickel chloride hexahydrate (NiCl2·6H2O), cobalt chloride hexahydrate (CoCl2·6H2O), sodium hypophosphite monohydrate (NaH2PO2·H2O), and commercial Pt/C (10 wt%) were bought from Aladdin Reagent (Shanghai, China). Ammonium hydroxide (NH3·H2O, 25–28 wt%), hydrochloric acid (HCl), and potassium hydroxide (KOH) were supplied via Sinopharm Chemical Reagent Co., Ltd. All chemical reagents were of the analytical variety and were employed without purifying. Bare nickel mesh with diamond-shaped holes (b-NM) and commercial nickel mesh with diamond-shaped holes (c-NM) were purchased from Hongze (Jiangsu) Technology Co., Ltd.

Synthesis of NC/NM, NCO/NM, and NCP/NM

To remove surface oxides and contaminants, b-NM (1 cm × 2 cm × 0.1 cm) was ultrasonically bathed in 4 wt% HCl solution and acetone, each for 20 min. All samples were washed with deionized water before that being immediately transferred for the next procedure. The 22 mL of deionized water was mixed with 0.1 mol each of NiCl2·6H2O and CoCl2·6H2O to dissolve them and stirred the mixture for a while. A total of 8 mL NH3·H2O as a complexing medium was added before the mixture was sonicated, forming a homogeneous solution. Subsequently, a 50 mL Teflon-lined stainless steel autoclave was loaded with the above solution and b-NM. After being maintained at 120 °C for 12 h, the hydrothermal reaction was left to spontaneously drop to ambient temperature. Later, a vacuum oven was used to dry the synthetic catalytic electrode overnight at 60 °C after it had been cleaned several times with deionized water. The above electrode loaded with an electrocatalyst was labeled as NC/NM. In the next step, NC/NM and 200 mg NaH2PO2·H2O were kept with quartz vessels in the downstream and upstream of a tube furnace, separately. The target electrodes were obtained by heating to 350 °C under a nitrogen atmosphere and holding for 3 h and then letting the temperature drop naturally to ambient temperature. The produced catalytic electrodes were cleaned with deionized water and ethanol several times before being vacuum-dried at 60 °C for the duration of the night and labeled as NCP/NM. To compare the effect of phosphorylation or not on the catalytic electrodes, contrasted electrodes were fabricated under the same requirements without the addition of NaH2PO2·H2O. The products obtained under the above conditions were labeled as NCO/NM. The loading mass of Ni-Co-P nanosheets on NM was calculated as about 5.0 mg cm−2 from the weight change. The loadings for NC/NM and NCO/NM were about 4.7 mg cm−2 and 5.0 mg cm−2, respectively.

For further comparison, commercial Pt/C was loaded on an NM electrode. In detail, 10 mg Pt/C was dispersed in a solution mixed with 250 μL isopropanol, 700 μL deionized water, and 50 μL 5% Nafion. The above mixture was sonicated for 60 min to form a homogeneous ink. Afterwards, all ink was evenly applied to NM and allowed to dry naturally to obtain Pt/C/NM.

Characterizations

An X’Pert Pro MPD diffractometer was utilized to obtain X-ray diffraction (XRD) data by using a Cu target (λ = 0.154 nm) with a scan range from 5 to 80°. Al Kα radiation was applied to conduct X-ray photoelectron spectroscopy (XPS) on a Thermo ESCALAB250Xi instrument. The scanning electron microscopy (SEM, Gemini 500) and the equipped energy dispersive spectroscopy (EDS, Aztec UltimMax 100) were used to investigate the structure and morphology of the produced catalytic electrodes. Transmission electron microscope (TEM) and high-resolution TEM (HRTEM) measurements were further analyzed for the morphology and phase structure of the electrocatalysts which were ultrasonically stripped.

