A comprehensive study of polyaniline-derived porous carbons via KOH activation

Abstract

High surface area porous carbons (up to 3768 m² g⁻¹) were prepared from as-synthesized polyaniline (PANI) by KOH activation. The structural characteristics of the resulting carbons have been comprehensively studied in terms of activation parameters such as activation temperature, KOH ratio and preheating temperature. It has been found that increasing activation temperature/KOH ratio improves pore development, but will result in pore widening effect and eventually decrease of Brunauer Emmett Teller (BET) surface area and micropore volume. It has also been determined that a proper preheating stage leads to the formation of network-like nanostructures through polymeric cross-linking. These nanostructures are beneficial to the pore development during subsequent activation at high temperatures and can lead to activated carbons with exceptionally high BET surface area and microporosity.

---

1. Introduction

Activated carbons (ACs) possess large specific surface area (SSA), well-developed porous structure and good thermal/chemical stability.^1–3 They are widely used in various applications such as gas adsorption,^4,5 water treatment^6,7 and energy storage.^8,9 When comparing two main activation methods (physical and chemical activation), chemical activation is usually preferred due to its lower activation temperature and higher yields, shorter activation time and higher SSA and pore volume.¹⁰ In particular, activation by KOH is of increasing interest recently as most of the ultrahigh-SSA (3100–3500 m² g⁻¹) AC materials^11–14 are produced by this method.

KOH activation is a well-known method to produce carbon materials made up of micropores and some small mesopores. However, considering a large number of experimental variables and precursor selections, the activation mechanisms have not been well understood albeit with intensive studies.^12,15–18 A generally accepted theory for KOH activation is KOH etching carbon framework by the global reaction:

2C + 6KOH → 2K + 3H₂ + 2K₂CO₃ (1)

which was proposed by Linares-Solano's group¹⁹ based on experimental results and theoretical calculations. Additionally, at temperatures higher than 700 °C, the decomposition of K₂CO₃ into K₂O and CO₂ brings up physical activation effect of CO₂.²⁰ The produced K in the global reaction also could intercalate into the carbon lattice and cause irreversible expansion. After washing off the potassium compound, these expansions cannot return to their non-porous structure.^17,21 In most occasions the ultrahigh SSA and the well-developed microporosity are resulted from the synergetic contributions of the chemical activation, the physical activation and the potassium intercalation.²²

Experimentally, the porous structure and chemical composition of the resulting ACs depend on activation conditions (activation temperature and KOH ratio), carbon precursor and preheating temperature. Most studies show that by controlling the activation temperature and KOH ratio, a maximum Brunauer Emmett Teller (BET) surface area value and optimized microporous structure can be obtained.^23–26 In 2011 Sevilla et al.²⁷ synthesized a high SSA porous carbon with activation temperature of 700 °C and KOH ratio = 4. The resulting carbon possessed a BET surface area of 3480 m² g⁻¹ and a micropore volume of 1.18 cm³ g⁻¹. However, further increasing temperature and KOH dosage leads to pore widening effect and consequently decrease of BET surface area and micropore volume.^22,23,28,29 Use of different precursors leads to ACs with various unique porous structures.^3,30 Recently, many synthesized polymers (such as polyacrylonitrile,^29,31 polypyrrole,¹¹ polyaniline³²) were employed as carbon precursors due to their controllable structures and composition. The preheating temperature has been considered to be another important parameter for the activation process and could be critical to pore development,^33,34 but the mechanisms behind this influence have not been discussed before.

