Pulse-Reversal Deposition of Nickel Sulﬁde Thin Film as an Efﬁcient Cathode Material for Hybrid Supercapacitors

In this study, nickel sulﬁde (Ni 3 S 2 ) thin ﬁlm was electrodeposited on Ni foam substrate by using a facile potentiodynamic (PD) and pulse-reversal (PR) method, respectively. After optimization of deposition parameters, the optimized Ni 3 S 2 electrode was prepared by using PR mode under − 1 V pulse-on potential, 0 V pulse-reversal potential, duty cycle of 0.5 and pulse frequency of 0.1 Hz. It delivered an impressive speciﬁc capacity up to 179.5 mAh g − 1 at current density of 2 A g − 1 .The hybrid supercapacitor (SC) was assembled by using the optimized Ni 3 S 2 electrode as cathode and activated carbon ﬁber cloth (ACFC) as anode, respectively. The Ni 3 S 2 //ACFC hybrid SC exhibited a high speciﬁc capacity of 26.7 mAh g − 1 at charge-discharge current density of 2 A g − 1 and displayed a maximum energy density of 26.4 Wh kg − 1 at 1978 W kg − 1 based on the total mass of active material of 9.2 mg. Furthermore, the Ni 3 S 2 //ACFC hybrid SC showed an impressive excellent cycling performance in 1 M KOH aqueous electrolyte at current density of 2 A g − 1 , with 97% speciﬁc capacitance retained after cycling of 400–5000 consecutive charge/discharge tests.

Supercapacitors (SCs) have been widely recognized in a wide range of applications in energy storage over the past decade, such as hybrid electric vehicles, mobile electronic devices, large industrial equipments, memory backup systems, and military devices. [1][2][3][4] It is due to their unique properties including higher specific power density (10 kW kg −1 ) than batteries and higher specific energy density (5-10 Wh kg −1 ) than conventional dielectric capacitors. SCs can be typically classified by the mechanisms of the charge storage/delivery, namely the electric double-layer capacitors (EDLCs) and psuedocapacitors. The EDLCs store energy by the separation of electronic and ionic charges at interface between high specific surface material and electrolyte. The EDLCs, which are generally with the carbonaceous materials as electroactive materials including activated carbon, carbon nanotube and graphene, have been recognized to be with low capacitance, especially in aqueous electrolyte. The pseudocapacitors based on the typical RuO 2 and MnO 2 electroactive materials usually show higher specific capacitance than EDLCs due to their fast and reversible faradic redox reaction, in which a linear dependence of the charge stored with the width of the potential window like EDLCs can be observed. In recent years, great deals of transition metal oxides, and hydroxides, such as Co 3 O 4 , 5 Co(OH) 2 , 6 NiO 7 and Ni(OH) 2 8 have been extensively investigated as electroactive materials for pseudocapacitors. However, they often display flat discharge plateaus during charge/discharge tests under constant current, which is distinctly different from those of RuO 2 and MnO 2 . Therefore, transition metal oxides or hydroxides should be recognized as battery-type electrodes and the corresponding capacity should be expressed in coulombs, C, or mAh. 9 Additionally, nickel sulfides have been widely studied in the fields of SCs, [10][11][12] hydrogen evolution, 13,14 photovoltaic cells, 15 and rechargeable lithium ion batteries 16 over the past decade. Zhu et al. 10 synthesized the hierarchical nickel sulfide (NiS) hollow spheres by hydrothermal method, which can deliver a relatively high capacitance of 927 F g −1 at 4.08 A g −1 . Yang et al. 11 successfully synthesized a hierarchical flower-like β-NiS architecture by via hydrothermal method with diethanolamine as the coordination agent and solvent. It also displayed high capacitance of 857.76 F g −1 at 2 A g −1 . In our previous study, a flaky Ni 3 S 2 nanostructure was directly electrodeposited on Ni foam by using a facile potentiodynamic (PD) deposition method. 12 The flaky Ni 3 S 2 nanostructure showed a high specific capacitance of 717 F g −1 at 2 A g −1 , and a promising specific capacitance of 411 F g −1 can be still delivered at 32 A g −1 .
