Supercapacitor Using Polypyrrole and Carbon Nanotube Composite as Electrodes

Abstract

Electrodes and electrolytes are critical components for the performance of supercapacitors. In this study, supercapacitors with different interfaces were assembled using polypyrrole (PPy) or a polypyrrole–carbon nanotube (PPy-CNT) composite as active materials, and dimethyl sulfoxide (DMSO) and sodium chloride (NaCl) were used as electrolytes. Electrochemical measurements were obtained by cyclic voltammetry (CV) at a scan rate of 20 mV/s and galvanostatic charge–discharge (GCD) measurements at a scan rate of 50 µA/s. The results suggest that the supercapacitor with a PPy-CNT composite and NaCl electrolyte has promising electrochemical characteristics.

1. Introduction

Renewable energy sources, such as hydroelectric, wind, solar, geothermal, and marine, among others, represent an alternative to the overexploitation of fossil energy sources. Nevertheless, electricity generation from renewable energy sources is intermittent, requiring the coupling of energy storage devices such as batteries, conventional capacitors, and supercapacitors (SCs) [1]. Batteries are used primarily for their ability to store substantial amounts of energy, but they are incapable of the rapid release of stored energy due to their low discharge and charge rates [1,2,3]. A promising alternative to batteries is SCs [1], devices with exceptional storage properties such as fast charge–discharge response rates [1,2,4], high power density, high energy density [5,6,7], and high cycle stability, and they are capable of withstanding high energy demands [1,2,8] while having low manufacturing costs and exhibiting environmental friendliness [5]. Supercapacitors can be used as power supplies in different applications such as medical electronics, power backup, consumer electronics, electric and hybrid vehicles, portable electronics, and digital communications, among others [9,10,11,12].

Two electrodes separated by a solid separator membrane [13] with non-conductive electric characteristics, chemical resistance to electrode materials and electrolytes, ion permeability, and simple wetting properties [14] compose the assembly of SCs. The electrode materials and electrolytes of SCs play a determining role in the performance of these storage devices. Active materials with promising capacitive characteristics such as conductive polymers, carbon nanostructures, and composites [9,15] can constitute the electrodes of SCs. Among conductive polymers, PPy is widely used due to its ease of synthesis [16], higher specific capacitance, steady performance, low toxicity, environmental stability [17,18,19,20], and low electrical resistivity [21]. In addition, carbon nanotubes are interesting carbon nanostructures with excellent electronic properties [8], as well as an affinity for most electrolytes and ability to store substantial amounts of energy, capable of improving capacitive properties in storage devices [21,22,23,24,25,26]. Both PPy and CNT can form PPy-CNT composites used as SC electrodes with a chemical–physical storing mechanism [14,27]. PPy-CNT composites have shown improved electrochemical properties in terms of cycle stability, power densities, and rate capabilities, and they can mitigate the drawbacks of the low specific capacitance of CNT and poor cycle stability of PPy [8].

The electrolytes used in supercapacitors are classified as aqueous, organic, and ionic liquids, which can be used in acidic, alkaline, or neutral media [14]. Electrolytes with high ionic conductivity and operational safety, such as those based on sodium ions, are particularly suitable for sustainable energy storage applications [28,29]. Nevertheless, organic electrolytes are used in SCs to achieve higher charge voltage and energy capacity as they have high working voltage (cell voltage) and directly influence the increase in capacitance [30]. DMSO is used as an organic electrolyte because it enhances the electrochemical stability window and suppresses the corrosion of the metal current collector [31,32]. The charge storage performance of a supercapacitor is improved by treatment with DMSO [33]. However, an additional advantage of NaCl ionic electrolytes is that they are nontoxic, abundant, and environmentally safe, while DMSO presents safety and handling considerations.

Some studies have used PPy, CNT, and PPy-CNT hybrids for their application as active electrode materials in supercapacitors [10,34,35,36]. Farbob et al. synthesized a PPy-CNT nanocomposite as an active material for supercapacitor electrodes, obtaining a capacitance of 123 F/g [36]. Saini et al. formed a composite of reduced graphene oxide and PPy, reaching a capacitance of 4.66 and 6.11 F/g for different proportions of PPy and graphene oxide [37]. Hernández-Cortés et al. obtained a capacitance of 450 F/g with a PPy-MWCNT composite electrode [38].

Energy storage in supercapacitors is based on two primary mechanisms, electrostatic and pseudocapacitive. The electrostatic mechanism, typically observed in carbon-based materials such as activated carbon, carbon nanotubes, and graphene [39], involves the accumulation of charges at the electrode/electrolyte double-layer interface, where the separation of charges occurs without any faradaic reactions. Conversely, the pseudocapacitive mechanism, commonly associated with conducting polymers and transition metal oxides, arises from fast and reversible redox processes at the electrode surface.

