Introduction

As potential energy storage units, supercapacitors (SCs) have attracted extensive attention and been widely applied in electric vehicles, medical devices and forklifts, due to their long cycle life, quick charge and discharge, high power density, and low maintenance cost [1,2,3]. Currently, commercial products are primarily based on porous carbon using ion adsorption, for example, carbon nanotubes [4], mesoporous carbon [5], activated carbon [6], graphitic carbon [7]. Meanwhile, carbon combined with conductive polymers or transition-metal hydroxides/oxides can greatly increase energy storage capacity over a period of time through offering pseudo-capacitance by reversible redox reactions on the interface between the electroactive material and electrolyte [8,9,10]. The inherent capacitance of the carbon component is very important due to the fact that most of the pseudo-capacitive materials have the problems of inferior conductivity, poor cyclic stability, and high cost [11]. Moreover, porous carbon-based SCs electrode materials are known to face sluggish electrode kinetics relating to internal pore ion transport and transfer [12,13,14]. For instance, activated carbon has a long diffusion path (> 5 μm) and a large electrolyte ion-transfer resistance, which result in a large voltage drop (IR) and a small electrolyte ion-approachable specific surface area (SSA) at the large current density [15, 16]. The carbon nanotubes have remarkable conductivity and high electrolyte ion-approachable SSA that can offer ideal rate stability [17]; however, the value of electrical double-layer capacitance is restricted. Therefore, there it is an enormous and urgent challenge to design and develop a new carbonaceous electrode material with high performances for SCs.

The energy storage of carbon-based material is mainly based on the adsorption of ions; meanwhile, the surface charges are equally distributed at the interfaces between the electrode and the electrolyte [18]. Hence, high SSA and abundant pore structures adapted to solute/solvent ion sizes are required for carbon-based SCs electrode materials [19,20,21]. Additionally, incorporation of surface functional groups and heteroatom species (e.g., nitrogen (N), sulfur (S), phosphorus (P) and boron (B)) can modify the surface wettability of carbonaceous materials and provide pseudo-capacitive behavior [22,23,24,25]. Such doping effects can obviously increase the performances of SCs [26, 27].

In recent years, the synthesis of carbonaceous materials derived from biomass for energy applications has become a research focus due to the increasing lack of mineral fuels [28]. Biomass materials, such as foods, agricultural wastes, animal bones, silks, and even natural polymers, have been used as carbon source. For example, activated carbons fabricated through pyrolysis of popcorn have a specific capacitance (Csp) value of 245 F g−1 [29]. High-performance plant leaves-derived carbon showed a Csp value of 400 F g−1 and an energy density (Ecell) value of 55 W h kg−1 in 1 mol L−1 H2SO4 aqueous solution [30]. Raymundo et al. [31] reported the synthesis of microporous carbon materials deriving from natural seaweeds. The obtained carbon materials showed a Csp value of 264 F g−1 in 1 mol L−1 H2SO4 aqueous solution. Compared with usual carbon precursors (e.g., coal, pitch and phenolic resins), the superiorities of non-pollution and low cost facilitate the progress of biochar-based high-performance SCs in practical applications.

Chitosan, as natural amino polysaccharide, recently has been applied extensively in the field of chemical industry and food processing because of its rich resources and favorable biocompatibility. Furthermore, chitosan can be modified chemically, due to the existence of massive amino group (–NH2) and hydroxyl group (-OH) which make it a potential carbonaceous precursor with good electrochemical performances. For example, Lota et al. [32] carbonized chitosan with KOH directly and the as-prepared activated carbon materials possess 295 F g−1 at 0.1 A g−1. At the same time, Zhang et al. [33] fabricated chitosan aerogel with 3D interconnected hierarchical porous structure by crosslinking of chitosan and glutaraldehyde with Csp value 246.5 F g−1 at 0.5 A g−1. Beyond that Ling et al. [34] reported a sheet-shaped chitosan-based carbon gel and the Csp value can be as high as 242 F g−1 at the current density of 0.1 A g−1. Therefore, we expect the renewable and inexpensive chitosan can be a useful carbon precursor to achieve novel carbon-based electrode materials with outstanding capacitive performance.

