Supercapacitive performance of hydrogenated TiO₂ nanotube arrays decorated with nickel oxide nanoparticles

Abstract

Highly ordered self-organized TiO₂ nanotube arrays (TNTAs) could not only be used as current collectors, but also adopted as highly ion-accessible and charge transfer-channels for the construction of supercapacitors. In this paper, hydrogenated TNTAs were obtained and then nickel oxide (NiO_x) nanoparticles were successfully deposited onto the inner surface and interface of HTNTAs through a cyclic voltammetry electrochemical deposition process (NiO_x/HTNTAs). The FESEM images of the samples showed that the diameter of the NiO_x nanoparticles ranged from 7 to 60 nm. The as-fabricated NiO_x/HTNTAs exhibited an obviously pseudocapacitive performance with a specific capacitance of 689.28 F g⁻¹ at a current density of 1.5 A g⁻¹ and 91.9% of the initial capacitance remaining after 5000 charge/discharge cycles at a current density of 3 A g⁻¹ in 1 M KOH. This work reveals a feasible and green method for the fabrication of TNTAs modified with electroactive metal oxide nanoparticles as functional electrode materials for supercapacitors.

---

Introduction

Supercapacitors (SCs), combining the merits of traditional capacitors and batteries, have attracted more and more attention in the field of energy storage and conversion. Compared with the rechargeable batteries, supercapacitors possess larger power density, longer cycling life, faster charging speed and wider operating temperature. The increasing power and energy demands of portable and flexible electronics such as laptops, wearable devices and roll-up displays, etc., have aroused considerable interest in the investigation of lightweight, flexible and environmentally benign supercapacitors.¹ Highly ordered TiO₂ nanotube arrays (TNTAs) anodized from pure titanium foils (here denoted as TNTAs/Ti) are studied extensively as appropriate supercapacitor electrode materials or co-material with electroactive materials due to their semiconducting performance, huge specific surface area, high ion-accessible channel, controllable tube structure, and relatively low cost.^2–5 Moreover, vertically oriented TNTAs standing directly on current collectors (Ti foil for current collector) have the advantage of eliminating contact impedance, and reducing additional weight arising from the addition of conductive agents and binders. However, the lower capacitance of TNTAs/Ti with areal capacitance of 50–911 μF cm⁻² impedes their applications on a large scale due to the poorer electrochemical activity and lower electrical conductivity.^6–9 Recently, two major methods were adopted to enhance the electrochemical capacitance of TNTAs/Ti electrode. The first strategy is to improve electrical conductivity by carbonization (C@TiO₂),^10–12 nitridation (TiN),¹³ or hydrogenation (H–TiO₂)^14,15 of the pristine TNTAs, and these modified nanotube arrays demonstrated remarkable capacitance improvement of 40 to 60 times higher than that of the sole TNTAs/Ti. Xihong Lu research group demonstrated that hydrogenation of TiO₂ NTAs could improve the electrical conductivity and electrochemical activity of TiO₂ nanostructure by controlled introduction of oxygen vacancy (Ti³⁺ sites) and hydroxyl groups, which could significantly increase its pseudocapacitance.¹⁴ The second route is by constructing hybrid arrays, utilizing the unique tubular channels of the TNTAs, which provide a regular architecture for feasible loading of various electroactive materials, such as conducting polymers (e.g., polyaniline) or insertion-type compounds (e.g., MnO₂, Co₃O₄, NiO, CeO₂, etc.).^16–23 This structure promotes the utilization of these electroactive materials because of available large spaces and effective ion diffusion path for electrochemical reactions. NiO is one of the pseudocapacitive materials with a large theoretical specific capacitance (≈2573 F g⁻¹). Nevertheless, NiO sustains the relatively low electrical conductivity which leads to the poor specific capacitance. One promising way is the incorporation of NiO nanoparticles into the electrically conductive skeleton to enhance its electrochemical performance. The recent report of synthesizing RuO₂–C/TiNTs received specific capacitances of 1089 F g⁻¹.²⁴ Xihong Lu research group prepared MnO₂/H–TiO₂ composite and obtained specific capacitance of 912 F g⁻¹.¹⁴ Zhida Gao synthesised the Ni(OH)₂/TiO_xC_y electrodes and investigated the application in non-enzymatic biosensors.²⁵ However, the investigation of TiO₂ nanotube decorated with NiO nanoparticles with enhanced conductivity as electrode material for supercapacitor has never been reported.

Therefore, in order to combine the merits of both nanomaterials and enhance the supercapacitor performance, we explored the application of hybrid nanostructures by integrating nickel oxide into hydrogenated TNTAs/Ti as supercapacitor electrode. In this work, the TNTAs/Ti electrode was annealed in hydrogen atmosphere to obtain HTNTAs, and then NiO_x (NiO–Ni₂O₃) nanoparticles were successfully deposited onto the inner and intertubular spaces of HTNTAs through cyclic voltammetry electrochemical deposition process. The NiO_x/HTNTAs exhibits essentially improved specific capacitance of 689.28 F g⁻¹ at 1.5 A g⁻¹, which is 3 times larger than that of the NiO_x/ATNTAs obtained from air-annealing treatment of TNTAs under the same conditions. Our results demonstrate the NiO_x/HTNTAs can be served as functional electrode materials for supercapacitors.

