Fabrication of carbon nanofiber–polyaniline composite flexible paper for supercapacitor

Abstract

In this work we report a low cost technique, via simple rapid-mixture polymerization of aniline using an electrospun carbon nanofiber (CNF) paper as substrate, to fabricate free-standing, flexible CNF–PANI (PANI = polyaniline) composite paper. The morphology and microstructure of the obtained products are characterized by FESEM, FTIR, Raman and XRD. As results, PANI nanoparticles are homogeneously deposited on the surface of each CNF, forming a thin, light-weight and flexible composite paper. The resulting composite paper displays remarkably enhanced electrochemical capacitance compared with the CNF paper, making it attractive for high-performance flexible capacitors.

---

Introduction

The flexible electrochemical supercapacitor, one of the most promising modern energy storage systems, has aroused broad interest due to its potential applications in portable electronic devices, hybrid electric vehicles and medical devices.^1,2 An ideal flexible supercapacitor should have a combination of excellent mechanical strength and large electrochemical capacitance.³ Although transition metal oxides and conducting polymers have been widely studied as supercapacitors, only carbon-based materials including carbon nanotubes (CNTs),⁴ carbon fibers⁵ and graphene^3,6 have displayed favorable flexibility and hence been promising as freestanding and flexible capacitors. However, the use of these individual carbon materials in high-performance flexible supercapacitors is limited due to their drawbacks, such as the high cost of CNTs, the preparation complexity of graphene, and the unsatisfactory performance of carbon fibers. Therefore, research aimed at increasing electrochemical capacitance as well as lowering fabrication cost is very necessary.

It is well known that polyaniline (PANI) is one of the most promising conducting polymers. Owing to high capacitance, good environmental stability, and low cost combined with the easiness of preparation, PANI materials have been widely studied as supercapacitors.⁷ Also, PANI has been considered to promote the electrochemical capacitance of flexible CNT and graphene papers.^3,8,9

In this paper, we use electrospinning to produce flexible carbon nanofiber (CNF) paper. This electrospinning process is simple and fast, and the CNFs are relatively inexpensive compared with CNTs or graphene. More importantly, we first demonstrate the preparation of a freestanding CNF–PANI composite paper by a simple rapid-mixture polymerization of aniline monomers on the CNF paper. The obtained composite paper maintains the pristine flexibility of the CNF paper and displays improved electrical conductivity and electrochemical performances compared with the CNF paper.

Experimental

An N,N-dimethyl formamide (DMF) solution of polyacrylonitride (PAN, 10 wt%) was spun into PAN fibers using electrospinning equipment. The parameters were as follows: the applied voltage was 15 kV, the distance between the positive and negative electrodes was 15 cm, and the solution feed rate was 0.5 ml h⁻¹. The electrospun fibers were collected as a paper wrapped on a graphite drum. After that, the PAN fibers were stabilized at 270 °C in air and then heat-treated at 900 °C under nitrogen atmosphere to form CNF paper.

Aniline (6.4 mmol) and ammonium peroxydisulfate (APS, 1.6 mmol) were dissolved respectively in two 20 mL aliquots of 1 M HCl solution. The two solutions were rapidly poured together and immediately stirred before the polymerization began. After 30 s, the stirring of the solution was stopped and the CNF paper was carefully immersed into the solution. Polymerization was carried out at room temperature for 24 h without stirring, and then the paper was collected, washed with water, and dried. Meantime, nanostructural PANI generated in the aqueous phase was also collected.

The samples were characterized by field emission scanning electron microscope (FESEM, JSM 6701F), Fourier transformation infrared spectrometer (FTIR, Bruker IFS66V), Raman spectroscopy (JY-HR800, 532nm) and X-ray diffraction (XRD, Philips X’Pert Pro.). The electrical conductivity was measured by a standard four-probe method.

