One-step molten salt carbonization (MSC) of firwood biomass for capacitive carbon

Abstract

Capacitive carbons are prepared by a molten salt carbonization (MSC) process from Chinese firwood biomass, and the effect of the biomass size (lengths ranging from 0.09 to 2 cm) on the carbon yield and the electrochemical capacitive performance of the carbonized samples is investigated. The mechanism of carbon treatment and structure–activity correlations are discussed. Compared with carbon prepared without the assistance of MS, MSC-derived carbon prepared from the same precursor size (i.e. 0.09 cm) shows a higher specific capacitance (189 vs. 165 F g⁻¹ at 0.2 A g⁻¹) and enhanced high-rate capability (85% vs. 70% capacitance retention upon increasing the charge–discharge current density from 0.2 to 2.0 A g⁻¹). Upon decreasing the precursor size from 2.0 to 0.09 cm, the specific capacitance increases from 142 to 189 F g⁻¹, with the high-rate capacitance retention increasing from 70% to 85%. These improvements highlight the merits of the MSC method and engineering the precursor sizes for the preparation and modification of enhanced capacitive carbon, which is promising for practical applications. It is also found that the production yield of the MSC method decreases upon decreasing the precursor size. Therefore, the precursor size should be prudently tailored to ensure a balance between the capacitive properties and production yield.

---

1. Introduction

The environmentally sound, affordable and effective utilization of biomass for energy generation and the production of functional materials is of great importance in tackling the challenges of declining fossil fuels and environmental issues.¹ Biomass is produced from photosynthesis by plants and thereby a huge amount of carbon dioxide is converted to hydrocarbons on the earth. In addition to being directly used as foods and chemical feedstocks, a large portion of this biomass is discarded because of the lack of effective and cost-affordable conversion/utilization technologies. Biomass could be a feedstock for producing liquid fuels, syngas, and chemicals. The conversion of biomass is highly dependent on its chemical properties. The main chemical components of forest biomass are cellulose, hemicellulose and lignin, which are difficult to convert into syngas and liquid fuels. As a consequence, straightforward methods for converting forest biomass to value-added materials are in high demand for carbon capture and waste management.

Although there is ongoing difficulty in efficiently converting forest woods to syngas and liquid fuels, the production of functional carbon materials from them is an alternative approach for the utilization of forest woods. Carbon-based materials have widely been applied in energy conversion and storage devices such as supercapacitors, lithium-ion batteries and fuel cells. In recent years, supercapacitors have gained a substantial amount of interest for their potential applications, ranging from mobile devices to electric vehicles,² mainly due to their high power density and superior cycling performance. In commercial markets, the carbon-based supercapacitor possesses a predominant share. Biomass materials are abundant, cheap and environmentally friendly carbon sources.³ Therefore, the preparation of capacitive carbon from biomass such as paulownia flowers,⁴ bamboo,⁵ rice husks,⁶ dead leaves,⁷ oil palm kernels,⁸ waste tea-leaves,⁹ potato starch,¹⁰ shrimp shells¹¹ and olive pits¹² has become a well-documented method.

Chinese firwood is an important wood as well as one of the fastest growing tree species in south China.¹³ Therefore, the effective utilization and reclamation of discarded firwood is in demand. Firwood wastes are rich in lignin and cellulose, which can be utilized as a carbon source to synthesize advanced carbon materials. Wu et al. reported a steam carbonization at 900 °C (ref. 14) and an alkaline carbonization^15,16 of firwood to harvest capacitive carbon powders. However, the reported capacitance is relatively low. Generally, the biomass precursors were thoroughly ground to fine powders and sieved prior to the following carbonizations. However, from the engineering and commercial points of view, the shape and size of the biomass feedstock are important because they affect the pretreatment and feeding procedures. Thus it is worthwhile to investigate the effect of biomass size on the carbonization process.^17–20 To date, the effect of the size of firwood on the electrochemical properties of the resultant carbon electrodes remains poorly understood.

