One-step synthesis of a sustainable carbon material for high performance supercapacitor and dye adsorption applications

Materials Science and Engineering B 297 (2023) 116766 Available online 29 July 2023 0921-5107/© 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). One-step synthesis of a sustainable carbon material for high performance supercapacitor and dye adsorption applications Sai Rashmi Manippady a , Monika Michalska b , Marcin Krajewski a , Kamil Bochenek a , Michał Basista a , Angelika Zaszczynska a , Tomasz Czeppe c , Lukasz Rogal c , Amrita Jain a , * a Institute of Fundamental Technological Research, Polish Academy of Sciences, Pawi´ nskiego 5B, 02-106 Warsaw, Poland b Department of Chemistry and Physico-Chemical Processes, Faculty of Materials Science and Technology, Vˇ SB-Technical University of Ostrava, 17. listopadu 2172/15, 708 00 Ostrava-Poruba, Czech Republic c Institute of Metallurgy and Materials Science, Polish Academy of Sciences, Reymonta 25, 30-259 Cracow, Poland ARTICLE INFO Keywords: Activated carbon Amorphous material Biomass Polymer gel electrolyte Supercapacitor Dye adsorption ABSTRACT The sustainable transformation of bio-waste into usable, material has gained great scientific interest. In this paper, we have presented preparation of an activated carbon material from a natural mushroom (Suillus boletus) and explor its properties for supercapacitor and dye adsorption applications. The produced cell exhibited a single electrode capacitance of ~247 F g −1 with the energy and power density of ~35 Wh kg −1 and 1.3 kW kg −1 , respectively. The cell worked well for ~20,000 cycles with ~30% initial declination in capacitance. Three cells connected in series glowed a 2.0 V LED for ~1.5 min. Moreover, ultrafast adsorption of methylene blue dye onto the prepared carbon as an adsorbent was recorded with ~100% removal efficiency in an equilibrium time of three minutes. The performed tests indicate that the mushroom-derived activated carbon has the potential to become a high-performance electrode material for supercapacitors and an adsorbent for real-time wastewater treatment applications. 1. Introduction Nowadays, people are suffering from many health problems because of the increase in contamination levels in the environment. One of the biggest challenges is to find a replacement for fossil fuels as raw materials in the energy and water sectors [1-4]. To overcome this challenge, biomass can be considered as a sustainable alternative to fossil fuels as it is natural, abundant, clean, and a renewable source. Out of all promising materials for power supply devices and dye adsorption, biomass-based porous carbon from renewable sources such as coconut shells, eggshells, peanut shells, bagasse, rice husk, etc. can serve as a potential candidate for multifunctional applications [5-9]. Supercapacitors are energy storage devices exhibiting high power density, fast charge–discharge rates, and robust cycling. The high power density results from fast ion motion between the electrode and electrolyte, while the high energy density is due to large surface area and some specific electrochemical properties of electrodes [10-12]. Different types of carbon materials, such as activated carbon, carbon aerogel, carbon nanotubes, carbide-derived carbons, and graphene have been used as potential candidates for supercapacitor application. Activated carbons, among all the different types of carbon materials available, are still considered as one of the most attractive materials for supercapacitor electrodes. These materials offer various advantages, including high active surface area, excellent conductivity, tunable porosity (optimum mixture of mesopores and micropores), facilitate for the transport of electrolyte ions, and excellent electrochemical stability. Biomassderived activated carbons offer all the properties which are required for an ideal electrode material. Their lignocellulosic contents in various proportions provide classified porosity with a good balance of micro and meso-porosities [13-17]. Most of the supercapacitors reported in the literature are tested with liquid electrolyte (aqueous and organic) e.g. aqueous solution of KOH, Na 2 SO 4 , H 2 SO 4 , or in propylene carbonate or acetone solution [18,19]. Liquid electrolytes have some drawbacks, such as transportation issues, difficulty in miniaturization, insufficient flexibility, electrolyte leakage, chemical/thermal/electrochemical instabilities, etc. [20-22]. Therefore, supercapacitors are often manufactured using gel polymer electrolytes (GPEs), where inorganic salts are trapped in the host polymer network * Corresponding author. E-mail address: [email protected] (A. Jain). Contents lists available at ScienceDirect Materials Science & Engineering B journal homepage: www.elsevier.com/locate/mseb https://doi.org/10.1016/j.mseb.2023.116766 Received 14 April 2023; Received in revised form 12 July 2023; Accepted 23 July 2023 Materials Science & Engineering B 297 (2023) 116766 2 [23-25]. GPEs combine quasi-solid state-like properties and liquid-like electrochemical properties. In