However, both of EDLCs and pseudocapacitors still reveal lower energy densities than those of batteries. To promote the commercial-ization of SCs in the near future, the advance SCs should possess high operating voltage, great energy density and power density, and excellent cycling life. 17 Consequently, hybrid SCs have been developed, which typically consist of battery-type faradic electrode as energy source and an EDLC-type electrode as power source. [18][19][20] Since the different potential windows of the two electrodes can increase the maximum operating voltage of the cell, therefore resulting in the enhancement of specific capacity and energy density. 19 Until now, several hybrid SCs have been reported to be with high specific capacity and energy density. [21][22][23][24] For example, Wang et al. 23 exhibited the NiO//carbon asymmetric SCs with high energy density (11 Wh kg −1 ) and excellent stability. More recently, Yu et al. 24 utilized hydrothermal method to synthesize Ni 3 S 2 nanoparticles and the fabricated Ni 3 S 2 /CNFs//CNFs hybrid SCs exhibited a high energy density (25.8 Wh kg −1 ) at power density of 425 W kg −1 . As a result, hybrid SCs could be considered as promising devices for applications in energy storage systems.
Although we previously reported that the Ni 3 S 2 thin films can be directly deposited on the Ni foams via the facile PD deposition mode, the required power-supply equipment for the PD deposition is relatively expansive and hard to be scaled-up for large-scale deposition. In this current work, the Ni 3 S 2 thin films were successfully deposited onto Ni foam by using pulse-reversal (PR) mode. It is worth noting that the PR deposition technique possesses lots of well-known advantages of controlling the composition, microstructure and porosity of deposits by adjusting the deposition parameters of PR mode. 25 It is found that the Ni 3 S 2 thin film prepared by the optimized PR parameters reveals relatively higher specific capacity and smaller charge-transfer resistance (R ct ) value than that by using the previous PD deposition mode. In addition, a hybrid SC is then assembled by using the optimized Ni 3 S 2 electrode prepared by PR mode as cathode and activated carbon fiber cloth (ACFC) as anode, respectively. The as-assembled hybrid SC delivers high specific capacity value of 26.7 mAh g −1 at 2 A g −1 with a wide potential window between 0 and 1.8 V. The corresponding energy density and power density can be achieved to 26.4 Wh kg −1 and 1978 W kg −1 , respectively. Furthermore, the hybrid SC displays excellent cycling stability with only 3% performance loss after cycling of 400-5000 consecutive charge/discharge tests at a relatively high current density of 2 A g −1 .

Experimental
Preparation of Ni 3 S 2 electrode.-The Ni 3 S 2 films were deposited on to the nickel foam (10 mm × 10 mm × 1.7 mm, 94 PPI) substrate by PD and PR deposition modes in the deposition bath containing 50 mM NiCl 2 · 6H 2 O (Arcos, 97%) and 1 M thiourea (TU) (Arcos, 99%). All electrodeposition were performed in a conventional threecompartment cell consisting of a Ni foam substrate as working electrode, a saturated silver/silver chloride (Ag/AgCl) as reference electrode, and a Pt wire as counter electrode using a computer-controlled CHI 627D (CH Instrument) electrochemical analyzer. Prior to the preparation of the Ni 3 S 2 electrodes, Ni foam substrates were ultrasonically washed with 6 M HCl, deionized water and ethanol for 15 min, respectively. After that, the pre-cleaned Ni foams were dried under vacuum at 60 • C for 12 h. The PD deposition of Ni 3 S 2 thin film was conducted within the potential range between −1.2 V and 0.2 V vs. Ag/AgCl at a scan rate of 5 mV s −1 for 6 sweep cycles. The Ni 3 S 2 electrodeposition by PR deposition mode was performed by adjusting a variety of deposition parameters including pulse-on potential, pulse-reversal potential and duty cycle. It should be noted that the pulse frequency of PR mode was set at 0.1 for all PR deposition in this work. The duty cycle and pulse frequency in this study are defined as following.