In this work polyaniline (PANI) was selected as the carbon precursor. PANI is a low-cost, environmentally stable, conductive polymer³⁵ and has thus been studied extensively. The well-studied synthesis process of PANI^36,37 enables a better control of chemical composition and pore morphology, which provides a promising opportunity to understand the structures of subsequent ACs from microscopic level. A comprehensive study was carried out in this investigation to determine how the properties of PANI-derived porous carbon are affected by activation temperature, KOH ratio and preheating temperature. By tuning these factors, the BET surface area of the obtained ACs could achieve ∼3800 m² g⁻¹. Particularly, the influence of preheating temperature was thoroughly investigated. By examining samples after preheating stage, a new mechanism was proposed to elucidate the formation of porous structure. It is believed that this mechanism could be applied to the activation of other carbon precursors achieving improved SSA and micropore volume. This high SSA porous carbon could be applied as potential electrode materials for such energy devices as supercapacitors.

2. Experimental

The method of synthesizing PANI can be found from our previous work.³⁸ Briefly, aniline (0.23 M, 350 mL) acid solution was slowly dropped into an ammonium persulfate (0.54 M, 150 mL) acid solution under ice bath (0–1 °C) and then the mixture was further magnetically stirred for 12 h. The resulting mixture was vacuum-filtrated and washed by copious deionized water until the pH was ∼7, which was followed by oven drying. The final PANI yield was around 100%. For the KOH activation of PANI, KOH (pellet, Fisher Scientific) and PANI were ground separately before being physically mixed with a weight ratio of KOH : PANI = 1–3 : 1, and put into a round alumina crucible. The crucible was placed into a horizontal quartz tube and purged with Ar flow (100 mL min⁻¹) for 1 h before the thermal treatment. The temperature–time program of activation process is illustrated in Fig. 1. A hold at 150/200/250 °C for 1 h was made before it was ramped to the activation temperature at a rate of 3 °C min⁻¹. The activation process took 1 h and the samples were allowed to cool in the furnace under argon protection. The resulting samples were washed first with 1 M HCl and then deionized water until near-neutral pH (∼7) was reached before being dried in an oven at 80 °C for at least 24 h. The after-dry activated carbon yield was 5% to 20%. It was found that the higher the activation temperature/KOH ratio, the lower the yield. This is consistent with previous work.^39–41 These carbon samples were denoted as C-x-y-z, where x represents KOH to PANI ratio, y represents the preheating temperature (in °C) and z is the activation temperature (in °C). In addition, samples after preheating stage were taken out and washed with deionized water until near-neutral pH (∼7) was reached. The samples were then oven dried at 80 °C for at least 24 h. The dried samples were denoted as P-150/200/250, corresponding to different preheating temperature (150 °C/200 °C/250 °C). The KOH to PANI ratio of these samples was kept at 2.

Fig. 1 Schematic of temperature program for KOH activation process.

The morphologies of the samples were examined by scanning electron microscopy (SEM) using a JAMP-9500F system (JEOL, Tokyo, Japan) under 15 kV and by transmission electron microscopy (TEM) using a JEOL-2010 system (JEOL, Tokyo, Japan) with an operating voltage of 200 kV. Low-temperature nitrogen adsorption test was performed at 77 K with Autosorb 1MP (Quantachrome). All the samples (20–30 mg) were outgassed under high vacuum at 130 °C for 4 h before the adsorption tests. The BET surface area data were calculated using data points at a relative pressure of 0.05–0.2. The total pore volume was obtained based on adsorption data at P/P₀ = 0.995. Pore size distributions (PSDs) were determined using quenched solid density functional theory (QSDFT) model, assuming a slit pore shape. For PANI/P-150/P-200/P250, PSDs were determined using Barrett–Joyner–Halenda (BJH) desorption model due to the sample's absence of micropores. In order to characterize chemical properties, infrared spectrum was recorded with a Nicolet 8700 Fourier transformation infrared (FTIR) spectrometer (Thermal Scientific), the scanning range was between 500 and 4000 cm⁻¹, with a resolution of 1.93 cm⁻¹ and 64 scans per test. Differential scanning calorimetry (DSC) was applied using Q1000 (TA Instrument). The sample (∼5 mg) was sealed in an aluminum pan with pinhole, and then placed in the machine and purged with 50 mL min⁻¹ N₂ (99.999% pure). The sample was first cooled to 0 °C and then ramped to 400 °C at a rate of 5 °C min⁻¹. Thermal gravimetric analysis (TGA) was carried out using SDT Q600 (TA Instrument), where the samples (5–10 mg) were placed in a platinum crucible pans under nitrogen atmosphere (99.999% pure, flow rate: 100 mL min⁻¹). The temperature was then ramped from room temperature to 600 °C at a rate of 10 °C min⁻¹.