The electrochemical properties of the paper-like samples were carried out using a CHI660C Electrochemical Working Station in a conventional Teflon electrochemistry cell with a three-electrode system in 1 M H₂SO₄ electrolyte: a working electrode (WE), a platinum wire counter electrode and an Ag/AgCl (sat. KCl) reference electrode. Each paper-like sample was directly used as the WE and the area exposed to the electrolyte was about 0.28 cm². Due to the fact that the diameters of all the papers were the same (39 mm), the mass of the exposed electrodes could be calculated after measuring the mass of the whole paper, which was 8.3 × 10⁻⁵ g for CNF paper and 1.4 × 10⁻⁴ g for CNF–PANI paper (41 wt% for PANI phase, 59 wt% for CNF phase). The cyclic voltammetry (CV) curves were collected from −0.2 to 1.0 V versus Ag/AgCl at varying scan rates from 5 to 40 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) measurements were recorded from 10 kHz to 10 mHz with an AC amplitude of 5 mV. Galvanostatic charge/discharge measurements were run on from 0 to 0.8 V at varying current densities (1–5 A g⁻¹), and open circuit potential. The capacitance was calculated from the slope of the discharge curve, according to the equation C =(I × Δt)/(ΔV × m), where C is the specific capacitance, I is the constant discharge current, Δt is the discharge time, ΔV is the voltage difference in discharge and m is the mass of the exposed WE.

Results and discussion

The obtained freestanding CNF paper is flexible and black (Fig. 1). SEM analyses reveal that the CNF paper is consisted of layer-by-layer overlapped and crossed nanofibers (Fig. 2a–c) and the diameter of the fibers is ranging from 50 to 120 nm. After the polymerization, the color of the CNF paper is changed to dark gold, indicating an obvious difference in the structure or morphology for the CNF paper before and after polymerization. Similar to the CNF paper, the CNF–PANI composite paper can be bent easily, suggesting good flexibility. SEM images (Fig. 2d and 2e) show that small-compact PANI nanoparticles with diameters of 30–50 nm are deposited on the surfaces of each CNF, as a result of the widening of the pristine CNFs. The fracture edge of the paper reveals that PANI nanoparticles deposit on every layer of CNFs through the entire cross-section (Fig. 2f). In addition, there is no obvious change in the thickness after the deposition of PANI nanoparticles. As shown in Fig. 3a and 3b, the CNF and CNF–PANI composite papers have similar thicknesses of approximately 11 μm.

Fig. 1 Digital camera images of the CNF paper (left) and the CNF–PANI composite paper (right). The inset images indicate the flexibility of the samples.

Fig. 2 Top-view SEM images of the CNF paper (a, b) and the CNF–PANI composite paper (d, e). Side-view SEM images of the CNF paper (c) and the CNF–PANI composite paper (f). The scale is 1 μm in (a) and (d), 100 nm in (b) and (e), and 200 nm in (c) and (f), respectively.

Fig. 3 Side-view SEM images of the CNF paper (a) and the CNF–PANI composite paper (b). The scale is 10 μm in (a) and (b).

As shown in Fig. 4a, in the FTIR spectrum of the CNF paper, the broad peak at 3100–3700 cm⁻¹ is attributed to N–H bonds, the peak at 1632 cm⁻¹ is assigned to the C=C bond,^10,11 and the peak at 1380 cm⁻¹ is associated with C–O bonds.¹¹ By comparison, the FTIR spectrum of the CNF–PANI composite paper presents four new peaks attributed to PANI: the C=C stretching deformation of the quinoid ring at 1565 cm⁻¹, the C=C stretching deformation of the benzenoid ring at 1484 cm⁻¹, the C–N stretching at 1302 cm⁻¹, and the aromatic C–H bending at 1120 cm⁻¹.¹² As shown in Fig. 4b, the Raman spectrum of the CNF paper displays two prominent peaks at 1380 and 1618 cm⁻¹, corresponding to the well-documented D and G bands, respectively. In the Raman spectrum of the CNF–PANI composite paper, apart from the D and G bands, five representative peaks arising from PANI can be indexed to C–H bending of the quinoid ring at 1173 cm⁻¹, C–H bending of the benzenoid ring at 1233 cm⁻¹, C–N⁺ stretching at 1333 cm⁻¹, and C–C stretching of the benzene ring at 1502 and 1611 cm⁻¹, respectively.¹³ Moreover, the XRD investigation results further confirm the generation of PANI nanoparticles on the CNFs. As shown in Fig. 5, the XRD pattern of the CNF paper exhibits one broad peak centered at 22.6°, indicating the amorphous nature. For comparison, there are three new peaks at 15.2, 21.0 and 25.6° existing in the XRD pattern of the CNF–PANI composite paper, which are ascribed to the reflection of PANI.¹⁴

Fig. 4 (a) FTIR and (b) Raman spectra of the CNF paper and the CNF–PANI composite paper.