duty cycle = t pulse-on t pulse-on + t pulse-reversal [1] pulse frequency = 1 t pulse-on + t pulse-reversal [2] where t pulse-on and t pulse-reversal represents the period at pulse-on potential and pulse-reversal potential, respectively. The corresponding parameters of PR deposition mode for preparing Ni 3 S 2 electrodes were listed in Table I. It should be noticed that the Ni 3 S 2 electrodes prepared by using PD and PR modes were designated as NSPD and NSPR-1∼ NSPR-5. After the electrodeposition, the as-deposited Ni 3 S 2 electrodes were rinsed with deionized water and subsequently dried in vacuum at 60 • C for 12 h. The mass loading of all deposited Ni 3 S 2 thin films were controlled at approximately 1.4 mg measured by a microbalance (MettlerToleDo) with an accuracy of 0.01 mg. Commercial activated carbon fiber cloth (ACFC) with a specific surface area of 2000 m 2 g −1 was purchased from Taiwan carbon technology company and directly used as anode material for hybrid SCs.
Assembly of hybrid SCs.-The sandwich-type hybrid SCs was fabricated by employing the Ni 3 S 2 prepared by PR deposition mode as cathode, ACFC as anode and filter paper as separator, respectively. The total mass loading of active materials in the hybrid SC was controlled at ca. 9.2 mg. The sandwich-type hybrid SC was further encapsulated by the film of aluminum plastic. Afterward, the 1 M KOH electrolyte was injected into the hybrid SCs.
Characterizations and electrochemical measurements.-The crystal structure and surface morphology of the deposits were characterized by using a grazing incident X-ray diffraction (GIXRD, Rigaku D/TTRAX III, Japan) and field-emission scanning electron microscope (FESEM, JSM-7000F).
The capacity performance of various Ni 3 S 2 electrodes was investigated by cyclic voltammetry (CV) and galvanostatic charge/ discharge tests in 1 M KOH aqueous electrolyte using the aforementioned electrochemical analyzer. The CVs were recorded within a potential interval ranging from −0.2 V to 0.6 V vs. Ag/AgCl at different scan rates from 5 mV s −1 to 50 mV s −1 . The galvanostatic charge/discharge tests were performed within a potential range of 0-0.5 V vs. Ag/AgCl at charge/discharge current densities ranging from 2 to 32 A g −1 . The electrochemical impedance spectroscopy (EIS) measurements were performed at a constant dc bias potential of 0.3 V vs. Ag/AgCl in the frequency range from 100 kHz to 10 mHz with an amplitude of 5 mV by using a potentiostate (IM 6, Zahner) equipped with a frequency analyzer (Thales). The resultant EIS spectra were then simulated by means of Zview software.