In current work, the hybrid porous carbon was successfully synthesized through activating the mixture of chitosan and gelatin, which were prepared by a simple solution blending and freeze-drying process. The gelatin contains many kinds of functional groups (such as –OH, –NH2, and –COOH) on the main chains that can directly bring heteroatoms into the carbon skeleton and thereby improve the electrochemical performance of chitosan-based carbon electrode materials. Furthermore, gelatin can not only change the microstructure of chitosan-based carbon materials, but also increase the porosity and SSA of chitosan-based carbon materials as a pore-forming agent. The as-prepared hybrid porous carbon exhibits a novel sheet with tangerine pith-like nanowires evenly distributed on its surface, interconnected porous structure and high SSA of 927.17 m2 g−1. Consequently, it shows excellent energy storage capacity of 331 F g−1 at 1 A g−1, good cycle stability with 90% retention after 10000 cycles at 10 A g−1, and high energy density of 34 W h kg−1 at the power density of 900 W kg−1 in 6 mol L−1 KOH electrolyte. Compared with most chitosan-based carbon materials, the current research work has greatly enhanced the structural characteristics and energy storage performances.

Experimental section

Chemicals

Chitosan (80–95%, degree of deacetylation), gelatin and hydrochloric acid (HCl, 36–38%) used were purchased from Sinopharm Chemical Reagent Co., Ltd. Potassium bicarbonate (KHCO3, ≥ 99.5%) and acetic acid (≥ 99.5%) were supplied by Tainjin Baishi Chemical Industry Co. Ltd. All the chemicals were of analytical reagent and used without further processing.

Synthesis of CHPCs

Chitosan (2.0 g) was initially added into 100 mL 0.3 M acetic acid aqueous solution and magnetically stirred at 50 °C for 0.5 h until completely dissolved. Subsequently, gelatin (0.50 g) was dissolved into the above solution and stirred for 1 h to form a homogenous solution. Then the well-mixed solution was frozen overnight and freeze-dried into a sponge-like composite which was marked as chitosan-derived carbonaceous precursor. The precursor was calcined at 300 °C in Ar flowing for 3 h with a heating rate of 2 °C min−1 and further ground into powder after cooling naturally. The powder (0.20 g) was immersed in 5 mL KHCO3 solution (mpowder:mKOH = 1:1) and dried at 40 °C to form the mixture of powder and KHCO3. Next, the mixture was carbonized at 600 °C with the same heating conditions and washed with 0.5 M HCl solution to remove the impurities. The obtained sample was washed several times with deionized water and ethanol and collected by centrifugation and dried at 40 °C for 6 h. For comparison systematically, the control samples were prepared following the same experiment strategy with different dosages of gelatin (0, 0.25, 0.50, 0.75, and 1.0 g) or without the addition of chitosan. The final products were denoted as CPC, CHPC-0.25, CHPC-0.50, CHPC-0.75, CHPC-1.0 and GPC, respectively.

Materials characterization

The scanning electron microscope (SEM, Zeiss G500) and transmission electron microscope (TEM, JEM 2100) were used to characterize the morphological changes of the samples. The N2 adsorption–desorption measurements were measured by the specific surface aperture analyzer (Geminiv 2380) after vacuum drying for 24 h at 180 °C. The pore-size distribution and the pore parameters were analyzed by original density functional theory (ODFT) and Brunauer–Emmett–Teller (BET), respectively. The X-ray diffraction patterns (XRD) were carried out on the X-ray powder diffractometer (XRD-6100, Cu source) at the scan rate of 5 min−1. And the Raman spectra were evaluated on the LabRAM HR Evolution with solid-state laser of 532 nm and ranged from 600 to 3600 cm−1. The thermos-gravimetric analysis (TGA) was tested on the SDTA851e with the temperature range of 45–800 °C in the atmosphere N2. The X-ray photoelectron spectroscopy (XPS) was analyzed by using ESCALABXi + with the Al/Kα as the monochromatic source.

