Stabilizing NiCo2O4 hybrid architectures by reduced graphene oxide interlayers for improved cycling stability of hybrid supercapacitors

Kyu Hyun Oh , Girish Sambhaji Gund and Ho Seok Park *
School of Chemical Engineering, Sungkyunkwan University (SKKU), Suwon, 440-746, Republic of Korea. E-mail: phs0727@skku.edu

Received 2nd May 2018 , Accepted 4th July 2018

First published on 4th July 2018


We demonstrate nickel cobaltite@reduced graphene oxide (NiCo2O4@rGO) hybrid architectures directly deposited on nickel-foam (NF). The rGO interlayers restrict the growth of NiCo2O4 nanoneedles into smaller and thinner dimensions compared to NiCo2O4 without rGO layers, providing kinetic and structural stability for hybrid architectures. Accordingly, the NiCo2O4@rGO hybrid on NF achieves a specific capacitance of 1427 F g−1 at 8 A g−1, a coulombic efficiency of 96.2%, and a capacitance retention of 83.8% over 10[thin space (1/6-em)]000 cycles, which are greater than 1036 F g−1, 89.1%, and 40.8% of NiCo2O4 on NF in an aqueous 2 M KOH electrolyte. In order to enlarge the potential window of the aqueous system, hybrid supercapacitors (HSCs) are configured using the NiCo2O4@rGO hybrid on NF as a positive electrode and rGO on NF as a negative electrode. The HSCs exhibit a good cycling stability of 81.1% over 10[thin space (1/6-em)]000 cycles, delivering maximum energy and power densities of 25.24 W h kg−1 and 21.42 kW kg−1.


1. Introduction

Electrochemical energy storage devices with high energy efficiency, power capability and long life have been intensively investigated due to global climate change, utilization of renewable energy, and emerging markets of electric vehicles.1–4 In particular, the electrochemical double layer capacitor (EDLC), a type of supercapacitor, which electrostatically stores charge at the electrode surface, is considered a promising energy storage device owing to its high power density, long-term cycling stability and high efficiency. However, its application is limited by low energy density. In order to remove this impediment,5 high capacity active materials have been developed and used through material hybridization. The redox-active materials with high capacity can be synergistically hybridized with EDLC materials with high electrical conductivity and large surface area on both material and electrode levels.6–10

Among the various high capacity materials, spinel-type NiCo2O4 is an attractive active material due to its high capacity, low cost, natural abundance, and low toxicity.11,12 In a similar manner to other transition metal oxides, a poor cycling stability of NiCo2O4 arising from drastic volume expansion remains a critical challenge.13 Moreover, the redox capability of NiCo2O4 is revealed in aqueous electrolytes, which results in limited energy density due to a narrow potential window. Various carbon nanomaterials such as reduced graphene oxide (rGO), carbon nanotubes, fullerene, activated carbon, and so on have been used to provide electrochemical stability and electronic conductivity to high capacity active materials.14,15 Despite several studies on the NiCo2O4 on conductive carbon nanomaterials,14,15 a systematic investigation of the effect of rGO interlayers on the growth of NiCo2O4 is yet to be explored. Moreover, the nanoarchitecturing and modification of NiCo2O4 by rGO interlayers is expected to achieve kinetic and structural stability of hybrid materials.

Herein, we demonstrate NiCo2O4@rGO hybrid architectures directly grown on nickel-foam (NF), where nanoscale rGO interlayers restrict the dimensions of NiCo2O4 nanoneedles into small length and diameter. Moreover, the strong adhesion of rGO interlayers between NiCo2O4 and NF leads to excellent electrochemical stability, while providing high capacitance and coulombic efficiency. Configuring the NiCo2O4@rGO hybrid on NF as a positive electrode with rGO on NF as a negative electrode, the potential window of the hybrid supercapacitor (HSC) can be enlarged in aqueous 2 M KOH electrolyte to further improve the energy density.

2. Experimental section

2.1 Materials

2% Graphene oxide (GO) was purchased from Angstron Materials Corporation, USA. Nickel nitrate hexahydrate (Ni(NO3)2·6H2O) and cobalt nitrate hexahydrate (Co(NO3)2·6H2O) were purchased from Junsei Corporation, Japan. Urea (NH2COHNH2) was purchased from Wako Corporation, Japan. Ethyl alcohol (C2H5OH) and potassium hydroxide (KOH) were purchased from Daejung Corporation, Republic of Korea. Lastly, NF was purchased from MTI Corporation, USA. De-ionized (DI) water was used as the solvent for all the experiments. All the chemicals were used as received without further purification. The NF substrate was cleaned to remove foreign substances on the surface through 3 M HCl acid treatment with ultra-sonication, followed by rinsing in DI water to remove acid content.

