The Preparation and Electrochemical Pseudocapacitive Performance of Mutual Nickel Phosphide Heterostructures

Abstract

Transition metal phosphide composite materials have become an excellent choice for use in supercapacitor electrodes due to their excellent conductivity and good catalytic activity. In our study, a series of nickel phosphide heterostructure composites was prepared using a temperature-programmed phosphating method, and their electrochemical performance was tested in 2 mol L⁻¹ KOH electrolyte. Because the interface effect can increase the catalytic active sites and improve the ion transmission, the prepared Ni₂P/Ni₃P/Ni (Ni/P = 7:3) had a specific capacity of 321 mAh g⁻¹ under 1 A g⁻¹ and the prepared Ni₂P/Ni₅P₄ (Ni/P = 5:4) had a specific capacity of 218 mAh g⁻¹ under 1 A g⁻¹. After the current density was increased from 0.5 A g⁻¹ to 5 A g⁻¹, 76% of the specific capacity was maintained. After 7000 cycles, the capacity retention rate was above 82%. Due to the phase recombination effect, the electrochemical performance of Ni₂P/Ni₃P/Ni and Ni₂P/Ni₅P₄ was much better than that of single-phase N₂P. After assembling the prepared composite and activated carbon into a supercapacitor, the Ni₂P/Ni₃P/Ni//AC had an energy density of 22 W h kg⁻¹ and a power density of 800 W kg⁻¹ and the Ni₂P/Ni₅P₄//AC had an energy density of 27 W h kg⁻¹ and a power density of 800 W kg⁻¹.

1. Introduction

With the rapid development of human society, the demand for energy has increased substantially. At the same time, the energy crisis is becoming increasingly more serious. In the future, exploring and developing clean and renewable energy such as wind and hydropower will become the mainstream [1,2,3,4,5,6]. However, due to the instability of power generation, this type of energy has not yet fully met the demand [7,8]. Through electrochemical energy storage devices, the electricity generated from these renewable energy sources can be stored for effective use [9,10]. Presently, there are two main types of commercial electrochemical energy storage devices: batteries, such as lithium-ion batteries [11,12], and supercapacitors, such as pseudocapacitance supercapacitors [13,14]. Supercapacitors have many advantages, such as higher power density, long cycle life, and fast charging and discharging, so they have attracted much attention [15,16,17]. However, there are still some challenges associated with supercapacitors, including how to increase their energy density while maintaining the above advantages [18,19].

Electrode material is the core factor of the supercapacitor system and affects the performance of the supercapacitor directly [20]. According to their characteristics, electrode materials can be divided into two categories [21]. The first is a variety of carbon materials including activated carbon, graphene, and carbon nanotubes; the energy storage mechanism is electrostatic adsorption and desorption [9]. These electrode materials provide good stability, porous characteristics, and good electrical conductivity [22], but the relatively low energy density associated with this group limits its wide application [23]. The second category is pseudocapacitive materials including transition metal oxides and transition metal phosphides; the energy storage mechanism is rapid reversible redox or intercalation reactions on or near the electrode surface [24,25]. Compared to transition metal oxides, transition metal phosphides have better electrical conductivity and a lower cost. They also have a specific capacity comparable to or even beyond that of transition metal oxides [26,27]. This is due to some of the characteristics of transition metal phosphides.

Phosphorus atoms can enter the transition metal crystals to form intermetallics [28]. The presence of phosphorus atoms pulls the electron delocalization of metal phosphide, which improves its catalytic activity. The higher the metal content, the more free electrons it contains and the better it is for conducting electricity [29]. Therefore, this compound with metallic properties exhibits high electrical conductivity and specific capacity [30,31]. Nickel phosphide has been a good choice for electrode materials for supercapacitors because of its high electrical conductivity, fast charge transfer ability, good reaction kinetics, and abundant earth reserves [32,33,34]. Single-phase nickel phosphide is difficult to prepare, and its electrochemical performance is unsatisfactory [35]. Therefore, many researchers have shown great interest in preparing nickel phosphide composites and applying them to supercapacitors [36,37]. Nickel phosphide compounds can be prepared using available nickel chemical plating methods on a nickel phosphide surface coated with a layer of amorphous nickel or by mechanically mixing graphene and nickel phosphide. It is also possible to compound nickel phosphide with other compounds through chemical precipitation to make composite materials [38,39,40]. Other types of composites used in the study of the electrochemical performance of supercapacitors include nickel-cobalt oxide modified with reduced graphene oxide, ZnFe₂O₄ nanorods on reduced graphene oxide, NiCo₂O₄/Ni₂P, nitrogen-doped Ni₂P/Ni₁₂P₅/Ni₃S₂, and MoS₂–ReS₂/rGO [41,42,43,44,45]. Due to the synergistic effect between different components and the interfacial effect, the electrochemical performance of such materials is satisfactory. The preparation of composite nickel phosphide shows good electrochemical performance, which provides inspiration for exploring such electrode materials.