Electrochemical measurements in 6 M KOH electrolyte

The preparation processes of the working electrode are as follows: the samples, acetylene black and polyvinylidene fluoride (PVDF) were mixed into slurry in the N-methyl-2-pyrrolidone at the weight ratio of 8:1:1, then the mixture was coated on the stainless steel mesh and dried at 45 °C for 10 h (1 cm × 1 cm, about 2.5 mg). All the electrochemical tests, such as galvanostatic charge–discharge (GCD), cyclic voltammetry (CV), and electrical impedance spectroscopy (EIS), were carried out on a CHI660E electrochemical workstation in the 6 mol L−1 KOH electrolyte. The GCD and CV texts varied the constant current from 0.5 to 10 A g−1 and the scan rate from 10 to 100 mV s−1 within the voltage range of − 0.2 to 1.2 V, respectively. The frequency range for the EIS was from 0.01 to 105 Hz with the open-circuit potential.

In the typical three-electrode system (half-cell), platinum plate and a standard mercuric oxide electrode were served as the counter electrode and reference electrode, respectively. The specific capacitance (Csp, F g−1) of the electrode can easily be deduced by GCD curves based on Eq. (1) [35]:

$$ C_{\text{sp}} = I \times \Delta t/\left( {m_{\text{sp}} \times \Delta V} \right) $$
(1)

In the symmetrical two-electrode system (full-cell), two electrodes with the same mass were back to back and separated by the polypropylene membrane. Similarly, the specific capacitance (Ccell, F g−1) was also calculated according to Eq. (2) [36]:

$$ C_{\text{cell}} = I \times \Delta t/\left( {m_{\text{cell}} \times \Delta V} \right) $$
(2)

The energy density (Ecell, W h kg−1) and power density (Pcell, W kg−1) were obtained through Eqs. (3) and (4) [35]:

$$ E_{\text{cell}} = C_{\text{cell}} \times \Delta V^{2} /\left( {2 \times 3.6} \right) $$
(3)
$$ P_{\text{cell}} = \left( {E_{\text{cell}} \times 3600} \right)/\Delta t $$
(4)

where I is the discharge current (A), Δt represents the discharge time (s), ΔV refers to the voltage window (V), and msp and mcell (g) are the mass of active materials on a single electrode and the two electrodes, respectively.

Results and discussion

The synthesis mechanism of CHPC composites

The detailed synthesis schematic of CHPC composites is displayed in Fig. 1. Due to the protonation of amino functional groups and the hydrogen bonding between chitosan molecules and solvents, the hydrogen bonding forces between or within the molecules of chitosan are destroyed, which make the chains stretch sufficiently and dissolve in low concentration of acetic acid aqueous solution. After adding gelatin, the charged carboxyl groups (COO−) of gelatin form strong ionic linkage and produce electrostatic interactions with charged amine groups (NH3+) on the chitosan chains. In addition, the polar groups in gelatin (such as –OH, –COOH and –NH2) interact with the solvents and the functional groups in chitosan (–NH2 and –OH), forming hydrogen bonds. The strong intermolecular forces, including electrostatic attraction and hydrogen bond, result in molecule chains of gelatin and chitosan entangling mutually and an interpenetrating or semi-interpenetrating network structure coming into being. Subsequently, the network structure is preserved by freeze-drying process to form sponge carbon precursor in macro-structure, and a large-scale lamellar structure with tangerine pith-like morphology in microscopic scales, as shown in Fig. S1b and S1f. Moreover, the carbonization process not only retains the special tangerine pith morphology, but also effectively develops abundant pores with the help of KHCO3. Accordingly, KHCO3 can be completely decomposed into K2CO3 at 200 °C and the K2CO3 is further pyrolyzed into K2O, as follows (1)–(2) [33]. Moreover, carbon is consumed by the reaction of carbon and H2O with the emission of H2 [Eq. (3)] [37], and the resulting CO2 can be further reduced by carbon to form CO at high temperature [Eq. (4)] [38]. Importantly, the reaction of carbon atoms and the overflow of large quantities of gases lead to the formation of the abundant micropore structure in the chitosan-derived carbon.