Molten salt carbonization (MSC) is a thermo-chemical conversion of biomass in which biomass is converted to carbon in a molten salt with high thermal stability, enhanced heat transfer characteristics and good dissolution ability. Importantly, molten salts show a catalytic effect in cracking the large molecules such as lignin and cellulose in biomass.²¹ It is therefore anticipated that MSC could be an effective and affordable method for harvesting capacitive carbon from biomass. Herein, we investigate the effect of firwood size on the carbon yields and capacitive performance of MSC-derived carbon. The mechanism of carbon treatment and structure–activity correlations are discussed and rationalized.

2. Material and methods

2.1 Synthesis of porous carbons from firwood

The Chinese firwood was obtained from a wood processing factory (Wuhan, Hubei province, China). The wood blocks were cut to two different lengths, 1.0 and 2.0 cm. For the preparation of biomass with a size of 0.09 cm, the wood blocks were ground and sieved to a powder with the desired size. After that, the firwood blocks (L = 1.0 or 2.0 cm) and powders (L = 0.09 cm) were washed with distilled water and dried at 105 °C for 12 h in a drying box. The molten salt carbonization was carried out in a vertical resistance furnace. Carbonate salts (250 g Na₂CO₃ and 250 g K₂CO₃, eutectic point: 709 °C), contained in an alumina crucible, were placed in a stainless steel reactor. The salts were dried at 250 °C for 2 days to remove moisture. After reaching the target temperature (850 °C), the salts melted and about 10 g of wood raw materials wrapped in a nickel foam basket were immersed into the molten salts and kept in the molten salts for 1 h. During the whole experiment of heating and carbonization, argon gas (>99.999%, Wuhan Iron and Steel (Group) Crop.) was continuously purged into the reactor to guarantee an anaerobic environment. Afterward, the basket was lifted out of the molten salts and cooled down to room temperature in an argon atmosphere. The products were washed with 0.1 M HCl and then distilled water several times to remove the salts within the products. After being dried at 105 °C in air for 12 h, the final obtained carbons were collected. The samples WDPC-1, 2, and 3 were prepared by the molten salt carbonization of powder (L = 0.09 mm), small wood blocks (L = 1.0 cm) and large wood blocks (L = 2.0 cm), respectively. The sample WDPC-0 was prepared by the carbonization of powder (L = 0.09 mm) in argon at 850 °C for 1 h without molten salts.

2.2 Characterization of the as-prepared carbons

The chemical compositions of the obtained carbons were characterized by CHN elemental analysis on an Elementar Analysensysteme Gmbh VarioEL V4.03 analyzer. The surface chemistry was investigated with X-ray photoelectron spectroscopy (XPS) using an ESCALAB 250Xi system (Thermo Fisher Scientific Inc.) equipped with an Al Kα (1254 eV) monochromatic X-ray source and a 180° hemispherical energy analyzer with a mean radius of 150 mm and an energy range of 0 to 5000 eV. The thermal decomposition process was studied by thermogravimetric (TG) and derivative thermogravimetric (DTG) analysis (NETZSCH STA 409 C/CD, Germany). The structural information of the carbon samples was analyzed with an X-ray diffractometer (XRD) (Shimadzu X-ray 6000 with Cu Kα radiation at 40 kV and 250 mA, λ = 0.154 nm) and Raman spectroscopy using an excitation wavelength of 514.5 nm from a diode pumped solid-state laser (Laser Confocal Raman Microspectroscopy, Renishaw, RM-1000, UK) at room temperature. The morphologies were characterized on a field emission scanning electron microscope (FE-SEM, Zeiss SIGMA). The textural properties of the obtained carbons were determined by nitrogen adsorption/desorption²² on a Micromeritics ASAP 2460 automatic analyzer. The Brunauer–Emmett–Teller (BET) surface areas,²³ pore volume and pore size distribution (by the Barrett–Joyner–Halenda (BJH) method)²⁴ were calculated.

2.3 Electrochemical measurements

The cyclic voltammetry (CV), galvanostatic charge–discharge and Electrochemical Impedance Spectroscopy (EIS, with an AC voltage of 5 mV amplitude at open circuit potentials) measurements were performed on a PARSTAT 2273 (Princeton).

