Nitrogen doped activated carbon derived from orange peel for supercapacitor application

OP was collected from Ambrosia Apiaries Pvt. Ltd located in Mohamadpur, Delhi, India. Melamine, ${\rm KOH}$ and ${\rm HCl}$ were acquired from Sigma Aldrich. All solutions were prepared using DI water.

Briefly, OP was torn into small pieces, subsequently washed with DI water and dried in an oven at $98\,{}^\circ {\rm C}$ for 12 h. The dried OP was then crumpled to fine powder. Obtained powder form and melamine were mixed in a weight ratio of 1:1 and thereafter the mixture was carbonized at $700\,{}^\circ {\rm C}$ for 2 h under inert atmosphere. The resulting product, subsequent to carbonization was marked nitrogen-doped carbon (NC). The ramp rate selected for increase in temperature was $5\,{}^\circ {\rm C}\cdot {\rm mi}{{{\rm n}}^{-1}}$. The synthesized NC was mixed with KOH in a mass ratio of 1:1 and thereafter subjected to heat in tube furnace at $700\,{}^\circ {\rm C}$ for 2 h, under inert atmosphere. The ramp rate selected for increase and decrease in temperature was $5\,{}^\circ {\rm C}\cdot {\rm mi}{{{\rm n}}^{-1}}$. As soon as the temperature of tube furnace matched room temperature, synthesized sample was removed from tube and washed with 10% HCl, followed by repeated washing with warm DI water, until filtrate shows neutral pH. The resultant product was marked as NAC-1. To evaluate the effect of activating agent, synthesized NC was mixed with KOH in varying mass ratios of 1:2 and 1:3. The resultant mixture was annealed and washed using the same process and resultant samples were marked as NAC-2 and NAC-3.

To study the pore structure parameters of the synthesized samples, surface area and pore size analyzer (GEMINI V, Micromeritics, USA) was employed. The content of nitrogen in NAC samples was investigated using elemental analyzer (Vario Micro Cube, Germany). The surface morphology of sample was studied employing scanning electron microscopy (FESEM, MIRA3 TESCAN). For determining crystal structure, synthesized sample was scanned in the range of $5{}^\circ$ to $60{}^\circ$ at 0.02 steps employing x-ray diffractometer (XRD, Bruker D8, USA). Raman spectrophotomter (Horiba Yvon) having wavelength of 488 nm was used for Raman measurements.

For preparation of working electrodes, active material NAC, conducting additive acetylene black (10%) and binder PVdf-HFP (5%) were mixed in the mass ratio of 85:10:5 and the resultant mixture was layered on flexible graphite sheets. This was followed by drying the graphite sheets at $100\,{}^\circ {\rm C}$ (in vacuum oven), overnight. The average value of mass loading was 1.6 mg. Aqueous solution of 6 M ${\rm KOH}$ was used as electrolyte. Symmetric supercapacitor cells were fabricated using two identical electrodes separated by electrolyte soaked filter paper. The performance of fabricated capacitor cells was studied using the techniques of electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD). Employing CH Instruments (CHI-600E, USA), CV and EIS measurements were performed. CV measurements were done for varying potential range as well as scan rates while EIS measurements were carried out in the frequency range of 10 mHz to 100 kHz; at an amplitude of 10 mV. Employing charge-discharge analyzer (BT-2000, Arbin Instruments, USA), GCD measurements were taken for different amplitude as well as different current density. The specific capacitance value of the fabricated capacitor cell for each measurement method was evaluated using the given formulas: $C={\left(2\times {{I}_{{\rm av}}} \right)}/{\left(m \times s \right)}\;$ for CV, $C={2}/{\left(\omega \times {{Z}^{\prime\prime }}\times m \right)}\;$ for EIS and $C={\left(2\times I\times \Delta t \right)}/{\left(\Delta V\times m \right)}\;$ for galvanostatic charge-discharge techniques. Here, ${{I}_{{\rm av}}}$, $s$, $m$, $\omega$, ${{Z}^{\prime\prime }}$, $I$, $\Delta t$, and $\Delta V$ denote the average current, scan rate, active mass in single electrode, angular frequency, imaginary impedance at 10 mHz, applied current, discharge time and discharge voltage, respectively.

The surface area and pore structure parameters of various NAC samples were evaluated from ${{{\rm N}}_{2}}$ adsorption-desorption isotherm at 77 K. As evident from figure 1(a), the nature of the isotherm for all the samples, is similar to the type I curve, as prescribed by IUPAC. This indicates the microporous nature of the prepared samples. Values of porous parameters of synthesized NAC samples (such as average pore size, total pore volume and BET surface area) are recorded in table 1.