Experimental section

TNTAs/Ti with an areal of 1 cm × 4 cm were prepared by a conventional potentiostatic anodization process as described in our previous work.^26,27 The as-prepared TNTAs/Ti were annealed at 450 °C for 60 min with heating rate of 1 °C min⁻¹ in air (named as ATNTAs/Ti) and hydrogen atmosphere (named as HTNTAs/Ti) respectively. After annealing treatment, the HTNTAs/Ti substrate was weighed by electronic analytical balance with accuracy of 0.01 mg (BT 125D, Sartorius, Germany). Prior to electrochemical deposition, the HTNTAs/Ti was impregnated in the saturation of nickel acetate solution for 30 min and rinsed with distilled water. Subsequently, three-dimensional NiO_x/HTNTAs were synthesized based on the HTNTAs/Ti by means of cycle voltammetry electrochemical deposition process performed on a CHI760E electrochemical workstation. A three-electrode electrochemical system with HTNTAs/Ti as working electrode, Pt sheet as counter electrode and Ag/AgCl as reference electrode was used for electrochemical preparation of NiO_x/HTNTAs electrode. Electrochemical deposition experiment was carried out in 0.1 M Na₂SO₄ solution containing Ni(AC)₂ with the potential range 0 to −1.3 V at a scan rate of 50 mV s⁻¹ for 10 cycles. Finally, the NiO_x/HTNTAs was dried in vacuum drying oven at 40 °C for several hours to get constant weight. The mass of NiO_x loaded onto the HTNTAs/Ti was calculated from the two weighting difference value between the HTNTAs/Ti and NiO_x/HTNTAs.²⁸ For comparison, NiO_x/ATNTAs electrode were also obtained by means of the same impregnation and CV electrochemical deposition process.

Materials characterization

The morphology, microstructure and distribution of NiO_x nanoparticles onto the TNTAs were characterized by field emission scanning electron microscopy (FESEM, SU8020, Hitachi, Japan) and high resolution transmission electron microscopy (HRTEM, JEM-2100F, JEOL, Japan). The content and the composition of element was identified by energy dispersive spectrometer (EDS, Inca, Oxford, UK). The phase structure of the samples were measured by X-ray diffraction (XRD, D/MAX2500V, Rigaku, Japan) with Cu-Kα radiation (0.15418 nm) operating at 40 kV, 40 mA. The elemental valance status of the samples were analyzed by X-ray photoelectron spectrum (XPS, ESCALAB250, Thermo, US) with a monochromatic Al Kα (1486.6 eV) X-ray source.

Electrochemical measurements

The electrochemical capacitive properties of the samples were evaluated on an electrochemical workstation (Autolab PGSTAT302N, Metrohm, Switzerland) using a conventional three-electrode system with Ag/AgCl as reference electrode and Pt sheet as the counter electrode in 1 M KOH aqueous solution. The electrochemical impedance spectrum (EIS) tests were performed in a frequency range from 0.01 Hz to 100 kHz at an open circuit potential with an AC-voltage amplitude of 5 mV. The specific capacitance of electrodes were calculated by the following equation:

C_sp = IΔt/(MΔV) (1)

where C_sp (F g⁻¹) represents specific capacitance. I (A) is the current of discharge, Δt (s) is the time of discharge, M (g) is the mass of active materials, and ΔV (V) is the potential window excluding the IR drop.

Results and discussion

The schematic diagram of the fabrication of NiO_x/ATNTAs and NiO_x/HTNTAs electrodes is shown in Fig. 1. Details of the synthesis are elaborated in the Experimental section. Highly ordered TNTAs are vertically grown on titanium foil. It should be noted that there are obvious color difference among ATNTAs, HTNTAs, NiO_x/ATNTAs and NiO_x/HTNTAs electrodes.

Fig. 1 Schematic diagram of fabrication of NiO_x/ATNTAs and NiO_x/HTNTAs electrodes on titanium foil.

Characterization of the NiO_x/HTNTAs nanocomposite

The crystal structure of the as-prepared NiO_x/ATNTAs and NiO_x/HTNTAs samples are identified by XRD analysis, and the diffraction patterns are shown in Fig. 2. Both the NiO_x/ATNTAs and NiO_x/HTNTAs samples show a similar diffraction pattern. The diffraction peak consists of peaks that correspond to the anatase TiO₂, cubic NiO, and Ni₂O₃. The clear peak of the both samples at 2θ = 25.2°, 37.8° and 48° are considered to the (101), (004) and (200) planes of the anatase TiO₂ (JCPDS 21-1272), respectively. The results indicate that heating treatment at 450 °C could transform the amorphous titanium dioxide into anatase phase, irrespective of the air or hydrogen annealing atmosphere. The characteristic peaks at 2θ = 37.3°, 62.9° and 75.4° represent the (111), (220), and (311) crystalline face of cubic NiO (JCPDS 47-1049),^29,30 and the peaks at 2θ = 44.8°, 51.6° correspond to the Ni₂O₃ (JCPDS 14-0481).³¹

Fig. 2 XRD patterns of the NiO_x/ATNTAs and NiO_x/HTNTAs samples fabricated by cyclic voltammetry electrochemical deposition.