addition, they offer high ionic conductivity, satisfactory mechanical properties, thermal stability and safety due to non-flammable nature and sufficient potential window [26]. The pollution released by textile dyes or any fabric industries can cause a carcinogenic effect that must be treated, otherwise it is harmful to aquatic and human lives. Among several water treatment techniques like photocatalytic degradation, adsorption, membrane filtration, ion exchange and coagulation [27-31], adsorption is considered as one of the most promising techniques because of low cost, easy handling, good efficiency, and fast dye recovery [32]. The most commonly used textile industry dye is methylene blue (MB), a cationic organic dye that is stable in air, water, and light and which therefore is difficult to degrade easily due to its complex structure [33]. Among various possible adsorbents, the activated carbons with higher specific surface area and well distributed porosity are considered as the promising candidates with enhanced efficiency and faster equilibrium time. In this work, Suillus boletus (wild mushrooms), were chosen as the biomass source which can be easily found in the autumn season in the forests of Central Europe, and in the present case it was collected from Poland [34]. Like other available bio-based materials, they immediately react with different types of chemical agents to obtain the microstructures which are required. Chemical activation at varying temperatures like 700 ◦C, 800 ◦C, 900 ◦C, and 1000 ◦C under a nitrogen atmosphere was employed to activate the sample. Potassium hydroxide (KOH) was used as an activating agent because it is very effective in creating micropores and mesopores [35]. Magnesium ion (magnesium perchlorate, Mg(ClO 4 ) 2 ) based GPE was used as an electrolyte with poly (vinylidenefluoride-co-hexafluoropropylene) (PVdF-HFP) as the host polymer. Supercapacitor cells were prepared using mushroom-based carbon powder with GPE consisting of PVdF-HFP-PC-Mg(ClO 4 ) 2 . Different electrochemical techniques, including cyclic voltammetry, electrochemical impedance spectroscopy, galvanostatic charge–discharge, and cyclic efficiency studies were performed to test the performance characteristics of the cells. The activated carbon synthesized in this work was applied as (i) an electrode material in a high-voltage supercapacitor device and (ii) ultrafast adsorption material for the Methylene Blue dye. 2. Experimental 2.1. Carbon material preparation and characterization details Porous activated carbon powders were prepared from a biomass source, namely wild mushrooms which were collected from a forest in Poland. Chemical activation was used to convert mushrooms into activated carbon with well-developed porosity and high specific surface area. The schematic preparation is shown in Fig. 1. Firstly, the mushrooms were properly washed and dried at room temperature. The asdried mushrooms were placed in a muffle furnace at ~300 ◦C for ~5 h to convert them into pyrolytic carbon. Then the obtained pyrolytic carbon was mechanically grounded, and the powdered samples were soaked in 25 wt% of calcium chloride (CaCl 2 ) (Chempur, Poland) for 18 h. The CaCl 2 soaked carbon samples were thoroughly and repeatedly washed with double distilled H 2 O to remove calcium ions. The samples were kept for drying in the oven for ~12 h at 110 ◦C. To activate the carbon, a chemical activation technique was used. Potassium hydroxide (KOH; Chempur, Poland) in the ratio of 1:1 was used as an activating agent. The chemical reaction between the activating agent and the carbon during activation is: 6KOH +2C → 2 K +3H 2 +2 K 2 CO 3 (1) The details of the further activation process are discussed elsewhere [36]. In the present study, the samples were activated at 700, 800, 900, and 1000 ◦C which are labeled, in the sequel, as M1, M2, M3, and M4, respectively. After that, the final samples were ground with a pestle and mortar and, then, stored in a tightly closed container. Morphological studies of M1-M4 samples were carried out with field emission scanning electron microscopy by using SEM/FIB-Zeiss Crossbeam 350, while EDX analyses were carried out using an Ametek EDAX, Octane Elite. The accelerating voltage used for the EDX elemental analysis was lowered to 7 kV which minimizes the penetration depth. The microstructure of the powder was investigated using transmission electron microscopy (TEM) both by FEI Tecnai G2 200 kV FEG (for the bright field (TEM/BF) and high-resolution observations (HRTEM) imaging) and ThermoFisher Themis G2 200 kV S-corrected X-FEG equipped with a Super X EDX energy dispersive spectroscopy (EDS) system (for analysis of the local chemical composition). The structure of the prepared carbon materials was characterized with X-Ray powder diffraction (XRD) by using a Bruker D8 Discover Diffractometer which was equipped with CuK α (λ XRD =1.542 Å), a confocal Raman spectrometer (Renishaw inVia) equipped with a charge-coupled device (CCD) camera and a continuous wave diode-pumped Nd:YAG laser working at λ =532 nm, respectively. X-ray photoelectron spectroscopic (XPS) measurements were carried out by an Omicron EA 125 electron spectrometer in the “Fixed Analyzer Transmission” mode with nonmonochromatized Mg K α (1253.6 eV) radiation with resolution around 1 eV. The energy scale of the electron spectrometer was calibrated according to the ISO 15472 standard. A thermal analyzer DSC SDTQ600, DSC TGA was used to study the thermal stability of prepared materials from 30 ◦C to 900 ◦C with a heating rate of 10 ◦C min −1 under argon flow of 20 ml min −1 . The specific surface areas of the carbon materials were measured with an AutoSorb IQ, Quantachrome, USA under nitrogen flow. The pore size distribution curves were obtained by using the DFT model. Zeta potential of the carbon materials were determined in a Nano ZS dynamic light scattering (DLS) apparatus (Malvern Instruments, Malvern, Worcestershire, UK) with a 4 mW HeFig. 1. Schematic diagram for the synthesis of activated carbon. S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 3 Ne laser source (λ =633 nm). Disposable zeta cells (DTS 1070) (Malvern Instruments, Malvern, Worcestershire, UK) were used to record the electrophoretic mobilities at 25 ±0.1 ◦C and the Smoluchowski equation was applied to calculate the zeta potentials. The accuracy of the measurements was ±5 mV, and the zeta-standard of Malvern (55 ±5 mV) was used for calibration. Absorbance values were measured using a Shimadzu 1050 +UV–Vis spectrophotometer. 2.2. Supercapacitor cell preparation and characterization Supercapacitor cells, were prepared in-house from the activated carbon and the gel polymer electrolyte described in the preceding sections. The GPE was composed of host polymer PVdF-HFP; poly(vinylidene fluoride-co-hexafluoropropylene), plasticizer; propylene carbonate (PC) and magnesium perchlorate (Mg(ClO 4 ) 2 ). All the materials for the GPE synthesis were commercial materials purchased from Merck company and used as-received. The details of the GPE preparation are discussed in [37]. For the electrodes, the active carbon materials (M1, M2, M3, and M4) and the polymer binder, PVdF in the weight ratio of 90:10, respectively, were mixed with the agate mortar and pestle. Acetone was added drop by drop to obtain the slurries of active material and binder, which were then deposited as layers on a carbon cloth received from AvCarb, USA. The prepared electrodes were dried in the oven at ~100 ◦C for 8 to 10 h before using them to prepare the capacitor cell. Supercapacitor cells were prepared in the form of sandwich consisting of the gel polymer electrolyte film put in between two symmetrical electrodes prepared from M1, M2, M3, and M4 activated carbons. The area of the electrodes was set to 1 cm 2 and the mass of the active carbon layers varied between 0.9 mg and 1.6 mg. Configurations of cells were as follows: Cell#A: M1|GPE film|M1. Cell#B: M2|GPE film|M2. Cell#C: M3|GPE film|M3. Cell#D: M4|GPE film|M4.where GPE is PVdF-HFP-PC-Mg(ClO 4 ) 2 film. The performance of the supercapacitor cells was characterized by different electrochemical techniques such as cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), galvanostatic chargingdischarging measurements, and also prolonged charge–discharge cycles for cyclic stability testing. All these measurements were done using the Biologic VMP3 electrochemical workstation. The EIS measurements were recorded in a frequency range from 200 kHz to 1 mHz. The formulas used to calculate specific capacitance (C sp ), power density, and energy density are provided in the Supplementary Information (S1-S5). 2.3. Adsorption studies To explore the removal of organic dyes, methylene blue (MB) was selected and the dye concentration was examined using a UV–vis spectrophotometer at a wavelength (λ max ) of 664 nm and water as a solvent. A calibration curve was plotted based on Beer-Lambert’s law, to study the concentration of samples. Detailed studies of the dosage, pH, adsorption kinetics, and isotherms were performed to explore the adsorption behavior of the adsorbent (studied sample) onto the adsorbate (MB). All the adsorption studies were performed using a 25 ml glass beaker at a speed of 150 rpm at STP. To fix the adsorbent quantity in adsorption studies, the optimal dose of the sample needs to be known. For this purpose, three different quantities of adsorbents were taken starting from 2 mg to 6 mg in a fixed dye quantity of 15 ml of 10 mg L -1 concentration. To understand the effect of pH on the dye solution in adsorption studies, pH-dependent studies were conducted. At first, the pH value of MB dye was varied from 2 to 10 using 0.1 M NaOH and 0.1 M HCl solution. Further, the optimized material was added to the prepared dye solutions of pH 2–10, and its adsorption behavior was analyzed. The kinetic experiments were performed using a 10 ppm concentration of MB dye with twice the amount of dye and adsorbent taken in the dosage study. After the addition of the adsorbent to MB dye, the prepared solution was sonicated for 30 s and then stirred magnetically at a speed of 150 rpm. Then, at each 1 min time interval, the dye solution was collected and kept to settle