Fig. 5 XRD patterns of the CNF paper and the CNF–PANI composite paper.

It is known that high quality PANI nanofibers can be prepared using rapid-mixture polymerization.¹⁵ In this process, since the monomer aniline and the initiator ammonium peroxydisulfate (APS) are mixed completely before the reaction, these reactants are almost consumed during the initial polymerization. Thus, the secondary growth of the newly generated PANI can be suppressed effectively.¹⁵ Therefore, we believe that, when the CNF paper is immersed into the aqueous mixture of aniline and APS, some aniline molecules will be preferentially adsorbed on the surfaces of the CNFs via the hydrogen bond effect. These adsorbed aniline molecules will act as the ‘seeding dots’ and react with adjacent aniline molecules to form PANI nanoparticles in the early stage of polymerization. Since no reactants will be available for further reaction, the diameter of the PANI nanoparticles cannot be increased remarkably during further slow polymerization, as a result of the small size of the PANI nanoparticles in the final CNF–PANI composite paper. Consequently, the rapid-mixture polymerization exhibits great advantages over the traditional chemical polymerization and electro-polymerization: (1) the process is quite simple without the need of costly apparatus or the assistance of cooling or heating; and (2) secondary growth of PANI is limited, resulting in small-size PANI nanoparticles with high specific surface area. This is favorable to the electrochemical activity of the CNF–PANI composite paper. The results of electrical conductivity measurements show that the CNF–PANI composite paper has enhanced conductivity: for instance, 2.43 S cm⁻¹ for the composite paper is higher than that of the CNF paper (0.95 S cm⁻¹). Since the electrical conductivity of pure PANI is approximately 0.6 S cm⁻¹, the enhanced conductivity for the CNF–PANI composite paper is mainly attributed to the interaction between PANI backbone and the CNFs.^9,16

The electrochemical performances of the paper-like samples are analyzed using CV, EIS and galvanostatic charge/discharge. From the CV curves shown in Fig. 6a and b, the remarkable difference in electrochemical surface activity between the CNF paper and the CNF–PANI composite paper can be easily recognized. In detail, as shown in Fig. 6a, the CV curve of the CNF paper presents one pair of weak redox peaks due to the transition between quinone/hydroquinone groups, which is typical for carbon materials.^3,17 However, two couples of redox peaks (C₁/A₁, C₂/A₂) appear in the CV curve of the CNF–PANI composite paper (Fig. 6b), which are attributed to the transition between a semiconducting state and a conducting state and the transformation of emeraldine–pernigraniline, respectively.^3,18 Also, the cathodic peaks (C₁/C₂) shift positively and the anodic peaks (A₁/A₂) shift negatively with the increase of the potential sweep rate from 5 to 40 mV s⁻¹. It is noteworthy that, due to the synergistic effect between CNF and PANI, the composite paper shows a larger current density response than the CNF paper. In addition, the shape of the CV curves of the CNF–PANI composite paper shown in Fig. 6b indicates that the capacitance characteristic of the PANI phase is distinct from that of the electric double-layer capacitance, which would produce a CV curve close to the ideal rectangular shape.¹⁸

Fig. 6 Electrochemical properties of the CNF paper and the CNF–PANI composite paper. (a and b) CV curves recorded from 5 to 40 mV s⁻¹ in 1 M H₂SO₄ solution. (c) Nyquist plots recorded from 10 kHz to 10 mHz with an AC amplitude of 5 mV. Inset is the enlarged plots of the high-frequency region. (d) Galvanostatic charge/discharge curves at 2 A g⁻¹.