Fig. 3(i)–(vi) show the top view and cross-sectional FESEM micrographs of the as-fabricated ATNTAs, HTNTAs, NiO_x/ATNTAs and NiO_x/HTNTAs electrodes. Apparently, Fig. 3(i) and (iii) show self-ordered nanotube arrays formed with the average diameter ranging from 100 to 120 nm and approximate wall thickness between 10 and 15 nm. The smooth tube wall with high orderly vertical arrangement can be seen in Fig. 3(v). As shown in Fig. 3(ii), (iv) and (vi), NiO_x nanoparticles were successfully loaded onto the outer and inner surface of the nanotubes through cycle voltammetry electrochemical deposition method. The diameter of the NiO_x particle size varies from 17 to 60 nm and a closer examination of the morphology reveals that the tube mouth is not blocked by the nanoparticles. Hence, these structural characteristics keeps the fluent path for the fast ionic/electronic migration to the current collector during the faradic redox reactions. In the initial stage of impregnation, Ni²⁺ are preferentially absorbed onto the mouth of tube. As the impregnation time proceeds, more and more nickel ions are also absorbed inside and outside of the tube wall. Thereby, abundant of the NiO_x nanoparticles are deposited on the tube wall and surface of the nanotube arrays by electrochemical deposition. And the as-obtained uniform distribution of the NiO_x nanoparticles exhibit extremely high specific surfaces, providing an interconnected electron conduction pathway. These structural features are important for enhancing electrochemical performance of electrode material. For TNTAs annealing in hydrogen atmosphere, it is found that the ordered morphology could be completely maintained after hydrogenation (Fig. 3(i) and (iii)). Identical morphology of the NiO_x/ATNTAs and NiO_x/HTNTAs (Fig. 3(ii) and (iv)) are obtained, suggesting that the ordered arrangement of tube structure and the uniform distribution of the nanoparticles were not affected by hydrogenation.

Fig. 3 FESEM images of the electrodes fabricated by anodization, hydrogenation and cycle voltammetry electrochemical deposition strategy: the surface of (i) ATNTAs, (ii) NiO_x/ATNTAs, (iii) HTNTAs, (iv) NiO_x/HTNTAs, the cross-sectional of (v) HTNTAs, (vi) NiO_x/HTNTAs.

Fig. 4(i) and (ii) are the TEM image and HRTEM of the NiO_x/HTNTAs, respectively. We could think the gray and black nanoparticles in TEM image (Fig. 4(i)) indicate the NiO_x nanoparticles on the inner and outer tube wall of HTNTAs, respectively. Further morphology comparison of the individual bare HTNTAs and NiO_x/HTNTAs (Fig. S3(i)†) reveals that NiO_x could also be deposited into the tube. EDX elemental scanning technique is employed to confirm compositional distribution of NiO_x/HTNTAs nanocomposite. As shown in Fig. S2 and S3,† Ti, O and Ni elements distribute in an individual nanotube and closely rounded particles adhere onto the tube wall. Fig. 4(ii) is the typical HRTEM image of the as-fabricated NiO_x/HTNTAs, in which the lattice spacing of 0.202 nm and 0.209 nm are corresponding to the Ni₂O₃ and (200) plane of NiO, respectively, indicating the co-existence of Ni₂O₃ phase and NiO phase.

Fig. 4 (i) TEM image of NiO_x/HTNTAs, (ii) HRTEM image of NiO_x/HTNTAs.