Electrochemical measurements

To minimize the effect of capillary phenomena, all samples were encapsulated with epoxy resin before electrochemical testing to achieve a 1 cm × 1 cm conductive section in solution. The CHI760E electrochemical workstation served as a controller for electrochemical tests in a typical three-electrode system to collect relevant electrochemical data. The measured samples, graphite rod, and Hg/HgO electrode were employed as the working electrode, counter electrode, and reference electrode in the system, respectively. Simultaneously, 1 M KOH solution saturated with nitrogen was used as the electrolyte. For direct comparison of standardized potentials, the Nernst equation was used to translate each test potential reported in this investigation to the reversible hydrogen electrode (RHE) [27]: \({{E}}\left({{RHE}}\right) \, {=} \, {{E}}\left({Hg/HgO}\right) \, {+} \, {0.098} \, {+} \, {0.059pH}\). With a scan speed of 5 mV/s, all samples were tested for their HER performance using linear sweep voltammetry (LSV), and data were collected by 100% manual iR correction. Tafel plots were obtained by fitting the data of LSV to the Tafel equation. By using cyclic voltammetry (CV) measurements in the non-Faraday current zone with scan rates of 20–80 mV/s, the electrochemical double-layer capacitances (Cdl) of the catalytic electrodes were evaluated. With an amplitude of 5 mV and an overpotential of 200 mV, electrochemical impedance spectroscopy (EIS) was conducted in the frequency range of 100 kHz to 0.1 Hz. Continuous electrolysis was implemented by chronoamperometry for 10 h to assess the durability of the optimal electrocatalyst. In addition, multi-step chronopotentiometry was used to evaluate electrocatalytic applicability and stability at different current densities.

Results and discussion

Synthesis principle

Figure 1 provides a clear illustration of the proposed method for synthesizing catalytic electrodes. During the initial phase of the hydrothermal reaction, unlike the typical synthesis routes of NiCo2O4 loading on a substrate [28,29,30], ammonia solution was used to replace urea and ammonium fluoride as a solvent for the hydrothermal reaction to obtain the electrocatalyst precursor, because it provided ammonium and hydroxyl groups that complex and generate metal hydroxides with metals. In the above step, the main reaction equation at the electrode could be illustrated below:

Fig. 1
figure 1

Schematic illustration of the synthesis steps of catalytic electrodes

Full size image
$${{Ni}}^{2+}+ {4} {{Co}}^{2+}+ {12} {{OH}}^{-}{+}{{O}}_{2}\to {{2Ni}}{{Co}}_{2}{{{O}}}_{4}+ {6} {{H}}_{2}{{O}}$$
(1)
$${{Co}}^{2+}+ {2} {{OH}}^{-}\to {{Co(OH)}}_{2}$$
(2)

The precursor was subsequently phosphorylated by chemical vapor deposition (CVD), while the metal hydroxide on the precursor was subjected to heat to produce metal oxides. After natural cooling, the hue of the electrode surface changed from rosy color to black (Fig. S1). The possible reaction on the electrode at this step was the following equation:

$${6}{{Co(OH)}}_{2}{+}{{O}}_{2}\to {2}{{Co}}_{3}{{{O}}}_{4}+ {6} {{H}}_{2}{{O}}$$
(3)

Physicochemical properties

To detect and analyze the phase composition and crystal structure of various catalytic electrodes, XRD technology was performed. In Fig. S2, the (111), (200), and (220) planes of Ni (ICDD:00–004-0850) can be matched exactly to the peaks of b-NM appearing at 2θ = 44.5°, 51.8°, and 76.4°, respectively. After the hydrothermal reaction process, the (001), (100), and (011) planes of Co(OH)2 (ICDD: 01–074-1057) can be precisely correlated to the strong diffraction peaks which appeared at 19.05°, 32.47°, and 37.95°, as illustrated in Fig. 2 (black). Simultaneously, 2θ = 31.2°, 36.7°, and 65.0° of the other main peaks are corresponded to (220), (311), and (440) planes of NiCo2O4 phase (ICDD: 01–073-1702). This indicates that a mixture of Co(OH)2 and NiCo2O4 can be obtained by hydrothermal reaction. The curve of phosphorylated NCP/NM displays three diffraction peaks at 18.9°, 31.2°, and 36.7°, which can be ascribed to the (111), (220), and (311) planes of NiCo2O4 or Co3O4 (ICDD: 01–071-4921), since the diffraction peaks of NiCo2O4 and Co3O4 at diffraction angle above are quite close [31]. The XRD pattern of NCO/NM is also the same as above. It can be observed that the intensity of the dominant diffraction peaks of NCP/NM and NCO/NM are diminished when compared with that of NC/NM, demonstrating a low degree of crystallization in the material after a period of calcination.

