Facile fabrication of well-defined polyaniline microtubes derived from natural kapok fibers for supercapacitors with long-term cycling stability

Abstract

Hollow micro-/nano-structured materials have been recognized as a promising material for applications in energy-related systems, especially supercapacitors. In this study, polyaniline (PANI) microtubes derived from low-cost natural kapok fibers were facilely fabricated as the electrode materials of supercapacitors. The kapok fiber templates were removed facilely using a NaOH solution after the in situ polymerization of aniline on the outer surface of the kapok fibers. The PANI microtubes etched using 6.0 M NaOH for 60 min exhibit the highest specific capacitance of 667 F g⁻¹ in 1.0 M H₂SO₄. Interestingly, the as-prepared PANI microtubes showed excellent cycle stability with capacitance retention of 60.7% compared to the origin capacitance after 10 000 cycles. Asymmetric supercapacitors, which were fabricated based on the positive electrode of the as-prepared PANI microtubes and negative electrode of the commercial activated carbon showed a high energy density of 14.1 W h kg⁻¹. The superior electrochemical performance of the PANI microtubes might be due to their hollow structure, which can facilitate the ion diffusion and improve the utilization of the electroactive PANI during the charge–discharge processes. In addition, the residual kapok fiber can effectively relieve the contraction/expansion of PANI during the doping/dedoping processes.

---

Introduction

Polyaniline (PANI) is a promising candidate in pseudo-capacitor electrode materials because of its low cost, facile synthesis, environmental stability, high conductivity and different redox states.¹ Therefore, a range of micro/nanostructured PANI has been synthesized as electrode materials for supercapacitors such as nanowires, nanobelts, nanoparticles, hollow microspheres, nanotubes, hollow spheres, nanoballs, nanosheets, hollow arrays, and hollow nanofibers.^2–9 Interestingly, hollow tubular structured PANI exhibit enhanced performance as supercapacitor electrode materials owing to the enlarged effective faradic reactive sites and easy penetration of electroactive substances into the interior of the electrodes, leading to a high coefficient of utilization of electrode materials.¹⁰ Therefore, the hollow tubular PANI has been synthesized for high-performance supercapacitor electrodes by soft and hard template methods.¹⁰ By contrast, the hard template is an efficient and common route due to the controllable diameter and length provided by templates.¹¹ However, the associated disadvantages of the hard-template method have restricted its utility such as the expensive template materials and the harsh chemical treatments for the removal of templates.¹² In addition, most hollow tubular PANI materials show a unsatisfactory cycling stability. Wang et al. synthesized hollow PANI arrays using the electrodeposited ZnO nanorods as template exhibiting a poor cycling stability with a capacitance retention ratio of 70% after 400 cycles.¹³ Miao et al. fabricated hollow PANI nanofibers by in situ polymerization of aniline based on the electrospun poly(amic acid) fiber membrane, and the capacitance retention ratio of the obtained PANI nanotubes is approximately 62% after 500 cycles.¹⁴ Thus, it remains a challenge to fabricate hollow tubular PANI for electrode materials of supercapacitors with high capacitance retention after long cycles via a facile, mild, and effective hard-template method.

Recently, the low-cost natural biosourced or waste-derived materials is used widely in the field of energy conversion and storage. Cellulose and its derivatives have been used successfully for the preparation of green, inexpensive, and highly efficient quasi-solid electrolytes for the bio-based electrodes of lithium batteries and supercapacitors.^15,16 Kapok fiber (KF) is a type of single-cell natural cellulose fiber with rather fine natural microtubules.¹⁷ This fiber has potential for various new applications, especially as the support for active nanoparticles and as a biotemplate to prepare a series of microtubes from organic or inorganic materials.¹⁸ Our groups have fabricated KF-based electroactive composites with hollow tubular structures as electrode materials for supercapacitors.^18,19 Taking the unique hollow structure of KF into account, it can be employed to design PANI microtubes as a green and low-cost template. In this study, the well-defined PANI microtubes are fabricated by etching KF using NaOH solution after the in situ polymerization of aniline on the surface of KF. The effect of the etching conditions on the structure and electrochemical properties of the PANI microtubes was evaluated in detail, including the concentration of NaOH and the etching time. In addition, the asymmetric supercapacitor is assembled based on the obtained PANI microtubes and the commercial activated carbon (AC) as the positive and negative, respectively. The results suggest that the low-cost KF templates can be employed to fabricate PANI microtubes for high-performance supercapacitor with long-term cycling stability.

Experiments

Materials

KF was purchased from Shanghai Pan-Da Co., Ltd., China. NaClO₂ (chemically pure) was provided by Beijing Hue-Wei Chemical Reagent Co., China. Acetic acid and other reagents were all of analytical reagent grade from Tianjin Chemical Co., China, and used without further purification. Ultrapure water (18.25 MΩ cm) was used throughout the study.

