Morphology and structural characterization
The morphology of the three activated carbon materials was analyzed by FESEM/EDS, as stated earlier, and the results are shown in Fig. 1. The AC samples exhibited irregular sheets with apparent porosities. This implies that the morphology could be from the volatile organics and swell expansion, as Xie et al. [10] and Wang et al [5] reported. The three-carbon samples exhibited rough surfaces with excellent porosity creating a perfect interface for diffusion and intercalation of the electrolytic ions during the charge-discharge process of the EDLC electrodes. The visible surface porous structure of the samples varied with the varying temperature. ANN6 (Fig. 1a) exhibited huge irregular pores similar to a honeycomb structure comparable to ANN7, as shown in Fig. 1b. However, the ANN8 (Fig. 1c) indicates a densely packed structure than the ANN6 and ANN7, which we refer to as better in-depth porosity.
The EDS (Energy dispersive spectroscopy) and the elemental mapping of all three (3) samples confirm the presence of carbon materials with 75–99%wt in the Activation with KOH. This creates high porosity with a high surface area due to redox reactions and the evolution of volatile organics as gases, increasing the porosity [20]. The EDS of ANN6 shown 99.48% carbon with some silica and sulphur as other elements. The ANN7 exhibited 89.47% carbon with silica and calcium as the other elements. ANN8 sample exhibited 75.22% carbon, as shown in Table 1 and the spectra in Fig. 1. with increasing temperature, the grains agglomerate and grow, the number of the open pores reduces, and the open pores locate at multi-grain boundaries. This degrades the carbonous materials as it further loses the lower weight carbon molecules at higher temperatures and creates increase in the ash contents which may be detected as silica content. The higher value of silica in ANN8 is due to increased activation temperature, leading to the formation of silica groups. The silica functional groups also improve the wettability between the electrode carbon surface and electrolytic ions by forming polar functional groups. When the surface has more functional groups concentrations, it increases the interaction between activated carbon surfaces and water in the electrolyte.
Table 1
EDS elemental composition for ANN6, ANN7, and ANN8 AC samples
Sample
|
C
|
Si
|
S
|
Ca
|
ANN6
|
99.48
|
0.48
|
0.04
|
-
|
ANN7
|
89.47
|
7.91
|
-
|
2.62
|
ANN8
|
75.22
|
18.43
|
-
|
6.35
|
The study carried out XRD analysis of the activated carbon was to confirm the phase composition, as shown in Fig. 2a. The spectra exhibited typical firm, broad peaks at 2θ = 23.3o and 43.1o, corresponding to (002) and (100) lattice planes for all the samples, respectively [12, 21–23]. The peak formation at 2θ = 23.2o with planes of (002) indicates a graphitic structure and amorphous morphological phase in the derived activated carbon materials [24]. The degree of graphitization is closely related to the conductivity of the carbon materials [25]. The JCPDS card no. 41-1487 show the profile of ANN6 and ANN7 at 2θ = 23 and 40–43o with a d-spacing of 4.435–4.456 Å and 2.214–2.228 Å, respectively. The shift in the 2θ value of ANN8 towards large angles at high-temperature sample leads to the change in the d-spacing at 2θ = 26.4o and 43.09o to 3.380 Å and 2.108 Å, respectively. This is also visible in their morphology, where ANN8 is different from the other two. In all three samples, the d -spacing at 2θ = 23o is greater than 3.35 Å (for graphite), indicating hard carbon formation [1]. The decrease in d-spacing was causing a reduction in non-crystallinity, as shown in Fig. 1 between ANN6, ANN7, and ANN8 morphology variation [8].
The FTIR spectra of derived activated carbon are shown in Fig. 2b. The fingerprint region (1500–600) indicated three prominent firm peaks of C-OH, C-O-C, C-H, C-N, and C-O for the skeleton of the materials. Also, the diagnostic region (4000–1500 cm-1) exhibited the primary vibration at 1737 cm-1 for the three samples as the most substantial peak. This can be attributed to the frequency vibration of the C = O stretching in the structure of lactone, carboxyl, or quinone groups stretching in aromatic which are more intense in the activated carbon. Other characteristic peaks present include sp3(C-H), sp2(C-H) stretching vibration (2990 to 3050 cm-1), and O-H/sp stretching vibration (3450 cm-1). The vibration stretching between 2990–3050 cm-1 is attributed to the C-H stretch and out of plane bending vibration. The fingerprint region exhibited bending vibration of C-H or C-N, which indicates the reaction of dehydrogenation. This is accelerated with the increase in temperature of Activation. The functional groups' information and frequency vibrations are consistent with the literature [3, 13, 26].