3. Results and discussion

3.1 Study of pore structure

Fig. 2 shows microscopic morphologies of as-synthesized PANI and PANI-derived activated carbon (C-2-200-800). The SEM image of PANI (Fig. 2a) exhibits a sponge-like texture with fibrous structures embedded inside. The TEM image shown in Fig. 2b indicates that PANI nanofibers were formed and surrounded by polymer aggregates. This typical structure is caused by intrinsic morphology of PANI (fibrous structure) and its heterogeneous nucleation (aggregates).^37,42 After activation, the carbon sample (Fig. 2c) presents a smooth surface with some large cavities under low magnification. However, a high-resolution TEM image (Fig. 2d) displays a representative microporous structure with worm-like graphitic layers.

Fig. 2 Morphologies of PANI and PANI-derived carbons (C-2-200-800): (a) SEM image of PANI; (b) TEM image of PANI; (c) SEM image of C-2-200-800; (d) high-resolution TEM image of C-2-200-800.

The porous structure was further characterized by N₂ adsorption/desorption analysis and the results obtained are shown in Fig. 3, with statistic pore characteristics listed in Table 1. From the isotherms of different carbon samples (Fig. 3a, c and e), it could be observed that all these curves present significant amount of adsorption at relatively low pressure (P/P₀ < 0.1), followed by a relatively flat plateau characteristics (except C-2-200-900 and C-3-200-800). This means that most activated carbons are microporous and lacks of mesopores (except C-2-200-900 and C-3-200-800). C-2-200-900 and C-3-200-800 were obtained at high activation temperature/KOH ratio, and their isotherms display an obvious increase of adsorption at high pressure and hysteresis, implying the existence of considerable mesopores.

Fig. 3 Isotherms (a, c and e) and pore size distributions (b, d and f) of PANI-derived carbons under various activation conditions. (a) and (b) Different activation temperature; (c) and (d) different KOH ratio; (e) and (f) different preheating temperature (C-2-800 represents the sample without preheating stage).

Table 1 Porous structure data of PANI-derived carbons under various activation conditions

Sample	BET surface area^a (m^2 g^−1)	Total pore volume^b (cm^3 g^−1)	Micropore volume^c (cm^3 g^−1)	<1 nm pore volume^c (cm^3 g^−1)	<0.7 nm pore volume^c (cm^3 g^−1)
a BET surface area obtained using BET method at P/P0 = 0.05–0.2.b Total pore volume calculated at P/P0 = 0.995.c Micropore, <1 nm pore volume and <0.7 nm pore volume obtained from QSDFT slit pore model.d C-2-800 represents the activated sample without preheating stage.
C-2-200-600	1711	0.92	0.74	0.59	0.47
C-2-200-700	2889	1.48	1.23	0.67	0.36
C-2-200-800	3768	2.03	1.37	0.49	0.16
C-2-200-900	2914	2.19	0.83	0.26	0.06
C-1-200-800	2575	1.48	1.05	0.45	0.19
C-2-200-800	3768	2.03	1.37	0.49	0.16
C-3-200-800	2901	2.85	0.75	0.20	0.07
C-2-800^d	2800	1.47	1.06	0.39	0.12
C-2-150-800	3059	1.60	1.16	0.40	0.22
C-2-200-800	3768	2.03	1.37	0.49	0.16
C-2-250-800	3115	1.59	1.22	0.46	0.15

