Home

Search

Collections

Journals

About

Contact us

My IOPscience

Freeze-drying for sustainable synthesis of nitrogen doped porous carbon cryogel with enhanced supercapacitor and lithium ion storage performance

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2015 Nanotechnology 26 374003 (http://iopscience.iop.org/0957-4484/26/37/374003) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 131.172.36.29 This content was downloaded on 25/01/2016 at 10:21

Please note that terms and conditions apply.

Nanotechnology Nanotechnology 26 (2015) 374003 (11pp)

doi:10.1088/0957-4484/26/37/374003

Freeze-drying for sustainable synthesis of nitrogen doped porous carbon cryogel with enhanced supercapacitor and lithium ion storage performance Zheng Ling, Chang Yu, Xiaoming Fan, Shaohong Liu, Juan Yang, Mengdi Zhang, Gang Wang, Nan Xiao and Jieshan Qiu Carbon Research Laboratory, Liaoning Key Lab for Energy Materials and Chemical Engineering, State Key Lab of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, Dalian 116024, People’s Republic of China E-mail: [email protected] Received 13 May 2015, revised 14 July 2015 Accepted for publication 3 August 2015 Published 27 August 2015 Abstract

A chitosan (CS) based nitrogen doped carbon cryogel with a high specific surface area (SSA) has been directly synthesized via a combined process of freeze-drying and high-temperature carbonization without adding any activation agents. The as-made carbon cryogel demonstrates an SSA up to 1025 m2 g−1 and a high nitrogen content of 5.98 wt%, while its counterpart derived from CS powder only shows an SSA of 26 m2 g−1. Freeze-drying is a determining factor for the formation of carbon cryogel with a high SSA, where the CS powder with a size of ca. 200 μm is transformed into the sheet-shaped cryogel with a thickness of 5–8 μm. The as-made carbon cryogel keeps the sheet-shaped structure and the abundant pores are formed in situ and decorated inside the sheets during carbonization. The carbon cryogel shows significantly enhanced performance as supercapacitor and lithium ion battery electrodes in terms of capacity and rate capability due to its quasi two-dimensional (2D) structure with reduced thickness. The proposed method may provide a simple approach to configure 2D biomass-derived advanced carbon materials for energy storage devices. S Online supplementary data available from stacks.iop.org/NANO/26/374003/mmedia Keywords: supercapacitor, carbon, biomass (Some figures may appear in colour only in the online journal) 1. Introduction

desired to explore advanced carbon electrode materials with tailored physical and chemical properties from an inexpensive carbon source to further increase energy density without sacrificing the rate capability and cycle life of SCs and LIBs. Up to now, various strategies for tuning the properties of carbon materials aimed at enhancing their electrochemical performance have been developed via optimizing the pore structure by template methods, activation or etching [5–13], introducing heteroatom and coupling with nanocarbons or electrochemical active materials [14–17]. Of the strategies available now, the heteroatom modification, especially doping

Electrical energy storage technology is a critical element in reducing dependence on fossil fuel and developing a sustainable society. Supercapacitors (SCs) and lithium ion batteries (LIBs) are two electrical energy storage devices with the most potential due to their intriguing merits in terms of rate capability, energy density and cycle life [1, 2]. Although these devices have benefited tremendously from novel electrode materials with well-developed porosity [3], especially for carbon-based electrode materials [4, 5], it is still highly 0957-4484/15/374003+11$33.00

1

© 2015 IOP Publishing Ltd Printed in the UK

Nanotechnology 26 (2015) 374003

Z Ling et al

with nitrogen species, is one of the most promising approaches, as it can improve the wettability and the affinity for electrolyte, and contribute to improved electrochemical performance [18–21]. Biomass materials as cheap, ubiquitous and renewable natural resources have been demonstrated to be the ideal precursors for synthesis of heteroatom-doped carbonaceous materials with excellent performance for SCs and LIBs [22–26]. The template methods are versatile and widely used for fabricating porous carbon with precisely controlled pore sizes [4]; nevertheless, it remains a major challenge to make the giant molecules, specifically the biomass-based molecules, into porous carbons using common hard templates with confined pore sizes, such as zeolites and silica, due to the steric effect of the pore sizes of the hard templates. Besides, the limited solubility of biomass-based molecules makes the versatile template methods infeasible and inefficient. The complex process and high cost also hinder their practical use on a large scale. Alternatively, activation is the most common approach to transform biomass and biomass-based molecules into carbon materials with well-developed porosity and extremely high SSA in the presence of the activation agents, such as KOH [5, 27]. However, low yield and poor economic efficiency will hinder their application except for the high corrosivity of the employed activating agents. It is also challenging and highly desirable to develop a simple, efficient, and activationfree environmentally friendly method to simultaneously control the pore structure and composition of as-made carbon materials. Compared with template methods and activation approaches, direct carbonization is obviously a superior process in terms of convenience, cost effectiveness, and sustainable chemistry. However, few porous carbon materials with high SSAs have been synthesized via direct carbonization [28]. Recently, it was reported that a specific functional group of precursor can be employed to produce the mesoporous structures of the resulting activated carbons [29], and a mesoporous activated carbon can be synthesized through creating interchain bonding in its precursor [30]. However, results about their performance as electrodes of SCs or LIBs have not been reported. Besides, the mesopores were not so effective as to improve the energy density of SCs as a micropore dose, and their contribution to the enhancement of the rate capability is also limited [31]. The pore size at ca. 0.5 nm was suggested to be most effective in a double layer formation in an aqueous electrolyte, due to the good matching with the size of hydrated ions in the electrolyte [21]. The effectiveness of ultra-micropores (less than 1 nm) in the formation of a double layer was reported and supported recently [32–36]. Therefore, it is desirable to design and fabricate microporous carbon by a simple and environmentally friendly method. In this work, we report a simple yet effective method for synthesizing nitrogen-doped carbon cryogel with high SSA using chitosan (CS) as the nitrogen-containing carbon precursor without any activation agents. The CS was dissolved and freeze-dried to build CS cryogel which was composed of

