A facile one-step route to synthesize the three-layer nanostructure of CuS/RGO/Ni₃S₂ and its high electrochemical performance

Abstract

A three-layer nanostructure of a CuS/reduced graphene oxide (RGO)/Ni₃S₂ composite was in situ grown on nickel foam (NF) through a one-step hydrothermal-assisted process. During this process, the bottom Ni₃S₂ layer and middle RGO layer were simultaneously formed through the redox reaction of the Ni element on the foam surface of NF with GO and subsequent vulcanization. The upper CuS layer, consisting of sphere and fiber-like blocks, was converted from Cu²⁺ adsorbed by electrostatic forces and then well anchored on the RGO surface. The binder-free CuS/RGO/Ni₃S₂ electrode delivered high specific capacitance (10 494.5 mF cm⁻² at a current density of 40 mA cm⁻², i.e., 1692.7 F g⁻¹ at 6.5 A g⁻¹). It also exhibited an excellent cycling stability with ca. 91.5% of the initial capacitance retention after 4000 charge–discharge cycles at a current density of 100 mA cm⁻². The good electrochemical performance and simple accessibility prove that this CuS/RGO/Ni₃S₂ composite is a promising material for supercapacitor applications.

---

1. Introduction

Supercapacitors, combined with secondary batteries and conventional dielectric capacitors, constitute the basic rechargeable electric storage device system to meet different demands in power density, energy density and lifetime. Nowadays, the main task for supercapacitors is still focused on the exploitation of new electrode materials with excellent electrochemical performance.

Although integrating various semiconductive materials in conventional conductive reagents is successful to make almost any insoluble substances exhibit primary supercapacitor capability, it seldom endows these electrode materials with comprehensive supercapacitor performance in practical applications. It is therefore still necessary to devote effort to enhance the capabilities of various electrode materials, especially composites.

Firstly, integrating pseudocapacitive materials with carbon is an effective strategy, in which carbon materials greatly enhance the electrical conductivity for the electrode.^1–3 Especially, when nano-sized carbon material is used, nano-sized and nanoporous structure for pseudocapacitive materials can be easily built, which results in high specific area and thus decreases the current density. Therefore, improved specific capacitance and rate capability are obtained^1,2 due to the low electrochemical polarization and complete utilization of nanoporous active materials during charging/discharging process. Graphene is usually a good candidate of carbon component in the composite^2,4 because of its high electronic conductivity (16 000 S m⁻¹), high specific surface area (2630 m² g⁻¹, theoretical value) and mechanical properties.^5,6

Secondly, designing and tailoring the pseudocapacitive materials is another prior method to further improve the supercapacitor performance.⁷ Based on the careful selection of potential pseudocapacitive material and its shapes, designing and synthesizing those electrode materials composed of two different metal cations, e.g., ternary or hybrid of two-component of oxides or sulfides, is another effective route to increase the capability, as these two metal cations-containing materials usually exhibit prior performance to the single component ones, which may be attributed to their richer redox reaction.^1,8–10

Thirdly, a binder-free design is necessary to further improve the capacitance, especially in specific capacitance and rate ability.¹¹ Also, since binder has no contribution to the capacitance and the insulating binder will decrease the conductivity of the electrode material, it negatively affects the electrochemical performance in both specific capacitance and rate ability. In addition, in order to compensate the deficiency of conductivity, a certain amount of conductivity agents are commonly added in the preparation of the electrode, which will decrease the specific capacitance and energy density due to the macro-sized conductive agent. Therefore, a binder-free route favors the electrochemical performance.

Fourthly, a high areal capacitance plays a key role in determining the practical applications. High capacitance is usually needed in practice, which depends not only on the high specific capacitance, but also on high specific volumetric capacitance (loading amount). A stable, nanoporous architecture is required for this high loading.

Recently, transition metal sulfides have been receiving great interest as electrode materials for supercapacitors in view of their richer redox valences and higher conductivity.^12,13 Among various metal sulfides, Ni₃S₂ is one of the most promising candidates, because of its low cost, environmentally-friendly nature and high theoretical capacitance.¹⁴ On the other hand, another sulfide, CuS, is also widely used as electrode materials of lithium-ion batteries,¹⁵ and supercapacitor.^16,17 However, the composite of two kinds of transition metal sulfides is rarely reported, and hybrid material of CuS/Ni₃S₂ has not been reported yet.

In this study, CuS/RGO/Ni₃S₂ composites, in which the upper CuS spheres and nanofibers (covered by a thin nanoflakes layer) supported on the middle RGO layer and bottom Ni₃S₂ layer, were synthesized on Ni foam. Ni foam acts not only as conducting substrate, but also as Ni source and reducing agent for GO during the preparation process. RGO, acting as a second electron transfer and soft support, well connects the CuS and Ni₃S₂ layers. The electrode of as-prepared CuS/RGO/Ni₃S₂/NF composite film exhibits improved supercapacitor performance, e.g., 10 494.5 mF cm⁻² at 40 mA cm⁻², 4930.9 mF cm⁻² at 200 mA cm⁻², and 105.8% capacitance retention after 1000 cycles at 100 mA cm⁻². Even at 4000th cycle the capacitance can still retain 91.5% of the initial value.

