Introduction

Electrical double-layer capacitor (EDLC) is one type of supercapacitors that consists of a pair of polarizable electrodes and does not exhibit Faradaic reactions over the potential range of operation [1]. The mechanism of EDLC is based on the phenomenon that an electric double-layer is formed at the boundary between the electrode and electrolyte [2]. Since no electrochemical reaction occurred, EDLC is an environmental friendly device. Although EDLCs are commercially available in the market, they use liquid electrolytes. Liquid electrolyte has many disadvantages such as being leak-prone, corrosive, and devices using liquid electrolyte may be bulky with high self-discharge rate. The use of liquid electrolytes also incurs high cost because of the need of a separator to prevent electrode-electrode contact.

To overcome the problem, high-conducting polymer electrolytes with good mechanical strength, electrochemical stability, low cost, and compatible with electrode materials are being developed. Poly(ethylene oxide), PEO, introduced by Wright and co-researchers in the seventies is the state of the art host polymer in polymer electrolyte research. PEO has high concentration of crystalline phases and thus when complexed with a salt exhibits low ionic conductivity. Hence, other polymers were investigated to replace PEO. Examples include poly(methyl methacrylate) (PMMA), poly(vinyl alcohol) (PVA), poly(acrylonitrile) (PAN), and poly(vinyl chloride) (PVC). Methyl cellulose (MC) has also been used as a polymer host in the preparation of polymer electrolytes [3–5]. MC is a water soluble polysaccharide that consists of ß (1 → 4) glucosidic units with methyl substituent in linear chains [6]. It can be prepared from the reaction of alkali-cellulose with dimethyl sulfate or methyl chloride [7]. According to some reports MC is an amorphous polymer [8, 9] while others claimed that MC is semi-crystalline [10]. Among biodegradable polymers, MC has good mechanical strength [11], and excellent film forming ability [12]. It is also cheap and non-toxic, but most importantly, it has lone pair electrons at its oxygen atoms where cation of a salt can loosely interact with the polymer. These features serve as motivation in the study of polymer electrolytes based on MC.

Based on our previous work [3–5], the sample 75 wt% MC-25 wt% NH4NO3 exhibits a room temperature conductivity of 2.1 × 10−6 S cm−1. This value, if not better, is comparable to other systems such as poly(vinyl pyrolidone)-ammonium acetate (1.5 × 10−6 S cm−1) [13], PEO-NH4F (1.1 × 10−6 S cm−1) [14], and PEG46NH4ClO4 (2.7 × 10−6 S cm−1) [15]. Although the conductivity is comparable with literature reports for solid polymer electrolytes (SPEs) doped with ammonium salt, ~10−6 S cm−1 but it is still insufficient for application in electrochemical devices such as EDLC. Hence, in this work we have plasticized 75 wt% MC-25 wt% NH4NO3 with poly(ethylene glycol) of molecular weight 200 g mol−1 (PEG200) to enhance conductivity value and use the membrane in EDLC fabrication. Results will be presented, discussed, and compared with reports found in the literature.

Experimental details

Polymer electrolyte preparation

MC-based polymer electrolytes were prepared using the solution casting technique. MC, procured from Sigma-Aldrich M0512 (viscosity 4 000 cP in 2 % aqueous solution at 20 °C), liquid poly(ethylene glycol) of molecular weight, 200 g mol−1 PEG200, was procured from Sigma-Aldrich P3015 and NH4NO3 from Ajax were all used as received. Details of sample preparation have been reported elsewhere [3–5]. To the highest conducting MC–NH4NO3 system, PEG200 was added and the composition optimized for highest room temperature conductivity.

X-ray diffraction (XRD)

X-ray diffraction patterns have been obtained using Siemens D5000 X-ray diffractometer in the 2θ range from 5° to 80° using Cu-Kα radiation. Operating current was 40 mA and operating voltage was 40 kV. The diffractogram of 75 wt% MC-25 wt% NH4NO3 electrolyte was compared with that of 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG electrolyte. These samples have the highest conducting electrolyte in the unplasticized and plasticized systems.