Preparation of PANI microtubes

First, KF was pretreated with NaClO₂ solution to remove the waxy coating and create a hydrophilic surface.¹⁹ KF (100 mg) was dispersed in 60 mL of a HCl solution (1.0 M) combining stirring with ultrasound, aniline (1.6 mL) was added to the abovementioned solution with stirring for 30 min in an ice bath, and 20 mL of aqueous solution containing APS was then added to initiate the polymerization of aniline. The obtained composite was washed with water and separated by centrifugation until a neutral pH was reached, and the resulting product was then dried under vacuum at 40 °C and marked as KF/PANI. The KF/PANI composite was soaked in 6.0 M NaOH solution for 60 min under ultrasound to remove KF template. The obtained PANI microtubes were separated by centrifugation and washed to neutral before being dried under vacuum at 80 °C. To investigate the effect of the etching conditions on the structure and electrochemical properties of the PANI microtubes, the concentration of NaOH solution and the etching time were studied. As a control, the bulk PANI was also synthesized using the same method without addition of KF.

Characterization

The morphologies of KF and PANI microtubes were characterized using an S-4800 field emission scanning electron microcopy (SEM) (HITACHI, Tokyo, Japan). A Bruker IFS 66v/s IR spectrometer (Bruker, Karlsruhe, Germany) was used for the Fourier transformed infrared (FTIR) analysis in the range of 400–4000 cm⁻¹ with the resolution of 4 cm⁻¹. X-ray diffraction spectrographs analysis (XRD) was conducted using an X-ray powder diffractometer with a Cu anode (PAN analytical Co. X'pert PRO), running at 40 kV and 30 mA. The chemical composition and distribution of the products were obtained by energy dispersive X-ray spectroscopy (EDS). Thermogravimetric analysis (TGA) was performed on a Perkin Elmer STA6000 thermogravimetric analyzer at a heating rate of 20 °C min⁻¹ under an oxygen purge with a flow rate of 200 mL min⁻¹. The X-ray photoemission spectroscopy (XPS) analyses are carried out using a X-ray Photoelectrometer (K-Alpha-surface Analysis, Thermon Scientific) with a hemispherical energy analyser and using a monochromatic Al K_α X-ray source (1361 eV).

Electrochemical analysis

All electrochemical experiments were carried out using a three-electrode, platinum counter electrode, and a standard calomel reference electrode (SCE) 1.0 M H₂SO₄ aqueous. The working electrodes were prepared by mixing the PANI microtubes, carbon black, and polytetrafluoroethylene at a weight ratio of 80 : 10 : 10 and pressed on stainless steel. The electrochemical behavior of the working electrodes was evaluated by cyclic voltammetry (CV), galvanostatic charge–discharge (GCD) and electrochemical impedance spectroscopy (EIS) using a CHI660E electrochemical working station. CV tests were performed in the potential window ranging from 0 to 0.8 V (vs. SCE) at 10, 20, 40, 60, 80 and 100 mV s⁻¹ in 1.0 M H₂SO₄ solution. The EIS measurements were carried out over the frequency range from 100 kHz to 0.005 Hz at the open circuit potential with a perturbation of 5 mV. The capacitance was calculated from the discharge curves according to eqn (1):²⁰where C_s, I, t, m and V are the specific capacitance, the constant discharge current, the discharge time, the mass of active materials in a single electrode, and the discharge voltage, respectively.

Preparation of the asymmetric supercapacitor (ASC)

To fabricate ASC device, PANI microtubes was used as the positive electrode and AC was served as the negative electrode with one piece of cellulose paper separator. Each electrode had a geometric surface area of 1 cm². The charge balance between the two electrodes will follow the relationship, q⁺ = q⁻, the voltammetric charges (q) were calculated based on the following eqn (2):

q = C × V × m (2)

where C is the specific capacitance (F g⁻¹) of each electrode measured in a three electrode setup (calculated from GCD curves), V is the potential window (V), and m is the mass of the electrode (g). To balance the charges stored on was two electrodes, the mass ratio between the positive and negative electrodes was obtained based on the following eqn (3):²¹where m, C and V are the mass of the active material, capacitance and potential window of the electrode, respectively. The “+” and “−” represent the positive and negative electrode, respectively. The capacitance of the assembled ASC device was calculated from eqn (1) based on the total mass of active materials on both two electrodes. The energy density and power density were calculated from eqn (4) and (5):where E (W h kg⁻¹) and P (W kg⁻¹) are the energy density and power density, respectively. The definitions of C, V and t are the same as those in eqn (1).

Result and discussion

Preparation of the PANI microtubes and structure characterization

The preparation protocol of PANI microtubes derived from KF are illustrated in Scheme 1. First, the coating of PANI on KF surface was completed by in situ chemical oxidative polymerization using aniline as the monomer and APS as the oxidant, and KF templates are then removed facilely using a common NaOH solution. Therefore, PANI with a well-defined microtube structure was successfully obtained, which can be anticipated to display improved cyclic capacity retention and rate capability.

Scheme 1 Schematic for the preparation of PANI microtubes derived from KF.