2.2 Preparation of the rGO electrode

rGO was coated on the pre-treated NF substrates through a binder-free hydrothermal method, as discussed later in detail. The diluted 0.5% GO solution (20 ml) with ethanol (1 ml) was stirred for 30 minutes and transferred into a Teflon-liner stainless steel autoclave, which was further heated at a rate of 5 K min−1 up to 393 K for 2 hours. The GO sheets were reduced through a hydrothermal treatment with ethanol and adsorbed directly on the NF surface. Furthermore, the loosely attached rGO sheets were removed through rinsing the rGO on NF electrodes in DI-water at least 10 times and dried at 353 K for 12 hours. These electrodes were used as a substrate to coat NiCo2O4 and negative electrodes. The concentration of GO solution was optimized for rGO loading and the excess rGO weakly adsorbed was completely removed to construct the hybrid device.

2.3 Preparation of the NiCo2O4/rGO hybrid on NF

NiCo2O4 was synthesized and directly coated on the as-obtained rGO on NF electrode using hydrothermal treatment as explained below. Specifically, an aqueous homogeneous solution of 0.1 M Ni(NO3)2·6H2O, 0.2 M Co(NO3)2·6H2O, and 0.9 M urea was prepared by stirring for 30 minutes. Then, the solution was transferred into a Teflon-liner and rGO on NF electrodes were dipped vertically in the solution using a support inside the liner. Furthermore, the Teflon-liner was carefully transferred into the stainless steel autoclave and heated at 398 K for 1 h. After cooling down the autoclave to room temperature, the NiCo2(OH)x/rGO NF electrodes were washed using DI water and ethanol, and then dried at 353 K for 12 h. Finally, these electrodes were heated at 548 K for 2 hours to convert the hydroxide phase to an oxide one. Additionally, NiCo2O4 was directly coated on NF using the same process to allow a comparative study.

2.4 Materials characterization

The morphologies of all electrodes were recorded using a high resolution field emission scanning electron microscope (FE-SEM, Carl Zeiss). Transmission electron microscopy (TEM) images were collected on a JEM-3010 high-resolution transmission electron microscope (HR-TEM, 300 kV). Elemental mapping in scanning TEM (STEM) mode was performed using a probe focused to 0.2 nm and a camera length of 20 cm. The scan raster was 512 × 512 points with a dwell time of 8.5 seconds per scan. An X-ray diffraction (XRD) study was conducted using a Miniflex 300/600 system (Rigaku) at a grazing angle using a Cu Kα (λ = 1.5406 Å) source, in the 2θ range of 5–80°. Raman spectroscopy investigation of all electrodes was performed with a Horiba Jobin Yvon dispersive-Raman spectrometer equipped with a CW Ar-ion laser (514.5 nm). X-ray photoelectron spectroscopy (XPS) measurements were performed on a monochromatic Al-Kα (1486.6 eV) source instrument with Ag 3d5/2 energy resolution. EC-Lab (NEO Science) was used to investigate electrochemical measurements. The electrochemical parameters were estimated through fitting electrochemical impedance spectroscopy (EIS) data in EC-Lab software.

2.5 Electrochemical measurements

The pristine NF, rGO, NiCo2O4 and hybrid NiCo2O4/rGO electrodes on NF were cut into 1 cm2 samples and then used as working electrodes. The electrochemical performances were measured in a three electrode system with 2 M KOH electrolyte using cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and EIS. Platinum wire and an Ag/AgCl electrode were used as the counter and reference electrodes, respectively. The specific capacitance (Cs) of the electrodes was calculated from CV and GCD measurements, using the following equations, respectively:
 
image file: c8ta04038a-t1.tif(1)
 
image file: c8ta04038a-t2.tif(2)
where v (mV s−1) is the scan rate, ΔV is the potential window, I(V) denotes the current response (mA), Id (mA) is the constant discharge current density, td is the discharge time and m is the loading mass of the electroactive material on the NF, which is estimated through the following equation:
 
m = m2m1(3)
where m2 is the mass of the NF with the electroactive material and m1 is the mass of the NF without the electroactive material.

The mass ratio between the positive (m+) and negative (m) electrodes was calculated to preserve a charge balance q+ = q with the intention of balancing the hybrid device using the following equation.

 
image file: c8ta04038a-t3.tif(4)
where q+/q, m+/m, C+/C, and ΔV+V are the charges, masses, specific capacitances, and potential windows of the positive (+)/negative (−) electrodes, respectively. Hence, the mass ratio of the positive-to-negative electrode was calculated to be 0.42[thin space (1/6-em)]:[thin space (1/6-em)]1 to balance the two electrodes of the HSC. Accordingly, the mass loadings of active materials in positive and negative electrodes were well adjusted to 0.74 and 1.77 mg to fabricate a coin-cell type HSC, respectively. Furthermore, we considered the total mass (2.51 mg) of the two electrodes to calculate the specific capacitance of the HSC.

The energy density (E, W h kg−1) and power density (P, W kg−1) of the HSC were calculated using the following equation.

 
image file: c8ta04038a-t4.tif(5)
 
image file: c8ta04038a-t5.tif(6)
where ΔV is the potential window of the HSC device.