In our study, nickel phosphide composite was prepared in one step by temperature programmed phosphating. Since no polymer binder was added during the preparation of the electrode, the electrode had maximum conductivity and catalytic activity. We used 2 mol L⁻¹ KOH aqueous solution as the electrolyte for electrochemical performance testing. The best comprehensive electrochemical performance was observed when the stoichiometric ratio was 5:4. At this stoichiometric ratio, the nickel phosphide composite formed a Ni₂P/Ni₅P₄ heterostructure with a specific capacity of 218 mAh g⁻¹ and a rate performance of 76%. After 7000 cycles, the capacity retention rate was above 82%. After combining it with activated carbon to form an asymmetric supercapacitor, the energy density was 27 W h kg⁻¹ while the power density was 800 W kg⁻¹, demonstrating good electrochemical performance. At the stoichiometric ratio of 7:3, it formed a Ni₂P/Ni₃P/Ni heterostructure with a specific capacity of 321 mAh g⁻¹ and a rate performance of 59%. After combining it with activated carbon to form an asymmetric supercapacitor, the energy density was 22 W h kg⁻¹ while the power density was 800 W kg⁻¹.

2. Experimental Section

2.1. Chemicals

All chemicals used were of analytical grade. C₂H₅OH, KOH, H₂SO₄, and acetone were purchased from Kelon Chemicals Co. Ltd., Chengdu, China. Nickel foam was purchased from Shanghai (China) Hesen Electric Co. Ltd. Nickel powder and red phosphorus were from Aladdin.

2.2. Electrode Material Synthesis

The nickel foam was cut into small pieces, each with a length and width of 1 cm, and was then pretreated with dilute hydrochloric acid solution and acetone solution (V_acid:V_acetone = 6:1) to remove oxides and organic pollutants. The surface was then rinsed with anhydrous ethanol and deionized water. The metal nickel powder was pretreated with dilute sulfuric acid to remove oxides and organic pollutants, and then rinsed with anhydrous ethanol and deionized water several times. It was put into the oven at a constant temperature of 60 °C until dried. About 5 g of treated metal nickel powder for separated for use. Metal nickel powder and red phosphorus were mixed according to the stoichiometric amounts; a 1.5% excess of red phosphorous was required.

To make electrodes, slightly more than 4 mg of mixed powder was pressed on the treated nickel foam; the applied pressure was 10 MPa. The foam nickel electrode pads loaded with mixed powder were put into a porcelain boat along with the remaining mixed powder. The porcelain boat was then put into a tubular furnace, washed to vacuum with nitrogen, gradually heated to 700 °C at 4 °C min⁻¹, and held for 6 h. The intermediate temperatures were 350 °C, 450 °C, and 550 °C. Each intermediate temperature was held for 1 h. According to the stoichiometric amount, we recorded them as (3:1), (5:2), (12:5), (7:3), (2:1), (5:4), and (1:1). Single-phase Ni₂P can be made from stoichiometry (5:2).

2.3. Material Characterization

The crystalline structures were confirmed by an X-ray diffraction (XRD, Dandong DX-2700, Dandong, China) with Cu-Kα radiation (2θ = 5~80°) operating at 40 kV. The morphologies and microstructures of the samples were characterized by field-emission scanning electron microscope (SEM, JEOL JSM-6701 F, Tokyo, Japan) and transmission electron microscope (TEM, JEOL JEM2010, Tokyo, Japan). The pore properties and Brunauer-Emmett-Teller specific surface area were investigated via N₂ adsorption–desorption (NAD, ASAPR 2020, Atlanta, GA, USA) test at −196 °C. The surface bonding state was tested by X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, New York, NY, USA).