2. Experiment

2.1 Materials and reagents

Pristine graphite powder, hydrochloric acid (HCl, 36.0–38.0 wt%), hydrogen peroxide (H₂O₂, 30 wt%), sulfuric acid (H₂SO₄, 95–98 wt%), potassium permanganate (KMnO₄, ≥99.5%), phosphorus pentoxide (P₂O₅, ≥98.0%), potassium persulfate (K₂S₂O₈, ≥99.5%), and ethanol (C₂H₅OH, >99.7 wt%) were obtained from the Shanghai Chemical Co. Copper nitrate trihydrate (Cu(NO₃)₂·3H₂O, 99.0–102.0%) and thiourea (CH₄N₂S, ≥99.0%) were purchased from the Sinopharm Chemical Reagent Co. Nickel foam (hereinafter referred to as NF) was obtained from Keliyuan (Hunan). All of the chemical reagents were used as received without any further purification. Water used was deionized.

2.2 Preparation of CuS/RGO/Ni₃S₂, CuS/Ni₃S₂, RGO/Ni₃S₂, and Ni₃S₂ on NF

Prior to synthesis, NF was carefully cleaned with acetone, ethanol, and deionized water in an ultrasound bath to remove surface impurities, respectively.

Graphene oxide (GO) was prepared from pristine graphite powder based on a modified Hummers method.¹⁸ The synthesis of CuS/RGO/Ni₃S₂ composite was carried out through a hydrothermal process by immersing the cleaned Ni foam in a mixture solution of GO, copper salt and thiourea. Typically, GO (30 mg) and Cu(NO₃)₂·3H₂O (1 mmol) were added in deionized water (50 ml) under ultrasonication for 30 min. Thiourea (2 mmol) was subsequently dissolved into this solution. The NF (1 × 2 cm²) with a bared area of 1 × 1 cm² was then immersed in this aqueous solution. The mixture was loaded into a Teflon-lined stainless steel autoclave (100 ml in volume) for hydrothermal reaction at various temperatures (i.e., 150 °C, 180 °C and 210 °C) for different duration (12 h, 24 h and 36 h). The final products were washed with water and ethanol in turn, and then dried in a vacuum oven at 80 °C for 12 h. The samples of CuS/RGO/Ni₃S₂/NF composites were denoted as CRNS-150-24, CRNS-180-24, CRNS-210-24, CRNS-180-12, and CRNS-180-36 respectively according to the hydrothermal treatment conditions. CuS/Ni₃S₂/NF (CNS-180-24), RGO/Ni₃S₂/NF (RNS-180-24) and Ni₃S₂/NF (NS-180-24) composites were prepared under identical conditions (180 °C, 24 h) as the compared samples.

2.3 Characterizations

The structure and the diffraction pattern of the as-prepared materials were performed by wide-angle (10°–80°, 40 kV/200 mA) powder X-ray diffraction (XRD) using an X-ray diffractometer with Cu Kα (λ = 0.15406 nm). X-ray photoelectron spectroscopy (XPS) was performed on ESCALAB 250Xi. The morphology and elementary composition of the samples were characterized using a field emission scanning electron microscope (FESEM, S4800), a transmission electron microscope (TEM, JEM 2100), and energy dispersive spectrometer (EDS, Bruker, AXS, Quantax 400-30), respectively.

2.4 Electrochemical measurements

The measurements were carried out in an aqueous solution of 6 mol L⁻¹ KOH using a three-electrode system. The composite samples with a geometric area of 1 × 1 cm², a Pt foil (2 × 3 cm²), and a saturated calomel electrode (SCE) were used as working electrodes, counter electrode, and reference electrode, respectively. The electrochemical performances of the as-prepared materials were measured using a CHI 660E workstation. The loading amount of CuS/RGO/Ni₃S₂ (weight of active material) was determined by the weight difference of the above electrode before testing and after ultrasonic treatment with hydrochloric acid using an electronic balance,^4,19 which was determined to be 6.2 mg.

3. Results and discussion

3.1 XPS, and XRD patterns of CuS/RGO/Ni₃S₂ nanocomposite

X-ray photoelectron spectroscopy (XPS) was shown in Fig. 1a–d. It is reasonable to divide the C 1s spectrum of CRNS (Fig. 1a) into three peaks at 284.6, 286.7, and 288.4 eV which correlated with C–C, C–O, and C=O, respectively.^20,21 The peak intensities of C–O and C=O for the composite are much smaller than those for pure GO powder in previous work,²² suggesting considerable deoxygenation after hydrothermal treatment. Furthermore, in Fig. 1b, two main peaks at 856.6 and 874.5 eV in the Ni 2p XPS spectrum are assigned to Ni 2p_3/2 and Ni 2p_1/2 of Ni₃S₂, respectively.^23–25 On the other hand, the Cu 2p XPS spectra of the composite (Fig. 1c) exhibits two peaks at 934.4 and 954.4 eV, which are corresponded to the Cu 2p_3/2 and Cu 2p_1/2 spin–orbit peaks of the CuS phase, respectively.²⁶ The Cu 2p_3/2–Cu 2p_1/2 spin-energy separation is 20 eV. Moreover, the existence of Cu²⁺ in the sample can be further confirmed by the two extra shake-up peaks at 942.8 eV and 964.6 eV, which are positioned at higher binding energies compared to the main peaks, implying the presence of an unfilled Cu 3d⁹ shell of Cu²⁺.²² The peak at 162.2 eV for S 2p is demonstrated in Fig. 1d, indicating the presence of S²⁻ in the composite.^27,28 Hence, XPS results indicate the presence of CuS, Ni₃S₂ and RGO component in the CuS/RGO/Ni₃S₂/NF.