All the electrochemical measurements were conducted using a three-electrode cell containing 1 M H₂SO₄ solution. The fabrication method for the working electrode was the same as that reported in the literature.²⁵ The mass ratio of as-prepared carbon, conducting agent (acetylene black) and polytetrafluoroethylene (PTFE, as a binder) was 8 : 1 : 1. To make a slurry, 3–4 drops of ethanol were added to the mixture and it was ground thoroughly. Then the slurry was pressed onto a titanium mesh current collector (the active material loading was ∼8 mg cm⁻²). A graphite rod served as a counter electrode, and a saturated calomel electrode (SCE, CHI150) was used as a reference electrode. The gravimetric specific capacitance was calculated from galvanostatic charge–discharge tests by eqn (1).

where C_cp is the specific capacitance (F g⁻¹), I is current (A), Δt is the discharge time (s), ΔV is the potential change during the discharge process (1 V in this study) and m represents the mass (g) of the active material in the working electrode.

3. Results and discussion

3.1 Composition and microstructure of the obtained carbons

All the firwood samples turned black after MSC at 850 °C for 1 h (Fig. 1), and the as-prepared carbons kept the original shapes of the precursors, indicating that the conversion process mainly involves solid to solid and solid to gas transitions.

Fig. 1 Digital photographs of the firwood precursors (upper) and as-prepared carbons (below).

The composition and surface functional groups of the as-prepared carbons were characterized by X-ray photoelectron spectroscopy (XPS). All the binding energies (BEs) in the XPS analysis were corrected with respect to the C 1s peak (set at 284.6 eV) from surface adventitious carbon. All the survey spectra of the WDPC samples (Fig. 2a) show two peaks at 284.8 and 531.9 eV, corresponding to C 1s and O 1s, respectively.

Fig. 2 The full scan spectra (a) and C 1s spectra (b) of the WDPCs and high resolution O 1s spectra of the obtained carbons: (c) WDPC-0; (d) WDPC-1; (e) WDPC-2; (f) WDPC-3.

As listed in Table S1 (see the ESI†), with the decrease in raw material size, the oxygen content decreases. To understand the surface functional groups of the samples, the high-resolution O 1s spectra were analyzed. As shown in Fig. 2c, the asymmetric O 1s signal can be de-convoluted into three distinguishable peaks at 531.2, 532.4 and 533.4 eV, which correspond to C=O, C–O–H, and H₂O, respectively.^26,27 According to previous reports,^28,29 the C=O group could contribute to the electrochemical capacitance by introducing faradaic reactions. As summarized in Table S1,† the C=O group content increases upon decreasing the firwood size. Moreover, the relative C=O content in the carbons obtained with molten salts is higher than that obtained in Ar. Therefore, the MSC process is capable of producing more capacitance-related functional groups on the carbons, and MSC with the smaller-sized feedstock corresponds to a higher surface area due to the interaction between the biomass and molten salts.

The TG and DTG curves of the firwood raw material were obtained in an argon atmosphere, and are shown in Fig. 3. Minor weight loss appears at temperatures below 200 °C, ascribed to moisture elimination. The main mass loss (about 60%) is observed at temperatures ranging from 200 to 450 °C, due to the decomposition of the raw materials and loss of volatile matter. Minor mass loss at 450 to 700 °C is assigned to surface restacking. The DTG curve exhibits a sharp peak in the 200–450 °C range, indicating that the decomposition of the firwood biomass mainly occurs in this temperature range. This result agrees with that for other lignocellulosic biomass.³⁰ The carbon yields from different sizes of firwood biomass are shown in Table S2.† The carbon yield (ca. 11.6%) from MSC is lower than that obtained in argon (ca. 35.9%), which could be due to the high dissolving ability of the alkaline carbonate molten salts. The inorganic components such as Si and Mg can dissolve in the molten salts,³¹ thus resulting in higher mass loss and lower carbon yield and probably more pores.²¹

Fig. 3 TG and DTG curves of the firwood raw material (L = 0.09 cm), recorded in an argon atmosphere (heating rate: 10 K min⁻¹).