Results and Discussion
Characterization of Ni 3 S 2 electrodes.- Fig. 1 shows the XRD patterns of theNi 3 S 2 electrodes prepared by using PD and PR deposition modes. It can be apparently found that all diffraction peaks of the deposits can be assigned to the pure phase of Ni 3 S 2 [JCPDS NO. 44-1418] except to the three strong signals (2θ = 44.4 • , 51.7 • and 76.4 • ) originating from the Ni foam substrate. It should be noted that no XRD pattern is presented for NSPR-4 since almost no mass change is observed after the deposition. This can be explained by the reason that the dissolution rate at the reversal period is almost equal to the deposition rate at the pulse-on period with such low duty cycle. Figs. 2a-2e displays the FESEM images of various Ni 3 S 2 deposits. Obviously, all Ni foam substrates are fully and uniformly covered by the Ni 3 S 2 thin films. The corresponding magnified FESEM images are depicted in the inset of Figs. 2a-2e to further display the morphology details of the resultant Ni 3 S 2 thin films. As expected, the surface morphology of the NSPD electrode (Fig. 2a) is observed to be constructed with nanoflakes, which is well consistence with our previous study. 12 On the basis of the preliminary assessment, only small amount deposits on Ni foam substrates were observed while using the PR deposition mode, in which the pulse-on potential was set at more positive potential than −1 V vs. Ag/AgCl while accompanying with a constant pulse-reverse potential of 0 V vs. Ag/AgCl. This result indicates their low mass loading of Ni 3 S 2 deposits on the Ni foam substrates. For instance, only ca. 0.5 mg of Ni 3 S 2 was deposited on the substrate while setting the pulse-on potential at −0.9 V vs. Ag/AgCl for 900 PR cycles. Although the resultant NSPR-1 electrode demonstrates the similar surface morphology to that of the NSPD electrode (see Fig. 2b), it was found that the mass loading of Ni 3 S 2 thin film just increased a little bit while even doubling the PR cycles, which cannot be increased to that of the NSPD electrode (ca. 1.4 mg). As for employing pulse-on potential of −1 V vs. Ag/AgCl for the PR deposition mode, the mass loading of the NSPR-2 electrode can easily reach the level of the NSPD electrode by turning the PR cycles. As can be seen in Fig. 2c, more uniform Ni 3 S 2 thin film with less formation of cracks is deposited on the Ni foam substrate and its microstructure is constructed with more ordered interlaced nanoflakes while compared with the NSPD and NSPR-1 electrodes. This nanostructure with ordered open pores would be beneficial to the electrolyte penetration within the electrodes. Consequently, the pulse-on potential is set as −1 V vs. Ag/AgCl for the optimized PR deposition mode. As adjusting the pulse-reverse potential from 0 V to −0.1 V vs. Ag/AgCl, it is observed that interlaced flake-like morphology disappears for the NSPR-3 electrode (Fig. 2d). While using the low duty cycle of 0.3 for the PR deposition (NSPR-4 electrode), the dissolution rate was found to be almost equal to the deposition rate and no mass loading before and after the deposition was presented. Nevertheless, the increase in duty cycle from 0.5 to 0.7, the resultant NSPR-5 electrode is composed of the obscure Ni 3 S 2 flakes due to the increased deposition rate.   at various scan rates from 5 mV s −1 to 50 mV s −1 . All curves exhibit similar shape, and the shape of the CV curves represents distinct battery-type characteristics, which is totally different from the closely ideal rectangular CV shape for EDLCs. In particular, only one pair of redox peaks is observed in the all CV curves within the potential range from −0.2 to 0.8 V vs. Ag/AgCl, indicating that the electrochemical capacity of the Ni 3 S 2 electrode mainly originates from the purely faradaic reaction. The electrochemical reaction corresponding to the redox peaks of Ni 3 S 2 electrodes in alkaline electrolyte can be expressed as follows: 10,11 Typically, a higher current density and larger area surrounded by a CV curve indicate higher electrochemical capacity. It can be apparently found that NSPR-1 possesses the highest current density and the largest area under the current-potential curve among all electrodes. This signifies that NSPR-1 may have the highest specific capacity among all Ni 3 S 2 electrodes. 26 Since the mass loading for the NSPR-1 electrode is much lower compared to other electrodes, it would be unfeasible for practical applications. 27 Except the NSPR-1 electrode, the NSPR-2 electrode exhibits a much larger area surrounded by the current-potential curve than those of NSPD, NSPR-3 and NSPR-5 electrodes, suggesting its higher specific capacity and superior capacitive behavior. On the basis of the CV curves, the corresponding specific capacity of NSPD, NSPR-2, NSPR-3 and NSPR-5 electrodes can be estimated as 122.1, 199.4, 69.0 and 72.5 mAh g −1 at the scan rate of 5 mV s −1 , respectively. It signifies that the NSPR-2 electrode can indeed deliver the higher capacity than those of the NSPD, NSPR-3 and NSPR-5 electrodes.