2.4. Electrochemical Evaluation

All electrochemical measurements were executed on the electrochemical station, and the electrolyte was 2 mol L⁻¹ aqueous KOH. Before the test, we soaked the undertested materials in 2 mol L⁻¹ of KOH electrolyte for 24 h to activate them. The mass of the active substance was obtained by weighing the nickel foam before and after the reaction. The mass of the active material on the nickel foam electrode was observed as 4 mg. All electrochemical measurements for the single electrode were executed in a three-electrode system with a platinum plate electrode as a counter electrode and a Hg/HgO electrode as a reference electrode.

The formula of mass specific capacitance is as follows:

$${\mathrm{C}}_{\mathrm{s}}=\frac{{\mathrm{I}}_{\mathrm{d}}\Delta \mathrm{t}}{3.6}$$

(1)

Calculated by integrating the slope of the discharge curve for the asymmetric device, the formula for energy density (W h kg⁻¹) and power density (W kg⁻¹) is as follows [46]:

$$\mathrm{E}=\frac{{\mathrm{I}}_{\mathrm{d}}}{3.6}{\displaystyle \int }\mathrm{E}\mathrm{d}\mathrm{t}$$

(2)

$$\mathrm{P}=\frac{3600\mathrm{E}}{\Delta \mathrm{t}}$$

(3)

The supercapacitor was assembled with activated carbon as a cathode and the prepared electrode as an anode. The mass balancing of cathode and anode will follow the equation:

$$\frac{{\mathrm{m}}^{+}}{{\mathrm{m}}^{-}}=\frac{{\mathrm{C}}_{\mathrm{s}-}\Delta {\mathrm{V}}_{-}}{{\mathrm{C}}_{\mathrm{s}+}\Delta {\mathrm{V}}_{+}}$$

(4)

where C_s (mAh g⁻¹) is the specific capacitance, I_d is the specific current, and Δt is the discharge time.

3. Results and Discussion

3.1. Characterization of Materials

The crystal structure of each Ni/P ratio and single-phase Ni₂P are shown in Figure 1. In Figure 1d, some diffraction peaks appear and correspond to the crystal planes of standard PDF card Ni₃P, Ni₂P, and Ni. This XRD pattern shows the formation of Ni₂P/Ni₃P/Ni heterostructures with a Ni/P ratio of 7:3. Due to the interaction between the different phases, the positions of the diffraction peaks are slightly shifted, which indicates the successful synthesis of the heterostructure and has a positive effect on the electrochemical performance of the electrode. Due to the presence of metallic nickel, the catalytic activity of this electrode was greatly improved. In Figure 1f, some diffraction peaks appear and correspond to the crystal planes of standard PDF card Ni₂P, and Ni₅P₄. This XRD pattern shows that Ni₂P/Ni₅P₄ heterostructures are formed at a Ni/P ratio of 5:4. The combination of different phases produces a certain lattice distortion, so the position of the corresponding diffraction peak is slightly shifted compared to the standard card. Although only two nickel phosphide phases exist, the interfacial effect between the phases can still improve the electrochemical performance of the electrode, and at the same time will improve its catalytic stability. It can be observed from Figure 1h that diffraction peaks appear at 17.463°, 26.330°, 30.489°, 31.771°, 35.350°, 40.714°, 44.611°, 47.362°, 54.196°, 54.998°, 66.371°, 72.719°, and 74.790°, corresponding to (100), (001), (110), (101), (200), (111), (201), (210), (300), (211), (310), (311), and (400) crystal planes of standard PDF card Ni₂P, respectively. This XRD pattern shows the formation of single-phase Ni₂P. It can be observed from Figure 1a–c,e,g that Ni₃P/Ni₂P/Ni, Ni₅P₄/Ni₂P, Ni₅P₄/Ni₂P/Ni, Ni₅P₄/NiP₂/Ni, and Ni₅P₄/Ni₂P heterostructures are formed at a Ni/P ratio of 3:1, 5:2, 12:5, 2:1, and 1:1, respectively. Such heterostructures will contribute to electrochemical performance.