sheet-like CS layers and interconnected pores between sheet layers. The thin layer and pores facilitate the escape of produced gas to produce micropores during the carbonization process. The structure, SSA and surface properties of the carbon materials can be significantly adjusted and tailored by this cryogel method. The as-produced material shows dramatic enhancement of electrochemical performance compared with that from direct carbonization of CS powder. The present results provide new insights into the synthesis of advanced carbon materials via a simple process for utilizing biomassbased carbon precursors.

2. Methods 2.1. Preparation of chitosan-based carbons

Chitosan (with purity over 99%, viscosity of 50–800 mPa·s and degree of deacetylation of 80–95%) and hydrochloric acid (37 wt%) were purchased from Sinopharm Chemical Reagent Co. Ltd. All the chemicals were analytical grade and used without any purification. In a typical run, chitosan (2 g) was added into deionized water (70 mL) under vigorous magnetic stirring for 10 min. Then HCl (10 mL of 1 mol L−1) was dropwise added into the turbid liquid under magnetically stirring. The powders of chitosan dissolved in the acidized solution forming a viscous solution after stirring for 180 min. The solution was frozen in a refrigerator for 120 min followed by a freeze-drying process in a refrigerant dryer to remove the water. The freeze-dried gel was placed into a horizon quartz tube and heated at 800 °C for 60 min at a heating rate of 5 °C min−1 in flowing N2. After the heat treatment, the product was cooled naturally in a N2 atmosphere. The obtained carbon cryogel was named as G-CS-8. Carbon cryogels carbonized at 700 and 900 °C were also made and labeled as G-CS-7 and G-CS-9. Pristine chitosan powder (2 g) was directly carbonized through the same carbonization procedure as described above, and the product was named P-CS-8. The as-made samples were ground into fine powders before characterization and electrochemical tests. For comparison, CS film was made by naturally drying CS solution with the same composition for making CS cryogel. The CS film was carbonized at 800 °C for 60 min to get carbon film (labeled as CS-film-8). 2.2. Characterization methods

The compositions of the samples were analyzed by a powder x-ray diffractometer (D/MAX-2400, Japan) using Cu-Kα radiation generated at 40 kV and 100 mA. The morphology was studied by a scanning electron microscope (SEM, QUANTA 450, FEI USA) and a transmission electron microscope (TEM, F30, FEI USA) operated at 200 kV. The pore structures of the obtained carbon materials were studied by N2 adsorption/desorption at −196 °C (ASAP2020, Micromeritics, USA). The BET equation was used to calculate the specific surface area using the adsorption branch. The total pore volume was calculated using the adsorbed volume of N2 at relative pressure (P/P0) of 0.995. DFT was used to 2

Nanotechnology 26 (2015) 374003

Z Ling et al

get the pore size distribution with a slit pore model being used for calculation. The nitrogen of the samples was identified by an elemental analyzer (Vario EL III, Elementar, Germany) using a combustion method. The surface composition was analyzed by an x-ray photoelectron spectrometer (XPS, ESCALAB 250, Thermo VG, USA) with Al Kα radiation (15 kV, 150 W). The Raman spectroscopy analysis was conducted on a confocal microprobe Raman system (Thermo Scientific DXR Raman microscope) with an excitation wavelength of 532 nm.

For the lithium ion battery test, the as-made carbon was mixed with conductive carbon black and polyvinylidene fluoride at a mass ratio of 80:10:10 with NMP as the solvent to form a slurry. The slurry was coated and dried on copper foil, then punched into discs with a diameter of 16 mm (around 0.4 mg active materials on each electrode). The asmade electrode was assembled with lithium foil and Celgard polypropylene separator into a 2016-type button cell filled with electrolyte (1 mol L−1 LiPF6 in a mixture of ethylene carbonate and dimethyl carbonate with a volume ratio of 1:1) in an Ar-filled glove box with water and oxygen containing less than 0.1 ppm. The GCC and CV were employed to study the lithium ion battery performance of the as-made carbon materials.