To further evaluate the capacity performance, series of galvanostatic charge/discharge tests were carried out in 1 M KOH aqueous electrolyte at various current densities from 2 Ag −1 to 32 Ag −1 in a potential window ranging from 0 to 0.5 V vs. Ag/AgCl. Fig. 4 presents the recorded charge/discharge curves of the NSPD, NSPR-2, NSPR-3 and NSPR-5 electrodes. The obvious plateaus in the charge/discharge curves reveal that the capacity is dominantly attributed to faradaic reactions. The specific capacities of the different Ni 3 S 2 electrodes at different current densities can be calculated according to the following equation, and the results are summarized in Fig. 5.
Where is the specific capacity (mAhg −1 ), I is the discharge current (mA), t is the discharge time (h), and m is the mass of the electroactive materials (g). The NSPD electrode shows specific capacity of 102.9 mAh g −1 at 2 A g −1 , 92.2 mAh g −1 at 4 A g −1 , 83.8 mAh g −1 at 8 A g −1 , 74.4 mAh g −1 at 16 A g −1 and 60.4 mAh g −1 at 32 A g −1 , respectively. The Q m values of all Ni 3 S 2 electrodes are found to decrease with increasing charge/discharge current density. Among all Ni 3 S 2 electrodes, the NSPR-2 electrode exhibits the highest specific capacity of 179.5 mAh g −1 at 2 A g −1 , while the capacity remains 105.9 mAh g −1 at 32 A g −1 , presenting its excellent capacity performance. Therefore, the NSPR-2 electrode could be regarded as the optimized one of the Ni 3 S 2 electrodes prepared by the PR deposition mode. To further compare the electrochemical stability of the optimized PR-deposited Ni 3 S 2 electrode (NSPR-2) with the NSPD electrode, the galvanostatic charge/discharge cycling tests were performed in 1 M KOH aqueous electrolyte at a constant current density of 8 A g −1 for 2500 consecutive cycles, as depicted in Fig. 6. Both electrodes display two-stage capacity decay, representing region I (1∼500 cycles) and region II (500∼2500 cycles), respectively. At the region I, the specific capacity of NSPD and NSPR-2 electrodes rapidly decreases, thus resulting in the decrease of nearly 32% and 45% of the corresponding initial discharge capacity. As for the region II, the relatively less decrease in capacity is observed with further cycling for another 2000 cycles, in which only approximately 17% and 28% of capacity loss for NSPD and NSPR-2 electrodes compared to that at the 500th cycle, respectively. After the consecutive 2500 cycles, the stable specific capacity of NSPD and NSPR-2 electrodes is found to be 51.2 and 54.5 mAh g −1 , respectively.
To gain insight into the electrochemical kinetics and aged capacity performance of the NSPD and NSPR-2 electrodes before/after the cycling tests, EIS analyses were carried out in the frequency range of 10 kHz to 0.01 Hz. Fig. 7 shows the Nyquist plots the NSPD and NSPR-2 electrodes before and after the long-term cycling tests. The Nyquist plots are mainly composed of a semicircle at the highfrequency region and a straight line at low-frequency region. The inset in Fig. 7a is the equivalent-circuit model employed to simulate the resultant EIS spectra. The real axis intercept in high-frequency region represents the bulk solution resistance (R s ). The diameter of the semicircle in the high-frequency region is associated with the Faradaic interfacial charge-transfer resistance (R ct ). The constant phase element (CPE) is used to account for a double-layer capacitance, Z W and C F represent the Warburg impedance and the Faradaic pseudocapacitor, respectively. Before the cycling tests, the R s values of NSPD (∼1.43 ) and NSPR-2 (∼1.32 ) electrodes are found to be similar. It is worthy noted that the NSPR-2 electrode possesses much lower R ct value (4.18 ) than that of the NSPD electrode (10.41 ), indicating that the NSPR-2 electrode shows a higher rate of charge transport than the NSPD electrode. [28][29][30] Therefore, the high initial discharge capacity of the NSPR-2 electrode can be ascribed to its excellent charge-transfer characteristics. After the cycling test, the NSPD and NSPR-2 electrodes still have similar R s values, (∼1.45 ) and (∼1.34 ), respectively. However, the R ct value of the NSPD and NSPR-2 The inset in Fig. 7a is the equivalent circuit model used to simulate the resultant Nyquist plots.  electrodes are increased from 10.41 to 12.38 and 4.18 to 11.44 , respectively. It signifies that the NSPD and NSPR-2 electrodes have comparable R s and R ct values after the long-term cycling tests, therefore resulting in the similar discharge capacity after 2500 charge/discharge cycles.