Fourier transform infrared (FTIR) spectroscopy

FTIR spectroscopy was carried out to determine polymer-salt complexation and to examine interaction(s) between component materials in the sample. The FTIR spectrum of pure MC, pure NH4NO3, pure PEG, MC–NH4NO3, MC-PEG, PEG-NH4NO3, and MC–NH4NO3-PEG were recorded using the Thermo Scientific (Nicolet iS10) spectrophotometer from 4,000 to 650 cm−1 with resolution 1 cm−1 at room temperature. Each spectrum runs 32 scans with correction against the background spectrum of air. Dry nitrogen was used for purging purposes in order to exclude the infrared active H2O and CO2 in the atmosphere.

EIS measurements

EIS measurements on the MC–NH4NO3 and MC–NH4NO3-PEG samples were carried out using the HIOKI 3531-01 LCR bridge in the frequency range 50 Hz to 1 MHz and in the temperature range from 298 to 373 K. The voltage amplitude was 10 mV. Before measurement, the samples were continuously dried in desiccators for a few days. Following Qian et al. [16] blocking electrodes were used in a two-electrode cell configuration to determine the bulk resistance of electrolytes. In this study, stainless steel (SS) electrodes were used as blocking electrodes and the samples were sandwiched between them. SS blocks ion transport and is an inert electrode for the polymer electrolyte. This insures the non-occurrence of passivation phenomenon [17]. Thickness of the polymer electrolyte samples is <100 μm. The impedance data were displayed in a Nyquist plot.

Linear sweep voltammetry (LSV)

Before applying the membrane in EDLC, the electrochemical stability window of the sample was investigated using the Autolab PGSTAT 12 potentiostat/galvanostat in conjuction with the General Purpose Electrochemical System (GPES) software Version 4.9.005. The sample was sandwiched between stainless steel electrodes. The scan rate was 1 mV s−1 and the potential applied was in the range from −2.5 to 2.5 V. The experiment was done at room temperature.

Electrode preparation

Normally the materials to prepare electrodes consist of a carbonaceous active material, binder and solvent. 81 wt% activated carbon (BP20) purchased from Kuraray Chemical Co., LTD was used as carbonaceous material. This carbon is based on phenol resin and has total surface area between 1,700 and 1,800 m2 g−1. The particle size is between 2 and 20 μm in diameter. 6 wt% carbon black (CB), which is also an activated carbon, was added to the mixture of BP20 and N-methyl-2-pyrrolidone (NMP). CB has been reported to reduce internal resistance of the electrodes [2]. According to Latham et al. [18], CB can also be used to improve the contact between electrode and electrolyte. In this study, MC was used as binder in the electrode following Chang et al. [19] and Bispo-Foncesca et al. [20]. 13 wt% MC was added in the electrode material but it was first dissolved in NMP. The mixture was stirred until homogeneous. In this study, the electrode was prepared using dip-coating method. Aluminum grid was used as current collector on which the electrode was coated. Aluminum grid was washed using ethanol prior to use. Apart from the activated carbon, binder and solvent, some polymer electrolyte was also added in liquid form to the electrode slurry.

Preparation of EDLC

The EDLC assembly is shown in Fig. 1. The electrode area was 4.5 cm2. The 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG electrolyte-cum-separator was sandwiched between two identical carbon electrodes. The electrode/electrolyte/electrode conjugate was clamped between rectangular Perspex plates to form the EDLC. Two EDLC cells were prepared. In one, the electrodes were coated with a PEG layer and in the other, the electrodes were without the PEG coating. The thickness of the PEG layer is at most ~70 μm.

Fig. 1
figure 1

a Photograph of the fabricated EDLC. b Schematic representation of the side view of EDLC

Full size image

Cyclic voltammetry (CV)

The capacitor performance was investigated using CV test. CV was performed using the Autolab PGSTAT 12 potentiostat/galvanostat in conjunction with the General Purpose Electrochemical System (GPES) software Version 4.9.005. The scan rate was 1 mV s−1 in the sweep region 0–0.85 V at room temperature.

Charge-discharge and self-discharge characteristics

The charge-discharge and self-discharge characteristics were carried out using the Neware battery cycler. For charge-discharge characteristics, the working voltage applied to EDLC was 0.85 V and the current was 1 mA. For self-discharge characteristics, the EDLC was first charged to 0.85 V, then discharged to 0 V and charged again to 0.85 V at a constant current 1 mA. The open circuit voltage was monitored for 2 h. All measurements were performed at room temperature.