To verify the effect of hydrogenation and electrodeposition on the chemical composition and the oxidation state of NiO_x/HTNTAs, high-resolution XPS investigation of the ATNTAs and the NiO_x/HTNTAs were carried out to confirm the chemical bonding states. Fig. 5(i) shows the fully scanned XPS spectra of the NiO_x/HTNTAs. Typical peaks of Ti 2p (450–470 eV), O 1s (520–540 eV), and Ni 2p (850–890 eV) locate at corresponding element region, which indicates that Ti, O, and Ni element exist in NiO_x/HTNTAs heterojunction structures. Peak fitting to the spectra was adopted using Gaussian–Lorentzian peak shape after subtraction of Shirley background, as shown in Fig. 5(ii) to (vi). The Ti 2p spectra of ATNTAs and NiO_x/HTNTAs samples are shown in Fig. 5(ii) and (iii). Two broad peaks located at ∼464.5 eV and ∼458.8 eV are corresponded to the characteristic of Ti 2p_1/2 and Ti 2p_3/2 peaks of Ti⁴⁺ for both the samples.^14,32–35 There are two extra peaks for the NiO_x/HTNTAs centered at ∼463.9 eV and ∼458.2 eV (Fig. 5(iii)), which are in line with the characteristic peaks of Ti 2p_1/2 and Ti 2p_3/2 of Ti³⁺,^14,32,36 confirming the presence of Ti³⁺ ions in the NiO_x/HTNTAs samples. Fig. 5(iv) and (v) compare the O 1s peak of ATNTAs and NiO_x/HTNTAs samples. The characteristic peaks of Ti–O–Ti and Ti–OH are simultaneously found in both of the two samples. The binding energy of ∼529.9 eV is related to the characteristic peak of oxygen atoms in Ti–O–Ti,^14,31 and the shoulder peak centered at higher binding energy of ∼531.8 eV is attributed to the hydroxyl group of Ti–OH, which has been reported to be centered at 1.5–1.8 eV higher than the peak of Ti–O–Ti bond.^33,36 The intensity of Ti–OH peak for the NiO_x/HTNTAs is prominently higher than that of ATNTAs sample. The peak located at highest bonding energy of ∼533.1 eV may be ascribed to the Ni–OH bond.³⁷ The two oxygen peaks centered at ∼530.4 eV and ∼531.5 eV are consistent with the oxygen bond to Ni in NiO and Ni₂O₃, respectively. From these XPS characterization results, we can conclude that the oxygen vacancies (Ti³⁺ sites) are generated and the hydroxyl groups introduced on the surface of NiO_x/HTNTAs during hydrogenation and electrodeposition process. In addition, the oxidation states of Ni is determined by Ni 2p_3/2 XPS spectra, as shown in Fig. 5(vi). The binding energy of ∼855.2 eV and ∼861.7 eV are in accordance with the chemical bonding states of NiO. ∼855.8 eV and ∼861.4 eV are suited for the chemical bonding states of Ni₂O₃.³⁸ These XPS analysis are in consistence with the XRD pattern and HRTEM results.

Fig. 5 The XPS spectra for (i) the survey spectra of the NiO_x/HTNTAs, the Ti 2p peak of (ii) the ATNTAs and (iii) the NiO_x/HTNTAs. The O 1s peak of (iv) the ATNTAs and (v) the NiO_x/HTNTAs. (vi) The Ni 2p peak.

Fig. 6 shows cyclic voltammograms of the NiO_x deposition on the HTNTAs electrode in the scanned potential range between −1.3 V and 0 V. Under this potential range, electrolysis of water firstly occurs at the surface of the HTNTAs electrode. As a result, both hydrogen gas and hydroxyl ions are generated, as expressed in eqn (2).³⁹ Meanwhile, electrochemical reduction of oxygen is possibly the reason for the formation of hydroxyl ions (eqn (3)). The first cycle of CV with anodic and cathodic potential is apparently different from the subsequent cycle. As shown from the first cycle in the inset of Fig. 6, a wide range of oxidation peak labeled O₁ is observed, corresponding to oxidation current loads at potential of around 0.8 V vs. Ag/AgCl, which is probably ascribed to oxidation reaction of Ni²⁺ to Ni³⁺ (eqn (4)). The peak potential of the O₁ negatively shifts with increase in scan cycles. A reduction peak labeled R₁ is observed with the exception of the first scan cycle, which probably arises from the reduction reaction of Ni³⁺ to Ni²⁺. The hydroxide ions in the solution react with the Ni³⁺ to form insoluble nickel oxy-hydroxyl NiOOH on the surface of HTNTAs (eqn (5)). The peak current of R₁ enhances with the increase of the scan cycles, which reveals that partial Ni³⁺ ions maybe uncompletely react with the hydroxide ions to form NiOOH on the surface of the electrode during anodic scan, and unreacted Ni³⁺ ions are reduced to Ni²⁺ in the electrode/electrolyte interface. An cathodic peak labeled R₂ probably is attributed to the reduction reaction of NiOOH to Ni(OH)₂ (eqn (6)), while the anodic peak labeled O₂ is the relevant oxidation reaction of Ni(OH)₂ to NiOOH. Subsequently, Ni(OH)₂ transforms to NiO in the presence of oxygen (eqn (7)) and some NiO also reacts with Ni(OH)₂ to form Ni₂O₃ (eqn (8)). Based on the aforementioned description, the redox couple of O₁ and R₁ correspond to oxidation and reduction reactions between Ni²⁺ and Ni³⁺. While, the redox couple of O₂ and R₂ come from the oxidation and reduction reactions between NiOOH and Ni(OH)₂. The main electrochemical reaction can be described as follows:

2H₂O + 2e⁻ = H₂ + 2OH⁻ (2)

O₂ + H₂O + 4e⁻ = 4OH⁻ (3)

Ni²⁺ ↔ Ni³⁺ + e⁻ (4)

Ni³⁺+ 3OH⁻ = NiOOH↓+ H₂O (5)

NiOOH + H⁺+ e⁻ ↔ Ni(OH)₂ (6)

2Ni(OH)₂ + O₂ = NiO + H₂O (7)

NiO + Ni(OH)₂ = Ni₂O₃ + 2H⁺ + 2e⁻ (8)

Fig. 6 Cyclic voltammograms of the NiO_x deposition on the HTNTAs substrate at the 50 mV s⁻¹ of scan rate in the scanned potential range between −1.3 V and 0 V. The deposition solution consisted of 0.01 M nickel acetate, 0.1 M sodium sulphate.