To elaborate the reasons for the decayed capacity and increased R ct value after the long-term cycling tests, the composition and surface morphology of the NSPD and NSPR-2 electrodes were examined. Fig. 8a shows the GIXRD patterns of the cycled NSPD and NSPR-2 electrodes. It represents that the XRD patterns of both cycled electrodes still retain as their initial XRD patterns (Fig. 1), indicating the composition of both electrodes is without any change after 2500 charge/discharge cycling tests. Figs. 8b and 8c present the FESEM image of NSPD and NSPR-2 electrodes after 2500 cyclic chargedischarge tests. It can be observed that the pore size of interlaced nanoflakes of both NSPD and NSPR-2 electrodes becomes relatively larger after the long-term cycling test, which could be possibly ascribed to the volume expansion for the Ni-based electrodes after consecutive charge/discharge tests. 31 Consequently, both cycled NSPD and NSPR-2 electrodes have similar surface morphology. The evolution in surface morphology for both electrodes could be responsible to the significantly decreased capacity and the increased R s and R ct values after the cycling test. This also suggests that the capacity decay and the increased R s and R ct values for the electrodeposited Ni 3 S 2 electrodes would significantly occur after long-term cycling tests regardless of deposition modes due to the surface evolution during charge/discharge cycling. The capacity of the electrodeposited Ni 3 S 2 electrodes would reach stable while their surface morphology retains unchanged.
In order to evaluate its practical application, a hybrid SC device was assembled by using the optimized Ni 3 S 2 prepared by the PR deposition mode (NSPR-2) as positive electrode and ACFC as anode, respectively. The electrochemical behavior of ACFC was firstly measured using CV method at different scan rates from 5 to 50 mV s −1 . It can be observed that the nearly rectangular-like shapes are observed in the operating potential range (−1∼0 V vs. Ag/AgCl), as shown in Fig. 9a. Furthermore, the galvanostatic charge/discharge curves of the ACFC electrode shows highly symmetric and linear curve at galvanostatic current densities ranging from 1 A g −1 to 8 A g −1 , as shown in Fig. 9b. The ACFC electrode shows specific capacity of 34.7 mAh g −1 at 1 Ag −1 , 32.6 mAh g −1 at 2 A g −1 , 30.4 mAh g −1 at 4 Ag −1 and 27.1 mAh g −1 at 8 A g −1 , respectively. As for fabrication of the hybrid SC, the charge of the NSPR-2 cathode and the ACFC anode should balance according to the following equation: where m (g) is the mass, I(A) is the discharge current for positive (+) and negative (−), and t is the discharge time (s) for positive (+) and negative (−) electrodes, respectively. The optimum mass loading ratio between NSPR-2 and ACFC should be m + /m − = 0.18 in the hybrid SCs device. Before evaluating hybrid SCs, the sum of the potential range of NSPR-2 cathode and ACFC anode were evaluated by using CV measurements with the potential range from −1 V to 0.8 V vs. Ag/AgCl, as shown in Fig. 10a. Fig. 10b shows the CV curves of NSPR-2//ACFC hybrid SCs at various potential windows in 1 M KOH electrolyte at 50 mV s −1 . It clearly depicts that the NSPR-2//ACFC hybrid SC can deliver a stable working voltage of 1.8 V. Furthermore, Fig. 10c displays the CV curves of NSPR-2//ACFC hybrid SCs measured at different scan rate of 5, 10, 20 