Results and discussion

X-ray diffraction (XRD)

X-ray diffractogram of pure MC, 75 wt% MC-25 wt% NH4NO3, 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG and pure NH4NO3 are shown in Fig. 2. The diffractogram of pure MC film and pure NH4NO3 did not show any peak at 2θ = 8º. After addition of various concentrations of NH4NO3 from 5 to 30 wt%, it can be observed that, a new peak appeared at 2θ = 8º. The diffractograms for all MC–NH4NO3 samples have been reported in our previous papers [3–5]. This peak at 2θ = 8º indicates interaction between polymer and salt and it also exists in the diffractogram of samples containing PEG200. Kolhe and Kannan [21] have also shown the diffractogram of PEG (molecular weight 10,000 g mol−1) which do not show any peak at 2θ = 8º.

Fig. 2
figure 2

X-ray diffractogram for a pure MC film, b 75 wt% MC-25 wt% NH4NO3, c 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG, and d pure NH4NO3

Full size image

FTIR analysis

The functional groups were identified by monitoring the vibrational modes that are essentially the fingerprint of the materials under study [22]. The vibrational modes of pure MC, pure NH4NO3, and pure PEG are listed in Table 1. The vibrational modes of pure MC agrees well with that reported in the literature [7, 9, 12, 23–25]. The assignments for pure NH4NO3 are also comparable with Ramya et al. [26], Majid [27], Hashmi et al. [28], Singh et al. [29], and Marinković et al. [30]. The positions of the bands for pure PEG are also located at wavenumber positions similar to that reported in the literature [31–33].

Table 1 Vibrational modes (obtained in this study) and the corresponding assignments for pure MC, pure NH4NO3, and pure PEG
Full size table

The interactions between host polymer (MC), salt (NH4NO3), and plasticizer (PEG) were also investigated using FTIR spectroscopy. Figure 3 displays the infrared spectra for pure MC, pure NH4NO3, binary system-containing 75 wt% MC-25 wt% NH4NO3 to determine interaction between polymer and salt, pure PEG, 75 wt% MC-25 wt% PEG to determine interaction between polymer and plasticizer, 95 wt% PEG-5 wt% NH4NO3 to determine interaction between salt and plasticizer and the plasticized sample-containing 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG in the wave numbers region from 650 to 1,310 cm−1. Pure MC, Fig. 3(i) shows the vibrational mode of C–O stretching at 1,053 cm−1. After addition of 25 wt% NH4NO3, the band shifted to 1,051 cm−1 indicating interaction between the NH4 + cation and oxygen atom in C–O bond of MC as shown in Fig. 3(iii). The oxygen atom acts as electron donor as it possesses lone pair electrons. This, in fact increases amorphousness of the polymer electrolyte enabling continuous motion of the charge carriers that led to a maximum conductivity of 2.1 × 10−6 S cm−1 for the 75 wt% MC-25 wt% NH4NO3 polymer-salt binary system.

Fig. 3
figure 3

FTIR spectrum at room temperature of (i) pure MC, (ii) pure NH4NO3, (iii) 75 wt% MC-25 wt% NH4NO3, (iv) pure PEG, (v) 75 wt% MC-25 wt% PEG, (vi) 95 wt% PEG-5 wt% NH4NO3, (vii) 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG

Full size image

Before studying the effect of PEG plasticizer on 75 wt% MC-25 wt% NH4NO3 system, interaction between PEG and polymer and between PEG and the salt, were investigated. As can be seen in Fig. 3(v), the spectrum of 75 wt% MC-25 wt% PEG, shows a peak at 1,057 cm−1. This peak could have been formed due to the merger of peaks at 1,053 cm−1 (in the spectrum of pure MC) and at 1,061 cm−1 (in the spectrum of pure PEG), which have disappeared.