Fig. 1 XPS spectrums of CRNS composite: (a) C 1s spectrum; (b) Ni 2p spectrum; (c) Cu 2p spectrum; and (d) the survey spectrum of the CRNS composite.

XRD was conducted in order to determine the crystalline properties of the CuS/RGO/Ni₃S₂, and Fig. 2 gives the XRD patterns of CuS/RGO/Ni₃S₂, RGO/Ni₃S₂ and Ni₃S₂ prepared at 180 °C for 24 h. It is clear that three peaks at 44.5°, 51.9°, and 76.4° appear in the patterns of all these three composite samples, which are ascribed to the (111), (200), and (220) planes of metallic nickel (JCPDS no. 04-0850), respectively.¹⁹ Furthermore, characteristic diffraction peaks of Ni₃S₂ (JCPDS no. 44-1418) also emerge in all the three composites.^29,30 Referring to CRNS, peaks corresponding to CuS are not found, although it is clearly determined in XPS and EDS, suggesting that CuS in the composite is in the amorphous form.³¹

Fig. 2 XRD patterns of CRNS-180-24, RNS-180-24, and NS-180-24.

3.2 FESEM images, EDS, and TEM images of the CuS/RGO/Ni₃S₂ composite

Fig. 3a–d show the FESEM images of CRNS-180-24 composites. Compared with the smooth surface of untreated Ni foam (inset of Fig. 3a), the surface of CRNS-180-24 is rough and covered with a coating, as shown in Fig. 3a. The upper layer consists of two kinds of building blocks: spheres and nanofibers. Furthermore, a thin layer network consisting of ultrathin sheets (several nanometers in thickness) with porous structures grows on the spheres and nanofibers building blocks (Fig. 3b and c and the insets). All these spherical, fabric and network-like structures can be attributed to the formation of CuS, as they disappear in the absence of Cu salt. On the other hand, the lower layer of Ni₃S₂ (Fig. 3d), nest-like film fabricated by the assembly of vertically arrayed nanoflakes is also obviously visible at the bared zone, which is consistent with the morphology of RGO/Ni₃S₂ in our previous work.²⁸ Both CuS spheres, nanofibers layer and Ni₃S₂ nanoflakes layer possess highly open and porous hierarchical structures, making most of the surface area readily accessible to liquid electrolyte, which is able to provide efficient channels for electron transport.³² These indicate that the presence of CuS strongly influences the arrangement and structure of the composite sheets, and facilitates the formation of layer-by-layer structure of CuS/RGO/Ni₃S₂/NF composite films. In addition, the RGO sheets are efficiently and uniformly distributed on the surface of nickel substrate, which provides a large surface area for the interconnected and uniform films. Thus, CuS/RGO/Ni₃S₂ composite with a layer-by-layer structure has formed through this one-pot hydrothermal route.³³ It is worth noting that the RGO in the composite cannot be clearly observed in the FESEM images (Fig. 3). The reason is that ultrathin RGO coated on the Ni₃S₂ nanosheet is difficult to be distinguished from lower Ni₃S₂ nanosheets layer. This conjecture will be proven in the TEM images.

Fig. 3 FESEM images of different area of CRNS-180-24 (a–d) and nickel foam (a inset).

Energy-dispersive X-ray spectroscopy (EDS) mappings are performed in order to further confirm the composition and elements distribution of the different layer. Fig. 4a–e show the elemental mapping of carbon (Fig. 4a), oxygen (Fig. 4b), copper (Fig. 4c), nickel (Fig. 4d), and sulfur (Fig. 4e) of CuS/RGO/Ni₃S₂ for the corresponding overlay image in Fig. 4f. Obviously, Ni signal is weak at those areas occupied by CuS spheres or fibers, while it is strong at those unoccupied zone (Fig. 4d), which is especially distinct in the sample with big sphere blocks (Fig. S1a, b† and the inset). These indicate that Ni₃S₂ is firstly generated at the lower layer close to the Ni surface, while the upper CuS layer is subsequently formed, which is consistent with the SEM images.

Fig. 4 EDS mapping of CRNS composite: (a) C elements (blue); (b) O elements (cyan); (c) Cu elements (red); (d) Ni elements (green); (e) S elements (purple); (f) corresponding overlay of C, O, Cu, Ni and S elements.

Fig. 5 shows the typical TEM images for the CRNS-180-24 composite. A thin layer is clearly observed which covers both the CuS fiber and spheres. In addition, networks of nanoflakes are also visible on the CuS fiber surface (within the rectangle line), which is expected to be combination of CuS and Ni₃S₂ nanosheets from the bottom layer. This structure leads to open and porous three-dimensional structures, which is beneficial to electrolyte access and electron transport during electrochemical reactions.³⁴ In the HRTEM image (Fig. 5c), the lattice of the nanoflakes is 0.18 nm which can be assigned to the (211) plane of Ni₃S₂. The concentric rings of selected area electron diffraction (SAED) pattern given in the inset of Fig. 5c confirms the polycrystallinity of the Ni₃S₂ nanoflakes.²⁹

Fig. 5 TEM and HRTEM images (a–c) and SAED image (c inset) of CRNS-180-24 composite.