The SEM images of each Ni/P ratio and single-phase Ni₂P are shown in Figure 2. In addition to the single-phase nickel phosphide, two distinct structures can be seen in each image, which confirms the successful preparation of the composite phase. Figure 2d is the SEM image of a Ni/P ratio of 7:3. It can be observed from this image that the linear structure is wrapped around the granular structure, increasing the contact area between the two. Figure 2f is the SEM image of a Ni/P ratio of 5:4. It can be observed from this image that the villous structure grows on the spherical particles, and there are small particles interspersed between them. Such a combination of different phases will produce lattice distortion at the interface, thereby improving the charge transfer efficiency and enhancing the catalytic activity.

Figure 2a is the SEM image of a Ni/P ratio of 3:1. It can be observed from this image that braided linear structures and granular structures intersected with each other are generated under such condition. Figure 2b is the SEM image of a Ni/P ratio of 5:2. It can be observed from this image that under this condition, acicular and granular structures are formed, and most of the acicular structures are embedded between large and small particles. Figure 2c is the SEM image of a Ni/P ratio of 12:5. It can be observed from this image that prismatic and granular structures form under this condition, with granular structures surrounding the prism. It can also be observed that the prism is formed by a combination of small acicular structures. Figure 2e is the SEM image of a Ni/P ratio of 2:1. It can be observed from this image that the acicular and granular structures are bound together. Figure 2g is the SEM image of a Ni/P ratio of 1:1. It can be observed from this image that the bean-sprout-shaped structure is mixed with the lamellar structure. Figure 2h is the SEM image of single-phase Ni₂P. It can be observed from this image that large and small particles are not evenly distributed.

In order to explore the structure of the prepared nickel phosphide composite, we performed the TEM test on Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4). Figure 3a,d are the TEM images of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4). Both images display different shapes of substances. This is the result of different phase recombinations of nickel phosphide, corresponding to the SEM image. Figure 3b is the HRTEM image of Ni₂P/Ni₃P/Ni (Ni/P = 7:3). We can observe the (111) crystal plane of Ni₂P, (112) crystal planes of Ni₃P and (111) crystal planes of Ni, which prove the formation of the nickel phosphide composite phase. Figure 3e is the HRTEM image of Ni₂P/Ni₅P₄ (Ni/P = 5:4). We can observe the (111) crystal plane of Ni₂P and (103) crystal planes of Ni₅P₄, which further proves the formation of the nickel phosphide composite phase. The formation of the composite phase can effectively improve the catalytic activity of electrode materials. Different phases have different crystal lattices. The interface at the grain boundary when recombined is an inhomogeneous interface, and this causes lattice distortion due to the Jahn-Teller effect. The electronic state at the grain boundary is changed, and unpaired electrons may appear, which facilitates the transfer of charge. At the same time, the accumulated electrons can also generate a built-in electric field to promote the transmission of ions. Some randomly arranged atoms (marked by green dots) were also observed in image 3b,e; these disordered atoms rearrange in order to balance when the external environment changes. This generates more interfaces, greatly increases the efficiency of ion transport, and improves electrochemical performance [47]. We can observe the (311), (300) crystal plane of Ni₂P and (301) crystal planes of Ni₃P in Figure 3c, (102), (213) crystal plane of Ni₅P₄ and (311) crystal planes of Ni₂P in Figure 3f. These crystal planes correspond to XRD test results, which proves the formation of the nickel phosphide composite phase.

Figure 4 displays the BET test and pore size distribution information. Figure 4a,c show the BET test information of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4), respectively. From that information we can conclude that the BET surface area of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4) are 1.32 m² g⁻¹ and 0.54 m² g⁻¹, respectively. Figure 4b,d show the pore size distribution information of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4), and their pore size distribution is between 2–50 nm, belonging to mesoporous materials. Such a porous structure can provide more ion channels for the catalytic reaction, thus having high-efficiency energy storage effects and good electrochemical performance.