In the spectrum of 95 wt% PEG-5 wt% NH4NO3, the 1,061 cm−1 band of C–O–C vibrational mode shifted to 1,063 cm−1 as shown in Fig. 3(vi). This is attributed to the interaction between the ether group oxygen in PEG and the NH4 + cation of the salt. Hence, from these results it can be inferred that ion transport occurs between the oxygen atoms in MC from the shift of the C–O band at 1,053–1,051 cm−1 and between the oxygen atoms in PEG due to the shift in the C–O–C band from 1,061 to 1,063 cm−1. This implies that PEG can serve as transit site for ion conduction in the plasticized MC–NH4NO3–PEG system. In this system, the ion will require lesser energy for conduction [4] and the conductivity enhanced to 1.14 × 10−4 S cm−1. More details will be explained in next section.

EIS analysis and equivalent circuit model

The room temperature ionic conductivity of unplasticized electrolytes as a function of salt concentration and MC–NH4NO3 polymer electrolytes as a function of PEG concentration are displayed in Fig. 4.

Fig. 4
figure 4

Room temperature ionic conductivity of plasticized MC–NH4NO3 polymer electrolyte. Inset for unplasticized MC–NH4NO3 polymer electrolyte

Full size image

The ionic conductivity increases with increase of NH4NO3 concentration up to 25 wt% and decreases on further increase in salt concentration. The increase in the ionic conductivity with increasing salt content can be related to the increase in the number density of mobile charge carriers as explained in our previous paper [4]. Since the conductivity is insufficient for device application, low molecular weight PEG, M r  = 200, was chosen as plasticizer in order to enhance the conductivity. According to Park and Ruckestein [11], among the many plasticizers for MC, PEG is the most suitable for the preparation of proton conducting electrolytes. Since it can retain moderate moisture in the polymer and improves membrane flexibility without decreasing its mechanical strength. According to Ayranci and Tunç [34], PEG can reduce brittleness and increase flexibility, toughness, and tear resistance of edible films. However, increasing PEG content also increases water vapor permeability of the films. PEG plasticizer has various molecular weights, i.e., 200, 400, and 600 g mol−1. Park et al. [35] investigated relative permittivity and viscosity effects of PEG on room temperature conductivity and showed that low molecular weight PEG has higher relative permittivity and lower viscosity, which makes low molecular weight PEG more suitable for conductivity enhancement. Subban and co-researchers [36] also reported that PEG200 is a better plasticizer compared to the heavier molecular weight PEGs. Pradhan et al. [37] reported that plasticizing 95 wt% (PEO)25NaI-5 wt% DMMT composite polymer electrolyte with PEG200 enhanced the electrical conductivity by one order of magnitude at room temperature. DMMT stands for dodecyl amine modified montmorillonite. Srivastava and Chandra [38] prepared proton conducting polymer system based on poly(ethylene succinate) doped with NH4ClO4 and its conductivity is higher for plasticized sample using PEG200. The presence of PEG has resulted in new conducting pathways for the mobile ions.

Figure 5 depicts the Nyquist plots for 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG sample at various temperatures. In this experiment, only a spike was observed in the Nyquist plots for all temperatures studied, indicating that the current carriers are ions and the total conductivity is mainly the result of ion conduction [22]. Wu et al. [39] stated that, a straight line parallel to the imaginary axis in the low frequency region is caused by the effect of blocking electrodes and is also attributed to the non-homogeneity or roughness of the electrode-electrolyte interface.

Fig. 5
figure 5

Nyquist plot for 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG sample at various temperatures

Full size image

When an a.c. signal is applied to the system, the impedance, Z is defined as the ratio of the voltage to the current at a given frequency and consists of a real part, Z′ and an imaginary part, Z′′ with phase angle, θ [40]. Since EIS is based on Fourier transformation and only defined in linear systems theory, the impedance is expressed as a complex number:

$$ Z_{{\left( {j\omega } \right)}} = Z^{\prime} - jZ^{\prime\prime} $$

where Z′ is the real component and Z′′ is the imaginary component. The parameter j is \( \sqrt { - 1} \). In an alternate form, \( Z_{{\left( {j\omega } \right)}} \, = \,\left| Z \right|\,e^{j\,\varphi } \), where \( \left| Z \right| = \sqrt {\left( {Z^{\prime}} \right)^{2} + \left( {Z^{\prime\prime}} \right)^{2} } \). \( \theta = \tan^{ - 1} \left( {\frac{{Z^{\prime}}}{{Z^{\prime\prime}}}} \right) \).