down to separate the adsorbent. Further, the concentration of the supernatant dye solution was examined by using a UV–Vis spectrophotometer apparatus. Further the formulas used to calculate percentage of the dye removal, equilibrium adsorption efficiency (q e , mg g −1 ), pseudo-first order, pseudo-second order kinetics, Langmuir isotherm and Freundlich isotherm models were provided in Supplementary Information, equations (S6)-(S9). To study the maximal adsorption capacity of the M3 material, isotherm studies were carried out with initial MB dye concentrations ranging from 100 to 500 ppm. For this purpose, 6 mg of sample was put in 15 ml of the dye at pH =8 for 24 hrs. Then, the supernatant dye was accumulated to measure the concentration of sample using the UV–Vis spectrophotometer. Finally, we fitted the obtained data with the Langmuir and Freundlich isotherm models using equations S10 and S11, respectively. In order to understand the practical applicability of the M3 adsorbent, recycling studies were carried out using 10 ppm MB dye with ethanol as solvent. After the dye adsorption, the dye was separated and M3 adsorbent was treated with ethanol and sonicated and further stirred at 200 rpm for 30 min to desorb the dye molecules. Further the M3 adsorbent was dried at 150◦to remove all the moisture. After drying, the material was reused for next cycle. The same procedure was followed for 5 consecutive cycles and dye adsorption efficiency was determined. 3. Results and discussion 3.1. Characteristics of prepared activated carbon The surface morphology of carbon materials derived from mushrooms was observed using scanning electron microscopy (SEM) and the images are shown in Fig. 2. From the images, it is evident that with the increase in temperature, the porosity increased and the produced CO 2 could easily diffuse into the material. This, in turn, enhanced the process of activating CO 2 on the material’s surface, resulting in the creation of pores. Interestingly, when the temperature increased to 900 ◦C, the M3 material (Fig. 2c) turned to be more porous and the mixed micro to mesoporous morphology could be the reason for enhanced electrochemical performance. With further increase in temperature, the pores started to agglomerate (Fig. 2d) which was in accordance with the BET studies. All carbon materials revealed a high and heterogeneous porosity on the surface, with the pore size ranging from 1 to 4 µm. Fig. 3a presents the X-ray diffraction results of the activated carbon derived from wild mushrooms. Two characteristic graphitic diffraction lines at around 2θ value of 26◦(002) and at 43◦(100) were observed for all analyzed samples [13,38,39]. In general, the reflection (002) was divided into three parts: a broad peak which was associated with disordered carbon (D), then a sharp peak at 2θ =26.0◦which might be due to the presence of a random turbostatic graphitic carbon (T), and one more sharp peak at 2θ =26.5◦which confirmed the existence of graphite (G) [38,40,41]. The carbon materials synthesized at 700 ◦C and 800 ◦C exhibited a broad and weak (002) reflection line confirming the existence of disordered carbon. When the temperature increased from 900 ◦C to 1000 ◦C, it was clearly visible that the analyzed (002) reflection line shifted to a 26.5◦value associated with the crystalline graphite (G carbon) [40,41]. The crystalline structure of graphite was also confirmed by two diffraction lines at around 2θ =50.7◦(101), and 54.6◦(004) [40,41]. Fig. 3b presents the FTIR spectra of all examined samples with representative surface functional groups. According to Bora et al. [42], when the oxygen-containing functional groups are present in activated carbons, they have a significant impact on wettability and electrochemical performance. From Fig. 3b, it is evident that the characteristic S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 4 absorption peaks for the activated carbon appeared at 3274, 1608, and 1049 cm −1 . The absorption peak at 3274 cm −1 was assigned to the O–H bond stretching vibration of the hydrogen bond (–OH) in phenols, carboxyl, alcohols from the active carbon material [43-46]. The peak intensity decreased with an increase in the heating temperature. The absorption peak which is present at 1608 cm −1 was ascribed to the aromatic stretching C=C vibrations from polar functional groups (carboxylic acids, lactones, carboxylic anhydrides) [43-45]. The absorption band at 1049 cm −1 can be ascribed to the C-O in ethers (stretching), phenolic groups: - C-OH (stretching), –OH [43-46]. Fast pyrolysis usually results in the creation of CO 2 during the pyrolysis and activation of the activated carbon derived from biomass material [47]. The peak intensity of the above-mentioned process was detected in the region 2250 – 2300 cm −1 [47]. The Raman spectra of all investigated mushroom-derived carbon materials are presented in Fig. 3c. The spectra collected for samples M1 and M2 are very similar. They consist of two broad partially overlapping bands which are characteristic of amorphous carbon materials [48]. These bands are located at ~1340 