2.3. Electrochemical measurements

The electrochemical measurements were performed using a CHI660D electrochemical workstation (Shanghai, China) at room temperature in a three-electrode system with 1 mol L−1 H2SO4 aqueous solution as electrolyte. A Pt foil (20 mm*20 mm) was used as the counter-electrode and Hg/ Hg2SO4 as the reference electrode. The working electrode was prepared by mixing the finely ground active matter, carbon black and polytetrafluoroethylene (PTFE, 0.6 wt% dispersion in water) at a mass ratio of 85:10:5. The paste was rolled into a sheet with uniform thickness and punched into 1 cm2 discs. A typical electrode had a weight between 6 and 8 mg after vacuum drying for 12 h at 100 °C. The Ti mesh was used as current collector. The electrochemical impedance spectroscopy was studied at different electrode potential (−0.6, −0.3, 0.0 and 0.3 V versus Hg/Hg2SO4) in the frequency range of 10 mHz to 100 kHz. The cycle stability of electrodes was studied in a Land battery test system (CHI2000, Wuhan, China). The cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) were employed to study the performance of electrodes. The specific capacitance can be calculated according the following equation C=

ò I dV vm

3. Results and discussion The synthesis strategy of CS-based carbon powder (labeled as P-CS-8, where the number stands for carbonization at 800 °C) and carbon cryogel (labeled as G-CS-8) is schematically illustrated in figure 1. The carbon powder made by direct carbonization of pristine CS powder (P-CS) is expected to be nonporous, due to the extensive stacking of the CS molecules, while the CS cryogel (G-CS) functions as opened folding fans, leading to much larger interfaces and interconnected pores between sheet layers being exposed, where these pores can be produced and kept by the ice templates during freezedrying, as evidenced by their greatly expanded volume (see figure S1(a)). The reduced stacking of G-CS and interconnected pores would facilitate the formation of a larger specific surface area during the carbonization process. The morphologies of P-CS, G-CS and their heat-treated products (P-CS-8 and G-CS-8) were studied using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As seen in figures 2(a) and (c), there are dramatic differences between the microstructures of the P-CS and G-CS. The P-CS is an irregularly shaped particle, and most of the particles are dense and the size is over 200 μm (figure 2(a)). After carbonization, the size of the CS particle has shrunk and no obvious changes are observed in its morphology (figures 2(b) and S1(b)). In the case of G-CS, smooth sheets (see figure 2(c)) rather than irregular particles are observed, with a large number of voids locating between the sheets. The thickness of the sheet ranges from 5 to 8 μm. The sheet-like shape remains intact after carbonization. Nevertheless, some new pores were formed inside the sheets (see figure S2(a)), which may have been created by the escaping molecules produced by pyrolysis of G-CS. The produced pores become more obvious when the sheets of G-CS-8 are crumbled, while there are no pores in P-CS-8 as can be seen in figure S2(b). Therefore, it is confirmed that the thin sheet of G-CS facilitates the escape of small molecules to some extent, producing open pores during the carbonization process. Highresolution transmission electron microscopy (HRTEM) reveals G-CS-8 and P-CS-8 are highly disordered carbons without obvious graphite ribbons (figures 2(e) and (f)). However, G-CS-8 has relatively larger graphitic

(1 )

where C is the gravimetric specific capacitance (F g−1) , I is the response current density (A g−1), v is the potential scanning rate (V s−1), m is the active mass in the electrode (g). The gravimetric capacitance (F g−1) was also calculated from constant charge/discharge curves using the following formula C=

Icons t mV

(2 )

where Icons is the constant discharge current (A), t is the discharge time (s), m is the mass of the active matter (g) and !V is the voltage range. The self-discharge was studied using a Land battery test system (CHI2000, Wuhan, China). The symmetric supercapacitor was assembled using two pieces of G-CS-8 electrode with identical mass. Non-woven fabrics were used as the separator and 1 M H2SO4 as the electrolyte. The supercapacitor was charged at a current density of 0.5 A g−1 to 0.8 V, then held at 0.8 V for 30 s. After that, the voltage change was recorded over time. 3

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 1. Schematic illustration of the formation of (a) P-CS-8 and (b) G-CS-8.