and 50 mV s −1 . Interestly, the NSPR-2//ACFC hybrid SC exhibits both of the EDLC and Faradic electrode properties within the potential window of 0-1.8 V. Fig. 11a shows the galvanostatic charge/discharge curves of the NSPR-2//ACFC hybrid SCs measured at different current densities. As can be seen in Fig. 11a, the charge/discharge curves of NSPR-2//ACFC hybrid SCs show the behavior of the EDLC capacitors and battery-type electrodes within a stable potential window of 0-1.8 V. Moreover, the Q m values of NSPR-2//ACFC hybrid SCs at 2, 4, 8 and 16 A g −1 are estimated to be 26.7, 23.8, 18.7 and 10.6 mAh g −1 , respectively. In Fig. 11b, the Ragone plot related to power densities (P) and energy densities (E) is further used to evaluate the performance of the NSPR-2//ACFC hybrid SCs. The energy density and power density of the hybrid SCs were calculated as follows: 32 where Q m is calculated based on the loading mass of active materials in anode and cathode, V is the voltage, t 1 , t 2 is the start time and end time during discharging and t equal to (t 2 -t 1 ) is the discharge duration, respectively. The corresponding energy density and power density at 2  33 To further evaluate the electrochemical stability of the NSPR-2//ACFC hybrid SCs, the galvanostatic charge/discharge cycling tests were performed at a constant current density of 2 Ag −1 for 5000 consecutive cycles, as depicted in Fig. 11c. The 40% loss in Q m value ofNSPR-2//ACFC was observed after cycling of 1-400 charge/discharge cycles, which can be ascribed to the surface evolution of NSPR-2 electrode. Nevertheless, the relatively less decrease in capacity is found with further cycling for another 4600 cycles, in which only approximately 3% loss of specific capacity compared to that at the 400th cycle. To display the practical application of the NSPR-2//ACFC hybrid SCs, two hybrid SCs were assembled in series for lightening up a green light-emitting diode (LED) of driving voltage 3.0 V. Fig. 11d depicts that two NSPR-2//ACFC hybrid SCs in series can successfully lighten up the green LED.

Conclusions
In summary, the electrodeposited Ni 3 S 2 thin films with excellent capacitive properties have been successfully prepared using a PR deposition mode. After the optimization of the deposition parameters, the optimized NSPR-2 electrode delivers remarkable specific capacity up to 179.5 mAh g −1 at 2 A g −1 and its discharge capacity even retains 105.9 mAh g −1 at high charge/discharge current density of 32 A g −1 in 1.0 M KOH aqueous electrolyte. Moreover, the specific capacity retention of the NSPR-2 electrode can still achieve ac.78% after cycling of 500-2500 cycles at a high current density of 8 A g −1 . Furthermore, the NSPR-2//ACFC hybrid SCs can be reversibly charged and discharged at a stable cell voltage of 1.8 V and generates an impressive specific capacity of 26.7 mAh g −1 at charge/discharge current density of 2 A g −1 , delivering a maximum energy density of 26.4 Wh kg −1 at a power density of 1978 Wkg −1 . More importantly, the NSPR-2//ACFC hybrid SC still retains 97% of 400th capacity after 5000 consecutive charge/discharge test at a charge/discharge current density of 2 A g −1 , indicating its great electrochemical stability. In views of the impressive capacitive performance for the NSPR-2//CFC hybrid SC, it can be considered as a promising potential low-cost device for energy storage.