cm −1 and ~1580 cm −1 and mostly marked as D and G band, respectively. The band D is connected with the vibrations of sp 3 -bonded carbon atoms of defects and/or disordered graphitic structures, whereas the band G corresponds to well-arranged sp 2 graphitic layers [49]. Considering the Raman data related to the M3 sample, one can see that this sample reveals quite a complex structure because it is composed of a mixture of amorphous (cf. Fig. 3c(a)) and partially graphitized carbon (cf. Fig. 3c(b)). The results from the Raman studies are in line with the previously discussed SEM and XRD results. In the partially graphitized M3 sample, the signal coming from the disordered and/or defected carbon still dominates, since the D-band is more intense than the G band but, at the same time, the other two specific bands located at ~1610 cm −1 (D’ band) and ~2670 cm −1 (2D band) for graphite appear. The D-band is most commonly referred to as an intra-valley resonance with the G band [50], and the band 2D refers to a vibration of carbon atoms which is sp 2 -bonded in a 2-D hexagonal lattice [50,51]. In turn, the Raman spectrum of the M4 sample indicates a well-graphitized structure, in which the G and 2D bands are very intense. The degree of graphitization for all the investigated samples is estimated using the formula I G /I D , where I G and I D are the relative intensities of G band and D bands, respectively. Among the examined materials, the lowest value is found for the M3 sample. This shows that the M3 sample is the most defected and might be considered as a “transition material” between amorphous and graphitized state. The N 2 -adsorption–desorption studies were carried out for the mushroom-derived carbon materials. The pore size distribution curves and the isotherms are depicted in Fig. 3(d-e). As can be seen, M1 and M2 show Type I isotherm which confirms the presence of micropores [52], whereas in the case of M3, there is a steady increment in the amount of the absorbed gas for a pressure range of P/P 0 >0.02 which indicates that M3 contains a mixture of micropores and mesopores. The balanced micropore to mesopore ratio in M3 is one probable reason for the high values of specific capacitance and also good power capability of the cell. The micropores in the material mainly come from the etching effect of Fig. 2. (a-d) SEM spectra of M1, M2, M3 and M4 samples synthesized at temperatures from 700 ◦C to 1000 ◦C, respectively. S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 5 potassium hydroxide from the walls of carbonized cell and are independent of the carbonization temperature. On the other hand, with increasing carbonization temperature, the aggressive etching effect is observed resulting in the accumulation of micropores. More and more mesopores are formed because the micropore walls are almost collapsed. In addition to the above phenomena, the rupturing of macropores also forms more mesopores and as a result of these phenomena, the overall contribution of mesopores to the total pore volume with the increase in carbonization temperature is rational and at a specific temperature this ratio is properly balanced with respect to the electrochemical application. As per the IUPAC nomenclature, M3 shows the mixed features of type I and IV isotherms [53], and in the case of M4 material, a significant H4-type hysteresis can be seen in the plot which confirms the existence of a big amount of mesoporosity in the M4 material. In the literature density functional theory (DFT) methods have been used to estimate the pore size distribution and average pore size of the carbon materials [54]. Fig. 3f shows the pore size distribution curves of the material. The average pore size of the prepared materials is between 2.07 nm and 5.2 nm. All the parameters: BET specific surface area (SSA), average pore size, micropore volume, total volume, and mesopore volume were estimated for the investigated samples and are collected in Table 1. The total pore volume (V tot ) was noted from the amount of gas adsorbed at a relative pressure of P/P 0 =0.99 and the micropore volume (V mic ) was found out by using the t-plot method. The mesopores volume (V meso ) was calculated by subtracting the micropores volume from the total pore volume. Fig. 3. (a) PXRD patterns of the synthesized materials from M1 to M4, (b) FT-IR spectra of all samples, (c) Raman spectra of M1-M4 samples; here M3(a) and M3(b) spectra were acquired for the different parts of the same M3 sample, (d) N 2 -adsorption–desorption isotherm and (e) Pore size distribution curves for M1-M4 samples. S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 6 The specific surface areas gradually increased from 700 ◦C to 900 ◦C and then suddenly decreased for 1000 ◦C. It was probably due to the graphitization of carbon which caused pores agglomeration and the specific surface area decreased. This assumption is fully in agreement with the Raman, XRD, and FTIR results which are discussed before. Here, it is also worth mentioning that the highest specific surface area is found for the M3 sample activated at 900 ◦C (1550 m 2 g −1 ). Based on the specific surface area values, porosity distribution as well as the particular morphological and structural properties, the M3 sample was chosen for the detailed electrochemical and dye adsorption investigations. But before the application studies, it was also characterized with high resolution transmission electron microscopy (HRTEM) and thermogravimetry analysis (TGA). The detailed structural analysis of the M3 material was performed using Transmission Electron Microscopy. Bright field image (Fig. 4a) shows spherical grains with a diameter of approximately 15–40 nm embedded within the matrix. The HRTEM was recorded from the area in Fig. 4a, and in Fig. 4b, inserted Fast Fourier Transform (FFT) is shown. It is confirmed that the carbon derived from wild mushrooms has crystalline (FFT – 1, where weak reflections are present) and amorphous (FFT-2, where the typical amorphous halo around the central directly transmitted beam is visible) character, for sphere and matrix respectively. These results are also aligned with the Raman and XRD measurements. Thus, the HRTEM analysis revealed that the carbon derived at 900 ◦C possesses a mixture of amorphous and partially graphitized phases. In addition, we also performed mapping element distribution from the area of BF-TEM analysis, confirming that O, C, Cl, and Si are present in the synthesized activated carbon (Fig. S1). The presence of Si in the sample could be due to the chemical activation of biomass, such as mushrooms at higher temperatures [55]. In order to further confirm the presence of other elements in M3 Table 1 BET specific surface area and other parameters of the mushroom-derived materials. Sample Activating temperature (◦C) S BET (m 2 g −1 ) V t (cm 3 g −1 ) V micro (cm 3 g −1 ) V meso (cm 3 g −1 ) Average pore size (nm) M1 700 1116 0.6 0.2 0.4 2.1 M2 800 1328 0.7 0.3 0.4 2.07 M3 900 1550 0.9 0.3 0.6 2.7 M4 1000 909 1.2 0.08 1.1 5.2 Fig. 4. Transmission electron microscopy (TEM) studies of M3 material; (a) Bright field image, (b) High resolution micrograph, (c) XPS full scan spectrum and (d) TGA curve for M3 sample. Table 2 XPS data of surface composition of M3 sample. Element/ peak Binding energy (eV) Chemical state, relative contribution Concentration (atomic %) C 1s O 1s Si 2p S 2p N 1s Na 1s Mn 2p 3/2 284.4 286.1 288.1 290.0 533.0 536.6 104.0 107.2 – 399.0 400.9 404.4 1071.8 – graphitic carbon (84%) –OH, epoxide C-O-C (9%) carbonyl groups (5%) carbonyls at strongly oxidized edges of graphite sheets (2%) broad peak mainly due to SiO 2 1 SiO 2 2 SiO 2 1 SiO 2 2 not detected nitrogen in organic/carbonbound environment oxidized N in organic environment or ammonium highly oxidized N (?) ionic Na not detected C:53.7 O:30.6 Si:13.8 S:– N:1.7 Na:0.2 Mn:– S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 7 sample, XPS investigations were done and the results are summarized in Table 2 which confirms the presence of C, Si and O and the full spectrum is shown in Fig. 4c. The C 1s spectrum of M3 sample (Fig. S2a) was analyzed and modeled by the asymmetric line shape derived from the spectrum of carbon black. Additional peaks needed for adequate fit were assigned to the functional groups. Thus, the main C 1s contribution always arose from graphitic carbon (binding energy of the most intense part of the line shape was at 284.4 eV). A smaller peak at 286.1 eV is due to carbon singly bound to oxygen. This contribution is generally interpreted as C-OH groups or C-O-C-like arrangements such as cyclic ethers at the edges of the graphene sheets or epoxide groups on the basal planes. Nevertheless, certain (especially isolated) carbonyl groups can also contribute to this peak. The estimated amount of these contributions was around 9% of the total carbon content. A higher binding energy component at 288.1 eV can be assigned to more highly oxidized carbon species such as carbonyls, carboxylic or anhydride functionalities or carbon atoms bound to more than one oxygen atoms in different configurations like lactones [56,57]. The relative amount of this type of contribution was around 5% of the total carbon content. A third, very high binding energy contribution around 290.0 eV may arise from carbonyl groups at strongly oxidized sites like oxidized edges of graphene sheets. The amount of these functionalities was around 2%. The Si 2p spectrum (Fig. S2b) consisted of two relatively broad, partially overlapping peaks. The lower binding energy one appeared at 104.0 eV binding energy. Although this value is slightly high for SiO 2 (the Si 2p peak of SiO 2 is usually observed around 103.5 eV), it still can be attributed to completely oxidized Si in silica. The higher binding energy peak was found around 107 eV; this value is too high for any reasonable Si chemical state, but can be interpreted as a SiO 2 fraction which is electrically not coupled to the carbon material and its (photoionization induced) charge is not well compensated. Thus, the observed splitting of the Si 2p spectrum suggests that SiO 2 is partly in electrical coupling with the carbon while its other part is insulated from the backbone. The