Fig. 2
figure 2

XRD patterns of NC/NM, NCO/NM, and NCP/NM

Full size image

As shown in Fig. 3a–d, the SEM technique was employed to detect the surface morphology characteristics of electrocatalysts. The SEM image of b-NM (Fig. 3a) shows concave and uneven structures, and there are no obvious external objects on the surface. Apparently, a number of rough microspheres (particle size of ~ 300–400 nm) are grown on b-NM after hydrothermal reaction, as shown in Fig. 3b. The microstructure of NCO/NM catalytic electrode is exhibited in Fig. 3c. When compared to the sample without calcination treatment, it is obvious that the micromorphological structure of the small spheres dispersed on the substrate is no significant change except for a reduction in particle size (particle size of ~ 200–300 nm). Interestingly, after phosphorylation by the CVD process, the surface structures of NC/NM collapse significantly and the microspheres on the substrate become nanosheet-like structures (Fig. 3d). This increases the number of active sites for hydrogen precipitation, undoubtedly. Meanwhile, to characterize the elemental composition of the NCP/NM electrode surface, an EDS analysis was conducted. Figure 3e presents Ni, Co, P, and O elemental mapping images, demonstrating their homogeneous distribution across the substrate surface. In addition, the effective synthesis of NCP/NM is validated by the elemental composition of Ni, Co, P, and O, which can be seen in Fig. S3. TEM analyses were conducted to further observe the microstructure and chemical composition of NCP/NM. Figure 4a distinctly displays several irregular nanosheets, which is consistent with the SEM image. In Fig. 4b, the lattice fringes of d = 0.24 nm were clearly depicted by the HRTEM technique, which is in accordance with the distances of the (311) planes of NiCo2O4 or Co3O4, implying metal oxides occupy the major components of electrocatalysts. Figure 4c confirms that Ni, Co, P, and O are evenly dispersed in nanosheets, which is in line with the result of SEM (Fig. 3e), further revealing their uniform distribution in the whole electrocatalyst.

Fig. 3
figure 3

SEM images of the a b-NM, b NC/NM, c NCO/NM, and d NCP/NM; e EDS mappings of NCP/NM catalytic electrode

Full size image
Fig. 4
figure 4

a TEM image, b HRTEM image, and c elemental mapping of NCP/NM

Full size image

XPS technique, as an effective method to investigate the chemical properties of material surfaces, was adopted to further analyze the chemical composition and elemental valence of fabricated electrodes surface. As depicted in Fig. 5a, XPS survey spectra of three electrodes reveal the presence of elements Ni, Co, C, and O on the electrode’s surface. Here, carbon indicates calibration and oxygen derives from the synthesized reagent, and the partial oxidation of the sample surface may also contribute to the presence of oxygen. Notably, the curve of NCP/NM additionally shows the presence of element P. Figure 5b–d is the high-resolution XPS spectra for electrodes. The Ni 2p spectrum of NCP/NM represents the binding energy peaks at 856.5 eV and 873.75 eV associated with Ni 2p3/2 and Ni 2p1/2, respectively [32]. Meanwhile, there is a 17.25 eV difference in their binding energies, which further proves the existence of Ni2+. Also, two extra binding energy peaks can be ascribed to the shake-up satellite peaks of Ni 2p3/2 and Ni 2p1/2, respectively, with positioned at 862.1 eV and 881.05 eV. As for the Co 2p spectrum of NCP/NM, the appearance of two main peaks at 781.45 eV and 797.7 eV can be assigned to Co 2p3/2 and Co 2p1/2, correspondingly [33]. Specifically, the appearance of peaks at the binding energy of 781.05 eV and 796.95 eV can be attributed to Co3+ cation. Similarly, it is reasonable to attribute the peaks appearing at 782.95 eV and 798.4 eV to the Co2+ cation. This demonstrates the coexistence of Co3+ and Co2+ on the electrode surface [34, 35]. The residual peaks at 786.7 eV and 803.4 eV can be explained by the satellite peaks arising from high spin Co shake-up excitation. In Fig. 5d, two dominant peaks at 129.75 eV and 133.65 eV are portrayed at the XPS spectrum of P 2p for NCP/NM, compared with NC/NM and NCO/NM, which correspond to metal phosphides (P-M) on the electrode surface and P-O bond of phosphates [36, 37], respectively. It reflects the successful doping of P onto the catalytic electrode surface after phosphorylation as well as its presence in two forms. Moreover, note that the peaks of Ni 2p3/2 and Co 2p3/2 for NCP/NM show shifts of higher binding energies for 0.62 eV and 0.67 eV compared with untreated NC/NM; however, the peaks of NCO/NM show no significant shift. The result indicates a decrease in the electron density and an increase in electronegativity of Ni and Co species after phosphorylation, which is expected to enhance the electrocatalytic activity [38].