The FTIR spectra of KF, KF/PANI and PANI microtubes are shown in Fig. 1a. A broad absorption band was observed at about 3393 cm⁻¹, which is characteristic of the stretching vibration of –OH in cellulose of KF. The absorption band at 2914 cm⁻¹ was assigned to asymmetric and symmetric stretching vibration in –CH₂ and –CH₃ group in the acetyl ester groups of cellulose in the KF.¹⁸ The three important ester bands at 1741, 1375 and 1255 cm⁻¹ were associated with the stretching and bending vibrations of carbonyl bonds (C=O) of the acetyl ester group in the KF.¹⁸ The stretching vibration of C–C in different substituted aromatic rings in lignin were observed at 1592 and 1463 cm⁻¹ in KF. The bands in the 1000–1450 cm⁻¹ region correspond to the C–O stretching (ester or ether of carbohydrate and polysaccharide) and O–H bending vibrations.¹⁸ The FTIR spectrum of the KF/PANI composite is similar to that of KF apart from several characteristic absorption bands originating from PANI. The absorption bands located at 1611 cm⁻¹ and 1492 cm⁻¹ are associated with the C=C stretching vibration of benzenoid ring.²² Two peaks at 1306 and 1241 cm⁻¹ were attributed to the C–N stretching mode of the aromatic amine,²² suggesting that PANI had been coated successfully on the surface of KF. In the case of PANI microtubes, the characteristic bands of KF at 1741, 1374 and 1244 cm⁻¹ almost disappeared. This may be due to the deacetylation by the cleavage of acetyl groups linked as an ester group to the celluloses in NaOH solution in KF.²³ The relative intensity of the absorption peaks at 2914 cm⁻¹, 1254 cm⁻¹ and 1000–1450 cm⁻¹ decrease but did not disappear, which may be due to the KF is partially etched by 6.0 M NaOH solution.

Fig. 1 (a) FTIR spectra, (b) XRD patterns, (c) TGA curves and (d) XPS survey spectra of KF, KF/PANI and PANI microtubes.

The X-ray diffraction patterns of KF, KF/PANI and PANI microtubes are depicted in Fig. 1b. The characteristic diffraction peaks of cellulose appear at 2θ = 15.59° and 22.60° for KF, which correspond to the (110) and (200) crystallographic planes, indicating a cellulose Iβ pattern.²⁴ When PANI is introduced to the surface of KF, the characteristic diffraction peaks of cellulose in KF/PANI composites decrease in relative intensity and become boarder. Furthermore, the diffraction peaks at 2θ = 15.1°, 20.7° and 25.5° can be attributed to the (011), (020) and (200) crystal planes of PANI in its emeraldine salt form, respectively.²⁵ In the case of PANI microtubes, it can be observed that the characteristic diffraction peak of cellulose at 2θ = 15.59° almost disappears with a concomitant decrease in the relative intensity of the peak at 2θ = 22.60°. It can be inferred that KF is not removed completely and a small amount is left in the PANI microtubes. This may be assigned to the interaction between KF and PANI layer, including hydrogen bonding and electrostatic attraction.²⁶ A pair of new peaks can be clearly observed at 2θ = 12.29° and 20.35°, which was ascribed to PANI in its emeraldine base form.²⁷

The thermal stability of KF, KF/PANI and PANI microtubes was investigated by TGA under an O₂ atmosphere (Fig. 1c). The initial weight loss in the range from 30 to 120 °C is caused by the evaporation of moisture for all samples. For the TGA curve of KF, the sudden steep weight losses started at 240 °C and ended at around 365 °C with almost 77.6% weight loss. This may be due to the decomposition and dehydration process of cellulose in KF.²⁸ In the case of the KF/PANI composite, the large weight loss with about 45.4% occurred in the temperature range of 220–320 °C is related to the removal of doping anions and the decomposition of KF.^28,29 This was followed by slow degradation with a weight loss of 11% between 330 and 580 °C, which can be assigned to the carbonization of KF and PANI.³⁰ The TGA curve of PANI microtubes is similar with that of the KF/PANI composites. Within the same temperature range, the weight losses of PANI microtubes were about 30% and 25%. The difference in the TGA profiles of the KF/PANI and PANI tube is may be assigned to the different content of KF in the two samples. This suggests that the KF is partially removed by the 6.0 M NaOH solution, which is also consistent with the result of XRD.

The elemental composition and surface chemical status of KF, KF/PANI and PANI microtubes were analyzed by XPS. The survey spectra of KF, KF/PANI, and PANI microtubes are presented in Fig. 1d. It is clearly observed that KF is composed mainly of carbon and oxygen, while the survey spectrum of KF/PANI composites showed the characteristic peaks of C, O and N, indicating the presence of PANI. After removing the KF template, the relative intensity of the C, O, and N characteristic peaks in the PANI microtubes is higher than that of the KF/PANI composites due to the partial removal of KF. As shown in Fig. 2a and c, the C 1s XPS spectra of KF/PANI and PANI microtubes showed five peaks with energies of 284.5 eV, 284.8 eV, 285.1 eV, 285.6 eV and 287.5 eV, which can be assigned to the C atoms coming from C–N/C=N, C=C/C–C, C–O, C=O and –O–C=O, respectively.³¹ As shown in Fig. 2b, the N 1s spectrum of the KF/PANI composite can be deconvoluted into four distinct curves associated with different nitrogen forms of imine (=N–) at 398.5 eV, amine (–NH–) at 399.5 eV, protonated amine (–NH₂⁺) at 401.1 eV and protonated imine (=NH⁺) at 402.2 eV.^32,33 This is in line with the emeraldine salt form of polyaniline. After being treated with a 6.0 M NaOH solution, it can be found that the peaks for protonated imine (=NH⁺) and protonated amine (–NH₂⁺) disappear and N 1s spectrum of the PANI microtubes decreases (Fig. 2d), indicating the dedoping of PANI. This is characteristic of the emeraldine base form of PANI.^34,35

Fig. 2 C 1s and N 1s XPS spectra of KF/PANI (a and b) and PANI microtubes (c and d).