O 1s spectrum was dominated by the SiO 2 -related peaks and the line shape reflected the above described charging behavior of silica. The binding energy of the fraction electrically coupled to carbon was in agreement with data available for silica. The strong SiO 2 -related oxygen peaks completely covered signals from oxygen-containing groups bound to carbon. Fig. 4d shows the thermal decomposition of the M3 sample. As can be seen from the plot, the first decomposition stage was seen up to 200 ◦C, which is probably related to the evaporation of residual H 2 O (around 5%) [58]. The second weight loss was between 300 ◦C and 700 ◦C which is associated to the heat degradation of materials and is also associated with chemical changes like dehydration, degradation, and condensation. In this stage, there is a loss of aliphatic character resulting in an increase of aromaticity and simultaneously releasing the gas [39]. Beyond 700 ◦C, a huge loss in weight is observed which shows the major decomposition of the material. 3.2. Electrochemical analysis of supercapacitor cells The CV studies of all the cells were carried out in the two-electrode system. Fig. 5a depicts the CV response of Cell#A - Cell#D at 5 mV s −1 in the potential range of 0–1.0 V. Evidently, the CV curves of Cell#C Fig. 5. (a) CV curves of capacitor Cells #A-#D at a scan rate of 5 mV s −1 , (b) cyclic voltametric responses of Cell#C for different scan rates, (c) variation of specific capacitance values of Cell #C versus different scan rates, (d) CV curves of capacitor cell #C with variation of voltages at a scan rate of 5 mV s −1 . S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 8 exhibit a symmetrical rectangular shape without any redox peak, which indicates a typical double layer capacitive behavior with a reversible non-faradic process. The CV curves of Cell #A and Cell #B have a slight deviation from the rectangular pattern, which may be because of the high internal resistance as compared to Cell #C. Furthermore, it can be seen that the area under the CV curve is larger for Cell#C as compared with other cells. This area is in direct proportion to the specific capacitance feature, thus it confirms the superior specific capacitive properties of Cell #C. The capacitance values calculated from the CV studies for Cell#A-Cell#D are 109.5, 77.7, 246.9, and 81.4F g −1 , respectively. The formula used to calculate capacitance from CV is provided in the Supplementary Information as Equation S1. The cyclic voltametric response of Cell#C was also recorded at varying scan rates from 5 to 100 mV s −1 at a potential range of 0 V-1.0 V as shown in Fig. 5b. The voltammograms exhibit proper reversibility which indicate the ideal capacitive properties of the mushroom-based electrodes at different scan rates. As it can be seen from Fig. 5b, up to 50 mV s −1 , the cell reflects a nearly rectangular shape and beyond that it is slightly tilted. This confirms the good compatibility at the electrode–electrolyte interface even at the higher scan rates and sufficiently high rate capability of the capacitor cell. Fig. 5c shows the variation of specific capacitance values calculated at different scan rates. It was observed that capacitance values slowly decreased with regard to the scan rate values which reflects a fast switching of the electrolyte ions (Mg 2+ and ClO 4 - in the present case) through different pores of the electrode material. In Fig. 5d cyclic voltammetric curves of Cell#C were recorded at 5 mV s −1 for different voltage ranges and a maximum was obtained for 1.6 V. Almost rectangular and mirror images pattern was observed up to 1.4 V. Beyond that value a significant deviation occurred which confirms that the capacitor cell was cycled sufficiently enough in the range of 0 – 1.4 V and is hence considered as a safe limit for the present cell. The electrochemical impedance spectroscopy responses of the capacitor cells (Cell#A-Cell#D) are shown in Fig. 6a. The impedance plots were measured from the frequency range of 10 5 Hz to 10 -3 Hz at room temperature. As it can be seen, the Cell#C possesses the smallest high-frequency semi-circular spur as compared to other cells and is then followed by a sharp increasing pattern parallel to the imaginary axis (Z”) (the inset shows the overall impedance curve of Cell#C). The impedance plot of Cell#C is very close to an ideal capacitor, whereas for other cells high values of the bulk resistance R b , charge-transfer resistance R ct , and overall resistance R at high frequency and towards low frequency were observed. Also, the line towards low frequency was not parallel to the imaginary impedance axis. These studies also confirm the superior behavior of the mushroom-based electrode material which was activated at 900 ◦C. The high/mid frequency region of impedance plots (inset of Fig. 6a) shows the well-defined semi-circular spur of all cells. From this, bulk and interfacial properties of the cells can be determined which also confirms that Cell#C has the lowest resistance and thereby shows the best capacitive results. In the impedance pattern, the large semicircle