Introduction

Solid polymer electrolytes are made in a polar polymer host by dissolving inorganic salt where metal ions from the salt form a coordinating bond with the polymer functional group of oxygen [1, 2]. Adding low molecular weight compounds to the polymer matrix can result in plasticized polymer electrolyte [3]. They have been proven to show a good performance in various modern electrochemical devices based on their morphological, electrical, and thermal properties [4]. Properties such as compatibility, flexibility, and capability of forming thin film media have made solid polymer electrolyte as an excellent alternative to liquid electrolyte due to no problem of leakage, thermally stable, and easy to be produced [5]. Moreover, natural polymers have a few important characteristics such as non-toxic and environmentally and economically friendly as well as abundant in nature [6, 7]. The biodegradable feature of these polymers makes it an ideal alternative to synthetic polymers that are not biodegradable. Natural polymers are commonly used in many sectors, including medical [8], textile [9], cosmetic [10], and food packaging [11] industries.

Gelatin is one of the natural polymers and traditionally obtained by using the hydrolysis process of skin and bones from cattle and pigs which will lead to discomfort for the halal dietary restrictions community as well as pesco-vegetarians [12]. To solve this problem, gelatin also can be produced from fish because the fish industries have reported to produce a great amount of waste materials such as fish heads and skin [13]. Chitosan and fish gelatin can be obtained from these wastes which can minimize the environmental impacts [14, 15]. There is a huge amount of collagen that can be found in the fish skins and it is being commercialized as a food-grade gelatin. However, collagen content in cold water fish skin has fewer amounts of hydroxyproline and proline residues [16]. Besides, gelatin from marine sources showed lower functional properties, such as gelling temperature and gel strength than that of mammalian [17, 18]. The number of studies reported on the fish skin gelatin–based polymer electrolyte is still low.

Cations and anions are the ones responsible for the process of adsorption and desorption at the surface of the electrodes in an EDLC. Thus, salt doping has been employed by many researchers to improve the conductivity of the electrode separator. Lithium perchlorate (LiClO4) has been chosen as the ion provider in this work for some reasons. As mentioned by Teoh et al. [19], LiClO4 has a high diffusion rate and is comprised of large anion and small cation. LiClO4 has good conductivity and ease of complex formation in the fabrication of solid polymer electrolytes. LiClO4 has lower lattice energy which 715 kJ mol-1 compared to other lithium salts e.g. lithium tetrafluoroborate (LiBF4) (749 kJ mol-1) and lithium acetate (LiOAc) (786 kJ mol-1) [20]. More free ions can be produced with low lattice energy salts. Basu et al. [21] reported that LiClO4 is compatible with gelatin-based polymers and the dielectric properties of the polymer electrolytes are enhanced with the presence of LiClO4.

The increase of demand for renewable and high-performance energy devices is one of the current requirements of the human lifestyle that are focusing on a clean environment. Electrochemical double-layer capacitor (EDLC) can be used to replace conventional batteries which are now serving as the main choice to the community [22]. The EDLC uses a non-Faradaic process as an energy storage mechanism where only charge accumulation occurred and double-layer will form between the electrode surfaces as well as the interfacial region [23, 24]. To fabricate an EDLC, few methods have been widely studied to produce the electrode such as using carbon aerogel [25], graphite [26], and activated carbon [27]. However, activated carbon shows better characteristics and also has been extensively used in the electrochemical device field. This is because activated carbon has a reasonable price, better chemical stability, and large surface area as well as high electronic conductivity [28]. Moreover, EDLC has been a main option among the other supercapacitors due to simple fabrication method and has high durability, thermal stability, and power densities [29,30,31].

Experimental

Preparation of electrolyte

An amount of 2 g gelatin from cold-water fish skin (FS) (SIGMA) was stirred in 100 mL of distilled water at 60 °C for 30 min. The FS solution was cooled to room temperature before various amounts of LiClO4 (SYSTERM) were added to produce salted electrolytes. Upon dissolution, FS-LiClO4 solution was poured into a plastic Petri dish and left to dry in a fume hood at room temperature (25 °C, RH ~ 50%). The formed film was placed in a dessicator filled with silica gel to remove excess moisture before any characterization. The thickness of the film ranged from 0.022 to 0.035 cm. The designation of each electrolyte depends on the weight percentage (wt.%) of LiClO4; e.g., the designation for electrolyte with 0, 5, 10, 15, 20, and 25 wt.% LiClO4 was FS0, FS5, FS10, FS15, FS20, and FS25, respectively. Figure 1 shows the image of the highest conducting electrolyte film.