The EIS data are analyzed using Nyquist plots. As shown in Fig. 6c, the Nyquist plots of the CNF paper and the CNF–PANI composite papers both display a small semicircle at high frequency followed by a transition to linearity at low frequency. An ideal polarizable capacitance will give rise to a straight line along the imaginary axis.¹⁹ In our system, the real line for each electrode with a series resistance has a finite slope, representing the diffusive resistivity of the electrolyte within the pores of the electrode. The intersection of the plots at the X-axis represents the (ESR) of the electrode which determines the charge/discharge rate of the electrode.^20,21 As seen from the inset of Fig. 6c, the ESR of the CNF paper and the CNF–PANI composite paper is 62.7 and 21.6 Ω, respectively. It should be mentioned that, the ESR of our paper-like electrodes is much larger than those of the reported CNF-based supercapacitors.^5,22–24 Chen et al. have proved that the structure of electrochemistry cell affects the ESR value of the supercapacitor.²¹ Therefore, we speculate that the high ESR might be due to the structure of our electrochemistry cell, such as high electrical contact between the electrode and the current collector. Meanwhile, the high ESR is perhaps due to the poor conductivity of CNFs, in which the Raman spectrum exhibits broad peaks and high intensity of D-band, indicating the presence of amorphous carbon. The FTIR results also show a peak at 1380 cm⁻¹ associated with C–O bonds. The existence of amorphous carbon and C–O bonds would increase the intrinsic electrical resistance of the paper-like electrodes.

Galvanostatic charge/discharge properties are performed at a constant current density of 2 A g⁻¹ as shown in Fig. 6d. According to the capacitance equation evaluated from the slopes of the discharge curves, the specific capacitances of the CNF paper and the CNF–PANI composite paper are calculated to be 317 and 638 F g⁻¹ at 2 A g⁻¹, respectively. This indicates that the specific capacitance of the CNF–PANI composite paper is remarkably enhanced compared with the pristine CNF paper. We believe that the high specific capacitance is probably due to the combination effect of the layer-by-layer structure of CNFs, the high specific surface area of PANI nanoparticles and the interaction between PANI nanoparticles and CNFs.

Fig. 7a displays the dependence of specific capacitance on current density in the range of 1–5 A g⁻¹. It is clear to see that, for both the CNF paper and the CNF–PANI composite paper, the specific capacitance decreases with the increase of discharge current density. It indicates that the paper-like electrodes allow rapid ion diffusion. The life cycle of the CNF–PANI electrode was characterized by the variation of specific capacitance with cycle number at a constant current density of 2 A g⁻¹ and the corresponding result was shown in Fig. 7. It can be clearly seen that the specific capacitance still remains above 580 F g⁻¹ after 1000 cycles (above 90% of the original value), illustrating that he CNF–PANI paper-like electrode possesses a good stability and lifetime as the supercapacitor. Furthermore, during the cycling process, the coulombic efficiency (charge capacitance/discharge capacitance) remains approximately at 100% (Fig. 7b).

Fig. 7 (a) Specific capacitance of the CNF paper and the CNF–PANI composite paper as a function of discharge current. (b) Charge/discharge cycle at a current density of 2 A g⁻¹ (left axis: specific capacitance, right axis: coulombic efficiency).

Conclusions

In summary, a combination of electrospinning and rapid-mixture polymerization techniques has been successfully utilized to prepare freestanding CNF–PANI composite paper. The obtained composite paper maintains the pristine flexibility of the CNF paper, and displays remarkably improved electrochemical performances. Thus, we believe that the CNF–PANI composite paper is a potential low-cost candidate for use as flexible supercapacitor.

Acknowledgements

The authors acknowledge support from the Top Hundred Talents Program of Chinese Academy of Sciences. The authors thank Dr Sheng Liu for electrochemical measurements.