SEM images of KF, KF/PAN and PANI microtubes are shown in Fig. 3. KF has a silky surface and presents a regular hollow tubular structure with outer and inner diameters of 20–25 and 16–23 μm, respectively (Fig. 3a and inset). After being coated with PANI, the surface of KF becomes coarse accompanied by large PANI aggregations (Fig. 3b and inset). This indicates that the two components (KF, PANI) are strongly coupled together during the in situ polymerization process. Furthermore, it is worth mentioning that the hollow structure of KF is retained well in the process of modification with PANI, and the coating of PANI is focused mainly on the outer surface of KF. In addition, the chemical composition of the KF/PANI composite was confirmed by EDX. The results are depicted in the Fig. 4. It can be found that the composite is composed mainly of carbon, oxygen, and nitrogen elements; this characteristic agrees with the XPS result. The nitrogen species in the KF/PANI composite is approximately 9.21%, which was derived from PANI. For the PANI microtubes, the SEM image showed that the PANI microtubes exhibit shrunken tubular shape, which may be ascribed to the removal of KF. The PANI microtubes are tested in the dry state (Fig. 3c). In addition, a small amount of free PANI particles were attached to the outer surface of the PANI microtubes. The inset image shows the complete outer surface of the PANI microtubes. It can be clearly observed that the hollow structure in PANI microtubes was maintained in the process of removal KF template. The complete nozzle of the PANI microtubes was also presented in the inset. It can be anticipated that this feature benefits the penetration of the electrolyte, which may help improve the electrochemical performance.

Fig. 3 SEM images of KF (a), KF/PANI (b) and PANI microtubes (c).

Fig. 4 Element mapping and EDX curve of the KF/PANI composite: (a) SEM image, (b) C, (c) O, (d) N, and (e) EDX curve.

Electrochemical analysis

Compared to bulk PANI particles, the obtained PANI microtubes would be expected to be a promising candidate for the construction of high-performance supercapacitors due to the unique properties arising from their hollow structure. Therefore, the obtained PANI microtubes was used to fabricate supercapacitor electrodes and characterized by CV, GCD, EIS, and cycle stability to investigate the potential application in electrochemical energy storage.

Fig. 5a shows the CV curves of PANI microtubes with a potential range from 0 V to 0.8 V (vs. SCE) at different scan rates in a 1.0 M H₂SO₄ aqueous solution. The CV curves exhibit two pairs of distinct redox peaks, which were ascribed to the transformation between the leucoemeraldine base and emeraldine salt state and between emeraldine salt state and pernigraniline base states, respectively.^36,37 The rate capability of the PANI microtubes was further evaluated by CV at various scan rates (10, 40, 80 and 100 mV s⁻¹). By increasing the scan rate from 10 to 100 mV s⁻¹, the response current density increases gradually, demonstrating good rate capability. Moreover, the shape of the CV curves remains unchanged with an increasing scan rate. This further confirms the excellent rate capability.

Fig. 5 (a) CV curves of the PANI microtubes in 1.0 M H₂SO₄ electrolyte at different scan rates, (b) GCD curves of PANI microtubes in 1.0 M H₂SO₄ electrolyte at various current densities, (c) Nyquist plot of the PANI microtubes in a 1.0 M H₂SO₄ electrolyte; the inset is the selective area magnification of the EIS spectra and (d) cycle stability of the PANI microtubes in 1.0 M H₂SO₄ electrolyte for 10 000 cycles (the inset shows the last GCD curves).

The GCD tests were then carried out to evaluate the capacitance of the PANI microtubes at a series of current densities using a potential window 0–0.8 V versus SCE. The GCD curves are presented in the Fig. 5b. The PANI microtubes show typical capacitance characteristics, as displayed in the charge–discharge curves. The curve of the PANI microtubes is not an ideal symmetric profile, exhibiting the following two voltage stages; the first voltage stage in the range of 0.8–0.45 V is ascribed to the electrical double-layer (EDL) capacitance. In contrast, the second voltage stage in the range from 0.45 to 0 V with a longer discharge period implies that the electrode materials possess EDL and Faradaic capacitances simultaneously.³⁸ In addition, can be found that the IR drop of the PANI microtubes is about 0.05 V, exhibiting a low internal resistance.³⁸ The specific capacitance can be calculated according to eqn (1) used the GCD curves and is about 667, 534, 430, 376 and 354 F g⁻¹ at a current density of 2.0, 4.0, 6.0, 8.0 and 10.0 A g⁻¹, respectively. It retains about 54% as the current density is increased to 10 A g⁻¹, suggesting good rate capability. The high specific capacitance ascribed to the unique hollow structure is more favorable for the redox reaction to simultaneously occur on the surface and in the interior of the materials, which can improve the electrochemical activity of PANI.