X-ray photoelectron spectroscopy (XPS) was used to test and analyze the electronic structure of single-phase Ni₂P and Ni₂P/Ni₅P₄ (Ni/P = 5:4). Figure 5a is the XPS spectrum of single-phase Ni₂P and Ni₂P/Ni₅P₄ (Ni/P = 5:4) on the Ni 2p orbital. From the figure, we can observe that single-phase Ni₂P has peaks at the positions where the binding energy is 870.62 eV and 853.41 eV, which correspond to the Ni²⁺ 2p_1/2 and Ni²⁺ 2p_3/2 orbitals, respectively. Ni₂P/Ni₅P₄ (Ni/P = 5:4) has peaks at the positions where the binding energy is 870.22 eV and 853.09 eV, which correspond to the Ni²⁺ 2p_1/2 and Ni²⁺ 2p_3/2 orbitals, respectively. Compared with single-phase Ni₂P, the binding energy of Ni₂P/Ni₅P₄ (Ni/P = 5:4) has a negative offset of 0.4 eV and 0.32 eV in these two orbitals, respectively. This shows that the formation of the Ni₂P/Ni₅P₄ (Ni/P = 5:4) led to the occurrence of electronic remodeling, and the electrons flow from Ni₅P₄ to Ni₂P. This is due to the difference in Fermi energy levels. The electron density around the Ni₂P increases, and the electronic interaction between the Ni₂P and the Ni₅P₄ may increase the catalytic activity [48]. At the same time, it can be observed that single-phase Ni₂P has peaks at the positions where the binding energy is 875.4 eV and 857.46 eV and Ni₂P/Ni₅P₄ (Ni/P = 5:4) has peaks at the positions where the binding energy is 875 eV and 857.16 eV, corresponding to Ni³⁺ 2p_1/2 and Ni³⁺ 2p_3/2 orbitals, respectively. This proves that Ni has multiple valence states, so the electrons can transition and cause Debye relaxation [49], which facilitates the occurrence of redox reactions. In addition, studies have shown that the redistribution of interface electrons has a positive effect on conductivity [50,51]. The other peaks in this image are satellite peaks. Figure 5b is the XPS spectrum of single-phase Ni₂P and Ni₂P/Ni₅P₄ (Ni/P = 5:4) on the P 2p orbital. Single-phase Ni₂P and Ni₂P/Ni₅P₄ (Ni/P = 5:4) have peaks at binding energy 129.92 eV and 129.39 eV, respectively. Compared with single-phase Ni₂P, the binding energy of Ni₂P/Ni₅P₄ (Ni/P = 5:4) has a negative offset of 0.53 eV; this can once again prove the existence of charge transfer. As for the presence of P-O bonds, this is related to the surface reaction that occurs when the sample comes in contact with air during the transfer process [11,12].

3.2. Electrochemical Performance

The electrochemical test graphs of different nickel–phosphorus ratios are shown in Figure 6. Figure 6a shows the galvanostatic charge–discharge curves of electrodes prepared with different Ni/P ratios at 1 A g⁻¹. The curve of each electrode material has the appearance of a platform, which indicates that the redox reaction occurs during the charging and discharging process and corresponds to the redox peak in the cyclic voltammetry curve [30,31]. From this figure, it can be clearly seen that when the ratio of nickel to phosphorus is 7:3, it has the highest specific capacity of 321 mAh g⁻¹. Secondly, when the ratio of nickel to phosphorus is 5:4, the specific capacity is 218 mAh g⁻¹. The specific capacitance of each electrode material can be calculated according to Equation (1). The composites prepared with different raw material ratios and their electrochemical performances are shown in

Table 1. The electrochemical performance of composites prepared with various raw material ratios.

Ratio of Raw (Ni:P)	Prepared Compounds	Specific Capacity (mAh g−1)	Rate Capability (%)
3:1	Ni3P/Ni2P/Ni	195	30
5:2	Ni2P/Ni5P4	143	73
12:5	Ni5P4/Ni2P/Ni	73	69
7:3	Ni2P/Ni3P/Ni	321	59
2:1	Ni5P4/NiP2/Ni	108	69
5:4	Ni2P/Ni5P4	218	76
1:1	Ni5P4/Ni2P	119	53

. The faradaic reaction corresponding to the redox peak is as follows [40]:

Ni²⁺ + 2OH⁻ → Ni(OH)₂

(5)

Ni(OH)₂ + OH⁻ ↔ NiOOH + H₂O + e⁻

(6)