According to Han and Choi [41], the equivalent circuit is an effective way to analyse the impedance data because it is fast, simple and provides the complete picture of the system. Since the spike in this study did not begin at the origin, the equivalent circuit is implied to consist of a resistor and constant phase element (CPE) in series connection as shown in Fig. 6.

Fig. 6
figure 6

Equivalent circuit

Full size image

CPE can be assumed as a leaky capacitor, to compensate for inhomogeneity in the sample [16]. CPE for the spike is due to the double-layer capacity of an inhomogeneous electrode surface. In this paper, the equivalent impedance for the 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG highest conducting plasticized sample can be expressed with the resistance, R (due to electrolyte) in series with the CPE impedance, Z cpe which can be expressed as:

$$ Z_{\text{total}} = R + \frac{k}{{\omega^{p} }}\left[ {\cos \left( {{\raise0.7ex\hbox{${p\pi }$} \!\mathord{\left/ {\vphantom {{p\pi } 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}} \right) - j\sin \left( {{\raise0.7ex\hbox{${p\pi }$} \!\mathord{\left/ {\vphantom {{p\pi } 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}} \right)} \right] $$

The second-term on the right-hand side represents the impedance of the constant phase element, Z cpe. Here k is C −1 the inverse of capacitance, C and p is related to the deviation of the spike from the horizontal axis. When p = 1, \( Z_{\text{total}} = R - \frac{j}{\omega C} \). When p = 0, the constant phase element is a perfect resistor, Z total is frequency independent. When 0 < p < 1, CPE acts in a way intermediate between a resistor and a capacitor. When p = 0.5, the impedance is known as Warburg impedance.

Using the equivalent circuit, the impedance data was analyzed by superimposing the best fit of the dispersion data by trial and error. As shown in Fig. 5, the impedance of the assumed equivalent circuit fits the experimental impedance data very well. Table 2 lists the value for the circuit elements of the sample together with its conductivity.

Table 2 Parameter for circuit elements
Full size table

The electrode-electrolyte interface can be considered as a double-layer if blocking electrodes were used in the impedance measurement [39]. The spike in this study is inclined at an angle <90° and the value of p lies between 0 and 1. This reveals that the samples have a resistive and capacitive behavior. From Table 2, it can be observed that, the value of p increases as temperature increases from 298 to 373 K. It is obvious that, the sample becomes more capacitive than resistive as temperature increases. The bulk resistance, which was obtained from the intercept of the plot with the real impedance axis decreases when temperature increases. The decrease in bulk resistance enhances conductivity and conductivity increases with temperature. Ramya et al. [26] reported that the spike is characteristic of a blocking double-layer capacitance and should be of the order of μF. The capacitance is due to electrode polarization. The value of capacitance obtained from simulation studies in this study is comparable with the results as reported in the literature [26]. The capacitance is also observed to increase with temperature up to 343 K implying suitability for use in EDLC.

Linear sweep voltammetry

Figure 7 depicts linear sweep voltammogram for the highest conducting plasticized MC–NH4NO3 polymer electrolyte. From the figure, it can be seen that the current onset is detected at 2.4 V for the plasticized sample, 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG. The current onset is assumed to be the polymer electrolyte film breakdown voltage [42]. To test performance of any device, the electrochemical stability of the sample must be known prior to charge-discharge cycling test. The breakdown voltage is important to prevent the electrolyte from being destroyed. Since in this study, the working voltage used was 0.85 V, the electrolytes should be electrochemically stable for use in the EDLCs.

Fig. 7
figure 7

Linear sweep voltammetry (LSV) curve of the 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG

Full size image

Cyclic voltammetry

To understand the performance of EDLC such as electrochemical behavior and the nature of charge storage at the individual interfaces in the anodic and cathodic regions, cyclic voltammetry was carried out [43]. Figure 8 depicts the cyclic voltammograms of the fifth scan for EDLC fabricated using uncoated and coated electrodes. The electrode size is 1.5 × 3 cm2 in the range 0–0.85 V and the scan rate is 1 mV s−1. The shape of the voltammograms for the EDLCs with coated and uncoated electrodes is close to that of a rectangle. The electrolyte and the electrical contacts are responsible for the resistance of the EDLC. In the present polymer electrolyte system, ionic transport takes place in liquid phase entrapped in the polymer matrix. Therefore, the resistance value is expected to be high. Inset in Fig. 8 shows efficiency of the EDLC with PEG-coated electrodes for 100 cycles.