The electrochemical performance of the PANI microtubes electrodes was investigated by EIS at the open circuit potential in a 1.0 M H₂SO₄ solution. The Nyquist plot is shown in Fig. 5c. It can be observed from the typical Nyquist plots that the impedance spectrum has two distinctive parts composed of a semicircular arc in the high frequency region and a straight line in the middle-to-low frequency region. The intercept for the real component at the beginning of the semicircle shows the combined series resistance (R_s) of the electrolyte, electrode, current collectors, and the electrode/current collector contact resistance.³⁹ It can be observed that R_s of the PANI microtubes is about 1.2 Ω. The small R_s of the sample was attributed to the unique hollow structure, which can facilitate the efficient access of the electrolyte ions to the electrode surface and shorten the ion diffusion path. The high frequency arc represents the charge transfer resistance (R_ct) caused by the electrochemical reactions at the contact interface between the electrode and electrolyte solution.⁴⁰ The R_ct obtained from the semicircular arc diameter is around 7.2 Ω for the PANI microtubes electrode. At the low frequencies, the straight line with a slope from 45° to 90° was attributed to the Warburg diffusion behavior. It can be observed that the PANI microtubes electrode exhibits a short and vertical line, which is representative of fast ion diffusion/transport in the electrode material.⁴¹ An enlarged scale has been provided in Fig. 5c and the Warburg resistance was calculated according to the intersection at the low frequency region in the EIS plot;⁴¹ the Warburg resistance was about 11.4 Ω.

The cycling stability is a key factor in determining the supercapacitor electrodes for many practical applications and the excellent cycling performance is crucial for real supercapacitor applications. The electrochemical stability of the PANI microtubes electrodes was investigated by GCD cycling in a 1.0 M H₂SO₄ solution at a current density of 4 A g⁻¹ for 10 000 cycles, and the capacitance retention is shown in Fig. 5d. The PANI microtubes exhibit a high retention of 60.7% compared to its initial capacitance after 10 000 cycles in this study, indicating long-term electrochemical stability. In general, PANI electrode materials suffer a serious disadvantage of poor cycling stability during the charge–discharge process because the redox sites in the polymer backbone are not sufficiently stable and they undergo swelling/shrinkage during the doping/dedoping process, which can cause volume changes and damage to the polymer backbone within a limited number of charge–discharge cycles.⁴² The enhanced electrochemical stability of the PANI microtubes electrode might be caused mainly by the hollow structure, which significantly enhances its mechanical properties and efficiently hinders the deformation of the conductive polymer during the long charge–discharge process. More importantly, the residual KF can effectively relieve the contraction/expansion of PANI during doping/dedoping processes. To explore the effects of the different quality of residual KF on the electrochemical performance of PANI microtubes, PANI microtubes were obtained using 2.0 M, 4.0 M, 6.0 M and 8.0 M NaOH solution for 60 min, and the electrochemical performance of the as-prepared PANI microtubes was evaluated using GCD measurements in 1.0 M H₂SO₄ aqueous at a current density of 2.0 A g⁻¹. The specific capacitance of these PANI microtubes is shown in Fig. S1a.† The specific capacitance is approximately 385, 435, 667 and 427 F g⁻¹, corresponding to the PANI microtubes, which were etched using the different concentration of 2.0 M, 4.0 M, 6.0 M and 8.0 M NaOH solution, respectively. The microtubes etched by 6.0 M NaOH showed the highest specific capacitance. This might be due to the complete hollow tube structure and appropriate residual KF in this microtube. SEM images of the four microtubes are presented in the Fig. S2.† The tubular morphology of the microtubes etched using a 2.0 M and 4.0 M NaOH solution was completely destroyed and the collapsed KF exhibits large sheet structure in the SEM images. A large of PANI fragments is clearly observed on the surface of the collapsed KF sheet. In addition, some free PANI was found in the SEM images, which may result in a decrease in the specific capacitance. However, the tubular structure of the microtubes etched using 6.0 M and 8.0 M NaOH solution was kept well. This may be because the solubility of KF depends on the concentration of the NaOH solution. The solubility of KF in different concentrations of NaOH solutions was explored and the dissolution process was recorded by a digital video (Video S1†). Fig. 6 shows digital images of the collected residual after being etched for 60 min in different concentrations of NaOH solution. The solubility of KF in NaOH solution decreases with increasing concentration of NaOH. This phenomenon may be due to the solubility of KF, and the main part of cellulose only can be dissolved in a narrow concentration range of NaOH solutions from 1.8 M to 6 M.^43,44 At low concentrations of NaOH (below 1.8 M), cellulose is hardly dissolved because the size of the hydrate is very large to penetrate into cellulose. At high concentrations of NaOH (higher than 6.0 M), NaOH prefers to stay close to the cellulose chains forming a Na/cellulose crystal. When the concentration of NaOH is between 1.8 M and 6.0 M, NaOH hydrates can penetrate the cellulose fibers and bind to each chain without forming a Na/cellulose crystal.^45–48 The inset shows a digital image of the remaining KF and the quality. The quality of the remaining KF increases with the concentration of NaOH solution. With increasing concentration of NaOH from 6.0 M to 8.0 M, the specific capacitance of the PANI microtubes decreases, which is due to the more KF residue. The excessive residue of KF reduces the conductivity of the corresponding PANI microtubes and the specific capacitance. As a control, bulk PANI was also synthesized and tested using the same method. As shown in Fig. S1a,† the bulk PANI exhibits the lowest capacitance of 252 F g⁻¹. This may be because the prepared PANI microtubes structure can improve the electrochemical property. The EIS spectra of the PANI microtube etched with the 2.0 M, 4.0 M and 8.0 M NaOH solution for 60 min was also tested and the result is depicted in Fig. S3.† The R_s value of the four microtubes was approximately 0.95–1.05 Ω. However, the R_ct increases with increasing concentration of NaOH solution. This result is agreement with the specific capacitance and solubility experiments. Although the microtubes etched using 2.0 M and 4.0 M NaOH exhibit low internal resistance, the collapsed structure may result in a lower specific capacitance. In addition, the effects of the etching time on the electrochemical properties of the PANI microtubes was also investigated. As shown in Fig. S1b,† the specific capacitances of these PANI microtubes were calculated to be 308, 325, 412, 644 and 677 F g⁻¹ corresponding to the etching time of 7.5 min, 15 min, 20 min, 30 min and 60 min, respectively. There was no significant change in specific capacitance after etching for 30 min and 60 min, which may be due to the residual KF quality remaining unchanged as the etching time increases from 30 min to 60 min.