Based on these charge storage mechanisms, the electrochemical performance of supercapacitors can be evaluated through key parameters such as capacitance ($C$), energy density ($E$), power density ($P$), cycle stability, and lifetime [29,40]. Energy density is defined as the total amount of charge per unit mass or volume that the device can store, while power density is related to the charge–discharge rate; both parameters are inversely proportional in energy storage devices. In this study, the electrochemical properties of supercapacitors employing PPy, CNT, or a PPy-CNT composite as electrodes and DMSO or NaCl as electrolytes are investigated. Their electrochemical behavior was characterized using CV and GCD measurements.

2. Materials and Methods

2.1. Construction of Supercapacitor

The supercapacitor consists of two electrodes connected to collectors and a separator between the electrodes and electrolyte. The supercapacitor assembly was composed of electrodes consisting of PPy or PPy-CNT and DMSO or NaCl as the electrolyte, and a polytetrafluoroethylene (PTFE) membrane with a pore size of 0.45 μm was used as the separator. In particular, the CNT were obtained by the electric arc discharge method [40], and the PPy film was synthesized by electrochemical polymerization [16]. Pyrrole (Py, 98%), DMSO (99.9%), ethanol (C₂H₅OH, 99.5%), DMSO (99.9%), and NaCl (99.5%) were purchased from Sigma Aldrich. Sodium sulfate (Na₂SO₄, 99.8%) was procured from Fermont PA Cert. PPy was synthesized using a 0.1 M solution of pyrrole (Py) as a monomer and 0.05 M sodium sulfate (Na₂SO₄) as an oxidant in an electrochemical polymerization cell containing two stainless steel electrodes. During synthesis, a voltage of 5 V and a current of 40 mA were applied at room temperature for 30 min. The PPy film was removed from the cathode and washed with ethanol and distilled water.

The PPy film was ground in an agate mortar until the formation of fine powder. SC interfaces were formed using a fine powder of PPy or the PPy-CNT composite, dissolved in an electrolyte. The PPy-CNT composite was formed by mixing CNT powder with PPy powder at different ratios. An organic electrolyte was formed with DMSO with ethanol (C₂H₅OH) at a volume ratio of 1:1, and the other electrolyte was 0.1 M NaCl. Each separator was immersed in the electrode/electrolyte interface, and then they were mixed for 60 min using a Cole Palmer 8892 ultrasonic bath. The separator was weighed before immersion and after agitation to determine the active material mass deposited on it. Subsequently, the supercapacitor was assembled by inserting the separator impregnated with the electrode/electrolyte interface between two 20 mm square copper sheets that were used as collector terminals. After, each assembly was introduced into a thermoplastic cell made by 3D printing, which functions as a support to build the supercapacitor.

Four electrode/electrolyte interfaces were then obtained: polypyrrole in DMSO (PPy/DMSO), PPy in NaCl (PPy/NaCl), carbon nanotube–PPy in DMSO (PPy-CNT/DMSO), and carbon nanotube–PPy in NaCl (PPy-CNT/NaCl).

Figure 1 shows the assembly sequence of the SCs consisting of the synthesis of the active materials as electrodes, formation of the electrode/electrolyte interfaces, impregnation of the separator with each of the electrode/electrolyte interfaces, placement of the separator impregnated with the electrode/electrolyte interface between two current collectors for assembly formation, and placement of the assemblies in a support cell [41].

2.2. Characterization of Materials

The morphology of CNT and PPy was analyzed using a JEOL JSM6610 LV, JEOL Ltd. Tokyo, Japan, scanning electron microscope (SEM). The interaction between the electrolyte, PPy, and carbon nanotubes was determined using a JASCO FT-IR-4X, JASCO Inc., Tokyo, Japan, Fourier Transform Infrared spectrophotometer (FTIR) in the wavenumber range of 4000–400 cm⁻¹ at 16 scans. The importance of FTIR lies in its ability to corroborate the correct bonding of the elements in the micromembrane and the active electrode/electrolyte interface, which ensures the correct electroconductivity, mechanical performance, ionic conductivity, and even good self-healing efficiency in the hydrogel. Galvanostat/potentiostat equipment was used to obtain its capacitance, energy and power densities, and charge–discharge cycle behavior. These parameters are crucial to determine the effectiveness of the supercapacitor in storing and delivering energy.