Electrochemical analysis

In order to investigate the influence of the NiO_x loading amount on the capacitive performance of NiO_x/HTNTAs electrode materials, we presented various concentration of Ni(AC)₂ during the electrochemical deposition. As a result, the loading amounts of Ni 2.7 at%, 9.0 at% and 21 at% of NiO_x/HTNTAs in 5 mM, 10 mM and 15 mM of the Ni(AC)₂ concentration (named 5-NiO_x/HTNTAs, 10-NiO_x/HTNTAs, 15-NiO_x/HTNTAs) were obtained, respectively. The EDX measurement and the corresponding top view FESEM micrographs of 5-NiO_x/HTNTAs, 10-NiO_x/HTNTAs, 15-NiO_x/HTNTAs are shown in Fig. S4(i)–(vi)†. As can be seen, the number and the size of NiO_x nanoparticles deposited onto the HTNTAs gradually increases with the increasing of the concentration of the Ni(AC)₂. The galvanostatic charge/discharge (GCD) provides a reliable method for evaluating the capacitive performance of the electrode material. The GCD curves of 2.7 at%, 9.0 at% and 21 at% Ni-containing NiO_x/HTNTAs electrode are measured at a current density of 1.5 A g⁻¹ within the voltage window between 0 and 0.4 V as shown in Fig. 7. The specific capacitance of the 10-NiO_x/HTNTAs is 689.28 F g⁻¹, which is about 1.5 times higher than that of 15-NiO_x/HTNTAs electrode (463.4 F g⁻¹), 2.3 times higher than that of 5-NiO_x/HTNTAs electrode (298.7 F g⁻¹). Hence, the 10-NiO_x/HTNTAs with 9.0 at% loading amounts of Ni exhibit a higher specific capacitance than the other of NiO_x/HTNTAs, which was mainly due to the effect of both appropriate loading amounts of NiO_x and fluent tube channel (see in Fig. S4(v)†). The rare loading amounts of NiO_x for 5-NiO_x/HTNTAs (see in Fig. S4(iv)†) would not provide sufficient active sites, therefore the direct electrochemical reaction areas of NiO_x/HTNTAs are not increased. The effective interface of at which the faradic pseudocapacitive reaction process take place would be decreased. It finally leads to a lower capacitive performance. For 15-NiO_x/HTNTAs, the larger NiO_x nanoparticles block the mouth of tube (see in Fig. S4(vi)†). As a result, the efficiency charge/ion transfer were reduced during charge/discharge process. While, more and more NiO_x nanoparticles were probably prevented from depositing into the outer and inside of tube wall with the increase of time. These would be unbeneficial to improve the capacitive performance.

Fig. 7 The comparison of GCD at current density of 1.5 A g⁻¹ for NiO_x/HTNTAs prepared by 5 mM, 10 mM and 15 mM Ni(AC)₂ solution.

The cyclic voltammetric (CV) curves of the ATNTAs, HTNTAs, 10-NiO_x/ATNTAs and 10-NiO_x/HTNTAs electrode materials are measured at a series of scan rates ranging from 5 mV s⁻¹ to 100 mV s⁻¹ within the voltage window between 0 and 0.6 V as shown in Fig. 8(i)–(iv). The CV curves of both ATNTAs and HTNTAs electrode materials show rectangular shape (Fig. 8(i) and (ii)) with no evidence for faradic reactions, which could infer that they serve as a pure electrochemical double layer capacitor.^40,41 Compared to the ATNTAs electrode, the HTNTAs electrode material exhibits an obvious capacitive characteristic curve more close to an ideal rectangular shape.^41,42 The CV shape of EDLC electrode material is affected by the RC time constant (τ) of the electrode system.⁴³ The ideal rectangular curve represents the current decayed within an ultrashort period after the scanning voltage switching, indicating quite smaller RC time constant of the HTNTAs electrode than that of the ATNTAs electrode, which is ascribed to the highly increased conductivity of the HTNTAs electrode since the C (in RC) has been highly enlarged.³² Therefore, faster ionic/electronic migration happen within the HTNTAs electrode material, resulting in a better capacitive behavior over the ATNTAs electrode material. Furthermore, the enhanced conductivity could be attributed to the generation of oxygen vacancies and the introduction of hydroxyl groups on TiO₂ nanotube arrays by hydrogenation treatment in H₂ atmosphere. As can be seen from Fig. 8(iii) and (iv), the CV shape of the both electrode materials obviously exhibit pseudocapacitive behavior of the electrode which is totally different from the double layer capacitance. A pair of remarkable redox peak are observed from the CV curves at every scan rate, which is attributed to the incorporative effect of NiO and Ni₂O₃ nanoparticles from the NiO_x/ATNTAs and the NiO_x/HTNTAs electrode materials. It is well known that the surface faradic pseudocapacitive behavior is originated from the surface redox mechanism between Ni²⁺ and Ni³⁺:^44–46

