All‐Solid‐State Cable‐Type Supercapacitors with Ultrahigh Rate Capability

Abstract

All-solid-state cable-type supercapacitors with ultrahigh rate capability were realized by utilizing self-doped TiO2 nanotubes grown on titanium wire via an in situ process. The fabricated all-solid-state cable-type supercapacitors can work very effectively even at an ultrahigh scan rate of 200 V s–1, which indicates the ultrafast charge-storage capability of the device. Advanced energy-storage techniques and devices are playing more and more important roles in the development of future smart and portable electronic devices. Besides the performance, architecture, safety features, and mechanical properties (including stability and flexibility) of the energy-storage devices also become critical for the innovation of the modern smart and portable electronic devices to a certain extent as they are the indispensable parts to power the electronic device.1-4 Owing to the fast charge–discharge rate, high power density, long cycle life, and good safety features, supercapacitors show promising prospects as an energy solution for future electronic devices.5-14 Especially flexible supercapacitors enable the design of novel flexible and wearable electronic devices. Recently, cable-type supercapacitors have attracted considerable interest due to their compact size and high mechanical flexibility.15-23 A cable-type supercapacitor is easy to weave as fiber into wearable electronics with arbitrary shapes and high flexibility.16, 19 Therefore, it is very meaningful to develop high-performance cable-type supercapacitors for designing future wearable electronics. In recent years, considerable progress was achieved in the field of cable-type supercapacitors, particularly in device architecture and performance (especially energy density).15-24 Nevertheless, it is still a significant challenge to achieve high rate performance in cable-type supercapacitors. Most previous studies related to cable-type supercapacitors mainly focus on optimizing the electrode structure to attain high capacitance and thereby enabling them to compete with other energy-storage systems in terms of energy density. However, the rate performance of most current cable-type supercapacitors is not satisfactory, which greatly limits their practical applications; for example, when compared with planar flexible supercapacitors with interdigitated electrodes, the rate capability of current cable-type supercapacitors is around two to three orders of magnitude lower25-29 To extend the applications of cable-type supercapacitor for the development of next-generation electronic devices, it is extremely important to enhance the rate capability of cable-type supercapacitors. High rate capability means ultrafast charge–discharge characteristics along with exceptional capacitance, thus not only could these cable-type supercapacitors rival planar flexible supercapacitors but they might also endow some unique functions upon wearable electronics. To obtain high rate performance, there are two stringent requirements for the design of supercapacitor electrodes, i.e., shorter ion-diffusion path and lower electron-transfer resistance. The present fabrication strategies of cable-type supercapacitor electrodes mainly include wet- or dry-spinning and dip-coating of electrochemically active materials on the surface of conductive wirelike substrates (i.e., current collectors). These techniques generally have the disadvantage of generating long ion diffusion paths in the obtained supercapacitor electrodes; this is a major reason for the poor rate performance of the current cable-type supercapacitors. In some cases, poor adhesion between the electrode materials and the wirelike current collectors, or the use of binders for electrode fabrication, also results in poor rate performance of the as-prepared cable-type supercapacitors. Therefore, the key challenge to enhance the rate performance of cable-type supercapacitors is to develop cable-type and binder-free supercapacitor electrodes that possess a short ion-diffusion path as well as easy electron transport. However, there is still no efficient method to fabricate such cable-type and binder-free supercapacitor electrodes with