Fig. 1
figure 1

The flexible highest conducting sample

Full size image

Characterization of electrolyte

The conductivity and dielectric properties of the electrolyte were obtained using electrical impedance spectroscopy (EIS) measurements. EIS was performed using HIOKI 3532–50 LCR HiTESTER with a frequency range between 50 Hz and 5 MHz at room temperature. The cell arrangement was stainless steel (SS)| electrolyte | SS. The bulk resistance (R) of the electrolyte was extracted from the Nyquist plots. The conductivity value was identified using the following equation:

$$ \sigma =\frac{t}{RA} $$
(1)

where A is the electrode-stainless steel contact area and t is the thickness of the electrolyte which was measured using Mitutoyo micrometer screw gauge. The real (εre) and imaginary (εim) parts of the dielectric permittivity (ε*), which were recognized as dielectric constant and dielectric loss respectively, are extracted from:

$$ {\varepsilon}_{re}=\frac{Z_{im}}{\omega {C}_o\left({\left({Z}_{re}\right)}^2+{\left({Z}_{im}\right)}^2\right)} $$
(2)
$$ {\varepsilon}_{im}=\frac{Z_{re}}{\omega {C}_o\left({\left({Z}_{re}\right)}^2+{\left({Z}_{im}\right)}^2\right)} $$
(3)

where Zre is the real part of impedance, Zim is the imaginary part of impedance, ω stands for the radial frequency, and vacuum capacitance is given as Co.

Fourier transform infrared (FTIR) spectroscopy was performed using a spectrometer (Spotlight 400 Perkin-Elmer) from 450 to 4000 cm-1 (resolution = 1 cm-1). The image of electrolyte surface was captured using the Hitachi SU8220 FESEM with ×10k magnification. Thermogravimetric analysis (TGA) was performed using STA 6000 Perkin Elmer from room temperature to 970 °C at a heating rate of 20 °C min−1. X-ray diffraction (XRD) analysis was conducted using a Siemens D5000 X-ray diffractometer (1.5406 Å) with a resolution of 0.1° (2θ = 5° to 80°). Deconvolution analysis was conducted using the Origin 9.0 software to separate amorphous and crystalline peaks from the raw XRD data where the degree of crystallinity (χc) and crystallite size (Dc) can be obtained using the following equations:

$$ {\chi}_c=\frac{A_c}{A_c+{A}_a}\times 100\% $$
(4)
$$ {D}_c=\frac{0.94\lambda }{FWHM\cos \theta } $$
(5)

where the areas of crystalline and amorphous peaks are given as Ac and Aa, respectively. λ is the X-ray wavelength (1.54 Å), and FWHM is the full-width half-maximum of the selected crystalline peak which was determined together during the deconvolution analysis.

V&A Instrument DP3003 digital DC power supply was used to perform the transference number (TNM) analysis via DC polarization method. The cell was polarized at 0.2 V. The DC current was monitored as a function of time at room temperature. The potential stability of the electrolyte was determined using linear sweep voltammetry (LSV) analysis. DY2300 potentiostat was used at 10 mV s-1 from 0 to 2.5 V. The cell arrangement for both TNM and LSV was the same as EIS analysis. The value of ionic (ti) and electronic (te) transference number can be obtained from the following equations:

$$ {t}_i=\frac{I_b-{I}_s}{I_b} $$
(6)
$$ {t}_e=1-{t}_i $$
(7)

Fabrication and characterization of the fabricated EDLC

Step 1 was the dry mix process where 6.5 g activated carbon (AC) (Kuraray) and 0.5 g carbon black (CB) (Magna value) was mixed using planetary ball miller at 5 revolutions min-1 for 20 min. Step 2 was to dissolve 1 g of polyvinylidene fluoride (PVdF) (Magna value) in 30 mL of N-methyl-2-pyrrolidone (NMP) (EMPLURA). Step 3 was to dissolve the AC-CB powder in PVdF-NMP solution for several hours until homogenous black slurry was obtained. Step 4 was to clean the aluminum foil with acetone (SYSTERM) and coat the slurry using a doctor blade. The thickness of the electrode was set as 25 mm. Step 5 was to dry the coated aluminum foil in the oven at ~ 60 °C for several hours. The last step of electrode preparation was to cut the dried electrode into a small circle shape with an area of 2.01 cm2 and place the electrodes in desiccators to remove excess moisture. The most conductive sample was sandwiched in between electrodes and packed in CR2032 coin cells. The illustration of the fabricated EDLC was shown in Fig. 2. Cyclic voltammetry (CV) was performed with potential ranged from 0 to 0.9 V (V2–V1) using Digi-IVY DY2300. The capacitance (C) of the EDLC can be expressed as:

$$ C={\int}_{V_1}^{V_2}\frac{I(V) dV}{2{mS}_r\left({V}_2-{V}_1\right)} $$
(8)
Fig. 2
figure 2

The fabricated fish skin gelatin–based EDLC

Full size image

here, I(V)dV is the area of CV plot that is extracted from the Origin 9.0 software, Sr is the scan rate, and m is the mass of the activated carbon which is an active material.

Results and discussion

XRD study

XRD analysis enables us to study the structure of a material. Figure 3 illustrates the effect of LiClO4 addition on the crystallinity of the fish skin film. Raut et al. [32] stated that amorphous regions are preferable for ionic conduction in semi-crystalline polymers. Only a broad peak centered at 19.9° can be observed in FS0 (pure fish skin gelatin film) XRD pattern. This XRD pattern is common for gelatin-based film and comparable with other gelatin-based film studies [33, 34]. Das et al. [35] reported that the gelatin XRD pattern usually has a small crystalline peak at 7.0–8.1° and broad amorphous peak 19°. The peak at 7.0–8.1° in gelatin typically assigned to the triple-helical crystalline structure [36]. The absence of this peak in FS0 XRD pattern could be due to the reduction of hydrogen bonding between polymer molecules with the presence of water and heat during sample preparation. This condition reduces the crystallinity of the FS film. The intensity of broad peak at 19.9° is reduced as 15 wt.% LiClO4 is dope to FS film. As the concentration of LiClO4 increases to 25 wt.%, the peak (19.9°) has become less sharp and broader compared to other compositions. It is reported that the addition of LiClO4 into chitosan-dextran host has changed the XRD plot from a sharp shape to a broader shape [37]. This result indicates that more ions have built complexation with functional groups of FS film which will indirectly improve the ionic conductivity. The addition of lithium tetrafluoroborate (LiBF4) into natural rubber polymer increased the amorphousness of the polymer electrolyte [38]. Furthermore, the absence of LiClO4 sharp crystalline peaks indicates that the salt is completely dissolved in FS film.

Fig. 3
figure 3

XRD patterns of FS-LiClO4 system at various concentration of LiClO4

Full size image

Figure 4 portrays the deconvoluted XRD patterns of salted electrolytes. In XRD pattern of FS0, two peaks can be observed at 2θ = 19.5° and 22.5°. The addition of 10 wt.% LiClO4 has reduced the intensity and increased the FWHM of the crystalline peaks where the pattern continues as 25 wt.% LiClO4 is added. The crystalline peaks of FS25 are broader compared to other concentrations. Based on this information, values of χc and Dc are obtained and tabulated in Table 1. Both χc and Dc of FS0 have reduced from 22.22% and 2.93 nm to 6.38% and 2.45 nm, respectively, as 25 wt.% LiClO4 is added. This is due to the disruption of the crystalline structure of dextran by the presence of salt. Free ions from the salt form interactions with the functional groups of the polymer host via a dative bond, thus increasing the amorphousness of the electrolyte [39].

Fig. 4
figure 4

Deconvoluted XRD plot of FS-LiClO4 system with experimental (black), fitted (red), amorphous peak (blue), and crystalline peak (pink)

Full size image
Table 1 Parameters obtained from the deconvolution analysis of XRD patterns.
Full size table