drying techniques employed to process precursors were also demonstrated to have an impact on the morphologies of asmade carbons for the created differences in the spatial distance between precursors [40, 41]. In the present system, this is also the case. The CS molecules, which were dissolved into the diluted acid solution, were partly set free by weakening and/or disrupting the inter-molecule bonds. Then, the disturbed molecules were frozen and preserved in ice during the freeze-drying process and separated from the extensively stacked CS molecules, yielding voids derived from sublimation of the ice, which is also beneficial to produce porous carbon materials. Figure 3(b) shows the XRD patterns of carbonized products of P-CS and G-CS, and the broad diffraction peaks locating at ca. 25° indicate the characteristic of amorphous carbon. In comparison to P-CS-8, the peak at 44°, which corresponds to the (100) reflection of graphite, is much more obvious for G-CS-8, evidencing a relatively improved ordering of the microstructure. The improved degree of graphitization is also supported by the Raman spectra (see figure S4). The intensity ratio (ID/IG) of the D band and G band is 1.0 for G-CS-8, which is slightly lower than that of P-CS-8 (ID/IG=1.1). In order to further reveal a significant impact of the cryogel structure on their porosities, the pore structures of the obtained carbon materials were studied by N2 adsorption at −196 °C. It can be clearly seen from figure 4(a) that the P-CS-8 made by direct carbonization of the CS powder shows a type-II sorption isotherm, indicative of its non-porous characteristics. It is interesting that the adsorption isotherm of G-CS-8 demonstrates a typical type-I sorption isotherm and the nitrogen adsorption amount is much higher than that of

microcrystallite compared with P-CS-8, which suggests a slightly higher degree of graphitization. The HRTEM images also reveal that G-CS-8 has a much more developed porosity than P-CS-8, where the black points in the HRTEM images represent the pores (figures 2(e) and (f) and S3). The differences in size of microcrystallite and porosity between G-CS-8 and P-CS-8 samples are believed to result from the open cryogel structure and 2D thin sheet characteristics. The thin sheets of the cryogel would facilitate the rearrangement of carbonaceous intermediates and the escape of small molecules during the carbonization process. The interactions of disturbed molecules in the G-CS could also make a contribution to the developed porosity. The as-made samples were characterized using the x-Ray diffraction (XRD) technique; the detailed results are shown in figure 3. It can be clearly seen from figure 3(a) that the characteristic peaks corresponding to the crystalline nature of CS broaden or disappear after dissolving and freeze-drying to some degree. The distinct differences in XRD patterns (figure 3(a)) indicate that the arrangement of CS molecules in G-CS is different from that of P-CS in terms of ordering. The arrangement of CS molecules in the as-received P-CS is much more regular and ordered than that of the cryogel, which may be attributed to the hydrogen bonds and other interaction between CS precursors. Nevertheless, these interactions would be weakened and/or disrupted by the mutual repulsion of positive charges carried on the CS molecules due to protonation in the process of dissolution for CS powders in dilute acid [37]. It has also been pointed out that the morphology of CS can be significantly altered by some ions due to disruption of the inter-polymer bonds of CS [38, 39]. Moreover, the 4

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 2. SEM images of (a) pristine chitosan powder (P-CS), (b) P-CS-8, (c) freeze-dried chitosan cryogel (G-CS), and (d) G-CS-8. HRTEM images of (e) G-CS-8 and (f) P-CS-8.

only a few pores larger than 2 nm (figure 4(b)). These sizes are ideal for forming double-layer capacitance using an aqueous electrolyte [21]. The much larger pore volume and SSA of G-CS-8 would benefit from its cryogel structure, where small molecules would easily escape from the thinner matrix of the G-CS in contrast with P-CS, as indicated in figures 2(a) and (c). Moreover, the thin characteristics of the carbon cryogel sheet would contribute to the increase in the SSA. The disturbed molecule interactions may create larger

P-CS-8, evidencing the rich micropore characteristics of G-CS-8. Table 1 shows the detailed pore characteristic parameters of the as-made carbons. The Brunauer–Emmett–Teller (BET) SSA of G-CS-8 is 1025 m2 g−1 and the corresponding total pore volume is 0.44 cm3 g−1, which is much higher than that of P-CS-8 (26 m2 g−1, 0.035 cm3 g−1). Such a high SSA of G-CS-8 is impressive considering it is produced by direct carbonization in the absence of any activation agents. The pore size of G-CS-8 virtually concentrates at ca. 0.5 nm with 5

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 3. XRD patterns of (a) P-CS and G-CS and (b) P-CS-8 and G-CS-8.

Figure 4. (a) N2 adsorption/desorption isotherms of P-CS-8 and G-CS-8, (b) density functional theory (DFT) pore size distribution of

G-CS-8. Table 1. The pore structural parameters of the chitosan based carbon materials.

Sample

SBET (m2 g−1)

Vtotal (cm3 g−1)

Vmicro (cm3 g−1)

Vmicro/ Vtotal

P-CS-8 G-CS-8

26 1025

0.035 0.44

0.0096 0.40

0.27 0.91

Table 2. The doped nitrogen and oxygen contents of chitosan-based carbon materials.