ultrahigh rate capability. To address the challenge of realizing high rate performance in cable-type supercapacitors, we demonstrate here an effective strategy for the fabrication of cable-type and binder-free supercapacitor electrodes to realize all-solid-state cable-type supercapacitors with ultrahigh rate capability. The cable-type and binder-free supercapacitor electrodes are self-doped TiO2 nanotube arrays that are synthesized in situ on titanium wire (D-TiO2 NTA/Ti) by the anodization of a Ti wire (this creates TiO2 nanotube arrays on Ti) and a subsequent cathodic polarization process (that results in the self-doping of TiO2 with Ti3+ to increase the conductivity of the TiO2 nanotubes). The tubular architecture of D-TiO2 NTA offers a short ion diffusion path while the large specific surface area together with the improved electrical conductivity of the D-TiO2 NTA helps to achieve outstanding ultrahigh rate performance. As a result, the assembled all-solid-state cable-type supercapacitors exhibit superior rate performance even at a scan rate of 200 V s–1. The ultrahigh rate performance reported here is two to three orders of magnitude higher than that of many of the current cable-type supercapacitors, and is even comparable to that of recently reported planar flexible supercapacitors. Furthermore, the all-solid-state cable-type supercapacitor displays an impressive device capacitance of 859.83 μF cm–1 at a scan rate of 1 V s–1. Additionally, the devices also exhibit an outstanding cyclic stability and ideal charge-storage characteristics even when bent. Figure 1a shows a schematic illustration of the preparation of D-TiO2 NTA based cable-type supercapacitor electrodes. First, TiO2 NTA was synthesized in situ on the Ti wire (depicted as TiO2 NTA/Ti) by anodizing Ti wire at 60 V and subsequent annealing at 450 °C. Second, the obtained TiO2 NTA/Ti was electrochemically reduced at –1.4 V (vs. Ag/AgCl) to generate Ti3+. The self-doped TiO2 NTA/Ti (depicted as D-TiO2 NTA/Ti) obtained after the electrochemical reduction process was directly utilized for the construction of cable-type supercapacitors. The tubular architecture of D-TiO2 NTA provides both a high surface area and a short ion-diffusion pathway for achieving an excellent charge-storage performance, especially at ultrahigh charge–discharge rates. The remaining Ti wire beneath the D-TiO2 NTA array acts not only as a support for maintaining the integrity of the nanotube arrays, but also an excellent current collector. With such an electrode, there is no need for a binder or other current collector while constructing cable-type supercapacitor devices. By tuning the cathodic polarization time, the capacitive performance of D-TiO2 NTA/Ti could be optimized. Figure S1 (Supporting Information) shows the cyclic voltammetry (CV) curves of D-TiO2 NTA/Ti obtained after 5, 10, and 15 min of cathodic polarization with a scan rate of 50 V s–1. CV curves can be used to estimate the capacitance; it is clear from Figure S1 that D-TiO2 NTA/Ti with 10 min of cathodic polarization shows the maximum capacitive behavior. The high capacitive current should be attributed to the optimized doping concentration of Ti3+ (creating oxygen vacancies), which results in improved electrochemical behavior of the electrodes. For further analysis and device fabrication, we utilized the D-TiO2/Ti samples prepared by cathodic polarization for 10 min. The morphology of the D-TiO2 NTA/Ti was characterized by using scanning electron microscopy (SEM). As shown in Figure 1b and 1c, D-TiO2 NTAs are uniformly formed on the surface of Ti wire. A magnified top view of a typical D-TiO2 NTA/Ti is given in Figure 1d, which confirms the ordered structure of the D-TiO2 nanotube array. Morphology and other structural features of the D-TiO2 nanotube were further analyzed by using transmission electron microscopy (TEM). Figure 2a,b show the TEM image of a TiO2 nanotube and D-TiO2 nanotube, respectively. The images clearly reveal the seamless hollow interior of the nanotube, which indicate that there were no structural changes of TiO2 nanotube after electrochemical reduction. The lattice distance of the TiO2, as determined from a high-resolution TEM (HR-TEM) image given in the