Table 1. Values of pore structure parameters and nitrogen content in synthesized NAC samples.

Composition (NC: KOH)	BET surface area ${\rm (}{{{\rm m}}^{{\rm 2}}}\cdot {{{\rm g}}^{-1}})$	Micropore surface area ${\rm (}{{{\rm m}}^{{\rm 2}}}\cdot {{{\rm g}}^{-1}})$	Mesopore surface area ${\rm (}{{{\rm m}}^{{\rm 2}}}\cdot {{{\rm g}}^{-1}})$	Micropore volume ${\rm (c}{{{\rm m}}^{3}}\cdot {{{\rm g}}^{-1}})$	Total pore volume ${\rm (c}{{{\rm m}}^{3}}\cdot {{{\rm g}}^{-1}})$	Mesopore volume ${\rm (c}{{{\rm m}}^{3}}\cdot {{{\rm g}}^{-1}})$	Average pore size (nm)	Nitrogen content (%)
NAC-1	1266.11	516.68	749.43	0.2679	0.6649	0.3970	2.10	3.92
NAC-2	1577.27	950.65	626.62	0.4867	0.8714	0.3847	2.07	3.11
NAC-3	1351.85	643.97	707.88	0.3326	0.7052	0.3726	2.09	2.74

Zoom In Zoom Out Reset image size

From the observed magnitudes of pore structure parameters, it was found that prepared NAC-2 sample offers high BET surface area and total pore volume. As large surface area can provide numerous sites for storage of electrolyte ions, such good pore structure makes synthesized sample attractive candidate for energy storage applications. It may be noted from table 1, that with the increase in impregnation ratio of ${\rm KOH}$ to NC from 1 to 2, increment in values of pore structure parameters is observed.

However, with further enhancement (i.e. from 2 to 3), the decrease in such parameters was observed. This clearly shows the dependence of porosity on impregnation ratio. Figure 1(b) shows the pore size variation among various NAC samples, calculated using BJH(Barrett-Joyner-Halenda) theory. It was observed that all NAC samples show pore size distribution in the range of 2  −  5 nm which is in accordance with the results, observed using sorption isotherm. The nitrogen content present in the synthesized NAC samples was evaluated from elemental analyzer. The percentage of the nitrogen present in the samples is listed in table 1. From the observed values, one may clearly note the dependence of nitrogen doping on impregnation ratio. Nitrogen doping in the synthesized sample decreases, with the increase in the ratio of activating agent. This is accordance with the reported literature [37]. Since, the sample NAC-2 offers the highest values of specific surface area and pore volume, and moderate nitrogen doping in comparison to the samples NAC-1 and NAC-3. Hence, further detailed study on its surface morphology, crystal structure and vibrational response, and capacitance properties was carried out.

Figure 2 presents the SEM images of carbon precursor and NAC recorded at varying magnification. The observed images of the raw material indicate crumpled sheet like structure with thin strips. In contrast, the images for NAC show the presence of porosity in the synthesized material. The observed image shows that the material has skeleton structure with various cavitations. Such morphology (i.e. presence of cavities) is significant, in relation to energy storage. The cavities can be helpful in storage of electrolyte ions.

Zoom In Zoom Out Reset image size

XRD pattern of NAC-2 is shown in figure 3(a). The observed pattern shows a couple of characteristic peaks centered nearly at $23{}^\circ$ and $43.1{}^\circ$. The peaks can be ascribed to (002) and (100) planes of graphitic carbon and reveal the non-crystalline nature of the sample NAC-2. Figure 3(b) presents Raman spectra for NAC-2. Two characteristic peaks near $1351\,{\rm c}{{{\rm m}}^{-1}}$ and $1594\,{\rm c}{{{\rm m}}^{-1}}$ are seen in the spectra. These peaks are popularly referred to as D-band and G-band. D-band is associated with disordered carbon structure while G-band refers to ordered lattice of carbon material. The D band indicates the suitability for charge storage, while the G band is responsible for good electrical conductivity. The calculated intensity of D-band to G-band (${{{I}_{D}}}/{{{I}_{G}}}\;$) ratio was found to be 0.91. This indicates the formation of structural defect graphitized carbon. The observed results are in consonance with conclusions obtained using XRD.