corresponds to the resistance dominant behavior at the interface of electrode–electrolyte, while the small semicircle corresponds to the behavior where the capacitive nature dominates [59]. The bulk resistance R b , charge transfer resistance R ct , overall resistance R, and capacitance values of the cells are collected in Table 3. The formula used to calculate the capacitance values is provided as Equation S2 in Supplementary Information. In the case of Cell#C, the lower resistance values indicate a possibility of higher values of the specific power of the device (confirmed from GCD analysis and discussed in the next section). The higher specific capacitance values confirm that there is a proper formation of the double layer and easy movements of Mg 2+ and ClO 4 - ions in the M3 electrode. Fig. 6. (a) Electrochemical impedance pattern of Cell #A - Cell#D at a frequency level of 1 mHz. Expanded impedance plots of all cells at a higher frequency are illustrated in the inset; for clarity the impedance plot of Cell #C is also shown in the inset; (b, c) Real and imaginary impedance (Z’/Z”) and capacitance (C’/C”) versus frequency respectively for Cell #C. S. Rashmi Manippady et al. Materials Science & Engineering B 297 (2023) 116766 9 Further, from the electrochemical impedance spectroscopy plots rate capability was also calculated and is shown in Fig. 6(b-c). To find out the rate performance of Cell#C, its response frequency (f 0 ) was obtained from the point where Z’ and Z’’, also called the Bode plots of complex impedance suggested by Miller intersect. Consequently, the response time ( τ 0 ) is calculated by using the formula τ 0 =1/f 0 [60]. The response time was also verified from the real capacitance and imaginary capacitance (C’ and C”) versus frequency, which are also called the Taberna plots. The τ 0 value for Cell #C is found to be of the order of 30 s which indicates a moderate rate capability, also confirmed by other studies. The GCD, (galvanostatic charge–discharge) measurements were carried out to understand the charging/ discharging behavior of the supercapacitor cells. Fig. 7a shows the GCD curves of Cell#A-Cell#D at a constant current density value of 1.0 mA cm −2 and a voltage between 0 V and 1.0 V. An ideal GCD curve of the double-layer capacitor shows a symmetric triangular-shaped profile. It can be seen from the figure that Cell#A and Cell#B have a slight deviation in the GCD curve from the ideal shape, whereas Cell#C shows an almost perfect triangular shape with negligible internal resistance (IR) drop. In the case of Cell#D, the discharging time is significantly lower with a slightly higher IR drop. The capacitance values, IR, energy density, and power density values of all the cells are collectively shown in Table 4. The specific capacitance, energy density, and power density values are calculated by using Equations S3-S5 provided in Supplementary Information. The values of specific capacitance and internal resistance are almost comparable (or equal) to the capacitance values and resistance values calculated by different techniques as discussed in the preceding sections. Cell#C shows the energy density and the power density of ~35 Wh kg −1 and ~1.3 kW kg −1 evaluated at 1.0 mA cm −2 respectively. The energy density value of the present cell comprising GPE are comparable or in some cases even higher than in several reported systems in literature using activated carbon as electrodes. For instance, Vijaykumar et al. [61] developed a symmetric supercapacitor cell using activated carbon electrode and LiClO 4 -based GPE, which offers energy and power density of 10 Wh kg −1 and ~29.0 kW kg −1 . A similar work was reported by Yadav et al. [62] using PVdF-HFP gel polymer electrolyte in which they achieved an energy density of the order of 17.7 Wh kg −1 . Many other works were reported in the literature with the same Table 3 Various electrical parameters of supercapacitor cells from EIS analysis. Cells R ct (Ω cm 2 ) R b (Ω cm 2 ) 1 mHz R (Ω cm 2 ) C (mF cm ¡2 ) *a (F g ¡1 ) *b #A 151.7 41.6 1056 73.8 196.7 #B 739.1 70.4 1356 64.7 161.8 #C 53.9 20.7 261.1 54.8 243.7 #D 78.7 26.9 861.4 20.4 74.2 *a is the overall capacitance, *b is the single electrode capacitance. Fig. 7. (a) GCD variation of capacitance Cell #A – Cell #D at a current density value of 1.0 mA cm −2 , (b) specific capacitance variation of Cell #C with respect to the current density, (c) Ragone plot of Cell #C, (d) specific capacitance variation of Cell #C with respect to cycle numbers, GCD curves of the first hundred cycles are illustrated in the inset of (d), (e1-e4) photos showing a green LED glow of three cells connected in series, recorded at different times. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Table 4 Summary of the various parameters obtained from the GCD studies of Cell#A - Cell#D at a current density of 1.0 mA cm −2. Cell Discharge capacitance (C d ) [F g −1 ] ESR Ω cm 2 Energy density (E d ) [Wh kg −1 ] Power density (P d ) [kW kg −1 ] #A 129.8 646 18.1 0.9 #B 247.0 607 34.3 0.94 #C 247.0 272 34.4 1.32 #D 70.4 477 9.8 1.2 S. Rashmi Manippady et al.