Heteroatom loading Sample

N (wt%)a

N (at%)b

O (at%)b

P-CS-8 G-CS-8

8.33 5.98

4.60 3.14

14.75 14.97

a

Calculation by elemental analysis using the combustion method. By XPS analysis.

distance between CS molecules and make the CS pyrolysis and the pore development occur more easily. The voids resulting from sublimation of the ice not only facilitate the escape of the produced small molecules, but also ensure that the developed pores can be easily accessed. The nitrogen contents were analyzed by element analysis, of which the detailed results are shown in table 2. It is noted that G-CS-8 keeps a relatively high N content of 5.98 wt%, although a slight decrease is demonstrated compared with P-CS-8 (8.33 wt%). The reason behind this may result from the open structure of G-CS-8, where produced small

b

molecules can escape more easily from the carbon matrix during the carbonization process. In addition, the disrupted molecule interaction and less dense arrangement may also contribute to the loss of nitrogen species for G-CS-8. The corresponding surface states and components of the as-made carbon samples were further analyzed by x-ray photoelectron spectroscopy (XPS). 6

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 5. XPS spectra of (a) C1s, (b) N1s and (c) O1s for P-CS-8 and G-CS-8, (d) schematic diagram of nitrogen and oxygen species on ideal

graphene.

The spectra of C1s could be fitted by several peaks corresponding to sp2 C-C (284.6 eV), C-OH (285.3 eV), C=O (286.3 eV) and COOH (288.9 eV) (figure 5(a)) [17]. The results for nitrogen- and oxygen-containing functional groups of P-CS-8 and G-CS-8 are presented in figures 5(b), (c) and table 2 and S1, and the corresponding chemical formulas within an ideal graphene are schematically presented in figure 5(d). The nitrogen-containing functional groups of G-CS-8 and P-CS-8 are predominantly pyridinic N (398.4 eV), pyrrolic N (400.3 eV), and quaternary N (401.1 eV) [16, 17]. However, the relative content of each nitrogen species shows a distinct difference between G-CS-8 and P-CS-8 (table S1). Pyrrolic N is the dominant nitrogen species for G-CS-8, while pyridinic N is the main one for P-CS-8. The exact mechanism for this variation is not clear now due to the complex of the carbonization process. The arrangement of the CS in G-CS and P-CS could play a critical role in this variation. The XPS analysis also indicates high oxygen doping (figure 5(c)); nevertheless, no obvious differences are observed in both the amount and the types (figure 5(c) and table S1). The as-produced carbon cryogel with high SSA is expected to show greatly enhanced performance for SCs and

LIBs. The electrochemical performance of G-CS-8 and P-CS8 as SC electrodes was studied using cyclic voltammetry (CV), galvanostatic charge–discharge (GCD) and electrochemical impedance spectroscopy (EIS). All the CV curves at a scan rate of 2 mV s−1 exhibit a quasi-rectangular shape (see figure 6(a)), indicating that the electric double layer contributes to most of the capacitances. The area surrounded by the CV curve of the G-CS-8 electrode is much larger than that of the P-CS-8, indicative of a much higher specific capacitance for CS cryogel–based carbon, due to the much larger SSA. The specific capacitance of G-CS-8 is 206 F g−1 at 2 mV s−1, which is almost 52 times larger than that of P-CS-8 (4 F g−1 at the same scan rate). The inconspicuous peaks appearing in the CV curve of G-CS-8 indicate the existence of redox reactions and/or pseudocapacitance, which may be due to the reactions of heteroatom-containing groups with electrolyte. Figure 6(b) shows the GCD curves of G-CS-8 at current densities of 0.1, 0.2, 0.5 and 1 A g−1. The slight variation of the slope suggests the contribution of redox reactions and/or pseudocapacitance, which is in agreement with the CV curve. The variation of the specific capacitances as a function of the current density is shown in figure 6(c). The specific 7

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 6. Electrochemical characterization of chitosan-based carbons in 1 M H2SO4: (a) CV cures at a scan rate of 2 mV s

−1

, (b) galvanostatic charge–discharge curves of G-CS-8 at different current densities, (c) specific capacitances of chitosan-based carbons as functions of current density, (d) Nyquist plot of G-CS-8; the upper inset is the equivalent circuit, (e) frequency response of G-CS-8, (f) cyclic stability of G-CS-8 at 1 A g−1. All the tests were conducted in a three-electrode system with Pt as the counter-electrode, Hg/Hg2SO4 as reference electrode and 1 M H2SO4 as electrolyte. The electrochemical impedance spectroscopy was tested at open circuit potential (∼0.0 V).

capacitance of G-CS-8 is as high as 242 F g−1 at 0.1 A g−1, which is impressive considering this simple and activationfree method using biomass-based CS as precursor. The specific capacitance slightly decreases as the increase of the current density, which is not uncommon for carbon materials with microporous characteristics [42]. The specific capacitance of G-CS-8 retains 161 F g−1 at 10 A g−1 corresponding to a retention rate of 67%, evidencing high accessibility of surface area and fast ion diffusion due to the unique 2D

structure of the carbon sheets in spite of the pore being less than 2 nm. The EIS was used to study the ion transport limitations within the structure of G-CS-8 (figures 6(d) and (e)). The Nyquist plot (figure 6(d)) shows a large phase angle at frequencies below 0.1 Hz, because the microporous carbon behaves like a planar electrode at the low-frequency range [11, 12]. The diffusion-like process is thought to result from the rate-limiting step at moderate frequencies (ca. 0.1 to 8