3.3 Mechanism for the formation of CuS/RGO/Ni₃S₂ composite

We propose a possible formation route to the CRNS composite based on the experimental results and our earlier works.^4,28,30 As illustrated in Fig. 6, during the hydrothermal process, Ni on the foam is directly oxidized into Ni(OH)₂, and simultaneously GO is reduced to RGO, which successfully covers the surface of the Ni(OH)₂ layer⁴ and further conversion into Ni₃S₂/RGO composite during the hydrothermal process.^28,30 At the same time, an array of highly porous CuS nanoflakes converted from Cu²⁺ in the solution are vertically packed on the top side of the RGO, and then make up CuS spheres and nanofibers, eventually forming a multilayer diverse structure CuS/RGO/Ni₃S₂ film on Ni foam substrate. GO with its good solubility and negative electrostatic charge also makes for the uniform anchoring of the CuS array onto the surface of the GO to fabricate a three-dimensional (3D) porous diverse structure,^35–37 which is consistent with the SEM images. Obviously, RGO plays a key role in assembling the layer-by-layer CuS/RGO/Ni₃S₂/NF structure.

Fig. 6 Schematic representation of the formation process of the RGO/Ni₃S₂/NF (RNS) and CuS/RGO/Ni₃S₂/NF (CRNS) composites.

3.4 Electrochemical performance

CV has been considered to be a suitable electrochemical technology for estimating the difference between the non-faradaic and faradaic reaction. Fig. 7a presents CV curves of the CRNS electrode materials synthesized at different hydrothermal conditions at the scan rate of 10 mV s⁻¹. Obviously, at the potential range of −0.1 to 0.6 V (vs. SCE), all of the CV curves exhibit a pair of strong redox peaks which are different from those of electric double-layer capacitors, implying the presence of a reversible faradic reaction behaviour. The current output in the CV curves throughout the scan range for the CRNS-180-24 composite is higher than that of the other composite electrodes at the same scan rate, indicating higher specific capacitance.

Fig. 7 CV curves of: (a) CRNS-150-24, CRNS-180-12, CRNS-180-24, CRNS-180-36, and CRNS-210-24 at 10 mV s⁻¹; (b) CRNS-180-24, CNS-180-24, RNS-180-24, and NS-180-24 at 10 mV s⁻¹; (c) CRNS-180-24 at various scan rates.

Additionally, it is observed that the CRNS-180-24 exhibits much better performance than CNS-180-24, RNS-180-24 and NS-180-24 (Fig. 7b), which may be attributed to the enhanced electrical conductivity, fast electron transport and the rapid ion diffusion of CRNS-180-24. This is because that the RGO acts as both a basal plane for growth of CuS and a connect linker for upper and bottom layers, and there exists synergistic interaction between the RGO and two metal sulfides, which probably comes from the linkage of thiol bonds between residual surface group of the RGO and two metal sulfides, and thus achieves a higher specific capacitance.^9,38,39

For CRNS-180-24, apart from the faradaic redox reactions related to Ni₃S₂/Ni₃S₂(OH)₃,³⁰ the conversion reaction of CuS/CuSOH is also involved in the potential range of −0.1 to 0.6 V (vs. SCE).^16,40,41 The reversible reactions in the alkaline electrolyte are suggested as follows.

Ni₃S₂ + 3OH⁻ ↔ Ni₃S₂(OH)₃ + 3e⁻ (1)

CuS + OH⁻ ↔ CuSOH + e⁻ (2)

These amorphous CuS spheres and fibers favor to supply a high specific surface area and thus enhance specific capacitance.^42,43

Fig. 7c shows CVs of the CRNS-180-24 with various scan rates. In the range of −0.1 to 0.6 V, no obvious distortion is observed in the CV curves with the increase of scan rates, which is an indication of good capacitive behavior. The obvious increase of current with scan rates indicates a good reversibility process for this electrode. With the increase of scan rate, the difference decreases between electrode surface and the diffusion layer, which results in increased flux to the electrode surface and achieves a higher current.⁹ For comparison, the peak position of the CV changes, which may be caused by a small equivalent series resistance and weak polarization of the electrodes.⁴⁴ Moreover, cyclic voltammetry curves expressed as specific capacitance vs. cell potential for CRNS-180-24 at various scan rates are shown in Fig. S2.†

Galvanostatic charge–discharge (GCD) measurements performed on the three electrodes in a potential window from −0.1 to 0.45 V provide a complementary measurement of the capacitance. The specific capacitance is calculated according to the following equation:

where I (A) is the discharge current, m (g) the mass of the active material, ΔV (V) the potential window; and Δt the discharge time. Fig. 8a shows the galvanostatic discharge curves of CRNS-180-24 at different current densities (40–200 mA cm⁻²) in the potential window of −0.1 to 0.45 V. The shapes of GCD curves also confirm their characteristics of faradaic performance.^9,45 As shown in Fig. 8a, the capacitance mainly corresponds to the pseudo-capacitance, which is in accordance with the CV results. When the discharge current density increases from 40 mA cm⁻² to 200 mA cm⁻², i.e. from 6.5 A g⁻¹ to 32.3 A g⁻¹, an excellent rate of supercapacitor performance is observed: 10 494.5 mF cm⁻² vs. 4930.9 mF cm⁻², i.e. 1692.7 F g⁻¹ to 795.3 F g⁻¹, suggesting this electrode material is suitable for working under high current density. To the best of our knowledge, this specific capacitance value is higher than those CuS materials as supercapacitor electrodes,^{16,17,40,46–50} and even higher than that of Ni₃S₂,^9,51–60 which is listed in Table 1. This great enhancement of specific capacitance is not only attributed to the synergistic effect resulting from the more active sites offered by RGO nanosheets and the quick electron transport of the highly interconnected hybrid nanoflakes, but also to the well-defined open porous nanostructure directly grown on the conductive substrate. This porous nanostructure allows easy access of electrolyte to all of the nanoflakes and thus facilitates charge transport and ion diffusion without any blocks of a binder.