---

A more direct observation could be made from PSDs in Fig. 3b, d and f. Fig. 3b and d show influences of activation temperature and KOH ratio on the pore development of the prepared carbons, respectively. It could be seen that at the activation temperature of 600 °C, the sample presents mainly micropores with pore sizes smaller than 0.7 nm. As the temperature increases, the amount of narrow micropores (<0.7 nm) decreases while large micropores (∼1 nm) and small mesopores are formed. At 800 °C, the major peaks are at ∼1 nm and ∼2 nm. At 900 °C, the pore size is even larger. This is evidently caused by micropore coalescence to form larger pores (such as mesopores) due to severe KOH–C reaction (over-activation). At the same time, new micropores fail to emerge at a satisfactory rate, resulting in a gradual decrease of narrow micropore (<1 nm and <0.7 nm) volume with the increase of activation temperature, as could be observed in Table 1. As a consequence, the BET surface area increases from C-2-200-600 (1711 m² g⁻¹) to C-2-200-800 (3768 m² g⁻¹), and then decreases for C-2-200-900 (2914 m² g⁻¹). Similar pore widening effect occurs when the KOH ratio increases. As could be seen in Fig. 3d, from C-1-200-800 to C-3-200-800, the largest pore size increases from ∼1.5 nm to ∼3 nm and the number of pores with sizes <1 nm decreases substantially. In Table 1, the BET values first increase from C-1-200-800 (2575 m² g⁻¹) to C-2-200-800 (3768 m² g⁻¹) and then decrease for C-3-200-800 (2901 m² g⁻¹). This is similar with the effect of temperature. These observations are also consistent with previous study.²² Particularly, at activation temperature of 800 °C and KOH ratio of 2, the activated carbon achieves an optimum micropore volume of 1.37 cm³ g⁻¹ and the highest BET surface area of 3768 m² g⁻¹. In Fig. 3f, however, preheating temperature has not caused pore widening as samples with different preheating temperature have similar pore size distributions, although the pore volume and SSA are different. This indicates that the preheating temperature influenced the pore structures of PANI-derived carbons through mechanisms other than KOH–C reaction.

3.2 Study of the preheating stage

In order to understand how the preheating temperature influences porous structure of the PANI-derived carbons, PANI/KOH mixtures were taken out after the preheating stage and KOH was washed off carefully to study the remaining materials (P150, P200 and P250). As could be observed in TEM images from Fig. 4, different preheating temperatures have resulted in different morphologies of the resulting materials. P-150 demonstrates a nanofiber/agglomerate structure similar to the original PANI (Fig. 2b), which means that no obvious morphological change occurred during preheating at 150 °C. However, P-200 shows a network-like nanostructure. Considering the fact that the width of the nanofibers in P-150 is similar to that of the network structures in P-200, it is suggested that the network of PANI was formed by nanofiber fragmentation, possibly through melting and cross-linking with each other. The size of the cavities in the PANI network is several tens of nanometers. For P-250, the TEM image shows that no network-like structure but some disordered agglomerates is present.

Fig. 4 Morphologies of PANI after preheating stage of different temperatures.