Fig. 5
figure 5

a XPS survey spectra of NC/NM, NCO/NM, and NCP/NM; High resolution XPS spectra of b Ni 2p, c Co 2p and d P 2p for fabricated electrodes

Full size image

HER performance

The electrocatalytic activities of the relevant electrodes were respectively evaluated in 1 M KOH. Figure 6a shows the LSV curves with 100% manual iR compensation of all samples. The sharp increasing current density for HER can be clearly seen at NCP/NM with the increasing driving voltage, indicating a favorable kinetic and thermodynamic activity [39]. To quantify their catalytic activities, the driving overpotentials required at current densities of 10 and 100 mA cm−2 are listed in Fig. 6b. The required overpotentials are 73 mV (Pt/C/NM), 126 mV (NCP/NM), 147 mV (c-NM), 269 mV (NCO/NM), 309 mV (NC/NM), and 345 mV (b-NM) at 10 mA cm−2, respectively. Similarly, the overpotentials of the above electrodes are 230, 208, 281, 353, 421, and 471 mV for 100 mA cm−2. From above the facts, except for Pt/C/NM, NCP/NM shows the lowest overpotentials for HER in an alkaline medium, which is competitive with some non-precious metal-supported electrodes reported in recent years (Table S1). It should be pointed out that the catalytic activity of NCP/NM at higher current densities even surpasses that of Pt/C/NM.

Fig. 6
figure 6

HER performance of different electrodes in 1 M KOH: a IR corrected LSV curves; b Required overpotential at cathode current densities of 10 and 100 mA cm−2; c Tafel plots; d Curves of capacitance current versus scan rate

Full size image

It is widely acknowledged that the Tafel slope is an essential parameter to understand the HER kinetic process. The Tafel slope is fitted by the LSV data with 100% manual iR compensation via the following Tafel equation [40]:

$$\upeta =\mathrm{a}+\mathrm{blog}\left|\mathrm{j}\right|$$
(4)

where, respectively, η, b, and j stand for overpotential, Tafel slope, and current density. The presented results are well-fitted linear segment in Fig. 6c. As expected, except for Pt/C/NM, NCP/NM exhibits the lowest Tafel slope (87.62 mV dec−1) compared with that of c-NM (88.17 mV dec−1), NCO/NM (93.08 mV dec−1), NC/NM (93.65 mV dec−1), and b-NM (94.22 mV dec−1), demonstrating the fastest charge transfer processes and kinetic processes in 1 M KOH. Generally, in an alkaline medium, the HER mechanism may involve two processes (Volmer-Heyrosky and Volmer-Tafel mechanism) by the following three basic steps [41]:

$${M} + {{H}}_{2}{O} + {{e}}^{-}\to {{M}}{{H}}_{{ads}}{+}{{OH}}^{-}{(}{{Volmer}}{)}$$
(5)
$${{M}}{{H}}_{{ads}}{+}{{H}}_{2}{O} + {{e}}^{-}\to {M} + {{H}}_{2}{+}{{OH}}^{-}{(}{{Heyrosky}}{)}$$
(6)
$${{M}}{{H}}_{{ads}}+ {M} {{H}}_{{ads}}\to {2M} + {{H}}_{2}{(}{{Tafel}}{)}$$
(7)

Correspondingly, the theoretical Tafel slopes in the above steps are 120, 40, and 30 mV dec−1, which are utilized to investigate the kinetic mechanism and rate control steps of the HER process. Because the values are situated in a region of 87 to 95 mV dec−1, HER reaction processes follow a classical Volmer-Heyrosky mechanism. Furthermore, it can be emphasized that the rate-determining step is the Volmer step.

Admittedly, improving the electrocatalytic performance in HER processes can be accomplished via increasing the active sites and the catalytic capacity at the electrode surface. Owing to the linear relationship between the electrochemical active surface area (ECSA) and Cdl, CV tests with different sweep rates in the non-Faraday interval are implemented to estimate the Cdl value by the following equation [42]:

$$\Delta j={2\mathrm{v}C}_{dl}$$
(8)

where Δj denotes the corresponding current density difference at the intermediate potential. The CV plots of different samples at different sweep speeds are shown in Fig. S4a–e. On this basis, the calculated results of Δj against ν are plotted in Fig. 6d. The calculated Cdl value of NCP/NM (15.28 mF cm−2) is higher than other electrodes, reflecting the fact that the sheet electrocatalytic substances which were presented on the electrode surface provided more sites for reaction with the electrolyte [43]. Unexpectedly, the Cdl value of NCO/NM (0.38 mF cm−2) lays between b-NM (0.27 mF cm−2) and NC/NM (0.55 mF cm−2). Corresponding to their SEM images, the situation can be related to the structural collapse of the electrocatalyst after calcination for a certain period, and the particle size of active material on the surface becomes smaller, resulting in the reduction of ECSA. In Fig. S4f, although the value of NC/NM is over half higher than that of NCO/NM, the current density differences are all lower than NCO/NM. Under the same circumstances, the current response of NCO/NM is more active than that of NC/NM, revealing the difference in their electrocatalytic ability. Besides that, when the current density was normalized to ECSA, as shown in Fig. S5, NCP/NM also exhibits a satisfactory intrinsic activity compared with NC/NM and NCO/NM, suggesting the good electrical conductivity and fast charge transfer kinetics of the intrinsic active species.