inset of Figure 2a is 3.52 Å, which can be attributed to anatase TiO2, while the lattice distance of the D-TiO2 is also 3.52 Å (inset of Figure 2b). The same lattice distance suggests the anatase structure of the D-TiO2. However, there are some disordered regions in the D-TiO2 nanotubes after the electrochemical reduction process, as shown in the inset of Figure 2b. This disorder arises from the generation of Ti3+ during the electrochemical reduction process. Such a disordered structure was also reported for hydrogen-treated TiO2 due to the doping process.30 The presence of the Ti and O is further ascertained by energy-dispersive X-ray spectroscopy (EDX) and elemental mapping, respectively. The EDX pattern and elemental mapping data of the D-TiO2 is given in Figure 2c. Both EDX and elemental mapping clearly show the presence of Ti and O in the samples. X-ray diffraction (XRD) measurements are performed to understand the crystalline properties of the TiO2 NTA/Ti and D-TiO2 NTA/Ti (Figure 2d). The intense peaks at 25.2° and 48° are representative of pure anatase TiO2. Diffraction peaks at 38.1° and 39.9° indicate the presence of Ti.31, 32 The XRD pattern of D-TiO2 NTA/Ti shows all peaks obtained in the case of TiO2, which indicates the existence of the anatase TiO2 after electrochemical reduction, which is consistent with the TEM results. The effect of cathodic polarization on the composition and oxidation states of the elements in the TiO2 nanotubes was analyzed by using X-ray photoelectron spectroscopy (XPS). The doping of Ti3+ in D-TiO2 NTA/Ti was confirmed by assessing the binding energies of Ti 2p1/2 and Ti 2p3/2 of TiO2 NTA/Ti and D-TiO2 NTA/Ti. The negative shift in the binding energies of Ti 2p1/2 and Ti 2p3/2 after the cathodic polarization indicates differing bonding situations due to the presence of Ti3+ (oxygen vacancy) in D-TiO2 NTA/Ti (Figure 2e). To clarify this point, the intensity of TiO2 NTA/Ti is subtracted from that of D-TiO2 NTA/Ti and plotted as the difference in intensity vs. the binding energy (Figure 2f). This plot clearly shows a peak at 457.9 eV and one at 463.6 eV that can be ascribed to Ti 2p1/2 and Ti 2p3/2 of Ti3+, which suggests the presence of Ti3+ on the surface of TiO2 nanotubes.33, 34 Symmetric cable-type supercapacitors were assembled by using two D-TiO2 NTA/Ti electrodes by following the steps illustrated in Figure S3a with poly(vinyl alcohol)–potassium hydroxide(PVA–KOH) as the solid-state electrolyte. A digital photograph of one of the cable-type supercapacitors prepared for the analysis is given in Figure S3b. CV studies were primarily used to understand the electrochemical performance of the cable-type supercapacitors. Electrodes containing D-TiO2 nanotubes with a length of about 1.6 μm were first utilized for the fabrication of cable-type supercapacitors (denoted as CTSC-A). The CV curves of the CTSC-A obtained at 1, 10, 50, 100, and 200 V s–1 are given in Figure 3a–e. The CTSC-A maintains a nearly rectangular shape even at an ultrahigh scan rate of 200 V s–1, which indicates the perfect charge-storage characteristics of the device. Figure 3f shows the variation of the discharge current at different scan rates, obtained for CTSC-A. The discharge current maintained a linear relationship until 100 V s–1, which illustrates the ideal capacitive characteristics of the CTSC-A and also the high instantaneous power characteristics of the device. The capacitance values of the device (CD) measured for CTSC-A at scan rates of 1 and 50 V s–1 are 71.89 (areal capacitance of the electrode (CA) = 0.91 mF cm–2) and 55.58 μF cm–1 (CA = 0.70 mF cm–2), respectively. To understand the influence of the length of the D-TiO2 nanotubes on the CTSC performance, CTSC-B and CTSC-C were fabricated by using electrodes with 7- and 26-μm-long D-TiO2 nanotubes, respectively. The cross-sectional images of the electrodes are given in Figure S4. The CV curves of CTSC-B measured at 1 and 10 V s–1 are respectively given in Figure 4a,b and those of CTSC-C are given in Figure 4d,e. Figure S5 shows the CV curves of CTSC-A and CTSC-B obtained at 200 V s–1. These CV curves clearly show that both CTSC-B and CTSC-C can also work at ultrahigh scan rates. The ultrahigh scan rates