Fig. 6 Digital images of KF (100 mg) dispersed in 5.0 mL of 2.0, 4.0, 6.0 and 8.0 M NaOH solution accompanied by agitation for 60 min (the inset shows the remained KF in the corresponding NaOH solution).

In addition, the cycling ability of the PANI microtubes etched with the 2.0 M, 4.0 M and 8.0 M NaOH solution for 60 min and bulk PANI was also evaluated and the result is summarized in the Fig. S4.† All samples showed good cycling ability. The capacitance retention of the four PANI microtubes is about 44.2%, 58.2%, 60.7% and 58.6% after 10 000 GCD cycles, respectively. The difference in the cycling stability performance may be due to the different quality of residual KF in the corresponding PANI microtubes. A comparison of the cycle stability of various PANI electrode materials presented in previous literature and this study is presented in Table S1.† It can be seen that the PANI microtubes show good cycling stability. Compared to the complicated preparation process of other micro/nano structured PANI materials, the PANI microtubes shown in the present study can be fabricated through a facile and mild process. Furthermore, compared to the high price of other templates, such as ZnO nanotubes and MnO₂ nanotubes, the present KF is cost-effective and it can be produced on a large-scale.

Building ASC

To test the real applications of the as-prepared PANI microtubes, an ASC device was fabricated using the PANI microtubes as a positive electrode and the AC as a negative electrode. According to the specific capacitance (175 F g⁻¹ for AC and 667 F g⁻¹ for PANI microtubes), the optimal mass ratio should be m₊/m₋ = 0.328 in the ASC device. Fig. S5a† shows the electrochemical behavior of the AC and the PANI microtubes electrodes measured independently in a three-electrode configuration in 1.0 M H₂SO₄ aqueous electrolyte. A stable CV curve was observed in the potential range from −1.0 to 0.0 V for the AC electrode and from 0.0 to 0.8 V for the PANI microtubes electrode at a scan rate 10 mV s⁻¹. The CV curve of the AC electrode exhibits a nearly ideal rectangular shape, and no peaks for oxidation and reduction were observed, indicating a typical characteristic of electrical double layer capacitor behavior. The CV curve of the PANI microtubes electrode with obvious redox peaks differs from that of the electrical double layer capacitor of AC. Fig. S5b† shows CV as a function of voltage for the ASC with an optimal mass ratio between the two electrodes. The supercapacitor shows ideal capacitive behavior with quasi-rectangular CV curves at a scan rate of 100 mV s⁻¹, even at the potential up to 1.6 V. The assembled ASC device exhibits quasirectangular-shaped CV curves coupled with a pair of redox peaks resulting from faradic reactions on PANI microtubes and AC electrodes. As the potential window is extended up to 1.6 V, an undesirable oxygen revolution reaction-induced peak is clearly observed within a potential range of 1.5–1.6 V. This suggests that the ASC device can deliver a maximum working voltage of 1.4 V, which was chosen as the default voltage value for further investigation. Fig. 7a shows the typical CV curves of the ASC device at various scan rates from 10 to 100 mV s⁻¹. The CV curves at different scan rates exhibit electrical double layer capacitance and pseudocapacitance from AC and PANI microtube electrodes, respectively. The shape of the CV curves was not significantly influenced by an increasing scan rate, indicating high rate capability and good reversibility of the supercapacitor. Fig. 7b shows the GCD curves of the ASC device at different current densities from 0.1 to 1 A g⁻¹ with a voltage window of 0–1.4 V. As shown, at a cell voltage as high as 1.4 V, both charge and discharge curves maintain very good symmetry, indicating the good capacitance characteristics and electrochemical reversibility. In addition, the ASC device can deliver a high cell capacitance of 51.79 F g⁻¹ at a current density of 0.1 A g⁻¹. As the discharge current densities increase to 0.25, 0.5, 1.0, 2.0 and 4.0 A g⁻¹, cell capacitances of 45.54, 36.22, 29.29, 21.57 and 13.42 F g⁻¹ were achieved, respectively, implying good rate capability. Power density and energy density are the two key parameters to characterize the performance of electrochemical supercapacitors. The energy density and corresponding power density of the ASC device were calculated based on eqn (4) and (5) and the results are demonstrated as a Ragone plot in Fig. 7c. An energy density of 14.1 W h kg⁻¹ was achieved at a power density of 70.0 W kg⁻¹. A comparison of the energy density between this supercapacitors and literature is presented in Table S2.† It can be observed that the value of the present supercapacitor was equal to those of many reported PANI composites devices. This may be related to the large scale hollow tubular structure. In addition, the PANI microtubes were fabricated through a simple etching process compared to the reported complicated preparation process of other similar PANI materials. Furthermore, considering the high price of other related methods and materials, the hollow PANI microtubes can be produced on a large-scale because KF is an agricultural product with a total annual production of two thousand tonnes, which is a potential template for fabricating large scale electroactive materials. The long-term cycle stability of the as-fabricated ASC device was also evaluated by repeating the GCD tests at a current density of 1.0 A g⁻¹ for 2000 cycles and the result is shown in Fig. 7d. The overall specific capacitance did not show clear degeneration, suggesting the as-fabricated device maintains acceptable cycling stability. Furthermore, two ASC devices were assembled in series and it was found that the device could power a red light-emitting diode (LED) well for about 5 min after charging for 30 s (Fig. 8 and Video S2†). This highlights the potential of the fabricated supercapacitor device in energy storage.