FTIR study

FTIR analysis is used by many researchers to observe interactions within the electrolyte where the shifting of peaks or changes in the intensity of a peak is considered as interactions. Figure 5 shows the FTIR spectra for the FS-LiClO4 system from 450 to 4000 cm-1. It is well known that any structure that is associated with protein will show several types of amide bands. The existence of these bands is from the vibrational motions of peptide bonds which connect together the amino acids that constitute the protein, in this case, the gelatin. The amide bands consisted of amide A (3500–3100 cm-1), amide B (2800–3000 cm-1), amide I (1630 cm-1), amide II (1522 cm-1), and amide III (1236 cm-1) [40]. Amide A indicates the interaction between N–H stretching with hydrogen bonding and free OH [41]. As observed, the peak of amide A of FS0 has shifted from 3270 to 3325 cm-1 as 25 wt.% LiClO4 is added. This situation means that cations (Li+) from the salt have formed complexation with oxygen and nitrogen atoms in amide A. Amide B is C–H antisymmetric and symmetric stretching in aliphatic compounds [42]. Amide I is associated with the stretching vibrations of the C=O bond of the amide bond [43] while amide II is related to the bending vibrations of the N–H bond [44]. Amide III is due to C–N stretching vibrations coupled to N–H in-bending vibrations of the bound amide with weak contributions from C–C stretching [45]. It is noticeable that the peak of vibrational motions of C–O and C–C moieties (1072 cm-1) increases with the addition of more LiClO4. Thus, salt-polymer interactions in FS-LiClO4 mostly happen in amide A and C–O regions.

Fig. 5
figure 5

FTIR spectra of the FS film with various amount of LiClO4

Full size image

Conductivity study

The segmental motion of the polymer chain depends on the free volume availability, glass transition temperature of the system, amount of salt, plasticizer, and fillers. These parameters affect the free volume of the polymer [46]. Thus, the effect of LiClO4 addition on ionic conductivity of FS film is portrayed in Fig. 6. A low conductivity of (9.29 ± 0.89) × 10-11 S cm-1 can be observed for pure FS film. This is due to the lack of ionic number density and polymers which are known to possess insulator behavior. The addition of 15 wt.% LiClO4 into the polymeric network has improved the conductivity up to 3 magnitude orders. This verifies that ions from LiClO4 have interacted with complexation site of FS film [47]. The conductivity value of FS film is further increased to (3.58 ± 0.58) × 10-6 S cm-1 with the presence of 25 wt.% LiClO4. More charge carriers are provided at a high salt concentration which indirectly improves the ionic conductivity. The pattern of conductivity is consistent with the result from the FTIR study. Sangeetha et al. [48] reported a conductivity of 1.2 × 10-9 S cm-1 as 20 wt.% LiBF4 is added to PVdF-HFP host. The presence of 30 wt.% LiOAc in PEO matrix provides a conductivity of 2.18 × 10-6 S cm-1 [49]. These reported conductivity values are lower than this work where one of the reasons is due to the difference in lattice energy values. Ion dissociation is easier for salt with low lattice energy where more free ions can be produced that will enhance the ionic conductivity. The conductivity trend is consistent with XRD analysis where as more LiClO4 is added, the polymer electrolyte becomes more amorphous and FS25 possesses the lowest crystallinity.

Fig. 6
figure 6

Various ionic conductivity values at different concentration of LiClO4

Full size image

FESEM study

FESEM micrographs of the selected electrolytes are shown in Fig. 7. It can be seen that the surface of FS0 (pure fish gelatin) film is smooth and homogeneous, without pores or cracks. This result is similar to pure fish gelatin film surface reported by Qiang et al. [50]. As 10 wt.% LiClO4, the surface turns to a rough structure with the presence of particles. Bhad and Sangawar [51] stated that these particles act as the channels for ionic conduction and considered evidence of salt-polymer interactions. However, as 25 wt.% LiClO4 is included, the surface has changed to a denser structure indicating that the electrolyte has become more amorphous. This outcome infers that the polymer segmental has been improved with the presence of salt where it aids in the conduction of ions. This result is harmonized with conductivity and XRD result for FS25 where it has the highest conductivity value and the XRD shape is the broadest.

Fig. 7
figure 7

Surface morphology of a FS0, b FS10, c FS20, and d FS25

Full size image

TGA study

The effect of salt doping on the thermal stability of a polymer film can be obtained using TGA analysis. Figure 8 shows the TGA thermograms for FS0 and FS25 from 25 to 970 °C. Two curves can be seen in each thermogram, one is the percentage of weight loss and the other one is the first derivative. The peak of the first derivative indicates the point of the greatest rate of change on the weight loss curve or sometimes called as an inflection point. The 1st weight loss below 100 °C in FS0 thermograms is due to the weight loss of water as polymer possesses hydroscopic nature [52]. The 2nd weight loss of FS0 (54 %) has a larger first derivative peak compared to the first weight loss. This means that most weight loss happens in this region which is from 210 to 464 °C in FS0. Shehab et al. [53] reported that decomposition in this region is attributed to the loss of gelatin molecules. As 25 wt.% LiClO4 is added as shown in Fig. 7b, the decomposition temperature has reduced to 150 °C where the weight loss is 7%. This is due to the disruption of hydrogen bond between polymer chains by the presence of salt [54]. As reported in Manap et al. [55], the decomposition temperature of poly(lactic acid)-poly(propylene glycol) has decreased with the addition of 15 to 30 wt.% LiClO4. The 2nd weight loss in the thermogram of FS25 is due to the degradation of the gelatin.