Fig. 5 and Table 2 further display the N₂ adsorption/desorption analysis of the PANIs after preheating at different temperatures. As can be seen in Fig. 5, isotherms of all samples are type II (according to IUPAC), revealing non-porous nature of the materials. PSDs (Fig. 5b) at small pore range (<10 nm) are rather random. However, it could be found in Table 2 that at pore sizes ranging between 10 nm and 100 nm, the pore volume increases slightly from 0.20 cm³ g⁻¹ (PANI) to 0.22 cm³ g⁻¹ (P-200). This range of pore sizes (10–100 nm) is very comparable to the range of cavity sizes observed in Fig. 4, and an increase of the BET surface area with increasing preheating temperature, from 27 m² g⁻¹(original PANI) to 34 m² g⁻¹ (200 °C) as listed in Table 2, is observed. In contrast, the BET surface area was decreased to 18 m² g⁻¹ at the preheating temperature of 250 °C, which should be related to the non-network-type structure of the PANI-derived carbons, as evidenced in Fig. 4. These observations also imply that the formation of network nanostructure increased the surface area, and the network morphology may offer a large interfacial area for the polymer to react with KOH, which is essentially important to achieve a homogeneous reaction and well-developed porosity. This also explains the reason that the trend of the PANI-derived carbons' BET values in Table 1 correlates well with the trend of the PANIs' listed in Table 2. Furthermore, the low BET surface areas and the absence of micropores after pre-heating also reveal that KOH–C reaction does not occur at temperatures below 250 °C. This further confirms that the different pore structure of PANI-derived carbons under different preheating temperature was not caused by KOH–C reaction. Instead, it is suggested that the morphologies of PANIs resulting from a non-activation process during preheating stage may play an important role.

Fig. 5 (a) Isotherms and (b) pore size distributions (based on BJH desorption model) of samples after preheating stage of different temperatures.

Table 2 Porous structure data of PANI after different preheating condition

Sample	BET surface area^a (m^2 g^−1)	Total pore volume^b (cm^3 g^−1)	10–100 nm pore volume^c
a BET surface area obtained using BET method at P/P0 = 0.05–0.2.b Total pore volume calculated at P/P0 = 0.995.c 10–100 nm pore volume obtained from BJH desorption model.
PANI	27	2.54	0.20
P-150	30	2.48	0.21
P-200	34	3.11	0.22
P-250	18	1.28	0.08

---

In order to reveal the causes of the morphology change in the preheating stage, FTIR was conducted after different preheating processes and the results obtained are shown in Fig. 6. Compared with the original PANI sample, the samples after preheating show intensity changes at 1585 cm⁻¹, 1504 cm⁻¹ and 1160 cm⁻¹ (peaks with arrows). Specifically, the decreased peak intensity at 1585 cm⁻¹ represents the C=C stretching vibration of quinoid rings, the increased peak intensity at 1504 cm⁻¹ represents the C=C stretching of benzenoid rings and the intensity of the 1160 cm⁻¹ peak representing N=Q=N (Q means quinoid rings) also decreases.^43,44 These changes indicate that at preheating stage a crosslinking reaction occurs among PANI nanofibers. As proposed in Fig. 6b, the crosslinking is established through a link of the imine nitrogen with its neighboring quinoid rings,⁴⁵ which is consistent with the decrease of N=Q=N and quinoid rings peaks in Fig. 6a. It is also proposed that the crosslinking reaction was triggered by thermal treatment^43,46,47 (in this case the preheating stage). Furthermore, P-200 presents the strongest intensity change, implying the most intense crosslinking activity. It is suggested that the PANI network morphology of P-200 in Fig. 4 is resulted from the crosslinking. This is consistent with the study of Huang et al.,⁴³ who melted and merged PANI nanofibres via crosslinking induced by high-power camera flash. Fig. 6a shows that P-150 exhibits changes similar to those of P-200, but the intensities of the 3 representative peaks (indicated with arrows) are weaker than P-200, implying the degree of cross-linking of PANI at 150 °C is less than at 200 °C. Indeed, previous study⁴⁸ showed that the crosslinking time required at 150 °C is 4 times more than that at 200 °C for the same conversion rate. Therefore, the fact that the morphology of P-150 (from Fig. 4) is not the same as that of P-200 is mainly caused by incomplete crosslinking at 150 °C. Thus it can be concluded that the network morphology of PANI at preheating stage was caused by more complete crosslinking reaction, and this in turn resulted in higher SSA and well-developed pore structure of PANI-derived carbon. Additionally, P-250 displays a decrease of all peak intensity compared with P-200 because of its non-cross-linked structure. Since the non-cross-linked structure led to a smaller SSA and undesirable structure of PANI-derived carbon, it is necessary to further study the behavior of PANIs at this preheating temperature (250 °C).