Fig. 7 (a) CV curves of the ASC device measured within the potential window 1.4 V at different scan rates, (b) GCD curves of the ASC device measured at different current densities between 0 and 1.4 V, (c) Ragone plots and (d) cycling stability of the ASC device.

Fig. 8 Digital images of a red-light-emitting diode (LED) lighted by the ASC device.

Conclusion

In summary, PANI microtubes were fabricated from the low-cost natural kapok fibers. The kapok fiber templates were removed facilely using a NaOH solution after the in situ polymerization of aniline on the surface of the kapok fiber. The electrochemical result shows that the PANI microtubes prepared using 6 M NaOH for 60 min has the highest specific capacitance of 667 F g⁻¹. The capacitance retention ratio is around 60.7% of the original capacitance after 10 000 cycles, suggesting excellent cycling stability. An ASC device fabricated based on the PANI microtubes as a positive electrode and the AC as a negative electrode shows high energy density of 14.1 W h kg⁻¹. The excellent electrochemical performance indicates that it may be a feasible approach to facilely prepare PANI microtubes with well-defined hollow structures and long term cycling stability derived from natural kapok fibers.

Acknowledgements

The authors gratefully acknowledge the support of the National Natural Science Foundation of China (No. 51303190).

Notes and references

1. J. W. Luo, W. B. Zhong, Y. B. Zou, C. L. Xiong and W. T. Yang, J. Power Sources, 2016, 319, 73–81.
2. N. Hui, F. L. Chai, P. P. Lin, Z. L. Song, X. T. Sun, Y. N. Li, S. Y. Niu and X. L. Luo, Electrochim. Acta, 2016, 199, 234–241.
3. J. Pedrós, A. Boscá, J. Martínez, S. Ruiz-Gómez, L. Pérez, V. Barranco and F. Calle, J. Power Sources, 2016, 317, 35–42.
4. S. Mondal, U. Rana and S. Malik, Chem. Commun., 2015, 51, 12365–12368.
5. Q. Zhou, L. Liu, Z. F. Huang, L. G. Yi, X. Y. Wang and G. Z. Cao, J. Mater. Chem. A, 2016, 4, 5505–5516.
6. Y. B. Li, Z. H. Li and F. Zheng, Mater. Lett., 2015, 148, 34–36.
7. H. Y. Mi, J. P. Zhou, Z. B. Zhao, C. Yu, X. Z. Wang and J. S. Qiu, RSC Adv., 2015, 5, 1016–1023.
8. L. J. Zhang, Y. Z. Long, Z. J. Chen and M. X. Wan, Adv. Funct. Mater., 2004, 14, 693–698.
9. K. Mallick, M. J. Witcomb, A. Dinsmore and M. S. Scurrell, Macromol. Rapid Commun., 2005, 26, 232–235.
10. L. F. Shen, L. Yu, X. Y. Yu, X. G. Zhao and X. W. Lou, Angew. Chem., Int. Ed., 2015, 54, 1868–1872.
11. P. Liu and L. Zhang, Crit. Rev. Solid State Mater. Sci., 2009, 34, 75–87.
12. P. Liu, X. Wang and Y. J. Wang, ACS Sustainable Chem. Eng., 2014, 2, 1795–1801.
13. Z. L. Wang, R. Guo, G. R. Li, H. L. Li, Z. Q. Liu, F. M. Xiao, M. Q. Zhang and Y. X. Tong, J. Mater. Chem., 2012, 22, 2401–2404.
14. Y. E. Miao, W. Fan, D. Chen and T. X. Liu, ACS Appl. Mater. Interfaces, 2013, 5, 4423–4428.
15. P. Hao, Z. H. Zhao, J. Tian, H. D. Li, Y. H. Sang, G. W. Yu, H. Q. Cai, H. Liu, C. P. Wong and A. Umar, Nanoscale, 2014, 6, 12120–12129.
16. Z. Weng, Y. Su, D. W. Wang, F. Li, J. H. Du and H. M. Cheng, Adv. Energy Mater., 2011, 1, 917–922.
17. Y. Liu, J. T. Wang, Y. A. Zheng and A. Q. Wang, Chem. Eng. J., 2012, 184, 248–255.
18. W. B. Xu, B. Mu, W. B. Zhang and A. Q. Wang, RSC Adv., 2015, 5, 64065–64075.
19. B. Mu, W. B. Zhang, W. B. Xu and A. Q. Wang, Electrochim. Acta, 2015, 178, 709–720.