The composition and functional groups of the samples were further examined by the XPS technique as shown in Fig. 3 and Figure S1 (supplementary information) with their binding energy and assignment of C1s, O1s, and Si2p. The samples exhibited no significant difference in their positions and shapes for different temperature treatments. The deconvolution of the C1s for sample ANN8 in Fig. 3, ANN7 and ANN6 spectra Fig. 1S (Supplementary information) was observed in five-carbon states as follows; C-C and sp2 at 284 eV, C-O, C-N at 285.5 eV, C-C at 287.8 eV, O-C = O, C = O at 289 eV. The Si2p exhibited one peak at 102 eV, which indicates Si-O and the O1s shown two oxygen states of C = O at 532 eV and C-O at 534 eV. The spectra states are consistent with the other literature as reported [4, 11, 27, 28]. The denoted functional groups of O = C, C-O-H, C-O-C, among others, are typical structures of phenolic, ester, quinine, lactonic, carboxyl group from the surface of the materials. Table 2 shows the atomic percentages from XPS analyses. This depicted a considerable decrease in carbon and oxygen percentage, attributed to an increase in temperature that caused an increase in the silica percentage at a higher temperature of Activation [9, 29].
Table 2
Atomic percentages as obtained from XPS analysis
Sample
|
Atomic concentration (%)
|
C 1s
|
O 1s
|
Si 2p
|
ANN8
|
51.86
|
37.68
|
10.46
|
ANN7
|
75.66
|
24.08
|
5.26
|
ANN6
|
83.46
|
14.48
|
2.05
|
The thermogravimetric analysis (TGA/DSC) was carried out to confirm the degradation and purity of the ANN6, ANN7, and ANN8 samples materials. The study collected TGA/DSC data by heating the three samples to 1000oC under a pure airflow as shown in Fig. 4a and the TGA data in Table 3.
Table 3
TGA/DSC material decomposition of the raw material and activated carbon samples
Decomposition stages
|
Mass fraction (%) lost at different stages
|
Raw material
|
ANN6
|
ANN7
|
ANN8
|
I
|
10.4
|
2.1
|
2.5
|
2.5
|
II
|
63
|
10.9
|
23.7
|
26.1
|
III
|
|
16.4
|
-
|
-
|
At 1000 oC
|
15.3
|
67.1
|
49.3
|
43.8
|
The TGA data collected on the raw material Zea mays cobs shown in Fig. 4a exhibited two distinct decomposition stages; the 1st stage is at 210 to 260 oC, and the 2nd stage is between 350 to 370 oC. The heat flow through the raw material exhibited exothermal and endothermal flow. The first endothermic phase at 100 oC was due to moisture in the sample. The second endothermic demonstrated loss of water of crystallization at 360 oC, after which the heat flow was uniform throughout up to 1000 oC, with the highest exothermal phase at 352 oC with 0.70 mW/mg. The material decomposition was up to 85% mass loss at 1000 oC. ANN6 sample material (Fig. 4b) exhibited three decomposition stages; the 1st and 2nd between 45 to 114 oC, the third stage at 286 to 298 oC and the mass loss as shown in Table 3. The heat flow through ANN6 exhibited endothermic and exothermic flow (Fig. 4b). The ANN7 and ANN8 sample materials exhibited two distinct similar decomposition stages. These are consistent with that of ANN6 for the first and second stages. Stage one and stage two decomposition may be due to organic volatile materials and moisture trapped after activation processes. The heat flow through the three materials exhibited the same trend with endothermic at 106 oC and peak exothermic at around 351 oC and 0.91 mW/mg.
Table 4
The pore characteristics of ANN6, ANN7, and ANN8 activated carbon samples
Sample
|
SBET (m2/g)
|
Smic (m2/g)
|
Smic/SBET (%)
|
Sext (m2/g)
|
Vmicro (cm3/g)
|
Vmeso (cm3/g)
|
Vtotal (cm3/g)
|
Vmic/Vtot (%)
|
ANN6
|
461.36
|
385.66
|
83.6
|
75.71
|
0.1889
|
0.05749
|
0.2464
|
76.7
|
ANN7
|
972.65
|
689.28
|
70.87
|
460.76
|
0.3628
|
0.1682
|
0.5310
|
68.3
|
ANN8
|
1443.94
|
1000.89
|
69.30
|
383.05
|
0.5167
|
0.27485
|
0.7915
|
65.3
|
The BET technique determined the porosity and the specific surface area (SSA) as shown in Fig. 5 of the Zea mays derived activated carbon. The N2 adsorption-desorption isotherm was performed to calculate the pore volume and specific surface area, as shown in Table 4 for the three samples. The isotherms of the samples show type I isotherm with micropores, and the hysteresis loop is mainly caused by the mesoporous structure [10]. The hysteresis loop at a relative pressure from 0.4 to 0.9 P/Po in the adsorption-desorption isotherms mainly because of the presence of mesopores with the formation of interparticle condensation of N2 at > 0.9P/Po. The calculated SSA for ANN8 exhibited the highest value of 1443.94 m2/g with a total pore volume of 0.7915 cm3/g. The samples ANN7 and ANN6 exhibited the SSA of 972.65 and 461.36 m2/g with a total pore volume of 0.5310 and 0.2464 cm3/g, respectively, as shown in Table 4. Figure 5b exhibited a pore size distribution of the three samples with an average mean pore diameter of 6.39, 7.71, and 7.91 nm for ANN6, ANN7, and ANN8, respectively [13]. The high surface area and averagely smaller pores are suitable for electrode materials for energy storage systems as its morphology can induce fast charge transfer which also contributes to an EDLC formation of surface interfaces for the electrode surfaces and electrolyte ions [4, 5, 13]. The specific surface area of ANN8 is higher than 1050 m3/g that was reported by Moses et al [1] which was performed without holding for 3 hours to allow intercalation of the carbon lattice expansion by K + ion before Activation.