3. Results and discussion

The binder-free electrodes of the NiCo2O4@rGO hybrid on NF were prepared by a two-step hydrothermal method, as schematically illustrated in Fig. 1a. Initially, the concentrations of GO solution in DI water were controlled at 0.05, 0.1, 0.2, 0.5, and 1.0%. In this work, 0.5% was optimum for a uniform coating of rGO on NF. In the first step, the GO sheets were reduced during the hydrothermal treatment with ethanol acting as a reducing agent at 393 K16 and the resultant rGO sheets were physically adsorbed onto the NF surface. In the second step, NiCo2O4 nanoneedles were grown and deposited on the NF under the second set of hydrothermal conditions of 398 K for 1 hour. Pristine rGO and NiCo2O4 electrodes were prepared for the control samples of rGO on NF and NiCo2O4 on NF, respectively, using the same hydrothermal treatment as the synthetic conditions of the NiCo2O4@rGO hybrid on NF.
image file: c8ta04038a-f1.tif
Fig. 1 (a) Schematic illustration of the synthesis of the binder-free NiCo2O4 and NiCo2O4@rGO hybrid on a NF surface. FE-SEM images of (b and c) the bare NF surface, (d and e) rGO coated on a NF surface, (f and g) NiCo2O4, and (h and i) the NiCo2O4@rGO hybrid on a NF surface.

The morphologies of the bare NF, rGO on NF, NiCo2O4 on NF, and NiCo2O4@rGO hybrid on NF were characterized using FE-SEM results, as shown in Fig. 1b–i. The bare NF showed a three-dimensional (3D) reticulated macroporous structure with a smooth surface (Fig. 1b–c), which provides fast ionic and electronic transportation and a large surface area for uniform distribution of the active material. Particularly, the pore size was in the range of 100 to 500 μm. The reticulated porous framework of NF was preserved, while rGO sheets were uniformly coated onto the smooth surface during the hydrothermal treatment as shown in Fig. 1d. As shown in Fig. 1e and its inset, the rough and crumpled rGO sheets on the NF surface exhibited an interlocked structure that can provide a large accessible surface area to deposit NiCo2O4 with different surface chemistries from the bare NF surface for the growth of nanoneedles. Accordingly, the NiCo2O4 spiky nanoneedles grown onto the rGO-coated NF were 30–40 nm in diameter and <1 μm in length as shown in Fig. 1f and g. These dimensions of the NiCo2O4@rGO hybrid on NF were thinner and shorter compared to >50 nm diameter and >1.5 μm length of NiCo2O4 on NF, which was grown directly onto the bare NF surface (Fig. 1h and i). These different dimensions of NiCo2O4 nanoneedles were attributed to distinctive surface chemistries; the rGO interlayers coated on NF provide hetero-epitaxial growth and the pristine NF surface offers homo-epitaxial growth of nanostructures as schematically shown in Fig. 1a. This hetero-epitaxial growth leads to a different evolution process of the nanoneedles driven by the rGO interlayers from the homo-epitaxial one as a function of reaction time. The rGO surface of the hybrid initiated the formation of nanoflakes at 15 minutes, became thick to jerk out thin trunks for nanoneedles at 30 minutes, and formed a forest of nanoneedles at 1 hour (Fig. S1). On the other hand, the bare NF surface started from the formation of trunks at 15 minutes, followed by constructing a forest of nanoneedles at 30 minutes, and then, resulted in the densification of the forest at 1 hour. The low surface energy offered by rGO interlayers resulted in restricting the growth of NiCo2O4 nanoneedles into smaller dimensions as confirmed by mathematical analysis and contact angle measurements (see ESI note 1 and Fig. S2), yet offering strong adhesion and electrochemical stability. Moreover, the short length and diameter of NiCo2O4 nanoneedles resulted in shortening the diffusion length of ions into active sites and enlarging the accessible surface area.

The microstructure of the NiCo2O4@rGO hybrid on NF was further characterized using STEM-EDS, TEM, HRTEM, and SAED results, as shown in Fig. 2a–d. For the characterization, powder was scratched and collected from the NF electrode, and then dispersed in ethanol and finally dropped on a copper grid. The STEM-EDS mapping revealed the uniform distribution of Ni, Co, O and C elements. The C signals are indicative of the existence of rGO interlayers between NiCo2O4 and NF, but the position of rGO could not be confirmed by the weak intensity and overlap with other signals. As shown in the TEM images (Fig. 2a and b), nanoneedles were aggregated with nanocrystals of around 10 nm and vertically assembled in an anisotropic manner. These nanoneedles contained sharp tips at their ends due to the consumption and depletion of the precursor during the hydrothermal process, showing a nanocrystalline nature as seen in the TEM, HR-TEM and SAED images (inset of Fig. 2b–d). The lattice fringe from the HR-TEM image was found to be 0.243 nm, which corresponds to the (311) plane of spinel structured NiCo2O4. The ring-type SAED pattern supported the polycrystalline feature of the NiCo2O4 nanoneedles as indexed in detail (see Fig. 2d).