Notes and references

1. B. E. Conway, Modern aspects of electrochemistry, Plenum Press, New York, 1999.
2. (a) M. J. Pitkethly, Nano Today, 2004, 7, 20; (b) V. L. Pushparaj, M. M. Shaijumon, A. Kumar, S. Murugesan, L. Ci, R. Vajtai, R. J. Linhardt, O. Nalamasu and P. M. Ajayan, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 13574.
3. D. W. Wang, F. Li, J. P. Zhao, W. C. Ren, Z. G. Chen, J. Tan, Z. S. Wu, L. Gentle, G. Q. Lu and H. M. Cheng, ACS Nano, 2009, 3, 1745.
4. (a) F. Beguin, K. Szostak, G. Lota and E. Frackowiak, Adv. Mater., 2005, 17, 2380; (b) D. N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y. Kakudate, O. Tanaike, H. Hatori, M. Yumura and S. Iijima, Nat. Mater., 2006, 5, 987; (c) M. Kaempgen, J. Ma, G. Gruner, G. Wee and S. G. Mhaisalkar, Appl. Phys. Lett., 2007, 90, 264104; (d) L. J. Ci, S. M. Manikoth, X. S. Li, R. Vajtai and P. M. Ajayan, Adv. Mater., 2007, 19, 3300; (e) J. Chen, A. I. Minett, Y. Liu, C. Lynam, P. Sherrell, C. Wang and G. G. Wallace, Adv. Mater., 2008, 20, 566.
5. (a) C. Kim, Y. O. Choi, W. J. Lee and K. S. Yang, Electrochim. Acta, 2004, 50, 883; (b) L. W. Ji, Z. H. Lin, A. J. Medford and X. W. Zhang, Carbon, 2009, 47, 3346; (c) E. J. Ra, E. Raymundo-Pinero, Y. H. Lee and F. Beguin, Carbon, 2009, 47, 2984.
6. H. Q. Chen, M. B. Müller, K. J. Gilmore, G. G. Wallace and D. Li, Adv. Mater., 2008, 20, 3557.
7. (a) A. G. MacDiarmid, Angew. Chem., Int. Ed., 2001, 40, 2581; (b) K. S. Ryu, S. K. Jeong, J. Joo and K. M. Kim, J. Phys. Chem. B, 2007, 111, 731; (c) J. Jyongsik, B. Joonwon, C. Moonjung and Y. Seong-Ho, Carbon, 2005, 43, 2730; (d) H. R. Ghenaatian, M. F. Mousavi, S. H. Kazemi and M. Shamsipur, Synth. Met., 2009, 159, 1717.
8. C. Z. Meng, C. H. Liu and S. S. Fan, Electrochem. Commun., 2009, 11, 186.
9. H. L. Wang, Q. L. Hao, X. J. Yang, L. D. Lu and X. Wang, Electrochem. Commun., 2009, 11, 1158.
10. M. S. A. Rahaman, A. F. Ismail and A. Mustafa, Polym. Degrad. Stab., 2007, 92, 1421.
11. A. C. boccara, D. Fournier, A. kumar and G. C. Pandey, J. Appl. Polym. Sci., 1998, 63, 1785.
12. (a) D. Y. Wang and F. Caruso, Adv. Mater., 2001, 13, 350; (b) E. Marie, R. Rothe, M. Antonietti and K. Landfester, Macromolecules, 2003, 36, 3967; (c) R. Cruz-Silva, J. Romero-Garcia, J. L. Angulo-Sanchez, E. Flores-Loyola, M. H. Farias, F. F. Castillon and J. A. Diaz, Polymer, 2004, 45, 4711.
13. (a) G. Louarn, M. Lapkowski, S. Quillard, A. Pron, J. P. Buisson and S. Lefrant, J. Phys. Chem., 1996, 100, 6998; (b) X. B. Yan, Z. J. Han, Y. Yang and B. K. Tay, Sens. Actuators, B, 2007, 123, 107.
14. N. R. Chiou and A. J. Epstein, Adv. Mater., 2005, 17, 1679.
15. J. Huang and R. B. Kaner, Angew. Chem., Int. Ed., 2004, 43, 5817.
16. J. Jang, J. Bae, M. Choi and S. H. Yoon, Carbon, 2005, 43, 2730.
17. Y. R. Nian and H. S. Teng, J. Electrochem. Soc., 2002, 149, A1008.
18. Y. G. Wang, H. Q. Li and Y. Y. Xia, Adv. Mater., 2006, 18, 2619.
19. C. Kim, Y. O. Choi, W. J. Lee and K. S. Yang, Electrochim. Acta, 2004, 50, 883.
20. M. D. Stoller, S. Park, Y. W. Zhu, J. H. An and R. S. Ruoff, Nano Lett., 2008, 8, 3498.
21. Y. Wang, Z. Q. Shi, Y. Huang, Y. F. Ma, C. Y. Wang, M. M. Chen and Y. S. Chen, J. Phys. Chem. C, 2009, 113, 13103.
22. C. Kim, B. T. N. Ngoc, K. S. Yang, M. Kojima, Y. A. Kim, Y. J. Kim, M. Endo and S. C. Yang, Adv. Mater., 2007, 19, 2341.
23. Q. H. Guo, X. P. Zhou, X. Y. Li, S. L. Chen, A. Seema, A. Greiner and H. Q. Hou, J. Mater. Chem., 2009, 19, 2810.
24. S. K. Mondal, K. Barai and N. Munichandraiah, Electrochim. Acta, 2007, 52, 3258.

---