Fig. 8
figure 8

Cyclic voltammogram of the EDLC fabricated using uncoated and coated electrodes. Inset efficiency of EDLC with PEG-coated electrodes for 100 cycles

Full size image

Specific capacitance, C can be determined from cyclic voltammogram using the equation [44]:

$$ C = \frac{i}{vm} $$

Here, i is the average current (A), v is the scan rate (V s−1), and m is the weight of the electrode (g). The specific capacitance is listed in Table 3 for the two EDLCs.

Table 3 Specific capacitance values for the EDLC obtained by cyclic voltammogram
Full size table

The capacitance obtained with the 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG electrolyte is higher than that fabricated EDLC using the unplasticized electrolyte as reported in our previous paper [3]. From CV test, the voltammograms imply that, the fabricated EDLC using the plasticized 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG sample as electrolyte and activated carbon electrode depicts the capacitive behavior but the performance of the EDLC can still be improved to make the EDLC store and deliver charge in a shorter time. It can be observed that the EDLC with PEG-coated electrode showed higher charge and discharge capacitance. This is attributed to better contact between the electrode and electrolyte and should lead to a lower equivalent series resistance (ESR) compared to the EDLC with uncoated electrode.

Charge-discharge characteristics

For EDLC performance, working voltage is largely dependent on the electrolyte breakdown voltage, while internal resistance and capacitance depends on the electrode resistance and conductivity of electrolyte. In this study, the voltage limits were set between 0 and 0.85 V and constant current of 1 mA was applied during charge and discharge. As reported in our previous paper [3], unplasticized MC-based electrolyte (75 wt% MC-25 wt% NH4NO3) with conductivity 2.10 × 10−6 S cm−1 when used in a capacitor delivers low capacity of ~1–2 F g−1. In this study, since the plasticized sample (63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG) has conductivity 1.10 × 10−4 S cm−1 (about two orders of magnitude higher), the capacitance of the fabricated EDLC should be higher. After studying the effect of PEG coating on the EDLC performance using CV, we now use the charge–discharge characteristics to evaluate the performance of the EDLC. Figure 9 shows the charge–discharge profile of the EDLCs fabricated.

Fig. 9
figure 9

Charge-discharge profile of the EDLCs for a uncoated and b coated electrodes

Full size image

From Fig. 9, it can be observed that, the discharge characteristics of the EDLC using coated electrodes is more linear compared to EDLC using uncoated electrodes. The linear curve shows the capacitive behavior of EDLC. The initial sudden jump/drop in the voltage while charging and discharging is due to ohmic loss across the internal resistance, equivalent circuit resistance (ESR) in electrode and electrolyte and represents the resistive behavior of EDLC [45]. ESR can be calculated from the voltage drop or IRdrop while discharging the EDLC as shown by the equation:

$$ {\text{ESR}} = \frac{{{\text{IR}}_{\text{drop}} }}{i} $$

Here i is the discharge current and IRdrop is the voltage drop upon discharge. The specific capacitance can be evaluated from the gradient of the linear portion of the charge and discharge characteristics using the equation:

$$ C_{\text{s}} = \frac{i \times \Updelta t}{m \times \Updelta V} = \frac{i}{{m \times \left( {{\raise0.7ex\hbox{${\Updelta V}$} \!\mathord{\left/ {\vphantom {{\Updelta V} {\Updelta t}}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{${\Updelta t}$}}} \right)}} $$