image file: c8ta04038a-f2.tif
Fig. 2 (a) Elemental mapping, (b) TEM (the inset is the the high magnification image), (c) HR-TEM, and (d) SAED pattern images of the NiCo2O4@rGO hybrid on NF. (e) XRD patterns of NiCo2O4 on NF and the NiCo2O4@rGO hybrid on NF. (f) Raman spectra of bare NF and rGO, NiCo2O4, and the NiCo2O4@rGO hybrid on NF. (g) Full-scan and (h) high-resolution O 1s XPS spectra of rGO, NiCo2O4 and the NiCo2O4@rGO hybrid on NF.

The XRD spectra of NiCo2O4 and the NiCo2O4@rGO hybrid collected from NF were measured to further confirm their crystalline structures as shown in Fig. 2e. All the diffraction peaks were assigned to the pure spinel crystalline structure of NiCo2O4 (JCPDS no. 73-1702) without additional impurity peaks of Ni- or Co- oxides. Especially, the diffraction peak of the NiCo2O4 and NiCo2O4@rGO hybrid samples at 36.7° corresponded to the (311) plane, suggesting the preferential growth of nanoneedles in this direction in good agreement with the results of HR-TEM and SAED. Additionally, the peak intensity of the (311) plane was changed due to the existence of rGO affecting the preferential growth and the atomic positions of NiCo2O4, confirming the control of nanocrystal growth by heteroepitaxy growth of nanoneedles as supported by FE-SEM. The average crystalline size of NiCo2O4 nanoneedles for the hybrid architecture was calculated to be around 10 nm from the (311) plane using the Debye–Scherer formula (see Fig. S3). These findings about the crystal size obtained from XRD, HR-TEM and SAED measurements matched well with each other, indicating a highly nanocrystalline structure of spinel NiCo2O4.

Fig. 2f shows the RAMAN spectra of the bare NF, rGO on NF, NiCo2O4 on NF, and NiCo2O4@rGO hybrid on NF. Two representative peaks of rGO and the NiCo2O4@rGO hybrid were captured at 1352.1 and 1579.3 cm−1 corresponding to D and G bands of graphene.17 By contrast, the peaks at 184.8, 457.9, 511.1, and 649.1 cm−1 were ascribed to the F2g, Eg, F2g, and A1g modes of spinel phase NiCo2O4, respectively. The additional peak for the NiCo2O4@rGO hybrid at 1096.3 cm−1 originated from the 2LO phonon mode of nanocrystalline NiCo2O4, being consistent with the results from HR-TEM and XRD measurements.18

XPS analysis was performed to analyze the surface chemical composition of the as-prepared samples as shown in Fig. 2g and h. Full survey XPS spectra of the NiCo2O4 and NiCo2O4@rGO hybrid coated on NF samples identified the presence of Ni 2p, Co 2p, O 1s and C 1s, while rGO coated on NF indicated the existence of Ni 2p (related to NF), O 1s and C 1s. As shown in Fig. 2h, the O 1s scan of the rGO sample revealed two peaks, including the most intense one at 530.9 eV and a shoulder at 532.8 eV, corresponding to C–O and C[double bond, length as m-dash]O bands, respectively. This band of oxygen-containing groups on the rGO sheets was attributed to a partial reduction and atmospheric moisture adsorption. On the other hand, both NiCo2O4 on NF and the NiCo2O4@rGO hybrid on NF showed a strong peak at 529.2 eV, where oxygen atoms are included in the metal–oxygen–metal lattice, and weakened peak intensities of oxygen-containing groups.19 The O 1s spectrum of the NiCo2O4@rGO hybrid was broader compared to those of NiCo2O4 and rGO samples due to the contribution of metal–oxygen–carbon interaction.20 A negative peak shifts of Ni 2p and Co 2p by 0.15 eV with respective to pristine NiCo2O4 and a positive peak shift of C 1s for C[double bond, length as m-dash]C bonding of sp2-hybridized C with respective to pristine rGO were observed for the NiCo2O4@rGO hybrid. These peak broadening and shifts were attributed to the interaction between rGO interlayers and NiCo2O4 (see ESI note 2 and Fig. S4).21 Thus, these high resolution and deconvolution fitted XPS spectra confirm the pure phase of NiCo2O4 and its successful hybridization with rGO.