Fig. 6 (a) FTIR spectra of PANI and samples after preheating stage of different temperatures; (b) schematic for thermal crosslinking of PANI (remake from ref. 36).

DSC and TGA were conducted to further characterize the effect of preheating at 250 °C and the results obtained are shown in Fig. 7a and b, respectively. In Fig. 7a, the endothermic peak at ∼100 °C was attributed to the evaporation of water. The following two broad exothermic peaks between ∼155 °C and ∼220 °C were assigned to crosslinking reaction,^45,49 which confirms the crosslinking effect found from the FTIR analysis shown in Fig. 6. Another strong endothermic peak appears at 240–350 °C. In the TGA curve of Fig. 7b it could be observed that a slow weight loss occurred at a low temperature (<250 °C). This is followed by a major weight loss at ∼300 °C. In the derivative weight curve in Fig. 7b, three peaks can be identified. The first two small peaks are located at 160–220 °C, in accord with the crosslinking peaks in Fig. 7a. The next broad but strong peak is located at 250–320 °C. Based on the obvious weight loss observed from the weight profile, this peak should be resulted from evolution of volatile species, which proves to be the degradation of PANI,⁵⁰ and explains the appearance of the peak at 240–350 °C in Fig. 7a. It is suggested that the degradation may have destroyed the favorable network structure if formed initially, and consequently produced PANI-derived carbons with lower SSA and decreased porosity. Since the degradation involves in decomposition of functional groups,^51,52 it should be the degradation of PANI at 250 °C that causes the decrease of peak intensities for P-250 in FTIR spectrum in Fig. 6a.

Fig. 7 (a) DSC curve of original PANI and (b) TGA and its derivative curves of the original PANI.

Fig. 4 to 7 illustrates how the preheating stage influences PANI and PANI-derived carbons. Briefly, at low preheating temperatures (150–200 °C), crosslinking reaction occurs and the degree of crosslinking increases as the preheating temperature. A complete crosslinked PANI will result in a network-like nanostructure. This structure increases the surface area of PANI and leads to a more homogeneous activation and a higher SSA/better developed porosity of the PANI-derived carbon. At high preheating temperature (>250 °C), degradation of PANI is triggered, which destroys the favorable structure of PANI, and thus decreases the surface area of PANI. In this case, the pore development of the PANI-derived carbon is restricted. Therefore, a well-tuned preheating stage is beneficial for obtaining high surface area and highly porous ACs.

4. Conclusion

In conclusion, a comprehensive study on KOH activation of PANI has been carried out.

(1) The PANI-derived carbons have BET surface area of 1711–3768 m² g⁻¹ and micropore volume of 0.74–1.37 cm³ g⁻¹. Specifically, C-2-200-800 has the highest BET surface area of 3768 m² g⁻¹ and the largest micropore volume of 1.37 cm³ g⁻¹.

(2) It has been found that the BET surface area and micropore volume increase as the activation temperature/KOH ratio increases, but start to decrease at 900 °C or at KOH : PANI ratios higher than 3 : 1. This is caused by widening of narrow micropores because of intensive reactions during activation.

(3) Varying preheating temperature does not widen the pore size. Characterizations of the after-preheating samples show that thermal crosslinking is maximized at ∼200 °C, which leads to the formation of network-like nanostructures. These nanostructures are beneficial to the pore development during subsequent activation at high temperatures and can lead to ACs with high BET surface area and microporosity.

Acknowledgements

The authors would like to acknowledge Carbon Management Canada for the financial support.