The morphologies of the firwood derived carbons are shown in Fig. 4. The carbons obtained in the molten salts (WDPC 1 to 3) show a sheet-like microstructure. In MSC, the ionized environment could prevent the product from agglomerating.³² Moreover, the decrease in firwood size leads to a decrease in the thickness of the carbon sheet, as shown in WDPC-3 to WDPC-1. A thinner carbon sheet could render better electrical conductivity and then facilitate ion transfer, which is beneficial for its application in supercapacitors.³³

Fig. 4 FE-SEM images of the carbons resulting from firwood: (a) WDPC-1; (b) WDPC-2; (c) WDPC-3; (d) WDPC-0.

The graphitization degree of the carbons WDPC-x is determined by XRD and Raman spectroscopy analysis. As shown in Fig. 5a, all the carbons display low crystallinity. The peaks located at a 2θ value of about 22° and 43° correspond to the (002) and (100) diffractions of a graphitic-type carbon. The inter-planar distance (d002) could be calculated from the position of the (002) peak by the Bragg equation (2d sin θ = λ).^34–36 The 2θ values of the (002) peak for WDPC-1, 2, 3 and 0 are 21.9°, 21.76°, 21.2° and 21.46°, and the corresponding values of d002 are 0.405, 0.408, 0.419 and 0.414 nm, respectively. As the size of the firwood decreases, the d002 value of the derived carbon decreases and becomes closer to that of natural graphite (0.335 nm).

Fig. 5 (a) XRD patterns and (b) Raman spectra of the obtained samples WDPC-x.

As shown in Fig. 5b, two peaks centered at about 1350 and 1600 cm⁻¹ represent the characteristic disordered (D) and graphitic (G) planes of a carbon material, respectively.^37,38 The graphitization degree can be estimated from the relative intensity of the D and G bands (I_D/I_G). The values of I_D/I_G for WDPC-1, 2, 3 and WDPC-0 are 0.84, 0.87, 0.92 and 0.89, respectively. Typically, the smaller the value of I_D/I_G, the higher the degree of graphitization. These results are consistent with the XRD data.

To understand the effect of the firwood size on the evolution of the pore structure in their derived carbons, the corresponding derived carbons were characterized by N₂ sorption analysis. N₂ adsorption–desorption isotherms and the corresponding BJH curves of the as-prepared carbons are presented in Fig. 6. All four carbons’ results resemble the type I isotherm with H4 hysteresis loops, which suggests the existence of different pore sizes from micro- to meso-pores.^30,37 The N₂ sorption data are summarized in Table S2.† It can be seen that the porosity of the resultant carbon materials is significantly related to the size of the raw materials. Upon decreasing the size of the raw materials from 2 to 0.09 cm, the specific surface area of the resultant carbons increases from 215 to 818 m² g⁻¹, and the pore volume increases from 0.10 to 0.44 cm³ g⁻¹. Moreover, compared with the carbons carbonized in pure Ar, the molten salt carbonized carbons possess a larger specific surface area and pore volume, indicating that the molten salt catalyzed the carbonization of the firwood biomass precursor.

Fig. 6 (a) Nitrogen adsorption–desorption isotherms and (b) pore size distribution curves of the obtained samples of WDPC-x.

3.2 Electrochemical performances of the obtained WDPC-x

As shown in Fig. 7a, the CV curves of WDPC-1 remain rectangular in shape for all scan rates. Such behavior is in accord with the typical characteristics of supercapacitors and implies superior high-rate capability for WDPC-1. The galvanostatic charge–discharge curves for WDPC-1 at a current density from 0.2 to 2 A g⁻¹ are shown in Fig. 7b. As can be seen, symmetric triangular shapes at all current densities indicate high electrical conductivity and low charge transfer resistance for the electrode. The specific capacitances calculated based on the discharge curves at 2, 1, 0.5, 0.2 A g⁻¹ are as high as 142, 145, 158, 189 F g⁻¹, respectively. All the CV curves of the different WDPC-x working electrodes at a scan rate of 20 mV s⁻¹ maintain similar quasi-rectangular shapes, suggesting typical capacitive behavior (Fig. 7c), and the areas of the rectangles and current responses obey the order WDPC-1 > WDPC-2 > WDPC-0 > WDPC-3. These results suggest that the capacitive performance of firwood derived porous carbon increases upon decreasing the size of the raw materials. Fig. 7d shows the charge–discharge curves of the WDPC-x working electrodes at a current density of 0.2 A g⁻¹. The lengths of the discharge times are in the order WDPC-1 > WDPC-2 > WDPC-0 > WDPC-3.