The electrochemical performance of the NiCo2O4@rGO hybrid on NF was measured in a three-electrode configuration in 2 M KOH electrolyte, as shown in Fig. 3. The CV curves of bare NF and NiCo2O4 on NF in the potential range of −0.4 to +0.5 vs. Ag/AgCl, and the NiCo2O4@rGO hybrid on NF in the potential range of −0.4 to +0.6 V vs. Ag/AgCl were measured at 5 mV s−1. The current density of bare NF was negligible not contributing to the capacitance of the hybrid architecture. By contrast, the CV curve of the NiCo2O4@rGO hybrid on NF showed obvious redox peaks over the longer range of potential window compared to NiCo2O4 on NF at different potentials. This means that the oxygen evolution reaction (OER) was inhibited by rGO interlayers and shifted to a higher potential. Since the hetero-type interface provides kinetic and electrochemical stability arising from interfacial resistance and intrinsic stability of rGO,22 the NiCo2O4@rGO hybrid on NF was capable of enlarging the stable potential window for higher energy density. These electrochemical reactions of NiCo2O4 nanoneedles in alkaline electrolyte can be described by the following equations:23

 
NiCo2O4 + OH + H2O ↔ NiOOH + 2CoOOH + e(7)
 
CoOOH + OH ↔ CoO2 + H2O + e(8)


image file: c8ta04038a-f3.tif
Fig. 3 (a) CV curves of the bare NF, NiCo2O4, and NiCo2O4@rGO hybrid on NF at 10 mVs−1, (b) CV curves of the NiCo2O4@rGO hybrid on NF at various scan rates from 5 to 100 mV s−1, and (c) GCD profiles of the NiCo2O4@rGO hybrid on NF at different current densities. (d) Variation of specific capacitance and coulombic efficiency as a function of current density. (e) Specific capacitance as a function of GCD cycle number at 16 A g−1, for NiCo2O4 and the NiCo2O4@rGO hybrid on NF. (f) Schematic illustration of the role of rGO interlayers to achieve stable electrochemical reactions within the hybrid material. (g) Nyquist plots of the bare NF, rGO, NiCo2O4, and NiCo2O4@rGO hybrid on NF (insets represent the high frequency region).

Fig. 3b shows the CV curves of the NiCo2O4@rGO hybrid on NF at varying scan rates from 5 to 100 mV s−1. A pair of redox peaks of the NiCo2O4@rGO hybrid on NF was observed for all sweep rates within the potential range of −0.4 to 0.6 V vs. Ag/AgCl (even for NiCo2O4 on NF also, see Fig. S5a), following the faradaic redox reactions of eqn (7) and (8). In addition, the CV profile was well maintained even at a high rate of 100 mV s−1 (Fig. 3b and see S5a), demonstrating a fast electrochemical response. The GCD curves of NiCo2O4 on NF and the NiCo2O4@rGO hybrid on NF at various current densities from 8 to 80 A g−1 were measured within the potential range of 0 to 0.45 V vs. Ag/AgCl are shown in Fig. 3c (see Fig. S5b). The non-linear sloped shape of the GCD profile was more prominent for the NiCo2O4@rGO hybrid on NF compared to NiCo2O4 on NF (see Fig. S5c). The GCD profile of NiCo2O4 on NF showed a flat plateau regime during charging, which is close to battery-type charge storage (see Fig. S5c). This GCD feature and limited potential window of NiCo2O4 on NF might be associated with lower OER potential compared to hybrid NiCo2O4@rGO. As shown in the GCD measurements of NiCo2O4 on NF and the NiCo2O4@rGO hybrid on NF (Fig. 3d), the specific capacitances and coulombic efficiencies of the NiCo2O4@rGO hybrid on NF were 1427 F g−1 and 96.2% at 8 A g−1, which were greater than 1036 F g−1 and 89.1% of NiCo2O4 on NF. At 64 A g−1, the NiCo2O4@rGO hybrid on NF achieved 938 F g−1 and 100.0%, which were also greater than 727 F g−1 and 99.5% of NiCo2O4 on NF. Thus, the coulombic efficiency was greater for the NiCo2O4@rGO hybrid on NF, while the rate capability of the NiCo2O4@rGO hybrid on NF was lower than that of NiCo2O4 on NF. The specific capacitances of our nanoneedles were compared with those of previous literature reporting different nanostructures of NiCo2O4 as summarized in Table 1.15,24–30 EIS data of the bare NF, rGO, NiCo2O4 and NiCo2O4@rGO hybrid were collected in the frequency range of 0.1 Hz to 100 kHz to analyze the charge transportation as shown in Fig. 3g. As shown in the Nyquist plots, RESR of the NiCo2O4@rGO hybrid electrode (0.91 Ω) was greater than that of the NiCo2O4 electrode (0.30 Ω) due to multiple hetero-contacts and thickness resistance of rGO interlayers. This finding indicates that the electrochemical faradaic reactions occurring on the NiCo2O4 electrode are fast but irreversible-type, due to its lower interfacial resistance and long length of nanoneedles. Despite a low rate capability as a consequence of interfacial resistance, the high specific capacitance and reversible feature of the NiCo2O4@rGO hybrid on NF were due to the larger accessible surface area and short diffusion length of nanoneedles. As shown by the 10[thin space (1/6-em)]000 cycle stability test at 16 A g−1 in Fig. 3e, the specific capacitance and coulombic efficiency of the NiCo2O4@rGO hybrid on NF decreased from 1244 to 1043 F g−1 (capacitance retention of 83.8%) and 99.1 to 99.0%. By contrast, NiCo2O4 on NF decreased from 1031 to 421 F g−1 (capacitance retention of 40.8%) and 98.2 to 97.2% (see Fig. S5d). The long-term cycling stability of the NiCo2O4@rGO hybrid on NF with good coulombic efficiency was associated with the existence of rGO interlayers, which acted as a buffering region to release localized stresses during a volume expansion and contraction and tight binding between NiCo2O4 and NF.13,31,32 In order to support these findings, the FE-SEM images of NiCo2O4 and the NiCo2O4@rGO hybrid on NF were analyzed before and after 10[thin space (1/6-em)]000 charging/discharging cycles (see Fig. S6). The destruction of the NiCo2O4@rGO hybrid on NF was almost negligible even after 10[thin space (1/6-em)]000 cycles, while the nanoneedle structure of NiCo2O4 on NF was significantly demolished and only the trunk part remained. This intact nanoneedle structure of the hybrid confirmed the role of rGO as a buffering interlayer as schematically represented in Fig. 3f.