$$ 2{\text{KHCO}}_{3} \to {\text{K}}_{2} {\text{CO}}_{3} + {\text{CO}}_{2} + {\text{H}}_{2} {\text{O}} $$
(5)
$$ {\text{K}}_{2} {\text{CO}}_{3} \to {\text{K}}_{2} {\text{O}} + {\text{CO}}_{2} $$
(6)
$$ {\text{H}}_{2} {\text{O}} + {\text{C}} \to {\text{H}}_{2} + {\text{CO}} $$
(7)
$$ {\text{CO}}_{2} + {\text{C}} \to 2{\text{CO}} $$
(8)

The surface and microscopic characterization of carbon materials with different dosage of gelatin and GPC are analyzed by SEM and TEM images in Fig. 2. Compared with that before carbonization (Fig. S1), the morphologies of CHPC series samples (Fig. 2) remain the large-sized lamellas, while the lamellas are broken into smaller sheets due to the activation of KHCO3 during the carbonization process. In sharp contrast, the morphology and structure of GPC change significantly after carbonization, from crisp lamella with tangerine pith-like morphology (Fig. S1d and h) into porous spongy structure (Fig. 2d and h). The difference in morphology before and after carbonization is mainly because the thermal stability of GPC is so poor that the massive porous structure formed easily after carbonization and activation. Thus, the blending of gelatin and chitosan solution not only alters the content of heteroatoms, but also introduces abundant pores into the lamella of chitosan which can dramatically promote the specific surface area. In Fig. 2a, the surface of the carbon sheet is smooth and flat. Interestingly, with the addition of gelatin, the novel tangerine pith-like nanowires form and well scatter on the surface of sheet, as shown in Fig. S1f and S1 g. It is obvious that the nanowires exist stably and become ever more evident after carbonization, as seen in Fig. 2f and g. Moreover, the distribution of the special tangerine pith-like nanowires in CHPC-0.50 (Fig. 2f) is more uniform than that of CHPC-1.0 (Fig. 2g). The difference in distribution is mainly because the aggregation of molecules caused by excessive gelatin. The TEM images of CHPC-0.50 are displayed in Fig. 2i–k. We can clearly observe the nanosheet structure from Fig. 2i, which is consistent with the SEM images. In Fig. 2j and k with larger magnification, lots of densely interconnected nanopores can be seen mainly caused by the activation of KHCO3. The interconnected nanopores not only enlarge the specific surface area of electrode materials for charge storage, but also greatly shorten the transportable distance of electrolyte ion, causing the improvement of capability performance [19, 39].

Figure 1
figure 1

The illustration for the synthesis of CHPC

Full size image
Figure 2
figure 2

SEM images of CPC (a, e), CHPC-0.50 (b, f), CHPC-1.0 (c, g), and GPC (d, h). TEM images of CHPC-0.50 (i–k)