Fig. 7 (a) CV curves of the WDPC-1 electrode at different scan rates from 2 to 50 mV s⁻¹. (b) Charge–discharge curves of WDPC-1 at current densities from 0.2 to 2 A g⁻¹. (c) CV curves of different electrodes at a scan rate of 20 mV s⁻¹. (d) Charge–discharge curves of WDPC-x at a current density of 0.2 A g⁻¹. (e) Specific capacitances of WDPC-x at various current densities. (f) Nyquist plots of WDPC-x. All the measurements were conducted using a three-electrode system in 1 M H₂SO₄ aqueous solution (the active material mass loading was about 8 mg cm⁻²).

Compared with the carbon prepared without the assistance of MS (WDPC-0), the MSC-derived carbon (WDPC-1) shows enhanced capacitive properties. As can be seen in Fig. 7, the specific capacitance of the sample prepared without MS is 165 F g⁻¹ at 0.2 A g⁻¹, which decreases to 115 F g⁻¹ at 2.0 A g⁻¹. The capacitance retention during high-rate charge/discharge is only 70%. A specific capacitance as high as 189 F g⁻¹ at 0.2 A g⁻¹ is seen for the MSC derived carbon (WDPC-1). This value remains as high as 160 F g⁻¹ at 2.0 A g⁻¹, with a capacitance retention of 85%. These improvements in the specific capacitance and high-rate capability highlight the merits of the MSC method for the preparation of capacitive carbons.

For practical applications, the areal capacitance (F cm⁻²) is equally as important as the mass capacitance (F g⁻¹). For the carbon derived using MSC from precursors with a size of 0.09 cm, the areal capacitance is as high as 1.5 F cm⁻² (189 F g⁻¹) at a current density of 0.2 A g⁻¹, with an areal capacitance of 1.3 F cm⁻² (160 F g⁻¹) remaining at a current density of 2.0 A g⁻¹. Upon increasing the precursor size to 2.0 cm, the areal capacitance of the obtained carbon decreases to 1.1 F cm⁻² (142 F g⁻¹) at a current density of 0.2 A g⁻¹. This value sharply decreases to 0.88 F cm⁻² (110 F g⁻¹) at a current density of 2.0 A g⁻¹. This result highlights the importance of the precursor size in the potential applicability of the MSC derived carbons, with enhanced capacitive capability achieved for MSC derived carbon prepared with a smaller precursor size.

It was previously reported²¹ that the specific capacitance of the carbons obtained via MSC of waste biomass such as peanut shells, chestnut shells, leaves of the phoenix tree, corn cores, wormwood, orange peel and sunflower shells without tuning the precursory size ranged from 78.5 to 148 F g⁻¹. The above values are much lower than those of the herein obtained carbons with rational tailoring of the precursor size. Such an enhancement highlights the merits of the present strategy for obtaining better capacitive carbon.

Although a smaller precursor size is beneficial to capacitor applications, the production yield of the MSC method also decreases with decreasing precursor size. Therefore, the precursor size should be prudently tailored to ensure a balance between the capacitive properties and production yield.