2.3. Electrochemical Characterization

CV and GCD, using Autolab PGSTAT302N, Metrohm Autolab B.V., Utrecht, Netherlands, galvanostat/potentiostat equipment, were evaluated for the electrochemical characterization of the SC with different assemblies. CV was evaluated in an electrochemical two-electrode system at a scan rate of 20 mV/s$(\nu$). CV areas were used by determining the $C_T$ and $C_{sg}$ of the SC with different electrode/electrolyte interfaces. $C_T$ was determined by the means of Equation (1), and $C_{sg}$ was calculated by Equation (2). The CV areas of the PPy/DMSO, PPy/NaCl, PPy-CNT/DMSO, and PPy-CNT/NaCl interfaces were calculated by integrating over the full CV curve and determined at a voltage scan rate of 20 mV/s.

$$C_T={\displaystyle \frac{2{\int }_{V_b}^{V_a}idV}{\nu \Delta V}}$$

(1)

$$C_{sg}={\displaystyle \frac{C_T}{m}}$$

(2)

where $C_T$ is the total capacitance, ${\int }_{V_b}^{V_a}idV$ is the cyclic voltammetry area, $\nu$ is the voltage scan rate, $\Delta V$ = $\left(V_a-V_b\right)$ is cell voltage, $C_{sg}$ is the specific gravimetric capacitance, and $m$ is the mass of the active material.

$C_T$ was used by calculating $E$, determined with Equation (3). Additionally, $R_S$ and $m$ were used by determining $P$, calculated by Equation (4), using an electrochemical two-electrode system. The $R_S$ value was obtained using the Electrochemical Impedance Spectroscopy (EIS) curve by considering the closest point to the real impedance axis.

$$E={\displaystyle \frac{1}{2}}\left(C_T\Delta V^2\right)$$

(3)

$$P={\displaystyle \frac{1}{4}}\left({\displaystyle \frac{\Delta V^2}{m{R}_S}}\right)$$

(4)

where $E$ is energy density, $C_T$ is the total capacitance, $\Delta V$ is cell voltage, $P$ is power density, $m$ is the weight of the active material, and $R_S$ is the equivalent series resistance or real impedance. The actual impedance was taken from the Frequency Response Analysis or Electrochemical Impedance Spectroscopy curve, which is the equivalent of the series resistance. In the case of PPy-CNT/NaCl interfaces, impedance was determined in the range of 0.5 to 2 Ω, depending on the percentage of PPy vs. CNT, while in PPy-CNT/DMSO, it was determined at 128 Ω, the value that is applied to calculate power density.

GCD measurements were performed to determine the cycling stability and SC degradation behavior in the $\Delta V$ sweep range from 0 to 1 V at a current scan rate of 1 A/s and a sampling of 50 cycles. The SC degradation time is important for determining working conditions in terms of the posterior coupling of the SC with other devices. Similarly, GCD was used to calculate $C_G$ using Equation (5). Likewise, GCD specific capacitances $C_{Gsg}$ and $C_{GsA}$ were calculated using Equations (6) and (7), respectively [42,43].

$$C_G={\displaystyle \frac{I\cdot \Delta t}{\Delta V}}$$

(5)

$$C_{Gsg}={\displaystyle \frac{C_G}{m}}$$

(6)

$$C_{GsA}={\displaystyle \frac{C_G}{A}}$$

(7)

where $C_G$ = galvanostatic charge–discharge capacitance; $I$ = applied current; $\Delta t$ = discharge time rate; $\Delta V$ = cell voltage; $C_{Gsg}$ = gravimetric specific galvanostatic charge–discharge capacitance; $m$ = mass of active material; $C_{GsA}$ = areal specific galvanostatic charge–discharge capacitance; $A$ = area of electrode.

3. Results

3.1. Structural Characterization

The morphology of the PPy film and CNT powder as active materials was realized by SEM. In Figure 2a, the morphology of the PPy film before the grinding process is presented, showing the characteristic hemispherical appearance of PPy of around 8 µm. The morphology of CNT powder is shown in Figure 2b, showing that it has a diameter of around 20 nm.

The identification of the functional groups of all samples by IR spectroscopy is shown in Figure 3. The IR spectrum of PPy-CNT/NaCl showed a band at 3300 cm⁻¹, characteristic of the N-H stretching vibration of the PPy ring, whereas the peak at 3250 cm⁻¹ corresponds to O-H stretching vibration due to atmospheric humidity. The bands at 2850 and 2800 cm⁻¹ correspond to the C-H of PPy rings. The characteristic stretching vibration of the C=C bond in PPy is observed at 1650 cm⁻¹. This absorption band overlaps with the characteristic peak of CNT at the same wavenumber, indicating strong interfacial interactions between both materials. Such overlapping suggests possible π-π interaction that could facilitate charge transfer across PPy-CNT/NaCl [44]. Additional vibrations in PPy are observed at 1450 cm⁻¹ and 1375 cm⁻¹, which correspond to C-C and C-N bonds, respectively. The interaction of the NaCl electrolyte could be depicted from the C–Cl and C–O links found at around 1000 cm^{- 1} and at 1100 cm⁻¹ [6,13,45,46].