Fig. 8
figure 8

TGA thermograms of a FS0 and b FS25

Full size image

Dielectric study

The dielectric study is a powerful tool to check the mechanism of ion transportation and the phase transitions within polymer electrolytes. Figure 9 a and b show the dielectric constant and dielectric loss, respectively, at various LiClO4 concentrations. The dielectric constant is observed to increase as more salt is added. There are several bonds in a polymer, e.g., hydrogen bonding (0.13–0.30 eV), dipolar interactions (0.07–0.13 eV), and dispersion interactions (0.002–0.09 eV) that can be disrupted with the presence of salt. The addition of salt affects the segmental motion of the polymer chain which can lead to more development of double-layer capacitance [56]. Dielectric loss in Fig. 9b also has the same increasing pattern. As more salt is added, more free ions undergo conduction towards the electrodes. During this transport time, ion collides with other ion and this collision cause some energy loss. Thus, when there are more ions, more energy loss will be produced [57]. Other than that, dielectric is higher at 50 Hz compared to 30,000 and 50,000 Hz. At low frequency, it is easier for ions to obey the changes of the electric field and form a proper charge double-layer. Meanwhile, incomplete polarization at high frequency occurs due to the rapid and unstable movement of ions [58]. The conductivity trend is consistent with the pattern of dielectric constant and loss.

Fig. 9
figure 9

Plot of the a real and b imaginary part of the dielectric

Full size image

TNM study

To use the polymer electrolyte in energy storage devices, it is crucial to examine the transference number of both ions and electrons. Some electrical devices are dependent on electrons flow, e.g., semiconductor and solar cell. Typically, in a conventional battery, cations are responsible for the process of intercalation and deintercalation. However, in EDLC, ions are the ones in charge of storing energy. Thus, high ionic contribution is a must in EDLC application (>0.90). It is noticeable in Fig. 10 that the current is large at the beginning which is at this point both ions and electrons are conducted. As time goes by, ions are blocked at the surface of the electrodes and at this point the cell is polarized where the current flows are due to the electrons [59]. ti for the best conducting sample of FS-LiClO4 system is found to be 0.97. This value is sufficient enough and near the unity value to make the electrolyte become useful in EDLC application. Gohel et al. [60] reported that ti in the PVdF-PMMA-LiClO4 system ranged from 0.96 to 0.99. A series of ti from 0.85 to 0.89 for PEO-LiClO4-ZnO system is reported by Patil et al. [61].

Fig. 10
figure 10

Polarization current versus time for F25 at 0.2 V

Full size image

LSV study

The potential stability (i.e., working cell potential range) of polymer electrolytes is a vital parameter to be determined for their implementation in energy storage devices. The potential stability window of F25 is evaluated using linear sweep voltammetry (LSV) analysis, and its corresponding voltammogram is depicted in Fig. 11. It can be seen that there is no current flows through the working electrode below ~ 1.50 V. The cell is polarized at this stage. At potential below ~ 1.50 V, the increment of current value is slow. However, as the potential increased to more than 1.88 V, the current increase drastically. A sharp increase in the current marks the starting of electrolyte decomposition. This outcome confirms that FS25 with the best conductivity in this work is potentially stable up to 1.88 V; thus, it is suitable for EDLC application. A work of chitosan-LiCO2CH3-glycerol system has been reported to possess potential stability of 1.80 V [62]. Chandra et al. [63] reported the potential stability of PVA-PMMA-LiCl is up to 1.69 V.