Table 1 Comparison of the as-designed nanoneedles with previously reported nanostructures of NiCo2O4
No. Morphology Synthesis method Specific capacitance (F g−1) Ref.
1 Agglomerated particles Chemical co-precipitation 351 (1 A g−1) 24
2 Nanoflat-like particles Chemical co-precipitation 764 (2 mV s−1) 25
3 Flower-structure Hydrothermal 658 (1 A g−1) 26
4 Nano-belts Solvothermal 238 (1 mV s−1) 27
5 Interconnected nano-sheets Solvothermal 315 (1 mV s−1) 27
6 Nano-chestnuts Solvothermal 675 (1 mV s−1) 27
7 Nano-sponges Solvothermal 832 (1 mV s−1) 27
8 Nanoneedles Hydrothermal 1118.6 (5.56 mA cm−2) 15
9 Nanowires Hydrothermal 571 (0.5 mA cm−2) 28
10 Nanosheets Chemical bath deposition 899 (1 A g−1) 29
11 Nanoflakes Hydrothermal 1270 (1 A g−1) 30
12 Nanoneedles Hydrothermal 1427 (8 A g−1) Current work


In order to configure the hybrid device, binder-free rGO was chosen as the counter negative electrode. As shown in the CV profiles in Fig. 4a, the potential window of the rGO electrode ranged from 0.0 to −1.0 V vs. Ag/AgCl in a 2 M KOH aqueous electrolyte. The CV curve features the rectangular shape of the EDLC with limited capacitance and 1 V operating potential. In order to improve the energy density of the rGO EDLC, the high capacitance NiCo2O4@rGO hybrid on NF with different operating potential ranges was configured into a single device, the so-called hybrid HSC. In order to balance the positive and negative electrodes, the specific capacitance of the rGO electrode was measured (see Fig. S7). The mass loadings of active materials on both electrodes were controlled following eqn (4) at a moderate rate of 50 mV s−1, as shown in Fig. 4b. A 2 M KOH aqueous electrolyte was soaked in a glass fiber separator, which was packed in a coin cell. As shown in the CV and GCD profiles of the as-fabricated HSCs in Fig. 4c and d, respectively, the operating potential window was enlarged up to +1.45 V, which is wider than those of both symmetric devices based on rGO and the NiCo2O4@rGO hybrid. The quasi-rectangular and non-linear symmetric shape of the CV and GCD curves was preserved at various rates from 5 to 200 mV s−1 and from 1 to 32 A g−1, indicating the successful hybridization of the EDLC and faradaic charge storage mechanisms with good reversibility. As shown in Fig. 4e, the specific capacitances of HSC were varied from 86.4 to 53.1 F g−1 with high coulombic efficiencies of >94.8%. For the real-world applications of HSC, the energy and power densities in a Ragone plot are presented in Fig. 4h. The as-fabricated HSCs showed the maximum energy density of 25.24 W h kg−1 with a power density of 727 W kg−1 at 1 A g−1, as well as a minimum energy density of 16.11 W h kg−1 with a power density of 21[thin space (1/6-em)]417 W kg−1 at 32 A g−1 (Fig. 4h). Even over 10[thin space (1/6-em)]000 cycles at 6 A g−1 in Fig. 4f, the GCD curves of the HSC were well maintained with little decrement of charge–discharge time. As shown in Fig. 4g, the specific capacitance decreases from 75.5 to 61.2 F g−1 representing 81.1% capacitance retention even after 10[thin space (1/6-em)]000 cycles. The coulombic efficiency was very minutely reduced from 98.6 to 98.3% after 10[thin space (1/6-em)]000 cycles. These results confirmed the long-term cycling stability of the as-fabricated HSC due to the synergistic hybridization of materials and electrodes.


image file: c8ta04038a-f4.tif
Fig. 4 (a) CV curves of the rGO electrode at different scan rates. (b) Comparative CV curves of rGO and the NiCo2O4@rGO hybrid on NF. (c) The CV and (d) GCD profiles of the HSC. (e) Variations of specific capacitance and coulombic efficiency with different current densities. (f) The change in GCD profiles of the HSC at different cycles. (g) Variations of specific capacitance and coulombic efficiency of the HSC at different cycles and at 6 A g−1. (h) Ragone plot of the HSC as a function of current density.