Figure 6b is the cyclic voltammetry curve of various Ni/P ratios at 3 mV s⁻¹. The area is the largest when the ratio of nickel to phosphorus is 7:3, followed by a ratio of 5:4, which corresponds to its specific capacity. From the platform of the galvanostatic charge–discharge curve and the obvious redox peak of the cyclic voltammetry curve, it can be judged that the electrode material is a typical pseudocapacitive material. Figure 6c is the electrochemical impedance spectroscopy of different nickel–phosphorus ratios, and Figure 6d is an enlarged view of their high frequency region. Each curve has an intersection point in the high frequency area with the real axis, representing the equivalent series resistance (R_es). The resistance of this part is related to the interface resistance, internal resistance, and electrolyte ion resistance. It can be seen from Figure 6d that the high-frequency region has a shape similar to a semicircle, and the diameter of the semicircle can essentially represent the charge transfer resistance (R_ct). The linear slope in the low-frequency region is related to the diffusion resistance, which is related to the characteristics of the electrolyte and the material itself. With a larger slope comes a faster ion diffusion and stronger capacitance. It can be seen from the curve that when the ratio of nickel to phosphorus is 7:3, its equivalent series resistance and charge transfer resistance are the smallest, and its slope is the largest; therefore, it has good electrochemical performance. Figure 6e is the Bode graph of different ratios of nickel to phosphorus. At the phase angle of −45°, the nickel–phosphorus ratio of 7:3 corresponds to the largest frequency; therefore, its time constant is the smallest, followed by the time constant of the nickel–phosphorus ratio of 5:4. This proves that these two ratios need less time to reach the steady state. There are two main reasons for the relatively high electrochemical performance of the composites: electron transfer can occur between different components and the presence of heterojunctions improves the catalytic activity.

The electrochemical test graphs of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P are shown in Figure 7. Figure 7a displays the galvanostatic charge–discharge curves of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P at 1 A g⁻¹. According to Equation (1), their specific capacities are 321 mAh g⁻¹, 218 mAh g⁻¹, and 58 mAh g⁻¹, respectively. This shows that compared with single-phase Ni₂P, the preparation of the composite greatly improves its specific capacity. Figure 7b shows the cyclic voltammetry curves of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P at sweep rate of 3 mV s⁻¹. The areas of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4) are much larger than that of the single-phase nickel phosphide, which corresponds to the specific capacity. Figure 7c shows the Nyquist plots of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P. The equivalent series resistances (R_es) of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P are 0.838 Ω, 1.473 Ω, and 0.862 Ω, respectively. The charge transfer resistances (R_ct) of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P are 0.056 Ω, 0.111 Ω, and 0.178 Ω, respectively. Figure 7d is the bode plot of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P. The time constants of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase Ni₂P are 0.87 s, 4.01 s, and 43 s, respectively. Comprehensive electrochemical test results showed that the performance of nickel phosphide composite was better than that of single-phase nickel phosphide. The electrochemical performances of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase nickel phosphide are listed in

Table 2. The electrochemical performance of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single-phase nickel phosphide.

Electrode Materials	Specific Capacity (mAh g−1)	Rate Capability (%)	Res	Rct	Time Constant(s)
Ni2P/Ni3P/Ni (Ni/P = 7:3)	321	59	0.838	0.056	0.87
Ni2P/Ni5P4 (Ni/P = 5:4)	218	76	1.473	0.111	4.01
Ni2P	58	33	0.862	0.178	43

.