reported here are two to three orders of magnitude higher than those used for conventional cable-type supercapacitors (Table S1). Figure 4c and 4f show the variation of the discharge current at different scan rates in CTSC-B and CTSC-C, respectively. The discharge currents of CTSC-B and CTSC-C maintain a linear relationship until 50 and 10 V s–1, respectively. The aforementioned point also indicates the superior charge-storage capability of the D-TiO2/Ti electrodes over conventional thin-film-based electrodes. The measured capacitance values of the device (CD) at a scan rate of 1 V s–1 are CTSC-B = 217.03 (CA = 2.74 mF cm–2) and CTSC-C = 859.83 μF cm–1 (CA = 10.82 mF cm–2). At a scan rate of 50 V s–1, the CD value of CTSC-B = 169.47 (CA = 2.14 mF cm–2) and CD of CTSC-C = 358.16 μF cm–1 (CA = 4.51 mF cm–2). The capacitance values obtained here are comparable with the capacitance of the recently reported cable-type supercapacitors (Table S1), which indicates the great potential of D-TiO2/Ti based cable-type supercapacitors for high-power-demanding applications. The charge-storage capabilities of the cable-type supercapacitors were further characterized by means of charge–discharge measurements at different current densities (Figure S6). The charge–discharge profiles of CTSC-A, CTSC-B, and CTSC-C measured at a current density of 40 μA cm–1 are shown in Figure 5a. All three devices display a perfect charge–discharge profile without a visible IR drop due to internal resistance. The perfect charge–discharge curves indicate the ideal capacitive behavior of the device due to the efficient charge-storage process occurring on D-TiO2 NTA/Ti based electrodes. The short ion-diffusion paths in the as-prepared electrodes are mainly responsible for achieving such an excellent charge-storage behavior. This result also indicates that the surface of the well-ordered 3D D-TiO2 NTA/Ti electrodes can be easily accessed by ions in the electrolyte even at extreme current densities, which is not possible in conventional cable-type supercapacitor electrodes. The CD values of the CTSC-A, CTSC-B, and CTSC-C measured at a current density of 13 μA cm–1 are 73.71 (CA = 0.93 mF cm–2), 232.17 (CA = 2.93 mF cm–2), and 823.09 μF cm–1 (CA = 10.37 mF cm–2), respectively. As shown in Figure 5b, all three devices maintained very high CD values even at high current densities. For example, CTSC-A can deliver 62.49 μF cm–1 (CA = 0.79 mF cm–2) at 267 μA cm–1, CTSC-B can deliver 209.64 μF cm–1 (CA = 2.64 mF cm–2) at 427 μA cm–1, and CTSC-C can deliver 769.07 μF cm–1 (CA = 9.69 mF cm–2) at 640 μA cm–1. These high capacitance values also indicate the excellent rate capability of the device, which should be favorable during practical applications. The high capacitance obtained from the thick electrode (26 μm) indicates the advantage of the present electrodes over thin-film electrodes. Electrochemical impedance spectroscopy is a very powerful tool that we used to understand the resistive and capacitive characteristics of the cable-type supercapacitors electrodes. Nyquist and Bode plots obtained from the impedance measurements are given in Figure 5c and 5d, respectively. The Nyquist plots clearly show nearly vertical graphs without a high-frequency resistor–capacitor loop for all the cable-type supercapacitor electrodes, which indicates a low charge-transfer resistance of the electrodes. The inset of Figure 5c shows the enlarged area of the high-frequency region, which gives information about the equivalent series resistance (ESR). The ESR (a combination of solution resistance, current collector resistance, and interfacial contact resistance at the current collector and active electrode material) of the devices is 1.1 Ω for CTSC-A, 1.7 Ω for CTSC-B, and 2.8 Ω for CTSC-C. Such low ESR values should be mainly attributed to the excellent electrical conductivity and ordered nanotubes in all electrodes, which is also the one of the main reasons for obtaining ideal CV and charge–discharge profile at higher rates. The Bode plots in Figure 5d show a phase angle close to −90° at low frequency, indicating the ideal capacitive behavior. To understand the cyclic stability of the of D-TiO2-based cable-type supercapacitor electrodes, the