20. B. Mu, W. B. Zhang, S. J. Shao and A. Q. Wang, Phys. Chem. Chem. Phys., 2014, 16, 7872–7880.
21. X. C. Li, J. J. Shen, W. Sun, X. D. Hong, R. T. Wang, X. H. Zhao and X. B. Yan, J. Mater. Chem. A, 2015, 3, 13244–13253.
22. H. Xu, J. L. Zhang, Y. Chen, H. L. Lu and J. X. Zhuang, RSC Adv., 2014, 4, 5547–5552.
23. B. F. Tjeerdsma and H. Militz, Holz als Roh-und Werkstoff, 2005, 63, 102–111.
24. A. D. French, Cellulose, 2014, 21, 885–896.
25. S. Mutyala, J. Mathiyarasu and A. Mulchandani, RSC Adv., 2015, 5, 92648–92655.
26. Y. A. Zheng, W. B. Wang, D. J. Huang and A. Q. Wang, Chem. Eng. J., 2012, 191, 154–161.
27. J. Yu, W. C. Guo, M. Yang, Y. Luan, J. Z. Tao and X. W. Zhang, Sci. China: Chem., 2014, 57, 1211–1217.
28. X. W. Shi, Y. L. Hu, F. Y. Fu, J. P. Zhou, Y. X. Wang, L. Y. Chen, H. M. Zhang, J. Li, X. H. Wang and L. N. Zhang, J. Mater. Chem. A, 2014, 2, 7669–7673.
29. B. Rajender and S. Palaniappan, New J. Chem., 2015, 39, 5382–5388.
30. R. S. Andre, A. Pavinatto, L. A. Mercante, E. C. Paris, L. H. Mattoso and D. S. Correa, RSC Adv., 2015, 5, 73875–73881.
31. W. B. Xu, B. Mu and A. Q. Wang, Electrochim. Acta, 2016, 194, 84–94.
32. R. B. Patil, A. A. Jatratkar, R. S. Devan, Y. R. Ma, R. K. Puri, V. Puri and J. B. Yadav, Appl. Surf. Sci., 2015, 327, 201–204.
33. S. Golczak, A. Kanciurzewska, M. Fahlman, K. Langer and J. J. Langer, Solid State Ionics, 2008, 179, 2234–2239.
34. S. Zhu, M. Wu, M. H. Ge, H. Zhang, S. K. Li and C. H. Li, J. Power Sources, 2016, 306, 593–601.
35. M. M. Mahat, D. Mawad, G. W. Nelson, S. Fearn, R. G. Palgrave, D. J. Payne and M. M. Stevens, J. Mater. Chem. C, 2015, 3, 7180–7186.
36. S. Y. Lee, J. I. Kim and S. J. Park, Energy, 2014, 78, 298–303.
37. Z. Shao, H. J. Li, M. J. Li, C. P. Li, C. Q. Qu and B. H. Yang, Energy, 2015, 87, 578–585.
38. J. Yan, T. Wei, B. Shao, Z. J. Fan, W. Z. Qian, M. L. Zhang and F. Wei, Carbon, 2010, 48, 487–493.
39. O. S. Kwon, T. J. Kim, J. S. Lee, S. J. Park, M. J. Kang, J. E. Lee, J. S. Jang and H. S. Yoon, Small, 2013, 9, 248–254.
40. Z. Gao, W. L. Yang, J. Wang, B. Wang, Z. S. Li, Q. Liu, M. L. Zhang and L. H. Liu, Energy Fuels, 2013, 27, 568–575.
41. Z. F. Li, H. Y. Zhang, Q. Liu, L. L. Sun, L. Stanciu and J. Xie, ACS Appl. Mater. Interfaces, 2013, 5, 2685–2691.
42. Y. Z. Li, X. Zhao, Q. Xu, Q. H. Zhang and D. J. Chen, Langmuir, 2011, 27, 6458–6463.
43. T. Budtova and P. Navard, Cellulose, 2015, 1, 1–51.
44. L. F. Yan and Z. J. Gao, Cellulose, 2008, 15, 789–796.
45. B. Xiong, P. P. Zhao, P. Cai, L. N. Zhang, K. Hu and G. Z. Cheng, Cellulose, 2013, 20, 613–621.
46. M. Egal, T. Budtova and P. Navard, Biomacromolecules, 2007, 8, 2282–2287.
47. B. Xiong, P. P. Zhao, K. Hu, L. N. Zhang and G. Z. Cheng, Cellulose, 2014, 21, 1183–1192.
48. A. Arias, M. C. Heuzey, M. A. Huneault, G. Ausias and A. Bendahou, Cellulose, 2015, 22, 483–498.

---

Footnote

† Electronic supplementary information (ESI) available: The specific capacitance of the PANI microtubes prepared using different NaOH concentration and different etching time. The SEM images, EIS spectra and cycling stability of PANI microtubes etched using different concentrations of NaOH solution. CV curves AC and PANI microtubes electrodes. CV curves of the PANI microtubes//AC asymmetric supercapacitor within different potential windows. The cycling stability and energy density comparison. See DOI: 10.1039/c6ra16899j

---