Again i is the discharge current in amperes and ΔV/Δt is the gradient of the linear curve from charge-discharge profile (V s−1). Gradient was obtained using trendline on the linear portion of the charge and discharge curves. The regression value was 0.99. Efficiency is one of the important parameters in EDLC performance. Efficiency, η was determined using the equation:

$$ \eta = \frac{{C_{\text{d}} }}{{C_{\text{c}} }} \times 100\% = \frac{{t_{\text{d}} }}{{t_{\text{c}} }} \times 100\% $$

where C d and C c are discharge and charge capacitance, respectively. t d represents the time for galvanostatic discharging and t c represents the time for galvanostatic charging. Specific energy, E delivered during the discharge was calculated using the equation [46]:

$$ E = \frac{1}{2}CV^{2} $$

Here, C is discharge capacitance and V is working voltage, 0.85 V. Power capability, P for EDLC can be determined using the equation:

$$ P = \frac{{V^{2} }}{4R} $$

where V is working voltage, 0.85 V and R is internal resistance or ESR. Table 4 lists the value of equivalent series resistance (ESR), charge and discharge capacitance, efficiency, specific energy and power capability from EDLC studied evaluated from charge-discharge profile.

Table 4 Characteristic parameters for EDLC cells at 4th cycle for EDLC using uncoated and coated electrode
Full size table

From Table 4, it can be deduced that, EDLC using coated electrodes give better performance compared to uncoated electrodes since, charge and discharge capacitance, efficiency, specific energy, and power delivered from EDLC using coated electrodes are higher than EDLC with uncoated electrodes. CV results showed that ESR of the EDLC with PEG-coated electrode is lower to that of EDLC with uncoated electrodes and this has been confirmed from charge-discharge studies. Table 5 lists the comparison of the EDLC in this study with other reported studies. Exact comparison between results from this study and those of literature may not be possible because of different carbon materials and polymer electrolytes used. The basis of comparison is that in all the works quoted, SPEs are used.

Table 5 Capacitance of EDLCs using different solid polymer electrolytes
Full size table

Self-discharge characteristics

Self-discharge is one of the most crucial factors to determine the performance of the EDLC. It can be determined by measuring the open circuit voltage of the EDLC as a function of time. The EDLC cell was charged, discharged, and charged again at a constant current. When charging current is stopped, voltage decreased. This is self-discharge [50].

Figure 10 demonstrates the voltage drop upon discharge at 1 mA compared to self-discharge (open circuit voltage) for EDLC using uncoated electrodes. It can be observed that the voltage initially dropped to 0.67 when the EDLC was discharged at 1 mA and 0.77 V for self-discharge.

Fig. 10
figure 10

Self-discharge profiles resulting from charging EDLC to 0.85 V using uncoated electrodes

Full size image

Depicted in Fig. 11 is voltage drop upon discharge at 1 mA and open circuit voltage (OCV) for EDLC cell fabricated using PEG-coated electrode. The instantaneous voltage drop is related to internal resistance in EDLC. It can be observed that, the initial voltage drop for this cell when EDLC was discharged at 1 mA is 0.79 and 0.82 V for self-discharge. The voltage retention is better than that of the uncoated electrode EDLC. This again shows that PEG coating can upgrade the EDLC performance.

Fig. 11
figure 11

Self-discharge profile resulting from charging EDLC to 0.85 V using coated electrodes

Full size image

It may be concluded that the MC-based electrolyte can be used as an electrolyte in an EDLC. However, more work needs to be carried out to improve performance of EDLC using MC-based electrolytes.

Conclusions

FTIR studies show that, MC, NH4NO3, and PEG have interacted from the shifting of C–O stretching mode. From EIS measurements, the 63.75 wt% MC-21.25 wt% NH4NO3-15 wt% PEG sample can be represented by the equation:

$$ Z_{\text{total}} = R + \frac{k}{{\omega^{p} }}\left[ {\cos \left( {{\raise0.7ex\hbox{${p\pi }$} \!\mathord{\left/ {\vphantom {{p\pi } 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}} \right) - j\sin \left( {{\raise0.7ex\hbox{${p\pi }$} \!\mathord{\left/ {\vphantom {{p\pi } 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}} \right)} \right] $$

The specific discharge capacitance is 38 F g−1 obtained from the EDLC with PEG-coated electrodes. From LSV test, it was found that MC-based polymer electrolytes have high-electrochemical stability, 2.4 V and safe to use as a membrane in EDLC that is charged to below the decomposition voltage. CV voltammogram shows that EDLC with PEG-coated electrode has higher capacitance compared to EDLC with uncoated electrodes. EDLC using coated electrode with PEG gives better performance compared to uncoated electrode implying that coating the electrode with PEG helps to improve electrode-electrolyte contact.