NiO_x + yOH⁻ ↔ NiO_x(OH)_y + ye⁻ (9)

Fig. 8 Cyclic voltammograms of the as-prepared (i) ATNTAs, (ii) HTNTAs, (iii) NiO_x/ATNTAs, (iv) NiO_x/HTNTAs electrode materials at various scan rates from 5 mV s⁻¹ to 100 mV s⁻¹ in a 1 M KOH solution. (v) Cyclic voltammograms of different TiO₂ nanotube arrays electrode materials at scan rate of 100 mV s⁻¹. (vi) Nyquist plots of four electrode materials after fitting in 1 M KOH solution.

In addition, the oxidation and reduction current peaks shift more towards both sides of the axes and increase with the increasing scan rate, indicating an increasing internal diffusion resistance within the pseudocapacitive material. As shown in Fig. 8(v), the cyclic voltammograms of different TiO₂ nanotube arrays electrodes are measured at a scan rate of 100 mV s⁻¹ under the identical testing condition. Both NiO_x/ATNTAs and NiO_x/HTNTAs electrode materials exhibit the prominent capacitance performance compared with the ATNTAs and HTNTAs electrode materials. The NiO_x/HTNTAs electrode material shows a much larger enclosing area than those of the NiO_x/ATNTAs, HTNTAs and ATNTAs electrodes. This may be ascribed to the effect by the deposition of the NiO_x nanoparticles and hydrogenation for TiO₂ nanotube arrays.

On the one hand, NiO_x nanoparticles further improve capacitance performance with good pseudocapacitive behavior for the nanocomposite electrode material. On the other hand, the enhanced conductivity of TiO₂ nanotube arrays themselves by hydrogenation reduce charge transfer resistance of TiO₂ nanotube arrays, resulting in faster ionic/electronic diffusion rate during the faradic oxidation–reduction within NiO_x active materials. And EIS plots for the ATNTAs, HTNTAs, NiO_x/ATNTAs, and NiO_x/HTNTAs electrode materials further confirm the results discussed above as shown in Fig. 8(vi). The corresponding equivalent electrical circuit for impedance analysis is shown in inset of Fig. 8(vi), which consists of solution resistance R_s, charge transfer resistance R_ct, Warburg impedance, double layer capacitance C_dl and pseudocapacitance C_ps. The slope of the impedance plot in the low frequency region represents the Warburg impedance, which reflects the electrolyte diffusion rate in the electrode material. The HTNTAs, NiO_x/ATNTAs and NiO_x/HTNTAs present similar slope, which means that electrolyte diffusion between electrolyte and three electrodes surface are similar. The larger slope of NiO_x/HTNTAs indicates higher electrolytic ion diffusion within the electrode structure. The semicircle in the high frequency reflects the charge-transfer resistance (R_ct) caused by the faradic reaction and the double-layer capacitance (C_dl) on the electrode. Based on the value of the R_ct from fitted data, NiO_x/HTNTAs (0.001 Ω) displays the lowest charge transfer resistance among the four electrode materials (the NiO_x/ATNTAs is for 0.0067 Ω, HTNTAs is for 1.175 Ω, ATNTAs is for 1.709 Ω). It can be confirmed that the lower R_ct contributes to the improvement of supercapacitive property of NiO_x/HTNTAs.

The GCD testing of the NiO_x/HTNTAs electrode was operated was operated between 0 and 0.4 V (vs. Ag/AgCl) as shown in Fig. 9(i). The potential–time plots exhibit nearly similar charge/discharge time and symmetrical charge/discharge shape, indicating a high reversibility of the faradaic reaction taking place on the NiO_x surface. Fig. 9(ii) presents the charge/discharge plots of the NiO_x/HTNTAs and NiO–NiO_x/ATNTAs electrode materials at a current density of 1.5 A g⁻¹. The specific capacitance of NiO_x/HTNTAs electrode is 689.28 F g⁻¹, much larger than that of NiO–Ni₂O₃/ATNTAs electrode (230.1 F g⁻¹) due to aforementioned reasons. Fig. 9(iii) shows the specific capacitance of the both electrode materials at various current densities. The specific capacitance of the NiO_x/ATNTAs reaches 234.7 F g⁻¹ at 0.75 A g⁻¹, 230.1 F g⁻¹ at 1.5 A g⁻¹, 225.3 F g⁻¹ at 3 A g⁻¹, 216.9 F g⁻¹ at 4.5 A g⁻¹, 206.54 F g⁻¹ at 6 A g⁻¹. The NiO_x/HTNTAs electrode material display the specific capacitance with 689.28 F g⁻¹ at 1.5 A g⁻¹, 673.1 F g⁻¹ at 3 A g⁻¹, 637.36 F g⁻¹ at 6 A g⁻¹, 650.36 F g⁻¹ at 9 A g⁻¹, 586.86 F g⁻¹ at 15 A g⁻¹, 555.34 F g⁻¹ at 20 A g⁻¹, which are calculated by the loading mass of NiO_x. In addition, the pseudocapacitance loss of both electrode materials are 12% and 19%, respectively. Compared with previously reported TiO₂ nanotube arrays or nickel oxide nanocomposites, the capacitance of the NiO_x/HTNTAs electrode material is larger than NiO-based (NiO/ITO (320 F g⁻¹ at 2.8 A g⁻¹))⁴⁷ (MnO₂/TiN (681 F g⁻¹ at 2 A g⁻¹))¹⁴ composite electrode. Fig. 9(iv) shows the cycling stabilities of both NiO_x/HTNTAs and NiO_x/ATNTAs electrodes at a current density of 3 A g⁻¹. Their specific capacitance after 5000 cycles retain 91.9% and 88.2% of the initial capacitance, respectively, illustrating an excellent long-term stability. Apparently the synthesis of NiO_x/HTNTAs electrode associated with hydrogenation and cyclic voltammetry electrochemical deposition method is simple and appropriate for the preparation of nickel oxide and titania dioxide nanocomposite electrode for supercapacitor.