charge–discharge measurement was carried out on CTSC-C for 20 000 cycles at a current density of 267 μA cm–1 (Figure 5e). CTSC-C retained around 95% of its initial capacitance after 20 000 cycles, which indicates the long-term charge-storage capability of the device. Since the mechanical flexibility of the device is also very important for practical applications, the CV performance of CTSC-C was measured at a scan rate of 50 V s–1 while bent, with a bending angle of 90°. The performances of the device in its normal state (with a bending angle of 0°), after bending (with a bending angle of 90°), and after straightening from the bent state (bending angle of 0°) are given in Figure 5f (A representative video of the measurement is given in the Supporting Information). It is obvious from Figure 5f that there is no significant change in the performance of the device while bent. The cyclic stability of the device in the normal state and after bending is given in Figure S7. The presence of PVA-based solid electrolyte also plays a key role in preventing damage to the electrode as it encapsulates the nanotube arrays. Together with the impressive capacitance and ultrahigh rate capability, the outstanding cyclic stability and mechanical flexibility of the devices clearly show the great potential of the D-TiO2 NTA/Ti-based cable-type supercapacitors for next-generation electronic-device applications. The outstanding ultrahigh scan-rate performance of D-TiO2 NTA/Ti-based cable-type supercapacitors can be explained by the collective effects of the short ion-diffusion path and the high electrical conductivity of the electrodes. The unique in situ growth structure of the D-TiO2/Ti electrode ensures an uninterrupted pathway for both ion and electron transport, to achieve an ultrahigh rate performance. The short ion-diffusion path of the ordered D-TiO2 nanotube arrays facilitates fast ion transport during the charge–discharge process, while providing an extremely high surface area. The short ion-diffusion path of the electrode is mainly helpful for achieving an outstanding ultrahigh rate performance. Further, the direct growth of the electrode structure from the Ti wire and the presence of Ti3+ ions in the electrode significantly contribute towards the achievement of low electrode resistance. The charge-transfer characteristics and ESR of the devices obtained from the impedance spectroscopic measurements confirm the low resistive behaviors of the electrodes. Also, the Ti beneath the nanotube arrays maintains the integrity of nanotube arrays at the same time as acting as an excellent current collector. In summary, we have introduced a very effective strategy for achieving ultrahigh rate performances in cable-type supercapacitors. Electrochemical measurements confirm that the D-TiO2 NTA/Ti electrode can provide a large surface area, an improved electrical conductivity, and a short ion-diffusion path for realizing an excellent ultrahigh rate performance in cable-type supercapacitors. Due to the unique structural features of D-TiO2 NTA/Ti electrodes, the fabricated all-solid-state cable-type supercapacitors can show ideal capacitive behavior and superior rate capability even at an ultrahigh scan rate of 200 V s–1. To our knowledge, the ultrahigh rate capability reported here is among the best in comparison with previously reported cable-type supercapacitors. The capacitance of the cable-type supercapacitors based on 26-μm-thick electrodes is 859.83 μF cm–1 at a scan rate of 1 V s–1. Besides the ultrahigh scan performance, the cable-type supercapacitors also have excellent cyclic stability. Additionally, the devices can also show comparable energy-storage characteristics even in the mechanically bent state. Accordingly, the ultrahigh rate capability of the D-TiO2 NTA/Ti-electrode-based cable-type supercapacitors should be beneficial for developing next-generation flexible and wearable electronic devices. Preparation of D-TiO2 NTA/Ti Nanotube Arrays: D-TiO2 NTA/Ti was prepared by a two-step process. In the first step, the Ti wire (diameter: 0.5 mm) was anodized and subsequently annealed to obtain TiO2 NTA/Ti. TiO2 NTA/Ti was fabricated by the anodization of the Ti wire at 60 V. A mixture of ethylene glycol, NH4F (0.5%) and water (3%) was utilized as the electrolyte for the anodization. A platinum foil was used as the counter electrode. After the anodization, the sample was subjected to mild ultrasonication in ethylene glycol for 15 min and subsequently in ethanol for 45 min to remove debris. The samples were washed with deionized water and then dried under nitrogen flow. Later, the sample was annealed at 450 °C for 1 h at a heating rate of 1°C min–1.35-37 In the second step, an electrochemical reduction (cathodic polarization) was performed in the second step to convert as prepared TiO2 NTA/Ti into D-TiO2 NTA/Ti. This process was executed in a three-electrode cell at room temperature by employing TiO2 NTA/Ti, Ag/AgCl electrode, and a platinum foil as the working, reference, and counter electrodes, respectively. 0.5 M Na2SO4 solution was used as the electrolyte. The cathodic reduction of the electrode was done at –1.4 V. The process was optimized by carrying out the cathodic polarization process at three different time periods 5, 10, and 15 min. After the reduction process, the D-TiO2 nanotubes on the Ti wire electrodes were washed with water and dried.31, 38 Fabrication of Symmetric Cable-Type Supercapacitors: The samples obtained after the cathodic polarization were directly used as binder-free electrodes for cable-type supercapacitor fabrication. In the first step, the electrolyte was coated onto two identical electrodes by the dip coating method. For this, the electrodes were dipped in PVA-KOH electrolyte (kept at 90 °C) and later dried at 60 °C by using an oven. During the coating process, only the area containing D-TiO2 NTA was exposed to the electrolyte. In the second step, the electrodes were combined together using a polyester thread (Wenco garn extra). While combining the electrodes, the D-TiO2-free area of the electrodes was placed at the opposite ends. In the third step, the electrolyte was coated on combined electrodes and then placed inside of a heat-shrunk tube (Conrad). Subsequently, the whole sample, containing tube and electrodes, was dried in an oven at 60 °C. Materials Characterization: SEM images were collected on a Hitachi S-4800 SEM equipped with an EDX detector (Thermo Fischer Scientific). The SEM images were taken at 5 kV and EDX measurements were done at 15 kV. TEM measurements were carried out by using a JEOL 200 CX at 200 kV. The XRD pattern of the samples was analyzed by using a Siemens D5000 diffractometer using Cu K-α radiation. The scan rate during the XRD measurement was 1° min–1. X-ray photoelectron spectroscopy (XPS) was employed to understand the doping of TiO2 with Ti3+. A PHI Quantum 2000 XPS system (Physical Electronics Inc., Eden Prairie, MN, USA) with monochromatic Al-Kα X-ray source (1.4866 keV) operating at 25 W (emission current of 1.66 mA and anode potential of 15 kV) was used to analyze the samples. The analyzer operated in the constant analyzer mode during the measurements and the pass energy was 23.5 eV for high-resolution scans. The angle between analyzer and source was 90° and the angle between the sample and source was 45°. The pressure in the analysis chamber was kept as 10–9 Torr. A charge neutralizer was utilized to prevent the charging of the sample. Cyclic voltammetry, galvanostatic charge–discharge and impedance spectroscopy measurements were carried out on cable-type supercapacitors by using a Biologic electrochemical work station. Electrochemical Characterization: Cyclic voltammetric measurements and charge–discharge measurements were carried out by using a Biologic potentiostat (VSP) at a potential window from 0–0.6 V at different scan rates and current densities. Impedance measurements were performed at open-circuit potential in the frequency range of 0.01 Hz to 1 MHz at an amplitude of 10 mV. The authors acknowledge funding from the European Research Council (ThreeDsurface: 240144), BMBF (ZIK-3DNanoDevice: 03Z1MN11), BMBF (Meta-ZIK-BioLithoMorphie: 03Z1M511), and German Research Foundation (DFG: LE 2249_4-1) for financial support of this work. The authors also acknowledge the help of Mr. Jens Schawohl and Dr. Henry Romanus for XRD and TEM characterizations, respectively. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.