Although the device performances of our HSCs were measured under different conditions, we compared them with previous literature based on the NiCo2O4 HSC (see Fig. 4h). Ding et al. fabricated a NiCo2O4//AC HSC with energy and power densities of 14.7 W h kg−1 and 175 W kg−1 at 0.5 A g−1 and a cycle life of 85% after 5000 cycles at 1.5 A g−1.24 Hsu et al. demonstrated energy and power densities of 17.72 W h kg−1 and 25.42 kW kg−1 for a NiCo2O4//AC HSC at 32 A g−1 with a capacitance retention of 100% after 2000 cycles at 5 A g−1.25 Chen et al. reported energy and power densities of 23.9 W h kg−1 and 650 W kg−1 at 1 A g−1 for a NiCo2O4//rGO HSC with a capacitance retention of 93.2% after 10[thin space (1/6-em)]000 cycles at 1.5 A g−1.26 A Co3(PO4)2//AC HSC was fabricated by Sankar et al., showing energy and power densities of 26.66 W h kg−1 and 750 W kg−1 at 1 A g−1, and a capacitance retention of 80% after 6000 cycles at 6 A g−1.33 Thus, our NiCo2O4/rGO hybrid architecture showed long-term cycling stability and high coulombic efficiency with comparable energy and power densities to previous studies.

4. Conclusions

We have demonstrated an aqueous HSC consisting of a NiCo2O4@rGO hybrid as a positive electrode and rGO as a negative electrode. In particular, the rGO interlayers played the role of controlling the growth of the NiCo2O4 nanostructure and providing a buffer layer for improved electrochemical performance. The growth of NiCo2O4 nanoneedles on rGO coated NF was influenced by a hetero-epitaxial growth mechanism, which controls the dimensions of the nanostructure (length < 1 μm and diameter < 30–40 nm). The NiCo2O4@rGO hybrid on NF exhibited higher specific capacitance (1427 F g−1) and excellent long-term cyclability (capacitance retention of 83.8% after 10[thin space (1/6-em)]000 cycles), but its rate capability was lower than that of NiCo2O4 nanoneedles on NF due to the interfacial resistance and tortuosity of rGO interlayers. On the other hand, these rGO interlayers provided kinetic and structural stability, which was associated with an enlarged potential window and enhanced coulombic efficiency and cycling stability. Thus, we could improve the potential window up to 1.45 V and coulombic efficiency up to 96%, with a capacitance retention of 81.1% over 10[thin space (1/6-em)]000 cycles, fabricating a coin-cell-type HSC of NiCo2O4@rGO hybrid//rGO. Owing to the high specific capacitance and enlarged potential window of the NiCo2O4@rGO hybrid, the HSCs achieved a maximum energy density of up to 25.24 W h kg−1 with a power density of 727 W kg−1 at 1 A g−1. Therefore, this interfacial engineering using rGO interlayers opens up a way to improve the stability of high capacity materials and the energy density for configuring hybrid devices.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Korea Electric Power Corporation (Grant number: R17XA05-52) and the Korea Research Fellowship Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (KRF project grant number: 2016H1D3A1937977).