Electrochemical characterization of materials
The performance of as-prepared electrode materials was first tested in the three-electrode system using 6 M KOH as the electrolyte. The CV curves for ANN6, ANN7, and ANN8 at a scan rate of 10 mV/s is shown in Fig. 6a. Sample ANN7 exhibited a slight discharge hump near 0.8 V due to some redox pseudocapacitance of the functional groups. All the CV curves exhibited a quasi-rectangular-like shape throughout the scanning range showing the excellent performance of EDLC [25, 30]. The GCD curves for all the materials, as shown in Fig. 6b, indicated an isosceles triangular shape for the GCD profile carried at 0.25 A/g. This confirmed the dominant EDLC behaviour for the electrode material. The specific capacitance was determined using the discharge time at different current densities and equation one for the three samples, as shown in Fig. 6c. The highest specific capacitance of ANN6, ANN7, and ANN8 are 537.3 F/g at 0.5 A/g, 517.5 F/g at 0.5 A/g, and 389.7 F/g at 0.25 A/g, respectively. The other GCD plot for the samples exhibited the same V-shape at different current density as shown in Figure S2 (Supplementary Information).
The samples exhibited an excellent specific surface area that matches performance. The performance is also influenced by the possible interaction of electrolyte ions and suitable porous surfaces of the materials and the heteroatomic function groups on the surface wettability. The sheet-like structure of the materials, as shown in Fig. 1, exposes the active sites for complete interface interaction for electrode surfaces and electrolyte ions. The results are comparable with the results reported in the literature [13, 31–33].
To objectively evaluate the electrochemical (EC) properties of the synthesized materials, including specific capacitance, energy density, power density, and stability, were determined. The study evaluated as-prepared materials with the two-electrode configuration system. The CV was carried out at different scanning rates from 5 mV/s to 200 mV/s with a potential range from 0 to 1.0 V exhibiting quasi-rectangular shapes, as shown in Fig. 7. The GCD of the materials was evaluated at different current densities from 0.25 to 2.5 A/g depicting V-shapes, as shown in Fig. 8. The CV and GCD curves estimated in the two-electrode system show no significant changes in shapes compared to those for the three-electrode system, as shown in Fig. 6(a & b). The CV for the as-prepared electrode materials in Fig. 7 all show similar quasi-rectangular shapes. The ANN6 (Fig. 7a) exhibited a redox reaction at the lower and upper potentials. This behaviour was also depicted in the ANN7 sample (Fig. 7b) but with fewer redox peaks. This may be due to the reversible redox reaction between the functional groups mainly originating from quinone or carbonyl (R) functional groups with the electrolyte, as illustrated in equation five created by the high concentration of oxygen functional group that tends to decompose at a sintered temperature of 750 oC as shown in Fig. 4b.
R- CX - O + H+ + e- ↔ R- CX - OH (5)
The ANN8 sample exhibited quasi-rectangular shapes without causing the Faradaic process. This may be attributed to the uniform porosity in the ANN8 sample, as shown in Fig. 1(ANN8), which is different from others. All the sample materials showed identical shapes at even higher rates of 200 mV/s without distorting the shapes by indicating the high stability of the materials. This also implies that the materials have excellent charge-discharge character [24, 34]. The exhibited symmetrical rectangular shapes showed better performance than most reported results in the literature [11, 25, 30, 31]. The CVs after stability demonstrated less or no exponential for the faradaic process, indicating no pseudocapacitance after holding the electrode materials at the high potential for a total of 130 h, as shown in Figure S3(a-f) (supplementary information) at a scan rate of 5 mV/s.