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 7. Self-discharge of supercapacitor made of G-CS-8, (a) the voltage change over time, (b) the voltage change in a square root time

scale.

natural drying (figure S6(a)), and were transformed into freestanding carbon films (labeled as CS-film-8, figure S6(b)) that could be directly used as electrodes without conductive additives and binders. However, the performance of the CSfilm-8 is far from competitive with that of G-CS-8 (figure S7) although the G-CS-8 electrode was made by adding a binder, evidencing the superiority of the freeze-drying method for making high-performance carbon materials. The normalized specific capacitance of G-CS-8 is 24 μF cm−2, which is larger than the conventional double-layer capacitance ranging from 5 to 20 μF cm−2 [43], evidencing the contribution of pseudocapacitance. The as-made G-CS-8 has a tap density of 0.51 g cm−3 which is slightly larger than the current commercially available activated carbons (less than 0.50 g cm−3) for supercapacitors [1], suggesting a larger volumetric capacitance for G-CS-8. In addition, the specific capacitance of G-CS-8 shows no degradation during 5000 cycles at 1 A g−1 (see figure 6(f)), demonstrating its superb electrochemical stability. The large specific capacitance and high cycle stability suggest that CS cryogel–based carbon is a promising candidate for the electrode materials of SCs. Most importantly, such carbon material can be synthesized using biomass-based CS by a simple and environmentally friendly method without consuming any added activating agents, and has the ability to significantly enhance the performance of the produced carbon. In addition, the SC performance of the G-CS–based carbons can be easily adjusted by changing the carbonization temperature (see the supercapacitor of G-CS-7 and G-CS-9 in figure S8). Self-discharge is an important issue for supercapacitor application. However, most of the reported research focuses on the synthesis and the maximum specific capacitance [27]; publications related to self-discharge of carbon materials, particularly ones derived from natural sources, are limited [27]. The self-discharge rate of the G-CS-8–based supercapacitor was determined by recording the change of the supercapacitor voltage over time. The as-made G-CS-8 shows a fast self-discharge rate (see figure 7(a)). The voltage drop follows a linear relationship with the square root of time (figure 7(b)), indicating diffusion processes controlling the mechanism of self-discharge [44].

50 Hz), due to the small current-penetration depth into the micorpores’ structure [11]. The small diameter of the semicircle at the high-frequency section indicates a small charge transfer resistance for G-CS-8 [8]. The equivalent circuit for fitting the EIS is shown in the inset of figure 6(d). The value of Rs (0.58 Ohm, the as-fitted value) represents the equivalent series resistance (ESR), which is the combination of the resistances of the electrolyte and the electrode. The value of W1 is the Warburg element (2.59, the as-fitted value), representing the ion diffusion into the porous G-CS-8 at the moderate frequencies (ca. 0.1 to 50 Hz). The value of Rct (0.31 Ohm, the as-fitted value) is the charge transfer resistance, measuring the redox reaction rate at the interface of electrode/electrolyte. The values of CPE (0.98, the as-fitted value) and Cdl (0.0007, the as-fitted value) represent the double-layer capacitance and the external double-layer capacitance, respectively. The Nyquist plots and the related equivalent circuit were obtained at different electrode potentials (−0.6 V, −0.3 V, 0.0 V and 0.3 V versus Hg/Hg2SO4, where 0.0 V almost equals the open circuit potential) (figure S5; the fitted equivalent circuit elements can be found in table S2). The shape of the Nyquist plots at different electrode potentials are similar in the study frequency range, while the one measured at 0.3 V (versus Hg/Hg2SO4) shows a slight deviation from the ideal capacitive behavior at low frequency. This phenomenon could be due to the reduced proton adsorption step caused by the positive potential. The as-made G-CS-8 also demonstrates a fast frequency response, providing 50% of its maximum capacitance at an operating frequency of 0.13 Hz. The small ESR and the small charge transfer resistance could contribute to the good rate performance. The superior specific capacitance of the G-CS-8 electrode is thought to be due to its high SSA and exposed heteroatoms and functional groups, which benefit from the freeze-drying method used to synthesize the CS cryogel. In order to further confirm the superiority of freeze-drying in producing higher-performance carbons, a controlled experiment was conducted by carbonization of a naturally dried chitosan solution with the same composition as the one for the CS cryogel. Transparent chitosan films were obtained after 9

Nanotechnology 26 (2015) 374003

Z Ling et al

Figure 8. Representative CV curves of electrodes based on (a) G-CS-8 and (b) P-CS-8 at a voltage range of 0.01 to 3.0 V (versus Li+/Li) and

a scan rate of 0.1 mV s−1; (c) rate capability of G-CS-8 and P-CS-8, (d) capacity retention and coulombic efficiency of G-CS-8 and P-CS-8 during 100 cycles at 0.1 A g−1.