Fig. 9 (i) Galvanostatic charge/discharge curves of the NiO_x/HTNTAs electrode at current density ranging from 1.5 A g⁻¹ to 20 A g⁻¹. The NiO_x/ATNTAs and NiO_x/HTNTAs electrode for (ii) galvanostatic charge/discharge curves at the current density of 1.5 A g⁻¹, (iii) specific capacitance at different specific current and (iv) cycling stability at the current density of 3 A g⁻¹.

Conclusions

In summary, we have developed a feasible and green method for the fabrication HTNTAs modified with electroactive nickel oxides nanoparticles as functional electrode for supercapacitors. The three-dimensional architecture NiO_x/HTNTAs nanocomposites significantly enhance the supercapacitive performance by uniformly distributed electrochemical active NiO_x nanoparticles and simultaneously incorporated Ti³⁺ and –OH groups introduced by hydrogenation treatment of TiO₂ nanotube array. The NiO_x/HTNTAs electrode exhibits good specific capacitance (689.28 F g⁻¹ at 1.5 A g⁻¹) and an excellent long-term stability (91.9% of the initial capacitance at 3 A g⁻¹) after 5000 cycles. The simple and cost-effective synthesis method combining excellent supercapacitive property makes this nanocomposite an excellent electrode in the construction of supercapacitor.

Acknowledgements

The work was financially supported by the National Natural Science Foundation of China (51302060, 51272062, 51502071) and the key research project of Beifang University of Nationalities (2015KJ16). L. H. Cui thanks the staff in the Analytical and Testing Center of HFUT for their assistance in the materials characterization. Dr Y. Wang gratefully acknowledges the financial support from the China Scholarsh Council during his visit to Rice University.