Full size image

Based on the nanoscale porous characteristics from the SEM and TEM images, the more accurate pore parameters of as-prepared carbon materials are further investigated by N2 adsorption–desorption method and the results are exhibited in Fig. 3 and Table 1. According to the adsorption–desorption isotherm (Fig. 3a), the CHPC series samples have the typical characteristic of type I which has a sharp rise in the adsorption capacity at the low relative pressure (P/P0 < 0.01) and is getting closer to the level at P/P0 > 0.05 [40]. The typical characteristic hints a large number of micropores which can be illustrated by the activation of KHCO3. It is consistent with the result of pore-size distribution (Fig. 3b) which mainly focuses on 0.79–0.89 nm, accompanying significant increasement in the specific surface area of the samples. In the details, the dosage of gelatin also has an obvious influence on the specific surface area and porous structure. Compared with CPC, the specific surface area (SBET) and pore volume (Vpore) of CHPC-0.50 have been greatly increased from 852.12 m2 g−1, 0.36 cm3 g−1 to 927.17 m2 g−1, 0.39 cm3 g−1, mainly in the promotion of micropores. Further, the SBET and Vpore of CHPC-1.0 are further raised to 969.67 m2 g−1, 0.41 cm3 g−1 with the increasement of gelatin dosage. In particular, the rise in number of mesopores is prominent, which leads to the increasement of average pore size with 1.70 nm. Though the average pore size of CHPC-0.50 with 1.67 nm is a little smaller than that of CPC with 1.69 nm and CHPC-1.0 with 1.70 nm, it is larger than that of electrolyte ions (0.36–0.42 nm) [41]. The appropriate average pore size has a positive effect on electrochemical performance and conductivity, especially in the reducing of resistance. The results of N2 adsorption–desorption method confirm the fact that gelatin has been incorporated into composite and effected the porous structures of the samples.

Figure 3
figure 3

N2 adsorption–desorption isotherms (a) and pore-size distributions (b) of the CPC, CHPC-0.50 and CHPC-1.0 samples

Full size image
Table 1 Pore parameters of the CPC, CHPC-0.50 and CHPC-1.0
Full size table

In order to explore the changes of structure characteristics caused by gelatin, the XRD and Raman spectroscopy is adopted and the results are shown in Fig. 4. From all the XRD curves (Fig. 4a), the broad characteristic diffraction peaks located at around 2θ = 24°–25° can be clearly observed, corresponding to the amorphous structure [42]. With the increment of gelatin dosage, the diffraction peaks are flattened, suggesting the presence of high-density micropores [43]. The changes of disordered degree are also confirmed by the Raman spectra, as shown in Fig. 4b. Two characteristic peaks represent the defects in atomic lattices (D-band) and crystallized graphitic carbon (G-band), respectively [44, 45], and the position and the area integral are listed in Table S1. The 2D band centers around 2874 cm−1 are the second order of the D-band, which is the response to the number of graphene layers [46]. As shown in Table S1, with the increase in gelatin dosage, the D and G peaks slightly move toward the shorter wavelength, indicating the increase in disorder degree caused by heteroatom doping. Furthermore, the value of ID/IG (the ratio between the area of the D-band and G-band) is employed to explain the disordered degree of the samples [47]. The CPC has the lowest ID/IG value, meaning the greatest the degree of graphitization among these samples because no gelatin is added. Thus, the ID/IG value is sensitive to the porous structures and the dosage of gelatin. And the values are increased regularly from 3.41 to 3.49 along with the improvement of gelatin content, which is related to more defects generated by the gelatin, and this trend is in good agreement with the analyses of the XRD. These analyses show that the gelatin improves the defects by tuning the porous structures.

Figure 4
figure 4

XRD patterns (a) and Raman spectra (b) of the CHPC-1.0, CHPC-0.75, CHPC-0.50, CHPC-0.25 and CPC