Fig. 11
figure 11

LSV plot of FS25 at a scan rate of 10 mV s-1

Full size image

CV study of the fabricated EDLC

The energy storage mechanism of an EDLC is through adsorption/desorption process instead of intercalation/deintercalation where this mechanism can be verified via CV analysis. Figure 12a illustrates the CV plot of the constructed EDLC at different scan rates. No peaks can be observed in CV plots of all scan rates. This verifies that the EDLC in this work obeys the typical energy storing mechanism of a capacitor which is non-Faradaic process [64]. The CV plot at 100 mV s-1 is a leaf-like shape where it changes to more rectangular shape as the scan rate reduces to 10 mV s-1. It is common that CV for a capacitor is scan rate–dependent. The migration of charge carriers occurs at a rapid rate at a high scan rate. In addition, due to carbon porosity and internal resistance, current dependence of voltage is produced [65]. As observed in Fig. 12b, the value of C achieved is 15.3 F g-1 for 100 mV s-1, meanwhile, 28.2 F g-1 for 10 mV s-1. Ions are able to move at a stable rate when a low scan rate is employed; thus, they can form more and properly charged double-layer. At a high scan rate, ions are migrated at a very fast rate; this situation decreases the formation of charge double-layer which reduces the capacitance value [66, 67].

Fig. 12
figure 12

a CV plot of the EDLC and b specific capacitance at different scan rates

Full size image

The cycleability of the EDLC is analyzed using a scan rate of 10 mV s-1 for 150 cycles. In Fig. 13a, the CV plot has a rectangular-like shape for the first 50 cycles where the shape has changed to a leaf-like shape from 75th to 150th cycles in Fig. 13b. As the cycle number increases, some of the free ions undergo a recombination process to form ion pairs. These ion pairs disturb the process of charge double-layer development and thus, the current is dependent on the voltage. Figure 13c shows the calculated value of C throughout the 150 cycles. The value of C at the 1st cycle is 28.1 F g-1 which is similar to the result in Fig. 13b where the value of C is 28.2 F g-1 at 10 mV s-1. C then increases to 34.7 F g-1 at the 15th cycle. Stabilization is achieved at an average of 39.0 F g-1 from 25th to 60th cycle. At this stage, ions experience a constant rate of double-layer development where it is correlated to the rectangular shape of the CV plot for cycles < 50. As the cycle number exceeded 60, the value decreases to an average of 34.0 F g-1 from 70th to 100th cycle. Beyond 100th cycles, the value of C reduces to 19.7 F g-1 at 150th cycles. The decrement in the C value is due to the increment in the ion recombination process that obstructs the pathway of free ions to perform polarization at the surface of activated carbon. Table 2 shows various electrolyte systems-based EDLCs reported by other researchers.

Fig. 13
figure 13

CV plot of the EDLC at 10 mV s-1 for a 1st to 50th cycle and b 75th to 150th cycle. c Specific capacitance of the EDLC at selected cycles

Full size image
Table 2 Reported EDLCs with their respective polymer electrolyte and specific capacitance
Full size table

Conclusion

Fish skin gelatin-LiClO4–based natural polymer electrolytes have been successfully prepared using the solution casting method. XRD analysis verifies that the presence of LiClO4 has improved the amorphousness of the electrolyte as the highest conducting sample possessed the lowest crystallinity and crystallite size. The proof of polymer-Li+ interaction has been proven as the inclusion of LiClO4 has further shifted the peak of amide A and C–O bands via FTIR analysis. The inclusion of 25 wt.% LiClO4 has increased the conductivity value from 10-11 to 10-6 S cm-1. From the dielectric study, it is identified that the existence of LiClO4 has increased the dielectric constant of the electrolyte where it is harmonized with the pattern of ionic conductivity. The thermal stability of the polymer electrolytes is reduced by the addition of LiClO4. Ions have been found to be the primary charge carriers as 0.97 was achieved for the ionic transference number (ti). FESEM study showed that the surface of the electrolyte has turned to a denser structure. FS-LiClO4 was found to be potentially safe below 1.88 V from LSV analysis. The shape of the CV plot was changed throughout the 150 cycles and the specific capacitance obtained at the 1st cycle was 28.2 F g-1. Many aspects have to be considered to improve the performance of the EDLC, e.g., electrodes and electrolytes. For electrolytes, plasticizer and nanofiller can be added to reduce the resistance of the electrolyte. Hence, more ions are available to form a charge double layer. Modification of electrodes such as using less binder that increases the resistance of the electrode can increase the conductivity of the electrodes. The employment of carbon electrodes like carbon nanotubes, carbon aerogel, and graphite might enhance the performance of the EDLC.