Fig. 8 (a) GCD curves at various current densities; and (b) cycling ability at 100 mA cm⁻² of CRNS-180-24.

Table 1 Electrochemical performances of different CuS- or Ni₃S₂-based electrode materials

Electrode material	Electrode preparation	Specific capacitance (electrolyte)	Capacitance retention	Ref.
CNT@CuS	Pressed on Ni foam	110 F g^−1 (2.9 A g^−1) (2 M KOH)	100% after 1000 cycles	16
CuS	On glassy carbon electrode	597 F g^−1 (1 A g^−1) (2 M KOH)	80% after 1000 cycles	17
CuS	On Cu foil	274 F g^−1 (5 mA cm^−2) (1 M KOH)	87% after 5000 cycles	40
CuS@ppy	—	427 F g^−1 (1 A g^−1) (1 M KCl)	88% after 1000 cycles	47
CuS	On FTO substrates	72.85 F g^−1 (5 mV s^−1) (1 M LiClO4)	86.09% after 100 cycles	48
CuS	Pressed on Ni foam	833.3 F g^−1 (1 A g^−1) (6 M KOH)	75.4% after 500 cycles	49
CuS	On stainless steel substrates	101.34 F g^−1 (5 mV s^−1) (1 M NaOH)	81% after 1000 cycles	50
Ni@rGO-Ni3S2	Grown on Ni foam	625 F g^−1 (12 A g^−1) (6 M KOH)	97.9% after 3000 cycles	9
Bacteria-RGO/Ni3S2	Grown on Ni foam	962 F g^−1 (15 A g^−1) (2 M KOH)	89.6% after 3000 cycles	51
Ni3S2/MWCNT	Pressed on Ni foam	806 F g^−1 (3.2 A g^−1) (2 M KOH)	80% after 1000 cycles	52
CNT@Ni3S2	Pressed on Ni foam	480 F g^−1 (5.3 A g^−1) (2 M KOH)	88% after 1500 cycles	53
Ni3S2/Ni	Grown on Ni foam	1293 F g^−1 (5 mA cm^−2) (1 M KOH)	69% after 1000 cycles	54
Carbon coated-Ni3S2-RGO	Dropped onto Ni foam	996.7 F g^−1 (5 A g^−1) (3 M KOH)	98.6% after 500 cycles	55
Ni3S2	Grown on Ni foam	639.2 F g^−1 (4 A g ^−1) (1 M KOH)	626.1 F g^−1 (5 A g ^−1) after 2000 cycles	56
Ni3S2@Ni(OH)2/RGO	Grown on Ni foam	1003 F g^−1 (5.9 A g^−1) (3 M KOH)	99.1% after 2000 cycles	57
Ni3S2/CNFs	Pressed on Ni foam	814 F g^−1 (4 A g^−1) (2 M KOH)	83.5% after 1000 cycles	58
Ni3S2/graphene	Grown on Ni foam	1420 F g^−1 (2 A g^−1) (1 M KOH)	About 99.4% after 2000 cycles	59
Ni filled-Ni3S2/rGO	Pressed on Ni foam	833.3 F g^−1 (10 A g^−1) (2 M KOH)	92% after 1000 cycles	60
CuS/RGO/Ni3S2/Ni	Grown on Ni foam	1692.7 F g^−1 (6.5 A g^−1) (6 M KOH)	91.5% after 4000 cycles	This work

---

Table 2 presents the specific capacitance of the samples synthesized at different hydrothermal conditions. The capacitance increases with the increase of preparation temperature from 150 to 180 °C, while it decreases when the temperature increases further from 180 °C to 210 °C. The capacitance is affected by the duration of hydrothermal treatment experiences and has the similar trend, i.e. C_s (CRNS-180-24) > C_s (CRNS-180-12) > C_s (CRNS-180-36). Therefore, the sample prepared at 180 °C for 24 h shows the best performance (10 494.5 mF cm⁻²), which is much better than CNS-180-24 (2728.7 mF cm⁻²) and RNS-180-24 (6407.3 mF cm⁻²). According to the charging and discharging capacitances, coulombic efficiencies are also calculated and listed in Table S1.† It is clear that the coulombic efficiencies are higher than 90% at the various current densities, suggesting the excellent reversibility.