As we know, electrochemical impedance spectroscopy (EIS) is used to analyze the charge transfer kinetics during HER. Figure 7a depicts the Nyquist plots of relative electrodes at 200 mV (vs. RHE) overpotential. The insets show the corresponding equivalent electric circuit (EEC) and the partial enlargement Nyquist plots. A parallel unit consists of a constant phase element (CPE) and a charge transfer resistor (Rct), which is connected in series after the solution resistor (Rs). Although the Rs values of all the electrodes are in a low range of 1–2 Ω due to their favorable electrical contact, there are significant differences in their Rct values. For example, among these electrodes, the Rct value for NCP/NM (1.3 Ω) is the lowest compared with that of NCO/NM (36.4 Ω), NC/NM (63.5 Ω), b-NM (170.4 Ω), and c-NM (2.1 Ω). It is illustrated that the charge transfer kinetics at NCP/NM is significantly improved by the combination of hydrothermal and phosphorylation reactions, suggesting a faster Faraday behavior. NCP/NM is comparable with the commercial c-NM electrode. Furthermore, this is in accordance with the measured LSV of the electrodes. NCP/NM absolutely shows a positive electrocatalytic effect on HER. Therefore, the present NCP/NM was used in the following experiments.

Fig. 7
figure 7

a Nyquist plots of all samples at an overpotential of 200 mV, the bottom inset presents the equivalent electric circuit. b Polarization curves for NCP/NM before and after 10 h electrolysis. c Chronoamperometry curve of NCP/NM conducted at an overpotential of 130 mV. d Multi-step chronopotentiometry curve of NCP/NM from 10 mA cm−2 to 100 mA cm−2

Full size image

The stability of NCP/NM was investigated for HER by the chronoamperometry method at an overpotential of 130 mV with 10 h in 1 M KOH. It is readily seen that the LSV curves at NCP/NM hardly change even after continuous electrolysis of 10 h, as provided in Fig. 7b, suggesting a promising stability. The result could also be proved in Fig. 7c, relatively stable current density could be seen at NCP/NM during 10 h, although a slight degradation of the current density was observable towards the end of the test, which could be mainly attributed to the slight shedding of electrocatalyst when hydrogen escapes at the cathode. The stability of NCP/NM was also investigated for HER by the multi-step chronopotentiometry method at a density range from 10 to 100 mA cm−2 in 1 M KOH. As shown in Fig. 7d, the potential curve at NCP/NM remains good stability in every interval with different current densities, which indicates that the electrode has good electrocatalytic stability in the HER process at different current densities.

After a continuous electrolysis reaction of 10 h, the electrode was characterized by XRD, SEM, TEM, and XPS techniques to further investigate its structural and morphological stability. The XRD curves display no significant variation in the physical phase after a long stability test as shown in Fig. S6a. The SEM image shows the same nanosheet structure as that of the original NCP/NM in Fig. S6b. Apparently, nanosheet electrocatalysts can be observed in the TEM image (Fig. S6c), which has no significant change compared with the initial sample. Besides, the survey spectrum in Fig. S7a confirms the perseverance of Ni, Co, and P signals after the long-term HER test. High-resolution XPS spectra result in Fig. S7c–d show that the dominant peak positions of Ni 2p, Co 2p, and P 2p for tested NCP/NM are nearly the same as those of the sample before the long-term stability test although the peak intensity changed to a certain extent, proving no change in the species distribution. Therefore, all these demonstrate the desirable stability of the synthesized electrode in terms of crystal structure and micromorphology.

Conclusions

In summary, macroporous NM was used as the substrate, and the NCP/NM electrode was successfully synthesized through a simple two-step process by hydrothermal reaction with ammonia as the solvent and followed by phosphorylation. According to electrochemical characterizations, NCP/NM electrode provided competitive HER performance in 1 M KOH, requiring an overpotential of only 126 mV to produce a current density of 10 mA cm−2. The electrode also exhibited good suitability and stability in long-term electrolysis. These pleasant electrocatalytic properties can be assigned to fast electron transfer between the electrocatalyst and NM, the synergistic effect of Ni and Co, the positive effect of P-element doping, and the high ECSA. Owing to the advantages of large pore size, rigid material, simple preparation method, and high electrocatalytic activity and durability of NCP/NM, the prepared electrode provides an attractive cathode for hydrogen production in alkaline medium and has potential applications for HER in large-scale industry.