The Nyquist plots for WDPC-x are composed of a semicircle in the high frequency region and a straight line in the low frequency region (Fig. 7f). The high frequency intercept of the semicircle with the real axis (Z′) corresponds to the equivalent series resistance (ESR), which includes electrolyte resistance (R_s), the contact resistance between the current collector and the active material, and the internal resistance of the electrode.³⁹ With the same working electrode fabrication procedure and electrolyte, the different ESRs could be ascribed to the internal resistances of the electrodes. The ESR values of WDPC-1, WDPC-2, WDPC-3 and WDPC-0 are 0.67, 1.16, 2.03 and 1.61 Ω, respectively, indicating that WDPC-1 has extremely low resistance and excellent electrical conductivity. The diameter of the semicircle referring to the charge transfer resistance also obeys similar rules to the ESR. These results confirm that the electrical conductivity of the as-prepared carbon increases upon decreasing the size of the firwood precursor in a MSC process.

The cycling performance of the WDPC-1 electrode was evaluated using galvanostatic charge–discharge tests at a current density of 1 A g⁻¹ in 1 M H₂SO₄. As shown in Fig. 8, the specific capacitance remains at 97% of the initial capacitance after 5000 cycles, demonstrating that the WDPC-1 electrode exhibits excellent cycling stability.

Fig. 8 Cycling performance of WDPC-1 at a current density of 1 A g⁻¹ recorded in 1 M H₂SO₄ in a three-electrode system.

4. Conclusions

Firwood was successfully converted to capacitive carbons by a Molten Salt Carbonization (MSC) method. Compared with the carbonization of firwood in an inert gas atmosphere under the same conditions, the MSC derived carbon shows higher specific capacitance and enhanced high-rate capability, highlighting the merits of MSC in the preparation of capacitive carbon from biomass. Importantly, engineering the size of the precursor biomass is proven to be effective in further enhancing the capacitive capability, with smaller-sized firwood corresponding to higher-capacitance carbon but a lower carbon yield. This result highlights the merits of the MSC of biomass in producing more porous carbon and introducing more capacitance-active functional groups. The maximum capacitance of the obtained carbon is 189.4 F g⁻¹ at 0.2 A g⁻¹ with 97% capacitance retention after 5000 cycles.

Acknowledgements

The authors are grateful to the Natural Science Foundation of China (51325102, 21673162 and 51674177), the International Science & Technology Cooperation Program of China (ISTCPC, 2015DFA90750) and the Program for Creation Team of Hubei Province (2015CFA017).