References

1. J. Y. Kim, S. H. Lee, Y. Yan, J. H. Oh and K. Zhu, RSC Adv., 2012, 2(22), 8281.
2. R. P. Vitiello, J. M. Macak, A. Ghicov, H. Tsuchiya, L. F. P. Dick and P. Schmuki, Electrochem. Commun., 2006, 8, 544.
3. H. Zhou, L. Liu, K. Yin, S. L. Liu and G. X. Li, Electrochem. Commun., 2006, 8, 1168.
4. J. R. Jennings, A. Ghicov, L. M. Peter, P. Schmuki and A. B. Walker, J. Am. Chem. Soc., 2008, 130, 13364.
5. Y. Jun, J. H. Park and M. G. Kang, Chem. Commun., 2012, 48, 6456.
6. Z. Endut, M. Hamdi and W. J. Basirun, Appl. Surf. Sci., 2013, 280, 962.
7. M. Salari, S. H. Aboutalebi, K. Konstantinov and H. K. Liu, Phys. Chem. Chem. Phys., 2011, 13, 5038.
8. M. Salari, K. Konstantinov and H. K. Liu, J. Mater. Chem., 2011, 21, 5128.
9. M. S. Kim, T. W. Lee and J. H. Park, J. Electrochem. Soc., 2009, 156, A584.
10. B. Gao, X. X. Li, Y. W. Ma, Y. Cao, Z. Y. Hu, X. M. Zhang, J. J. Fu, K. F. Huo and P. K. Chu, Thin Solid Films, 2015, 584, 61.
11. F. Mole, J. Wang, D. A. Clayton, C. L. Xu and S. L. Pan, Langmuir, 2012, 28, 10610.
12. Z. D. Gao, X. Zhu, Y. H. Li, X. M. Zhou, Y. Y. Song and P. Schmuki, Chem. Commun., 2015, 51, 7614–7617.
13. S. M. Dong, X. Chen, L. Gu, X. H. Zhou, L. F. Li, Z. H. Liu, P. X. Han, H. X. Xu, J. H. Yao, H. B. Wang, X. Y. Zhang, C. Q. Shang, G. L. Cui and L. Q. Chen, Energy Environ. Sci., 2011, 4, 3502.
14. X. H. Lu, G. M. Wang, T. Zhai, M. H. Yu, J. Y. Gan, Y. X. Tong and Y. Li, Nano Lett., 2012, 2, 1690.
15. H. Wu, D. D. Li, X. F. Zhu, C. Y. Yang, D. F. Liu, X. Y. Chen, Y. Song and L. F. Lu, Electrochim. Acta, 2014, 116, 129.
16. H. Zhou and Y. R. Zhang, J. Phys. Chem., 2014, 3, 5626.
17. W. J. Wang, Q. Hao, W. Lei, X. F. Xia and X. Wang, RSC Adv., 2012, 2(27), 10268.
18. R. S. Ray, B. Sarma, A. L. Jurovitzki and M. Misra, Chem. Eng. J., 2015, 260, 671.
19. Y. B. Xie, C. J. Huang, L. M. Zhou, Y. Liu and H. T. Huang, Compos. Sci. Technol., 2009, 69, 2108.
20. Y. B. Xie, L. M. Zhou, C. J. Huang, H. T. Huang and J. Lu, Electrochim. Acta, 2008, 53, 3643.
21. S. H. Mujawar, S. B. Ambade, T. Battumur, R. B. Ambade and S. Lee, Electrochim. Acta, 2011, 56, 4462.
22. H. F. Su, T. Wang, S. Y. Zhang, J. M. Song, C. J. Mao, H. J. Niu, B. K. Jin, J. Y. Wu and Y. P. Tian, Solid State Sci., 2012, 14, 677.
23. H. Wen, Z. F. Liu, Q. B. Yang, Y. X. Li and J. Yu, Electrochim. Acta, 2011, 56, 2914.
24. Y. Y. Song, Y. H. Li and J. Guo, J. Mater. Chem. A, 2015, 3, 23754–23759.
25. Z. D. Gao, J. Guo, N. K. Shrestha, R. Hahn, Y. Y. Song and P. Schmuki, Chem.–Eur. J., 2013, 19, 15530–15534.
26. B. Yuan, Y. Wang, H. D. Bian, T. K. Shen, Y. C. Wu and Z. Chen, Appl. Surf. Sci., 2013, 280, 523.
27. H. D. Bian, X. Shu, J. F. Zhang, B. Yuan, Y. Wang, L. J. Liu, G. Q. Xu, Z. Chen and Y. C. Wu, Chem.–Asian J., 2013, 8, 2746.
28. F. Yang, J. Y. Yao and F. L. Liu, J. Mater. Chem. A, 2013, 1, 594–601.
29. M. S. Wu and H. Hsieh, Electrochim. Acta, 2008, 53, 3427.
30. M. M. Yao, Z. H. Hu, Z. J. Xu, Y. F. Liu, P. P. Liu and Q. Zhang, J. Power Sources, 2015, 273, 914.
31. X. Wen, X. C. Chen, N. N. Tian, J. Gong, J. Liu, M. H. R. Mmeli, P. K. Chu, E. Mijiwska and T. Tang, Environ. Sci. Technol., 2014, 48, 4048.
32. H. Zhou and Y. R. Zhang, J. Power Sources, 2013, 239, 128.
33. X. B. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746.
34. M. S. Lazarus, Chem. Phys. Lett., 1982, 92, 670.
35. Y. Y. Song, P. Roy, I. Paramasivam and P. Schmuki, Angew. Chem., Int. Ed., 2010, 49, 351–354.
36. E. McCafferty and J. P. Wightman, Surf. Interface Anal., 1998, 26, 549.
37. A. K. Singh, D. Sarkar, G. G. Khan and K. Mandal, ACS Appl. Mater. Interfaces, 2014, 6, 4684.
38. G. T. Liu, H. F. Chen, G. M. Lin, P. P. Ye, X. P. Wang, Y. Z. Jiao, X. Y. Guo, Y. Wen and H. F. Yang, Biosens. Bioelectron., 2014, 56, 26.
39. C. Z. Zhu, S. Guo, Y. Fang, L. Han, E. Wang and S. Dong, Nano Res., 2011, 47, 648.
40. Q. Li, Z. Li, L. Lin, X. Y. Wang, Y. Wang, C. Zhang and H. Wang, Chem. Eng. J., 2010, 156, 500.
41. P. Simon and Y. Gogotsi, Nat. Mater., 2008, 7, 845.
42. E. Frackowiak and F. Beguin, Carbon, 2001, 39, 937.
43. S. Yoon, J. W. Lee, T. Hyeon and S. M. Oh, J. Electrochem. Soc., 2000, 147, 2507.
44. K. Nam, K. Kim, E. Lee, W. Yoon, X. Yang and K. B. Kim, J. Power Sources, 2008, 182, 642.
45. K. C. Liu and M. A. Anderson, J. Electrochem. Soc., 1996, 143, 124.
46. V. Srinivasan and J. W. Weidner, J. Electrochem. Soc., 1997, 144, L210.
47. M. S. Wu, C. H. Yang and M. J. Wang, Electrochim. Acta, 2008, 54, 155.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra25581c

---