Zoom In Zoom Out Reset image size

As discussed in section 2.4, the two-electrode supercapacitive performance of the synthesized NAC-2 was tested in 6 M KOH aqueous solution using standard electrochemical avenues of analysis. For evaluating the operating potential of the capacitor cell, cyclic voltammetry plots of fabricated capacitor cell in varying potential range $(0-1.0\,{\rm V)}$ at a constant scan rate of ${\rm 10}\,{\rm mV}\cdot {{{\rm s}}^{-1}}$ were recorded.

They are depicted in figure 4(a). It was observed that the trend of CV curves exhibits almost rectangular behavior up to 0.8 V, indicating that the device can be safely operated till 0.8 V. However, for potential exceeding 0.8 V, deviation from rectangular shape was observed for CV curve. This limited operating potential of cell is primarily due to limited ESW(electrochemical stability window) of aqueous electrolyte. The functioning of the capacitor cell was assessed in the operating potential range $(0-0.8\,{\rm V)}$ with increasing scan rate (as shown in figure 4(b)). The shape of the CV plots remained almost rectangular, even for high scan rates of ${\rm 500}\,{\rm mV}\cdot {{{\rm s}}^{-1}}$, indicating appreciable rate capability of the cell. This high rate capability is an indication of good electrical double layer capacitor behavior and excellent ions responsitivity. The values of specific capacitance $({{C}_{sp}})$ of cell were evaluated by varying scan rates and are shown in figure 4(c). It has been observed from the calculated values that as-synthesized NAC sample offers high specific capacitance of $164.8\,{\rm F}\cdot {{{\rm g}}^{-1}}$ at ${\rm 10}\,{\rm mV}\cdot {{{\rm s}}^{-1}}$, which is more in comparison to the results found in literature [21]. This high capacitance value is due to high surface area as well as nitrogen doping. These features of NAC not only enhance its conductivity and ion diffusion capacity but also helpful in achieving fast ion transfer pathway and pseudocapacitance [38]. It was also observed that with the increase in scan rate, there was decrement in the value of ${{C}_{sp}}$. This common observation has been previously marked and is primarily due to limited access of active electrode sites by electrolyte ions, with increase in scan rate [12].

Zoom In Zoom Out Reset image size

Electrochemical behavior of fabricator capacitor cell and role of electrolyte was assessed using EIS techniques. Figure 5(a) depicts the Nyquist plot for the capacitor cell. On the basis of frequency variation, the observed plot can be divided into three regions, namely: low, mid and high. In low frequency region, fabricated cell exhibits almost similar steep rise (almost approaching $90{}^\circ$). This is an indication of good capacitive behavior of the cell, which is due effective contact between electrode and electrolyte. Expanded view of the plot in high frequency region has been depicted as inset. Crucial electrical characteristics of the cells related with electrolyte as well as electrode-electrolyte interface were investigated in this region and are given in table 2.

Table 2. Various electrochemical parameters of the capacitor cell evaluated from impedance analysis.

${{R}_{b}}$ $(\Omega \cdot {\rm c}{{{\rm m}}^{2}})$	${{R}_{ct}}$ $(\Omega \cdot {\rm c}{{{\rm m}}^{2}})$	${{R}^{\ast }}$ $(\Omega \cdot {\rm c}{{{\rm m}}^{2}})$	${{C}_{sp}}$ ${\rm (F}\cdot {{{\rm g}}^{-1}})$	Response frequency, ${{f}_{0}}({\rm Hz})$	Response time, ${{\tau }_{0}}({\rm s})$	Available specific energy, ${{E}_{0}}$ ${\rm (Wh}\cdot {\rm k}{{{\rm g}}^{-1}})$	Pulse power, ${{P}_{0}}$ ${\rm (kW}\cdot {\rm k}{{{\rm g}}^{-1}})$
1.53	1.87	19.26	149.62	0.51	1.96	4.59	8.43

Zoom In Zoom Out Reset image size

Low magnitude of ${{R}_{b}}$ and ${{R}_{ct}}$ shows the proper formation of electrode-electrolyte contact, and the diminished resistive route followed by electrolyte ions. The magnitude of specific capacitance $({{C}_{sp}})$ calculated 10 mHz frequency, was found to be in agreement with CV results.

The values of complex capacitance (${{C}^{\prime }}$ and ${{C}^{\prime\prime }}$) were found using the following equation:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {{C}^{\prime }}(\omega)=\frac{-{{Z}^{\prime\prime }}(\omega)}{\omega {{\left| Z(\omega) \right|}^{2}}},\nonumber \end{align} \tag{ 1 }$$

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {{C}^{\prime\prime }}(\omega)=\frac{{{Z}^{\prime }}(\omega)}{\omega {{\left| Z(\omega) \right|}^{2}}},\nonumber \end{align} \tag{ 2 }$$

where, ${{Z}^{\prime }}$ and ${{Z}^{\prime\prime }}$ are respectively real and imaginary part of impedance and $\omega$ is angular frequency.