References

1. L. L. Zhang and X. S. Zhao, Chem. Soc. Rev., 2009, 38, 2520–2531.
2. J. Lee, J. Kim and T. Hyeon, Adv Mater, 2006, 18, 2073–2094.
3. S. P. Wang, J. N. Zhang, P. Shang, Y. Y. Li, Z. M. Chen and Q. Xu, Chem. Commun., 2014, 50, 12091–12094.
4. R. E. Morris and P. S. Wheatley, Angew Chem Int Edit, 2008, 47, 4966–4981.
5. Y. Q. Li, T. Ben, B. Y. Zhang, Y. Fu and S. L. Qiu, Sci Rep-Uk, 2013, 3,  DOI:10.1038/srep02420.
6. J. Rivera-Utrilla, M. Sanchez-Polo, V. Gomez-Serrano, P. M. Alvarez, M. C. M. Alvim-Ferraz and J. M. Dias, J. Hazard. Mater., 2011, 187, 1–23.
7. T. Otowa, Y. Nojima and T. Miyazaki, Carbon, 1997, 35, 1315–1319.
8. M. J. Li, C. M. Liu, H. B. Cao, H. Zhao, Y. Zhang and Z. J. Fan, J Mater Chem A, 2014, 2, 14844–14851.
9. Y. W. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach and R. S. Ruoff, Science, 2011, 332, 1537–1541.
10. J. C. Wang and S. Kaskel, J. Mater. Chem., 2012, 22, 23710–23725.
11. M. Sevilla, P. Valle-Vigon and A. B. Fuertes, Adv. Funct. Mater., 2011, 21, 2781–2787.
12. H. L. Wang, Q. M. Gao and J. Hu, J. Am. Chem. Soc., 2009, 131, 7016–7022.
13. L. Zhang, F. Zhang, X. Yang, G. K. Long, Y. P. Wu, T. F. Zhang, K. Leng, Y. Huang, Y. F. Ma, A. Yu and Y. S. Chen, Sci Rep, 2013, 3,  DOI:10.1038/srep01408.
14. A. Almasoudi and R. Mokaya, J. Mater. Chem., 2012, 22, 146–152.
15. T. Otowa, R. Tanibata and M. Itoh, Gas Sep. Purif., 1993, 7, 241–245.
16. D. Lozano-Castello, J. M. Calo, D. Cazorla-Amoros and A. Linares-Solano, Carbon, 2007, 45, 2529–2536.
17. E. Raymundo-Piñero, P. Azaïs, T. Cacciaguerra, D. Cazorla-Amorós, A. Linares-Solano and F. Béguin, Carbon, 2005, 43, 786–795.
18. W. M. Qiao, S. H. Yoon and I. Mochida, Energy Fuels, 2006, 20, 1680–1684.
19. M. A. Lillo-Rodenas, D. Cazorla-Amoros and A. Linares-Solano, Carbon, 2003, 41, 267–275.
20. A. Swiatkowski, Studies in Surface Science and Catalysis, Elsevier, 1999.
21. R. S. Xue and Z. M. Shen, Carbon, 2003, 41, 1862–1864.
22. J. C. Wang, I. Senkovska, S. Kaskel and Q. Liu, Carbon, 2014, 75, 372–380.
23. J. C. Wang, A. Heerwig, M. R. Lohe, M. Oschatz, L. Borchardt and S. Kaskel, J. Mater. Chem., 2012, 22, 13911–13913.
24. Z. J. Zhang, P. Cui, X. Y. Chen and J. W. Liu, J. Solid State Electrochem., 2013, 17, 1749–1758.
25. H. Cong, M. Zhang, Y. Chen, K. Chen, Y. Hao, Y. Zhao and L. Feng, Carbon, 2015, 92, 297–304.
26. J. Romanos, M. Beckner, T. Rash, L. Firlej, B. Kuchta, P. Yu, G. Suppes, C. Wexler and P. Pfeifer, Nanotechnology, 2012, 23,  DOI:10.1088/0957-4484/23/1/015401.