Table 2 Specific capacitance of all the samples synthesized at different conditions at various current rates

Samples	Specific capacitance (mF cm^−2)
	40 mA cm^−2	60 mA cm^−2	80 mA cm^−2	100 mA cm^−2	150 mA cm^−2	200 mA cm^−2
CRNS-180-24	10 494.5	9016.4	7979.6	6356.4	5694.5	4930.9
CRNS-180-12	7003.6	6007.6	5329.5	4705.5	3466.4	2838.2
CRNS-180-36	2062.5	1533.8	1237.8	1021.1	711.3	581.5
CRNS-150-24	3524.4	2745.8	2289.5	1898.2	1195.4	872.4
CRNS-210-24	2726.5	2038.9	1643.6	1367.3	900	764.4
CNS-180-24	2728.7	2253.8	1917.1	1712.7	1364.5	1200.7
RNS-180-24	6407.3	4946.2	4007.3	3518.2	2072.2	1088.7

---

Long-term cycling stability was tested by performing continuous charge–discharge cycles at a constant discharge current density of 100 mA cm⁻². As shown in Fig. 8b, the CRNS composite electrode also exhibits good cycling stability, with the specific capacitance gradually rising from 6356.4 mF cm⁻² to 6727.3 mF cm⁻² after 1000 cycles. The capacitance retention is 105.8%, which is much higher than our previous work (RGO/Ni₃S₂: 90.98% capacitance retention after 1000 cycles).²⁸ Additionally, it has 91.5% capacitance retention even after 4000 cycles (Fig. 8b inset). Compared to other reports about Ni₃S₂- or CuS-based materials,^{9,16,17,40,47–60} the cyclability of CRNS is greatly improved, which can be seen in Table 1. Moreover, the increased effective interfacial area among CuS, Ni₃S₂, and electrolyte also improves the stability. In fact, the integrated multilayer films of CuS/RGO/Ni₃S₂ in situ grown on NF substrate provide a shortened diffusion path for both electrons and ions, which will improve charge–discharge efficiency, and restrain the stress caused by the volume change during the process of charging/discharging,⁶¹ leading to better cycling performance. It is worth noting the role of RGO in the CuS/RGO/Ni₃S₂ composites: RGO connects CuS and Ni₃S₂ well. Moreover, due to its soft and high mechanical properties, RGO will buffer the volume changes of metal sulfides during the consecutive charging/discharging process. In addition, the strongly bonding may be ascribed to chemical covalent bonding and Van der Waals interaction between the RGO and the metal sulfides. Thus, all components in the multilayer structure are close-knit connected with each other, which minimizes the possibility of degradation of the electrode material.⁹

Fig. 9a presents the Nyquist plots of CRNS at different potentials. All the Nyquist plots consist of one semicircle at the high-middle frequency region, which is related to the charge transfer process occurring at the electrode/electrolyte interface, and one straight line at the low frequency region, which corresponds to the electrochemical process and mass transfer process, respectively.⁶² An equivalent circuit (inset in Fig. 9a) fitting the EIS plots is composed of an equivalent series resistance (R_s), a charge transfer resistance (R_ct), a double layer (C_dl), a capacitive element (C_ps) and Warburg impedance (W). The X-intercept of the Nyquist plot at high frequency represents the equivalent series resistance (R_s) of the electrodes, whereas, the diameter of the semicircle corresponds to the resistance of charge transfer (R_ct) at the contact interface between the electrodes and electrolyte solution.^8,63 The key factors determining high energy and power density are a maximum value of C_dl and a minimum value of R_s.⁶⁴ According to Fig. 9a, Nyquist plots of CRNS and RNS electrodes were recorded at 0.2 V in the frequency range from 100 kHz to 0.01 Hz (Fig. 9b). The interconnected porous structure can provide higher specific surface area, which can not only increase the number of electrochemically active sites for the redox reaction, but also enhance sufficient contact between the electrolyte and electrode. Besides, the enhanced accessible surface can improve the electronic conductivity, and thus reduce the charge transfer resistance.^45,65 In the plots, it can be confirmed that the CRNS has smaller R_s value (0.50 Ω) than that of RNS (0.86 Ω) and the R_ct value for the CRNS sample are smaller than 0.1 Ω, which is also lower than the RNS (0.24 Ω). In addition, the CRNS composite electrode exhibits a line that is close to vertical at the low frequency region, indicating that the CRNS composite is suitable for being used as an electrode material for supercapacitors. Due to the low resistance of CuS/RGO/Ni₃S₂ and the contact resistance between CuS/RGO/Ni₃S₂ and substrate NF, it is expected to raise the upper limit of the high charge–discharge rate of the supercapacitor.

Fig. 9 Nyquist plots: CRNS-180-24 under various potentials (a) and the corresponding equivalent circuit (inset in a), and CRNS-180-24, RNS-180-24 electrode materials at 0.2 V (b).

4. Conclusions

Layer-by-layer CuS/RGO/Ni₃S₂ nanocomposite has been synthesized on Ni foam through a facial one-pot hydrothermal method. In the composite, RGO acts as a conductive agent and a soft support which efficiently connects CuS and Ni₃S₂. Based on the synergistic effects of CuS, Ni₃S₂ and RGO, the electrochemical performance is greatly improved with a high capacitance of 10 494.5 mF cm⁻² (1692.7 F g⁻¹) at 40 mA cm⁻² (6.5 A g⁻¹) and still maintains 4930.9 mF cm⁻² (795.3 F g⁻¹) at the current density of 200 mA cm⁻² (32.3 A g⁻¹). In addition, it exhibits excellent capacitance retention of 91.5% after 4000 cycle tests.