Full size image

The changes of element content and bonding configuration caused by the quantity of gelatin are systematically analyzed by XPS measurements, which are shown in Fig. 5. Obviously, three characteristic peaks are observed from the full-scale spectra (Fig. 5a) of the samples, which correspond to C1s (~ 286.1 eV), N1s (~ 400.8 eV) and O1s (~ 530.5 eV) [48]. According to Table 2, the content of nitrogen increases gradually from 5.6 to 5.93 at.% with the addition of gelatin, and the content of oxygen is reduced. And it is noteworthy that the content of heteroatoms (nitrogen and oxygen) of CHPC-0.50 is the highest among the samples. And the existence of heteroatoms helps to improve effectively the specific capacitance by introducing pseudo-capacitance [49]. The high-resolution O1s spectrums (Fig. 5b and Fig. S2a, 2d) display that three peaks at 531.5, 532.4, and 533.4 eV are associated with C=O, C–O, and O=C–O groups [50], and these three groups also can be found in the corresponding C1s spectrum (Fig. S2c) which is decomposed into four peaks corresponding to C–OH (285.3 eV), C=O (286.5 eV), O–C=O (288.8 eV) and C–C (284.7 eV) [51]. The oxygen-containing functional group can increase the wetting capability of carbon electrode in electrolyte and benefit for the ion-penetration diffusion process, fortifying the electrochemical performance [52]. And the N1 s (Fig. 5c and Fig. S2b, 2e) reveals three distinct types of nitrogen, including pyridinic-N (398.5 eV), pyrrolic-N (400.3 eV) and graphitic-N (401.1 eV) [53]. The pyridinic-N and pyrrolic-N impel the carbon electrode to obtain more electrochemical active sites, which contributes to the enhancement of wettability, and the graphitic-N can make positive contribution to the improvement of conductive ability and thermal stability [53, 54]. Deeply, the relative nitrogen contents of each type are shown in Fig. 5d, which are calculated by means of the area of these peaks. Accordingly, the CHPC-0.50 exhibits the high content of graphitic-N (46.32%) which is much higher than that of CPC (31.89%) and CHPC-1.0 (33.09%), while the content of pyridinic-N and pyrrolic-N is slightly lower. With the suitable gelatin dosage, partial pyridinic-N and pyrrolic-N were converted to graphitic-N, leading to the maximum content of graphitic-N in CHPC-0.50. Further increasing the amount of gelatin, the more pyridinic-N and pyrrolic-N come into being that induces the decrease of proportion of the graphitic-N. Thus, graphitic-N helps CHPC-0.50 show great advantages in conductivity and thermal stability and these results can also be verified by TGA curves (Fig. S2f).

Figure 5
figure 5

Full-scale XPS spectra (a) of the samples, O1s (b) and N1s (c) peaks of CHPC-0.50, and relative nitrogen contents (d) in the samples

Full size image
Table 2 Contents (at.%) of carbon, nitrogen and oxygen in samples obtained from XPS
Full size table

Electrochemical performance

The effects of the quantity of gelatin on electrochemical properties are shown in Fig. 6 and Fig. S3. The CV curves (Fig. 6a and Fig. S3a) of the samples show approximate rectangular shape, indicating that electrical double-layer capacitor is predominantly contributed to capacitance. However, wider redox peaks at around − 0.5 to − 0.9 V on the CV curves cannot be ignored, which is related to the pseudo-capacitance provided by heteroatom doping and functional groups [55]. More obviously, the area encircled by the CV curve of CHPC-0.50 reaches the maximum, corresponding to the best capacitance, which is also exemplified by the GCD curve with the longest charge–discharge time (Fig. 6b) and the capacitance performances curves at different current densities (Fig. 6d). The GCD curves (Fig. 6b and Fig. S3b) of all the samples show symmetrical triangular shapes at the current density of 1 A g−1, suggesting good reversible charge–discharge characteristics [56]. Furthermore, the value of specific capacitance can reflect electrochemical performance effectively and visually, which is calculated according to the GCD curves. As displayed in Fig. 6d and Fig. S3d, the specific capacitance of CHPC-0.50 is up to 331 F g−1 at 1 A g−1, which is significantly better than that of CPC (244 F g−1), CHPC-0.25 (301 F g−1), CHPC-0.75 (319 F g−1) and CHPC-1.0 (309 F g−1). And the specific capacitance of CHPC-0.50 can still remain 241 F g−1 even at 10 A g−1, and the capacitance retention can reach up to 73% which is superior to other as-prepared samples, as shown in Tab. S2. The improvement of rate stability is mostly because the layer and porous structure with appropriate pore size provide convenience for the fast and efficient electrolyte ion transportation [57]. The resistances of ion transportation and diffusion are further studied by the EIS test. And the equivalent circuit (Fig. 6c, insert) is used to analyze the EIS curves, which consists of the internal resistance (Rs) of electrolyte and electrode materials, the charge-transfer resistance (Rct) which occurs in the interface of electrode and electrolyte [58, 59], the additional impedance (Zw) caused by concentration and electrochemical polarization at the interface, and the capacitance (C) formed on the surface of electrode [60]. As seen in Fig. 6c and Fig. S3c, a half semicircle at high frequency, corresponding to the Rct, and near vertical linear curve at low frequency are observed. The intercept at the real axis represents the Rs which has little difference for all samples [61]. By contrast, the curve of CHPC-0.50 has smallest semicircle diameter and largest gradient, revealing the optimal capacitive behavior. And the analyses prove again that the layer morphology with hybrid porous structure can provide more storage space and shorten the distance of ion diffusion, and the existence of more heteroatoms can improve the wettability of the electrode materials in electrolyte and introduce the pseudo-capacitance, which have a positive impact on the improvement of electrochemical behavior [62].