27. M. Sevilla, R. Mokaya and A. B. Fuertes, Energy Environ. Sci., 2011, 4, 2930–2936.
28. M. B. Wu, Q. F. Zha, J. S. Qiu, Y. S. Guo, H. Y. Shang and A. J. Yuan, Carbon, 2004, 42, 205–210.
29. Y. J. Lee, J. H. Kim, J. Kim, D. B. Lee, J. C. Lee, Y. J. Chung and Y. S. Lim, Designing, Processing and Properties of Advanced Engineering Materials, Pts 1 and 2, 2004, vol. 449–4, pp. 217–220.
30. Y. Y. Lv, F. Zhang, Y. Q. Dou, Y. P. Zhai, J. X. Wang, H. J. Liu, Y. Y. Xia, B. Tu and D. Y. Zhao, J. Mater. Chem., 2012, 22, 93–99.
31. W. Z. Shen, S. C. Zhang, Y. He, J. F. Li and W. B. Fan, J. Mater. Chem., 2011, 21, 14036–14040.
32. Z. S. Zhang, J. Zhou, W. Xing, Q. Z. Xue, Z. F. Yan, S. P. Zhuo and S. Z. Qiao, Phys. Chem. Chem. Phys., 2013, 15, 2523–2529.
33. X. P. Zhu, Y. Fu, G. S. Hu, Y. Shen, W. Dai and X. Hu, Water, Air, Soil Pollut., 2013, 224,  DOI:10.1007/s11270-012-1387-y.
34. Z. H. Hu and M. P. Srinivasan, Micropor Mesopor Mater., 1999, 27, 11–18.
35. S. H. Weng, Z. H. Lin, Y. Zhang, L. X. Chen and J. Z. Zhou, React. Funct. Polym., 2009, 69, 130–136.
36. J. X. Huang and R. B. Kaner, Chem. Commun., 2006, 367–376.
37. D. Li, J. X. Huang and R. B. Kaner, Acc. Chem. Res., 2009, 42, 135–145.
38. D. H. Lin, X. T. Zhang, X. W. Cui and W. X. Chen, RSC Adv., 2014, 4, 27414–27421.
39. N. Yoshizawa, K. Maruyama, Y. Yamada, E. Ishikawa, M. Kobayashi, Y. Toda and M. Shiraishi, Fuel, 2002, 81, 1717–1722.
40. F. C. Wu, R. L. Tseng and R. S. Juang, Sep. Purif. Technol., 2005, 47, 10–19.
41. T. Tay, S. Ucar and S. Karagoz, J. Hazard. Mater., 2009, 165, 481–485.
42. D. Li and R. B. Kaner, J. Mater. Chem., 2007, 17, 2279–2282.
43. J. X. Huang and R. B. Kaner, Nat. Mater., 2004, 3, 783–786.
44. L. L. Ding, X. W. Wang and R. V. Gregory, Synthetic Met, 1999, 104, 73–78.
45. R. Mathew, B. R. Mattes and M. P. Espe, Synthetic Met, 2002, 131, 141–147.
46. I. Sedenkova, M. Trchova and J. Stejskal, Polym. Degrad. Stab., 2008, 93, 2147–2157.
47. C. H. Chen, J. Appl. Polym. Sci., 2003, 89, 2142–2148.
48. S. S. Pandey, M. Gerard, A. L. Sharma and B. D. Malhotra, J. Appl. Polym. Sci., 2000, 75, 149–155.
49. J. A. Conklin, S. C. Huang, S. M. Huang, T. L. Wen and R. B. Kaner, Macromolecules, 1995, 28, 6522–6527.
50. Y. Kieffel, J. P. Travers, A. Ermolieff and D. Rouchon, J. Appl. Polym. Sci., 2002, 86, 395–404.
51. S. F. Patil, A. G. Bedekar and C. Agashe, Mater. Lett., 1992, 14, 307–312.
52. B. Sreedhar, M. Sairam, D. K. Chattopadhyay, P. P. Mitra and D. V. M. Rao, J. Appl. Polym. Sci., 2006, 101, 499–508.

---