Additionally, self-discharge is also related to the metal impurities and surface carboxylic acid groups which are inevitable for the natural source–based activated carbons without special cleaning treatment. Therefore, extra work is still needed before the as-made materials can be put into practical application. As expected, the enhanced electrochemical performance of G-CS-8 was also found for LIBs. The CS-based carbons as the LIB anode materials were investigated in a half-cell configuration with lithium foil as a counter-electrode in 1 mol L−1 LiPF6 electrolyte with ethylene carbonate-dimethyl carbonate (1:1 in volume) as solvent. Both of the CV curves of G-CS-8 and P-CS-8 (see figures 8(a) and (b)) show pronounced cathodic peaks locating at 0–1 V (versus Li+/Li) in their first cycles and at 0–0.3 V (versus Li+/Li) during the second and third cycles, which are the typical features of carbon anode materials [45]. It is worth noting that G-CS-8 exhibits a larger current density compared with P-CS-8 in all three cycles, especially for the first one. Its larger SSA and much more active sites are thought to be the major contributors to the enhanced current density for G-CS-8, suggesting a larger capacity. Figure 8(c) shows the capacities of CS-based carbons at various current densities. It can be seen that G-CS-8 exhibits higher capacity at all the tested current densities, which is believed to benefit from the enhanced SSA and the quasi 2D structure with

thinner thickness for G-CS-8, increasing the lithium storage and shortening the path lengths for ion diffusion [3]. The initial discharge curves of G-CS-8 and P-CS-8 show much higher capacities (1148 and 820 mAh g−1, respectively; figure S9) compared with the theoretical capacity of graphite (372 mAh g−1) [45]. The formation of the solid electrolyte interphase layer was believed to be related to the significant initial capacities caused by the irreversible consumption of Li ions and charge [45]. The reduced tap densities of G-CS-8 and P-CS-8 (0.51 and 0.56 g cm−3, respectively) also contribute to the enhanced specific capacitance. The cyclic stability of CS-based carbons is shown in figure 8(d) accompanied by their coulombic efficiency. After 50 cycles at 0.1 A g−1, the reversible capacity of G-CS-8 gradually increased to 395 mAh g-1 at the 100th cycle with a coulombic efficiency of 100%, while the reversible capacity of P-CS-8 shows a gradual decrease during the cycle with a capacity of 323 mAh g−1 and a coulombic efficiency of 99.4% at the 100th cycle. The difference in stability results from the distinct structures, where G-CS-8 has a much more open structure facilitating the lithium ion intercalation and deintercalation reversibly. The enhanced performances in capacity, rate capability and cyclic stability for G-CS-8 benefit from the control over the structure and composition by this facile freeze-drying approach. The enhancement in LIB performance once again demonstrates that the present strategy 10

Nanotechnology 26 (2015) 374003

Z Ling et al

[9] Wang D W, Li F, Liu M, Lu G Q and Cheng H M 2008 Angew. Chem. Int. Ed. 47 373–6 [10] Thomberg T, Tooming T, Romann T, Palm R, Jänes A and Lust E 2013 J. Electrochem. Soc. 160 A1834–41 [11] Laheäär A, Jänes A and Lust E 2011 Electrochim. Acta 56 9048–55 [12] Laheäär A, Kurig H, Jänes A and Lust E 2009 Electrochim. Acta 54 4587–94 [13] Simon P and Gogotsi Y 2008 Nat. Mater. 7 845–54 [14] Zhao L, Fan L Z, Zhou M Q, Guan H, Qiao S Y, Antonietti M and Titirici M M 2010 Adv. Mater. 22 5202–6 [15] Fan X, Yu C, Ling Z, Yang J and Qiu J 2013 ACS Appl. Mater. Interfaces 5 2104–10 [16] Fan X, Yu C, Yang J, Ling Z and Qiu J 2014 Carbon 70 130–41 [17] Cong H-P, Ren X-C, Wang P and Yu S-H 2013 Energy Environ. Sci. 6 1185–91 [18] Kwon T, Nishihara H, Itoi H, Yang Q-H and Kyotani T 2009 Langmuir 25 11961–8 [19] Hao L et al 2012 Energy Environ. Sci. 5 9747–51 [20] Wang D-W, Li F, Yin L-C, Lu X, Chen Z-G, Gentle I R, Lu G Q and Cheng H-M 2012 Chem. Eur. J. 18 5345–51 [21] Hulicova-Jurcakova D, Seredych M, Lu G Q and Bandosz T J 2009 Adv. Funct. Mater. 19 438–47 [22] Titirici M M, White R J, Falco C and Sevilla M 2012 Energy Environ. Sci. 5 6796–822 [23] Wang X et al 2013 Nat. Commun. 4 2905 [24] White R J, Budarin V, Luque R, Clark J H and Macquarrie D J 2009 Chem. Soc. Rev. 38 3401–18 [25] Dutta S, Bhaumik A and Wu K C W 2014 Energy Environ. Sci. 7 3574–92 [26] Mao Y, Duan H, Xu B, Zhang L, Hu Y, Zhao C, Wang Z, Chen L and Yang Y 2012 Energy Environ. Sci. 5 7950–5 [27] Wei L and Yushin G 2012 Nano Energy 1 552–65 [28] Budarin V, Clark J H, Hardy J J E, Luque R, Milkowski K, Tavener S J and Wilson A J 2006 Angew. Chem. 118 3866–70 [29] Sun M and Hong L 2011 Carbon 49 2173–80 [30] Sun M, Hong L and Tan G M 2012 J. Phys. Chem. C 116 352–60 [31] Fernández J A, Tennison S, Kozynchenko O, Rubiera F, Stoeckli F and Centeno T A 2009 Carbon 47 1598–604 [32] Chmiola J, Yushin G, Gogotsi Y, Portet C, Simon P and Taberna P L 2006 Science 313 1760–3 [33] Raymundo-Piñero E, Kierzek K, Machnikowski J and Béguin F 2006 Carbon 44 2498–507 [34] Chmiola J, Largeot C, Taberna P-L, Simon P and Gogotsi Y 2008 Angew. Chem. Int. Ed. 47 3392–5 [35] Huang J, Sumpter B G and Meunier V 2008 Chem. Eur. J. 14 6614–26 [36] Largeot C, Portet C, Chmiola J, Taberna P-L, Gogotsi Y and Simon P 2008 J. Am. Chem. Soc. 130 2730–1 [37] Ladet S, David L and Domard A 2008 Nature 452 76–U76 [38] Trimukhe K D and Varma A J 2008 Carbohydr. Polym. 71 698–702 [39] Trimukhe K D and Varma A J 2008 Carbohydr. Polym. 71 66–73 [40] Ishida O, Kim D Y, Kuga S, Nishiyama Y and Brown R M 2004 Cellulose 11 475–80 [41] Jazaeri E, Zhang L, Wang X and Tsuzuki T 2011 Cellulose 18 1481–5 [42] Hasegawa G, Kanamori K, Nakanishi K and Abe T 2012 J. Phys. Chem. C 116 26197–203 [43] Pandolfo A G and Hollenkamp A F 2006 J. Power Sources 157 11–27 [44] Béguin F and Frackowiak E 2013 Supercapacitor: Materials, Systems, and Applications (New York: Wiley) p 380–1 [45] Li Z, Xu Z, Tan X, Wang H, Holt C M B, Stephenson T, Olsen B C and Mitlin D 2013 Energy Environ. Sci. 6 871–8