Figure 6
figure 6

Electrochemical capacitive behaviors of the samples. The CV curves (a) at a scan rate of 10 mV s−1, the GCD curves (b) at a current density of 1 A g−1, the Nyquist plots (c), and the dependence of capacitance as a function of current density (d)

Full size image

The detailed electrochemical performances of CHPC-0.50 are shown in Fig. 7. The shapes of the CV curves (Fig. 7a) transform to spindle with the increasing scan rate, caused by the pseudo-capacitance and the internal resistance which makes it impossible to perform the charge–discharge process in a short time [63]. As shown in Fig. 7b, all the GCD curves at different current densities are almost symmetric, confirming the wonderful capacitive reversibility [64]. The long-term electrochemical stability of CHPC-0.50 is estimated by GCD curves at 10 A g−1 (Fig. 7c). After 10000 cycles, the capacity still remains 90%, demonstrating the excellent cycle stability and the small reduction in capacitance are mainly due to the reduction of diffusion time for ions into the electrode surface. The Ragone plot (Fig. 7d) shows that energy density is about 34 W h kg−1 (with the power density of 900 W kg−1) at 0.5 A g−1, whereas it remains as high as 9 W h kg−1 (with the power density of 18000 W kg−1) at 10 A g−1 for CHPC-0.50, much superior to the most chitosan-based electrode materials [65,66,67,68] and the commercial devices (3–5 W h kg−1) [69].

Figure 7
figure 7

Electrochemical characterization of the CHPC-0.50 in 6 mol L−1 KOH aqueous electrolyte. a The CV curves at various scan rates, b GCD curves with different current densities, c cyclic stability at a current density of 10 A g−1 over 10000 cycles, d the Ragone plot

Full size image

Conclusions

In summary, we have successfully fabricated a novel hybrid porous carbon sheets from chitosan as carbon source and gelatin as nitrogen source through facile solution blending. The introduction of gelatin develops a novel tangerine pith-like structure distributed on the surface of carbon sheets and produces abundant interconnected micropores with the help of KHCO3, which greatly increase the specific surface area of the as-prepared carbon materials. Based on high heteroatom content (5.75 at.% of nitrogen and 15.04 at.% of oxygen), specific surface area (927.17 m2 g−1) and appropriate average pore size (1.67 nm), the CHPC-0.5 exhibits excellent electrochemical performances with ideal capacitive performance (331 F g−1 at 1 A g−1), excellent rate capability (241 F g−1 at 10 A g−1) and cycling stability (remains 90% after 10000 cycles). In addition, the energy density can reach 34 W h kg−1 when the power density is 900 W kg−1. The wonderful performances of the obtained carbon materials predict that using renewable and environmentally friendly bio-derived carbon as precursor to produce electrode materials is a potential strategy for the application in energy storage systems.