In order to further understand the electrochemical performance of nickel phosphide composites, further tests were done on materials with better performance. Figure 8 is the electrochemical test graph of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) and Ni₂P/Ni₅P₄ (Ni/P = 5:4). Figure 8a,b are the galvanostatic charge–discharge curves of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) under different current densities and cyclic voltammetry curves of Ni₂P/Ni₃P/Ni (Ni/P = 7:3) at different scan rates. As the current density increases, the specific capacitance shows a downward trend, which is caused by the penetration of electrons and ions into the electrode surface [35]. The current density increased from 0.5 A g⁻¹ to 5 A g⁻¹ and its rate capability is 59%. The appearance of the redox peak is related to the surface Faraday reaction. As the sweep speed increases, the area of the cyclic voltammetry curve also increases, and the redox peaks move to higher and lower windows, respectively. This is related to the diffusion kinetics and internal resistance [30,31]. After 2000 cycles, the capacity retention rate was above 66% as shown in the Figure 8f. Figure 8d,e are the galvanostatic charge–discharge curves of Ni₂P/Ni₅P₄ (Ni/P = 5:4) under various current densities and the cyclic voltammetry curves of Ni₂P/Ni₅P₄ (Ni/P = 5:4) at various scan rates. The current density increased from 0.5 A g⁻¹ to 5 A g⁻¹, and its rate capability was 76%. As the sweep speed increased, the area of the cyclic voltammetry curve also increased, and the oxidation peak gradually disappeared. This may be because the sweep speed was too fast, and the changes caused by the redox reaction were not recorded in time. After 7000 cycles, the capacity retention rate was above 82% as shown in the Figure 8f. For scientific analysis, we compared the electrochemical properties of our prepared electrodes with those of the electrode materials reported in the literature, as shown in

Table 3. Comparison of the electrochemical performance of our prepared electrodes with other reported electrode materials in the literature.

Electrode	Substrate	Electrolyte	Current Density	Specific Capacity	Refs.
NixCo1-xOy/Er-Go	Stainless steel	1 M KOH	1 A g−1	180 mAh g−1	[42]
CoP/Ni2P	NF	6 M KOH	1 A g−1	1557 C g−1	[30]
Ni-CoP	NF	6 M KOH	1 A g−1	578 C g−1	[30]
NiCo2O4/Ni2P	NF	3 M KOH	8 mA cm−2	2900 F g−1	[41]
N-Ni2P/Ni12P5/Ni3S2	NF	2 M KOH	20 mA cm−2	12.71 F cm−2	[44]
Ni2P@N-C	NF	3 M KOH	10 A g−1	1320.4 F g−1	[27]
Co-Ni2P	NF	3 M KOH	1 A g−1	864 F g−1	[35]
Fe-Ni2P	NF	3 M KOH	1 A g−1	856 F g−1	[35]
NiCoP/CoP	NF	2 M KOH	1 A g−1	152 mAh g−1	[37]
Ni2P/Ni3P/Ni	NF	2 M KOH	1 A g−1	321 mAh g−1	This work
Ni2P/Ni5P4	NF	2 M KOH	1 A g−1	218 mAh g−1	This work

. The electrochemical performance of our electrode material was relatively good.

In order to deeply evaluate the electrochemical performance of the prepared materials in practical applications, we assembled a supercapacitor with Ni₂P/Ni₃P/Ni (Ni/P = 7:3) as the cathode and activated carbon as the anode. We also assembled a supercapacitor with Ni₂P/Ni₅P₄ (Ni/P = 5:4) as the cathode and activated carbon as the anode. Figure 9a shows the galvanostatic charge–discharge curves of the Ni₂P/Ni₃P/Ni//AC and Ni₂P/Ni₅P₄//AC supercapacitors at a current density of 1 A g⁻¹. According to Equation (1), the specific capacity of supercapacitor Ni₂P/Ni₃P/Ni//AC at 1 A g⁻¹ is 27.72 mAh g⁻¹, and according to Equations (2) and (3), it has an energy density of 22 W h kg⁻¹ while its power density is 800 W kg⁻¹. The specific capacity of supercapacitor Ni₂P/Ni₅P₄//AC at 1 A g⁻¹ is 33.39 mAh g⁻¹, and it has an energy density of 27 W h kg⁻¹ while its power density is 800 W kg⁻¹. Figure 9b displays the cyclic voltammetry curves of the Ni₂P/Ni₃P/Ni//AC and Ni₂P/Ni₅P₄//AC supercapacitors at a scan rate of 100 mV s⁻¹. The area of Ni₂P/Ni₅P₄//AC is larger than that of Ni₂P/Ni₃P/Ni//AC, which corresponds to the specific capacity of the two supercapacitors. From the comprehensive test results, the electrochemical performance of Ni₂P/Ni₅P₄//AC was better than that of Ni₂P/Ni₃P/Ni//AC; therefore, we further explored Ni₂P/Ni₅P₄//AC. Figure 9c is the galvanostatic charge–discharge curve of the Ni₂P/Ni₅P₄//AC supercapacitor at various current densities. We noticed that there are no obvious charging or discharging platforms in the galvanostatic charge–discharge curves, which indicates that the assembled device has good capacitance characteristics and a good electron transfer rate. Figure 9d shows the cyclic voltammetry curves of the Ni₂P/Ni₅P₄//AC supercapacitor at various scan rates. We noticed that the CV curve has no obvious redox peak, indicating that it is suitable for high-power output, and the shape of the curve does not change with the scanning speed, indicating that the assembled supercapacitor has a rate capability. The CV and GCD curves correspond to each other. With a larger specific capacitance comes a longer charge and discharge time in the GCD and a larger CV area. The relatively large specific capacitance of Ni₂P/Ni₅P₄//AC may be mainly due to its better stability.