The study carried out GCD at different specific current to examine the specific capacitance performance of the materials with the two-electrode system configuration. The GCD curves exhibited almost symmetrical triangular V-shapes indicating a typical character for good EDLC properties, as shown in Fig. 8. This shows a no Faradaic process in the materials, which usually leads to poor stability performance. The GCD features correspond very well with the CV performance in Fig. 7. There was consistency in the charge-discharge character for the three-electrode system (Fig. 6b) and the two-electrode configuration. The electrochemical (EC) evaluation for specific capacitance was estimated by Eq. (2) for ANN6, ANN7, and ANN8 materials, considering the mass loading of the two electrodes as total mass [3, 4]. There was a variation in the calculated specific capacitance values of the devices, as shown in Figure S4a (supplementary information) with the applied current densities. The ANN7 sample material exhibited a maximum specific capacitance of 358.7 F/g at 0.5 A/g. The specific capacitance of the assembled devices decreased with the increasing current densities for all sample materials (Figure S4a supplementary information). This is mainly due to limited electrolyte ion transfer and limited interaction at the electrode interface at higher current density [35]. The ANN8 exhibited a maximum specific capacitance of 347.8 F/g at 0.5 A/g current density. Also, ANN6 presented its highest specific capacitance of 241.3 F/g 0.25 A/g current density.
The energy density and the power density of the assembled devices were calculated according to equations (3) and (4), respectively, as presented in the Ragone plot shown in Figure S4b (supplementary information) for all the sample materials. The device assembled from ANN7 gave the highest energy density of 12.45 Wh/kg for a current density of 0.5 A/g with a corresponding power density of 250 W/kg. This result is comparable and even higher than the symmetric assembled EDLC SCs reported in the literature [4, 5, 8, 9]. The ANN6 sample material exhibited an energy density of 8.38 Wh/kg for a power density of 125 W/kg at a current density of 0.25A/g. The ANN8 sample material also presented an energy density of 12.08 Wh/kg with a corresponding power density of 250 W/kg at 0.5 A/g current density. The energy density is approximately the same as that of the ANN7 sample, which can further be tested for commercial applications of different supercapacitors.
The electrochemical impedance spectroscopy (EIS) was performed on all sample materials to determine the resistive behaviour of the electrode materials for supercapacitor application. The EIS was determined in the frequency range from 10 kHz to 10 mHz at 10mV. The Nyquist plot of the sample materials are shown in Fig. 9 with the zoomed at high frequency in the insert and Figure S5 (Supplementary information) to expose the semicircle. The small semicircles at the real axis indicate low charge transfer resistance (RCT) at the sample interface, leading to high electrochemical performance [3]. The ANN7 sample material shown an RCT of 2.5 Ω as the highest resistance, with ANN6 and ANN8 exhibiting an RCT of 1.1 and 1.3 Ω, respectively. The vertical behaviour at low frequency with the imaginary axis indicates low ion diffusion into the pores of the electrode materials shown in Fig. 9. This suggests high cycling stability and the ideal capacitive performance of the materials. The sample materials were also tested at 20 mV to check the behaviour of the plots at higher potential, as shown in Figure S5. This suggested a similar behaviour at high frequency with a slight variation at a lower frequency. This indicates that at higher potential, there is little ion diffusion into the sample pores.
The study evaluated the stability test of the as-prepared devices using the voltage holding / floating method. The three sample materials exhibited almost no degradation for the electrochemical performance of the devices showing uniform behaviour, as shown in Figure S7 (Supplementary Information). The stability was carried out by holding the assembled device at a maximum potential of 1.0 V for 10 h after every three charge-discharge cycles repeated for 130 h at 0.5 A/g. The capacitance of the assembled devices remained almost constant throughout 130 h with no primary decrease with increased time and cycling. This indicates uniform wettability throughout the electrode/electrolyte ion interaction, and there were no trapped ions in the pores of the large porosity of the as-prepared electrode material [2]. The SCs showed excellent stability with 99% retention of the initial specific capacitance value even after 130 h holding. This superb stability of the assembled devices is attributed to the microporous nature of the prepared sample material that keeps the uniform wettability and easy access to electrolytic ion movement. The self-discharge test was carried out on the assembled devices after voltage holding to determine the life span. In the test, the device was charged fully to a maximum potential of 1.0 V at 0.5 A/g, then hold for 5 min before undergoing a self-discharge in an open circuit. Figure S8 (Supplementary information) exhibited an immediate drop for ANN6 sample material and lost 70% of the potential within 20 min. ANN7 and ANN8 showed a linear drop of potential losing 70% approximately after 4 h and 3 h. The possible reduction may be due to the decomposition of water used in the electrolyte at higher potential since ionic strength depends on the electrolyte. The water splitting requires a minimum potential difference of approximately 1.23V.