References

  1. M. Winter and R. J. Brodd, Chem. Rev., 2004, 104, 4245–4269 CrossRef PubMed.
  2. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359–367 CrossRef PubMed.
  3. J. R. Miller and P. Simon, Science, 2008, 321, 651–652 CrossRef PubMed.
  4. C. Z. Yuan, L. Yang, L. Hou, L. Shen, X. Zhang and X. W. Lou, Energy Environ. Sci., 2012, 7, 7883–7887 RSC.
  5. M. R. Lukatskaya, B. Dunn and Y. Gogotsi, Nat. Commun., 2016, 7, 12647 CrossRef PubMed.
  6. D. P. Dubal, O. Ayyad, V. Ruiz and P. Gomez-Romero, Chem. Soc. Rev., 2015, 44, 1777–1790 RSC.
  7. S. Shi, Z. Li, Y. Sun, B. Wang, Q. Liu, Y. Hou, S. Huang, J. Huang and Y. Zhao, Nano Energy, 2018, 48, 510–517 CrossRef.
  8. H. Ma, J. He, D. B. Xiong, J. Wu, Q. Li, V. Dravid and Y. Zhao, ACS Appl. Mater. Interfaces, 2016, 8, 1992–2000 CrossRef PubMed.
  9. M. Wang, F. Jin, X. Zhang, J. Wang, S. Huang, X. Zhang, S. Mu, Y. Zhao and Y. Zhao, ACS Sustainable Chem. Eng., 2017, 5, 5679–5685 CrossRef.
  10. Z. Chen, D. B. Xiong, X. Zhang, H. Ma, M. Xia and Y. Zhao, Nanoscale, 2016, 8, 6636–6645 RSC.
  11. T. Brousse, D. Belanger and J. W. Long, J. Electrochem. Soc., 2015, 162, A5185–A5189 CrossRef.
  12. Z. Wu, Y. Zhu and X. Ji, J. Mater. Chem. A, 2014, 2, 14759–14772 RSC.
  13. D. P. Dubal, P. Gomez-Romero, B. R. Sankapal and R. Holz, Nano Energy, 2015, 11, 377–399 CrossRef.
  14. G. S. Gund, D. P. Dubal, S. S. Shinde and C. D. Lokhande, ACS Appl. Mater. Interfaces, 2014, 6, 3176–3188 CrossRef PubMed.
  15. G. Q. Zhang, H. B. Wu, H. E. Hoster, M. B. Chan-Park and X. W. (David) Lou, Energy Environ. Sci., 2012, 5, 9453–9456 RSC.
  16. C. Gong, M. Acik, R. M. Abolfath, Y. Chabal and K. Cho, J. Phys. Chem. C, 2012, 116, 9969–9979 CrossRef.
  17. S. Gurunathan, J. W. Han, A. A. Dayem, V. Eppakayala and J. H. Kim, Int. J. Nanomed., 2012, 7, 5901–5914 CrossRef PubMed.
  18. L. Ma, X. Shen, Z. Ji, X. Cai, G. Zhu and K. Chen, J. Colloid Interface Sci., 2015, 440, 211–218 CrossRef PubMed.
  19. D. U. Lee, B. J. Kim and Z. Chen, J. Mater. Chem. A, 2013, 1, 4754–4762 RSC.
  20. J. Liu, J. Liu, W. Song, F. Wang and Y. Song, J. Mater. Chem. A, 2014, 2, 17477–17488 RSC.
  21. M. Magnuson, J. Halim and L. A. Naslund, J. Electron Spectrosc. Relat. Phenom., 2018, 224, 27 CrossRef.
  22. Q. Liu, Q. Fang, W. Chu, Y. Wan, X. Li, W. Xu, M. Habib, S. Tao, Y. Zhou, D. Liu, T. Xiang, A. Khalil, X. Wu, M. Chhowalla, P. M Ajayan and L. Song, Chem. Mater., 2017, 29, 4738–4744 CrossRef.
  23. C. Wang, E. Zhou, W. He, X. Deng, J. Huang, M. Ding, X. Wei, X. Liu and X. Xu, Nanomaterials, 2017, 7, 41 CrossRef PubMed.
  24. R. Ding, L. Qi, M. Jia and H. Wang, Electrochim. Acta, 2013, 107, 494–502 CrossRef.
  25. C. T. Hsu and C. C. Hu, J. Power Sources, 2013, 242, 662–671 CrossRef.
  26. H. Chen, J. Jiang, L. Zhang, T. Qi, D. Xia and H. Wan, J. Power Sources, 2014, 248, 28–36 CrossRef.
  27. T. Zhu, E. R. Koo and G. W. Ho, RSC Adv., 2015, 5, 1697–1704 RSC.
  28. Y. Li, F. Tang, R. Wang, C. Wang and J. Liu, ACS Appl. Mater. Interfaces, 2016, 8, 30232–30238 CrossRef PubMed.
  29. G. Gao, H. B. Wu, S. Ding, L. M. Liu and X. W. Lou, Small, 2015, 11, 804–808 CrossRef PubMed.
  30. Y. Zhao, X. Zhang, J. He, L. Zhang, M. Xia and F. Gao, Electrochim. Acta, 2015, 174, 51–56 CrossRef.
  31. T. M. Bond, J. C. Burns, D. A. Stevens, H. M. Dahn and J. R. Dahn, J. Electrochem. Soc., 2013, 160, A521–A527 CrossRef.
  32. G. S. Gund, D. P. Dubal, B. H. Patil, S. S. Shinde and C. D. Lokhande, Electrochim. Acta, 2013, 92, 205–215 CrossRef.
  33. K. V. Sankar, S. C. Lee, Y. Seo, C. Ray, S. Liu, A. Kundu and S. C. Jun, J. Power Sources, 2018, 373, 211–219 CrossRef.

Footnotes

Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ta04038a
Kyu Hyun Oh and Girish Sambhaji Gund contributed equally to this work.

This journal is © The Royal Society of Chemistry 2018