Acknowledgements

We are grateful for the support of the National Natural Science Foundation of China (No. 20504026), the Shanghai Natural Science Foundation (No. 13ZR1411900), the Shanghai Leading Academic Discipline Project (No. B502), the Shanghai Key Laboratory Project (No. 08DZ2230500), and the EU FP7 Staff Exchange program (No. PIRSES-GA-2012-318990-ELEKTRONANOMAT).

References

1. M. Sun, J. Tie, G. Cheng, T. Lin, S. Peng, F. Deng, F. Ye and L. Yu, J. Mater. Chem. A, 2015, 3, 1730–1736.
2. J. W. Lee, T. Ahn, D. Soundararajan, J. M. Ko and J.-D. Kim, Chem. Commun., 2011, 47, 6305–6307.
3. H. Jiang, D. Ren, H. Wang, Y. Hu, S. Guo, H. Yuan, P. Hu, L. Zhang and C. Li, Adv. Mater., 2015, 27, 3687–3695.
4. S. Min, C. Zhao, G. Chen and X. Qian, Electrochim. Acta, 2014, 115, 155–164.
5. M. D. Stoller, S. Park, Y. Zhu, J. An and R. S. Ruoff, Nano Lett., 2008, 8, 3498–3502.
6. C.-H. Park, F. Giustino, C. D. Spataru, M. L. Cohen and S. G. Louie, Nano Lett., 2009, 9, 4234–4239.
7. W. Li, J. Shao, Q. Liu, X. Liu, X. Zhou and J. Hu, Electrochim. Acta, 2015, 157, 108–114.
8. Y. Li, L. Cao, L. Qiao, M. Zhou, Y. Yang, P. Xiao and Y. Zhang, J. Mater. Chem. A, 2014, 2, 6540–6548.
9. D. Ghosh and C. K. Das, ACS Appl. Mater. Interfaces, 2015, 7, 1122–1131.
10. W. Li, G. Li, J. Sun, R. Zou, K. Xu, Y. Sun, Z. Chen, J. Yang and J. Hu, Nanoscale, 2013, 5, 2901–2908.
11. W. Li, K. Xu, L. An, F. Jiang, X. Zhou, J. Yang, Z. Chen, R. Zou and J. Hu, J. Mater. Chem. A, 2014, 2, 1443–1447.
12. Z. Stević and M. Rajčić-Vujasinović, J. Power Sources, 2006, 160, 1511–1517.
13. F. Tao, Y.-Q. Zhao, G.-Q. Zhang and H.-L. Li, Electrochem. Commun., 2007, 9, 1282–1287.
14. J.-S. Kim, H.-J. Ahn, H.-S. Ryu, D.-J. Kim, G.-B. Cho, K.-W. Kim, T.-H. Nam and J. H. Ahn, J. Power Sources, 2008, 178, 852–856.
15. J. S. Chung and H. J. Sohn, J. Power Sources, 2002, 108, 226–231.
16. T. Zhu, B. Xia, L. Zhou and X. Wen Lou, J. Mater. Chem., 2012, 22, 7851–7855.
17. H. Peng, G. Ma, J. Mu, K. Sun and Z. Lei, Mater. Lett., 2014, 122, 25–28.
18. W. S. Hummers Jr and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
19. K. Wang, C. Zhao, S. Min and X. Qian, Electrochim. Acta, 2015, 165, 314–322.
20. S. Pei and H.-M. Cheng, Carbon, 2012, 50, 3210–3228.
21. J. Yan, W. Sun, T. Wei, Q. Zhang, Z. Fan and F. Wei, J. Mater. Chem., 2012, 22, 11494–11502.
22. K. Wang, X. Dong, C. Zhao, X. Qian and Y. Xu, Electrochim. Acta, 2015, 152, 433–442.
23. J. Zhu, Y. Li, S. Kang, X.-L. Wei and P. K. Shen, J. Mater. Chem. A, 2014, 2, 3142–3147.
24. B. Zhang, X. Ye, W. Dai, W. Hou and Y. Xie, Chem.–Eur. J., 2006, 12, 2337–2342.
25. S. Pan, J. Zhu and X. Liu, New J. Chem., 2013, 37, 654–662.
26. J. Qian, K. Wang, Q. Guan, H. Li, H. Xu, Q. Liu, W. Liu and B. Qiu, Appl. Surf. Sci., 2014, 288, 633–640.
27. W. Wei, L. Mi, Y. Gao, Z. Zheng, W. Chen and X. Guan, Chem. Mater., 2014, 26, 3418–3426.
28. Z. Zhang, C. Zhao, S. Min and X. Qian, Electrochim. Acta, 2014, 144, 100–110.
29. S.-W. Chou and J.-Y. Lin, J. Electrochem. Soc., 2013, 160, D178–D182.
30. Z. Zhang, Q. Wang, C. Zhao, S. Min and X. Qian, ACS Appl. Mater. Interfaces, 2015, 7, 4861–4868.
31. P. V. Quintana-Ramirez, M. C. Arenas-Arrocena, J. Santos-Cruz, M. Vega-Gonzalez, O. Martinez-Alvarez, V. M. Castano-Meneses, L. S. Acosta-Torres and J. Fuente-Hernandez, Beilstein J. Nanotechnol., 2014, 5, 1542–1552.