Notes and references

1. P. Kalyani and A. Anitha, Int. J. Hydrogen Energy, 2013, 38, 4034–4045.
2. S. W. Lee, B. M. Gallant, H. R. Byon, P. T. Hammond and Y. Shao-Horn, Energy Environ. Sci., 2011, 4, 1972–1985.
3. A. Bridgwater and G. Peacocke, Renewable Sustainable Energy Rev., 2000, 4, 1–73.
4. J. L. Chang, Z. Y. Gao, X. R. Wang, D. P. Wu, F. Xu, X. Wang, Y. M. Guo and K. Jiang, Electrochim. Acta, 2015, 157, 290–298.
5. H. Chen, D. Liu, Z. H. Shen, B. F. Bao, S. Y. Zhao and L. M. Wu, Electrochim. Acta, 2015, 180, 241–251.
6. X. J. He, P. H. Ling, M. X. Yu, X. T. Wang, X. Y. Zhang and M. D. Zheng, Electrochim. Acta, 2013, 105, 635–641.
7. M. Biswal, A. Banerjee, M. Deo and S. Ogale, Energy Environ. Sci., 2013, 6, 1249–1259.
8. I. I. Misnon, N. K. M. Zain, R. Abd Aziz, B. Vidyadharan and R. Jose, Electrochim. Acta, 2015, 174, 78–86.
9. C. Peng, X. B. Yan, R. T. Wang, J. W. Lang, Y. J. Ou and Q. J. Xue, Electrochim. Acta, 2013, 87, 401–408.
10. R. B. Qiang, Z. G. Hu, Y. Y. Yang, Z. M. Li, N. An, X. Y. Ren, H. X. Hu and H. Y. Wu, Electrochim. Acta, 2015, 167, 303–310.
11. J. Y. Qu, C. Geng, S. Y. Lv, G. H. Shao, S. Y. Ma and M. B. Wu, Electrochim. Acta, 2015, 176, 982–988.
12. E. Redondo, J. Carretero-Gonzalez, E. Goikolea, J. Segalini and R. Mysyk, Electrochim. Acta, 2015, 160, 178–184.
13. B. Huang, X. Chen, M. Jiang and S. Gao, Spectrosc. Spectral Anal., 2006, 26, 455–459.
14. F. C. Wu, R. L. Tseng, C. C. Hu and C. C. Wang, J. Power Sources, 2004, 138, 351–359.
15. F. C. Wu, R. L. Tseng, C. C. Hu and C. C. Wang, J. Power Sources, 2006, 159, 1532–1542.
16. F. C. Wu, R. L. Tseng, C. C. Hu and C. C. Wang, J. Power Sources, 2005, 144, 302–309.
17. H. Bennadji, K. Smith, M. J. Serapiglia and E. M. Fisher, Energy Fuels, 2014, 28, 7527–7537.
18. C. Di Blasi, C. Branca, V. Lombardi, P. Ciappa and C. Di Giacomo, Energy Fuels, 2013, 27, 6781–6791.
19. H. Lu, E. Ip, J. Scott, P. Foster, M. Vickers and L. L. Baxter, Fuel, 2010, 89, 1156–1168.
20. T. Pattanotai, H. Watanabe and K. Okazaki, Fuel, 2014, 117, 331–339.
21. H. Y. Yin, B. H. Lu, Y. Xu, D. Y. Tang, X. H. Mao, W. Xiao, D. H. Wang and A. N. Alshawabkeh, Environ. Sci. Technol., 2014, 48, 8101–8108.
22. G. Leofanti, M. Padovan, G. Tozzola and B. Venturelli, Catal. Today, 1998, 41, 207–219.
23. S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc., 1938, 60, 309–319.
24. E. P. Barrett, L. G. Joyner and P. P. Halenda, J. Am. Ceram. Soc., 1951, 73, 373–380.
25. B. Lu, Z. Xiao, H. Zhu, W. Xiao, W. Wu and D. Wang, J. Power Sources, 2015, 298, 74–82.
26. S. Wang, Z. Ren, J. Li, Y. Ren, L. Zhao and J. Yu, RSC Adv., 2014, 4, 31300.
27. L. Zhao, Y. Qiu, J. Yu, X. Deng, C. Dai and X. Bai, Nanoscale, 2013, 5, 4902–4909.
28. L. Z. Fan, S. Qiao, W. Song, M. Wu, X. He and X. Qu, Electrochim. Acta, 2013, 105, 299–304.
29. C. T. Hsieh and H. Teng, Carbon, 2002, 40, 667–674.
30. T. E. Rufford, D. Hulicova-Jurcakova, K. Khosla, Z. Zhu and G. Q. Lu, J. Power Sources, 2010, 195, 912–918.
31. J. S. Le Blond, S. Strekopytov, C. Unsworth and B. J. Williamson, Anal. Methods, 2011, 3, 1752–1758.
32. B. He, W. C. Li and A. H. Lu, J. Mater. Chem. A, 2015, 3, 579–585.
33. Y. Li, Z. Li and P. K. Shen, Adv. Mater., 2013, 25, 2474–2480.
34. Z. J. Zhang, P. Cui, C. Chen, X. Y. Chen and J. W. Liu, J. Solid State Electrochem., 2014, 18, 59–67.
35. L. H. Zhu, Q. M. Gao, Y. L. Tan, W. Q. Tian, J. D. Xu, K. Yang and C. X. Yang, Microporous Mesoporous Mater., 2015, 210, 1–9.
36. C. I. Su, W. C. Shih, C. M. Wang, Y. S. Liu and S. P. Wu, Fibers Polym., 2013, 14, 1808–1816.
37. G. Ma, Q. Yang, K. Sun, H. Peng, F. Ran, X. Zhao and Z. Lei, Bioresour. Technol., 2015, 197, 137–142.
38. A. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095.
39. A. Ghosh and Y. H. Lee, ChemSusChem, 2012, 5, 480–499.

---

Footnote

† Electronic supplementary information (ESI) available: More characterizations. See DOI: 10.1039/c6ra22191b

---