In order to further study the internal reasons for the difference in performance of the different electrode materials prepared, we performed the work shown in Figure 10 including the linear sweep volt–ampere curve, schematic diagram of heterogeneous interface dangling bonds (unpaired electronics) generation, and energy band diagrams of metal and semiconductors. Figure 10a is the current–voltage curve of Ni₂P/Ni₃P/Ni (Ni/P = 7:3), Ni₂P/Ni₅P₄ (Ni/P = 5:4), and single phase Ni₂P tested by linear sweep voltammetry. Clearly, the conductivity of electrode Ni₂P/Ni₃P/Ni (Ni/P = 7:3) is the largest, followed by electrode Ni₂P/Ni₅P₄ (Ni/P = 5:4), and the conductivity of electrode single-phase Ni₂P is the worst. We recognize that there is an interface between the two-phase composite. Due to the difference in the phase structure, the lattice mismatch at the interface is inevitable. At this time, a part of the unsaturated bond will appear in the semiconductor material with a smaller crystal lattice at the interface as shown in the Figure 10b. These unsaturated bonds are dangling bonds, which form unpaired electrons. The bond density is affected by different semiconductor lattice constants and the crystal plane as the interface, which is determined by the following formula:

∆N_s = N_s1 − N_s2

(7)

N_s1, N_s2 are the bond density of the two semiconductor materials at the interface, respectively.

These dangling bonds, i.e., the unpaired electrons, are in a relatively free state and are easily excited to become free electrons, which provides excellent conditions for high-efficiency ion transmission and more active sites, thereby improving the electrochemical performance of the electrode. The prepared working electrodes with heterostructures all have such an interface as Ni₂P/Ni₅P₄ (Ni/P = 5:4). As shown in Figure 10c, the work function of the metal is greater than the work function of the semiconductor. The Mott-Schottky interface shown in Figure 10d is formed after the metal is in contact with the semiconductor. Due to the difference in work function, electrons flow from the metal to the semiconductor, carriers flow from the semiconductor to the metal, and the accumulation of electrons in the semiconductor forms an internal potential field, which provides a high-speed channel for the continuous transmission of free electrons [52]. This improves the utilization of active sites, promotes the occurrence of redox reactions, and makes the working electrode have better capacitance characteristics. The prepared working electrodes with metallic nickel all have such an effect as Ni₂P/Ni₃P/Ni (Ni/P = 7:3).

4. Conclusions

In summary, the nickel phosphide composites were prepared in one step using the temperature-programmed phosphating method, and the prepared nickel phosphide composite showed good electrochemical performance. The prepared compound Ni₂P/Ni₃P/Ni (Ni/P = 7:3) had a specific capacity of 321 mAh g⁻¹ under 1 A g⁻¹, and the prepared compound Ni₂P/Ni₅P₄ (Ni/P = 5:4) had a specific capacity of 218 mAh g⁻¹ under 1 A g⁻¹ and a 76% rate performance. After 7000 cycles, the capacity retention rate was above 82%. After assembling the prepared composite and activated carbon into a supercapacitor, Ni₂P/Ni₃P/Ni//AC was found to have an energy density of 22 W h kg⁻¹ and a power density of 800 W kg⁻¹, while Ni₂P/Ni₅P₄ //AC had an energy density of 27 W h kg⁻¹ and a power density of 800 W kg⁻¹. These results provide ideas for further research.