for fabricating biomass-based carbon is very unique and has potential for configuring the electrode with high-performance energy storage devices. 4. Conclusions In summary, we have reported a simple yet effective approach to produce nitrogen-doped carbon materials with high SSA from biomass-based CS in the absence of any activation agents. The freeze-drying process causes CS molecules to be arranged irregularly and in a disordered formation, thus leading to open channels and abundant pore structures within the carbon matrix. The potential of the as-made carbon cryogel as electrode material for the SC and the LIB is explored. The results show that the carbon cryogel with reduced thickness of matrix and enhanced SSA exhibits significantly enhanced electrochemical performance in terms of capacity, rate capability and cycle stability compared with the derived carbon by direct carbonization of CS powders. This facile freeze-drying method will provide a new way to synthesize other biomass-based advanced carbon materials. Acknowledgments This work was partly supported by the National Natural Science Foundation of China (nos. U1203292 and 21336001), the Education Department of Liaoning Province of China (T2013001), and the Fundamental Research Funds for the Central Universities (DUT14LAB06). Supplementary information Includes digital photos of P-CS, G-CS, P-CS-8 and G-CS-8; SEM images of G-CS-8 and P-CS-8; HRTEM image of G-CS-8; Raman spectra of P-CS-8 and G-CS-8; EIS of G-CS8 at different electrode potentials and the equivalent circuit, fitted equivalent circuit elements of G-CS-8; digital photos of CS film and CS-film-8; CV curve of CS-film-8, specific capacitances of CS-film-8 and G-CS-8 at different scan rate; CV curves of G-CS-7 and G-CS-9, specific capacitances of G-CS-7 and G-CS-9 at different current densities, relative surface concentration of nitrogen and oxygen species. References [1] Simon P and Gogotsi Y 2012 Acc. Chem. Res. 46 1094–103 [2] Bruce P G, Scrosati B and Tarascon J-M 2008 Angew. Chem. Int. Ed. 47 2930–46 [3] Vu A, Qian Y and Stein A 2012 Adv. Energy Mater. 2 1056–85 [4] Nishihara H and Kyotani T 2012 Adv. Mater. 24 4473–98 [5] Wang J and Kaskel S 2012 J. Mater. Chem. 22 23710–25 [6] Fechler N, Fellinger T-P and Antonietti M 2012 Adv. Mater. 25 75–9 [7] Romanos J, Beckner M, Rash T, Firlej L, Kuchta B, Yu P, Suppes G, Wexler C and Pfeifer P 2012 Nanotechnology 23 015401 [8] Fan X, Yu C, Yang J, Ling Z, Hu C, Zhang M and Qiu J 2014 Adv. Energy Mater. 5 11