Figure 5(b) shows the complex capacitance (${{C}^{\prime }}$ and ${{C}^{\prime\prime }}$) plots varying with logarithm of frequency. As frequency increases, decrement in value of ${{C}^{\prime }}(\omega)$ was observed. At very high frequency, supercapacitor shows the characteristic of pure resistor. Many parameters (such as the nature of the electrolyte used and electrode thickness) are responsible for this transition, in the value of ${{C}^{\prime }}(\omega)$. In the context of energy dissipation, ${{C}^{\prime\prime }}(\omega)$ changed with frequency and then reached its maximum value. It may be inferred from the plots that the magnitude of ${{C}^{\prime }}(\omega)$ become greater with decrease in frequency, whereas the value move towards higher frequency in the case of ${{C}^{\prime\prime }}(\omega)$. This indicates good capacitive behavior of the capacitor cell. The point at which the plots of real and imaginary parts of capacitance (i.e. ${{C}^{\prime }}(\omega)$ and ${{C}^{\prime\prime }}(\omega)$) cross each other, provides the value of response frequency $({{f}_{0}})$ [39]. The characteristic response time $({{\tau }_{0}})$ is inverse of response frequency and determines the rate capability of device. Employing characteristic response time, the pulse power density for capacitor cell was evaluated employing the given expression: ${{P}_{0}}={{{E}_{0}}}/{{{\tau }_{0}}}\;$ . Here, ${{E}_{0}}$ is the specific energy calculated at ${{f}_{0}}$ [40]. The calculated values of ${{f}_{0}}, {{\tau }_{0}}, {{E}_{0}}$ and ${{P}_{0}}$ for the cell are given in table 2. The observed values of pulse power density are high, indicating good rate capability of cells. Presence of good rate capability has also been seen during CV studies. Figure 5(c) presents the Bode phase angle plots for the capacitor cell. At low frequency, the capacitor cell phase angle is $\sim 83{}^\circ$ which is near $90{}^\circ$ (characteristic of idealized capacitor). This indicates capacitive behavior of cell.

As the measurements from CV and EIS are sometimes unreliable. Hence, for more accurate capacitive measurements, galvanostatic charge-discharge curves were noted, at room temperature, for varying values of current density. Figure 6(a) shows the typical charge-discharge curves for the capacitor cell, with varying voltage amplitude. Standard capacitive (nearly triangular shaped) charge-discharge characteristics for the cell indicate its capacitive nature. Subsequent to an early ohmic drop, nearly linear discharge patterns were observed. Performance of the cell was also evaluated at varying values of current density. It was noted that usual capacitive charge-discharge pattern were preserved, even when current density was large, indicating good rate capability of cell. Using the linear discharge curve, equivalent series resistance (ESR) and specific discharge capacitance $({{C}_{d}})$ values for the cells were calculated and are recorded in table 3.

Table 3. Charge-discharge characteristics of the capacitor cell at the current load of ${\rm 0}{\rm .74}\,{\rm A}\cdot {{{\rm g}}^{-1}}$.

ESR $(\Omega \cdot {\rm c}{{{\rm m}}^{2}})$	Specific capacitance ${\rm (F}\cdot {{{\rm g}}^{-1}})$	Specific energy ${\rm (Wh}\cdot {\rm k}{{{\rm g}}^{-1}})$	Specific power ${\rm (kW}\cdot {\rm k}{{{\rm g}}^{-1}})$
9.75	166.83	5.79	9.50

Zoom In Zoom Out Reset image size

The noted values of ${{C}_{d}}$ are in consonance with those found from CV and EIS results. It was noted that even with increase in current density, capacitor cell possess sufficient capacitance, showing good rate capability of cell. CV and EIS studies, discussed earlier also showed good rate capability. The magnitude of specific energy $(E)$ and specific power $(P)$ for the cell were calculated by the equations: $E={1}/{8}\;{{C}_{d}}{{V}^{2}}$ and $P={{{V}^{2}}}/{(8m\times {\rm ESR})}\;$ . The noted values are recorded in table 3. Figure 6(d) shows the comparative specific energy variation with specific power (i.e. Ragone plots) for the capacitor cell. The observed shape of Ragone plot is quite similar to the shape reported in literature [10, 12] and the values are much higher than previously reported values [21]. These characteristic of NAC render it a potential possibility to be employed as an efficient electrode for storing energy.