32. Q. Wang, R. Gao and J. Li, Appl. Phys. Lett., 2007, 90, 143107/1–143107/3.
33. Q. Pan, J. Xie, S. Liu, G. Cao, T. Zhu and X. Zhao, RSC Adv., 2013, 3, 3899–3906.
34. J. Pu, F. Cui, S. Chu, T. Wang, E. Sheng and Z. Wang, ACS Sustainable Chem. Eng., 2014, 2, 809–815.
35. S. Min, C. Zhao, Z. Zhang, G. Chen, X. Qian and Z. Guo, J. Mater. Chem. A, 2015, 3, 3641–3650.
36. L. Wang, X. Li, T. Guo, X. Yan and B. K. Tay, Int. J. Hydrogen Energy, 2014, 39, 7876–7884.
37. J. T. Zhang, S. Liu, G. L. Pan, G. R. Li and X. P. Gao, J. Mater. Chem. A, 2014, 2, 1524–1529.
38. S. Peng, L. Li, C. Li, H. Tan, R. Cai, H. Yu, S. Mhaisalkar, M. Srinivasan, S. Ramakrishna and Q. Yan, Chem. Commun., 2013, 49, 10178–10180.
39. H. Geng, S. F. Kong and Y. Wang, J. Mater. Chem. A, 2014, 2, 15152–15158.
40. Y.-K. Hsu, Y.-C. Chen and Y.-G. Lin, Electrochim. Acta, 2014, 139, 401–407.
41. L. Qian, X. Tian, L. Yang, J. Mao, H. Yuan and D. Xiao, RSC Adv., 2013, 3, 1703–1708.
42. G. Wang, L. Zhang and J. Zhang, Chem. Soc. Rev., 2012, 41, 797–828.
43. W. Wei, X. Cui, W. Chen and D. G. Ivey, Chem. Soc. Rev., 2011, 40, 1697–1721.
44. D. Wang, Y. Li, Q. Wang and T. Wang, J. Solid State Electrochem., 2011, 16, 2095–2102.
45. J. Zhang, H. Feng, J. Yang, Q. Qin, H. Fan, C. Wei and W. Zheng, ACS Appl. Mater. Interfaces, 2015, 7, 21735–21744.
46. X.-Y. Yu, L. Yu, L. Shen, X. Song, H. Chen and X. W. D. Lou, Adv. Funct. Mater., 2014, 24, 7440–7446.
47. H. Peng, G. Ma, K. Sun, J. Mu, H. Wang and Z. Lei, J. Mater. Chem. A, 2014, 2, 3303–3307.
48. C. Justin Raj, B. C. Kim, W.-J. Cho, W.-G. Lee, Y. Seo and K.-H. Yu, J. Alloys Compd., 2014, 586, 191–196.
49. K.-J. Huang, J.-Z. Zhang and Y. Fan, J. Alloys Compd., 2015, 625, 158–163.
50. K. Krishnamoorthy, G. K. Veerasubramani, A. N. Rao and S. Jae Kim, Mater. Res. Express, 2014, 1, 035006.
51. H. Zhang, X. Yu, D. Guo, B. Qu, M. Zhang, Q. Li and T. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 7335–7340.
52. C. S. Dai, P. Y. Chien, J. Y. Lin, S. W. Chou, W. K. Wu, P. H. Li, K. Y. Wu and T. W. Lin, ACS Appl. Mater. Interfaces, 2013, 5, 12168–12174.
53. T. Zhu, H. B. Wu, Y. Wang, R. Xu and X. W. D. Lou, Adv. Energy Mater., 2012, 2, 1497–1502.
54. K. Krishnamoorthy, G. K. Veerasubramani, S. Radhakrishnan and S. J. Kim, Chem. Eng. J., 2014, 251, 116–122.
55. L. Ma, X. Shen, Z. Ji, S. Wang, H. Zhou and G. Zhu, Electrochim. Acta, 2014, 146, 525–532.
56. Z. Zhang, Z. Huang, L. Ren, Y. Shen, X. Qi and J. Zhong, Electrochim. Acta, 2014, 149, 316–323.
57. W. Zhou, X. Cao, Z. Zeng, W. Shi, Y. Zhu, Q. Yan, H. Liu, J. Wang and H. Zhang, Energy Environ. Sci., 2013, 6, 2216–2221.
58. W. Yu, W. Lin, X. Shao, Z. Hu, R. Li and D. Yuan, J. Power Sources, 2014, 272, 137–143.
59. Z. Zhang, X. Liu, X. Qi, Z. Huang, L. Ren and J. Zhong, RSC Adv., 2014, 4, 37278–37283.
60. X. Ou, L. Gan and Z. Luo, J. Mater. Chem. A, 2014, 2, 19214–19220.
61. Z.-S. Wu, W. Ren, L. Wen, L. Gao, J. Zhao, Z. Chen, G. Zhou, F. Li and H.-M. Cheng, ACS Nano, 2010, 4, 3187–3194.
62. Y. Li, S. Chang, X. Liu, J. Huang, J. Yin, G. Wang and D. Cao, Electrochim. Acta, 2012, 85, 393–398.
63. Y. Dall’Agnese, P.-L. Taberna, Y. Gogotsi and P. Simon, J. Phys. Chem. Lett., 2015, 6, 2305–2309.
64. E. Frackowiak and F. Béguin, Carbon, 2001, 39, 937–950.
65. M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263–275.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra26428f

---