Likewise, the IR spectrum of PPy/DMSO exhibits a broad band at 3250 cm⁻¹, corresponding to the O-H stretching vibration of the hydroxyl group (Figure 3). The band at 2900 cm⁻¹ is attributed to C-H stretching vibration, which could indicate strong bonds between PPy and the electrolyte. The bands at 1650 cm⁻¹ and 1450 cm⁻¹ are assigned to C=C and C-C bonds, respectively. Meanwhile, the peak at 1300 cm⁻¹ is assigned to C-N stretching. Additionally, adsorption bands are observed at 1020 cm⁻¹, associated with the S=O bond of DMSO [45]. The spectrum of the PPy-CNT/DMSO interface shows adsorption bands that coincide with those of the IR spectra of PPy/DMSO, indicating that the incorporation of CNT does not significantly alter the vibrational modes of PPy or DMSO. This similarity suggests the good dispersion of the CNT within PPy, which is essential for efficient charge transport.

3.2. Electrochemical Characterization of Supercapacitor

Figure 4 shows the CV curves for the PPy/DMSO, PPy-CNT/DMSO, PPy/NaCl, and PPy/NaCl interfaces obtained at a scan rate ($\nu$) of 20 mV/s. The PPy/DMSO interface (Figure 4a) exhibits a narrow and slightly asymmetrical CV curve, characteristic of electrostatic behavior with limited pseudocapacitive contribution [47]. Conducting polymers, such as PPy, are not fully compatible with organic electrolytes, like DMSO, which impede efficient charge transfer through the polymer backbone [39].

In contrast, the PPy-CNT/DMSO interface (Figure 4b) displays a broader and more symmetrical curve, suggesting an enhanced capacitive response. The CNT contribute primarily to the electrostatic process, while the PPy chains provide minor pseudocapacitive contributions. Both the PPy/DMSO and PPy-CNT/DMSO interfaces reached a maximum current of 3.5 × 10⁻⁴ A. The PPy/NaCl interface (Figure 4c) presents an asymmetrical curve with a stronger capacitive response than PPy/DMSO, owing to the higher mobility of NaCl within the polymer matrix. This interface achieved a maximum current of 1.6 × 10⁻³ A. The PPy-CNT/NaCl interface (Figure 4d) shows a wide asymmetrical CV curve, indicative of combined electrostatic and pseudocapacitive charge storage mechanisms. This response is attributed to the interaction between PPy, CNT, and NaCl, reaching a maximum current of 6 × 10⁻³ A, which is superior to the other interfaces. The behavior of PPy-CNT/NaCl resembles that of the pseudocapacitive mechanism, as conducting polymer electrodes respond better to inorganic electrolytes, specifically PPy with NaCl [39].

In addition, the effect of the ratio of PPy to CNT as active materials was analyzed. The PPy-CNT/NaCl interface was determined with different rates of weight (75:25, 50:50, 25:75% wt.), while the PPy-CNT/DMSO interface remained constant (50:50% wt.). The electrochemical characteristics of interfaces are presented in

Table 1. Electrochemical characteristics of SC at 20 mV/s.

Electrode	Electrolyte	${\mathit{C}}_{\mathit{T}}$ (F)	$\mathit{W}$ (mg)	${\mathit{C}}_{\mathit{s}\mathit{g}}$ (F/g)	$\mathit{E}$ (W·h/kg)	$\mathit{P}$ (W/kg)
PPy	DMSO	0.0196	1.83	1.1	9.8	0.2
PPy50-CNT50	DMSO	0.0180	1.08	1.7	9.0	1.8
CNT	NaCl	0.0649	1.4	46.4	32.5	357.1
PPy25-CNT75	NaCl	0.0583	1.5	40.2	29.2	229.9
PPy50-CNT50	NaCl	0.0876	1.5	58.4	43.8	166.7
PPy75-CNT25	NaCl	0.0456	1.6	39.2	22.8	129
PPy	NaCl	0.0318	1.6	19.9	15.9	78.1

. The $C_T$, $C_{sg}$, and $E$ values were higher at the PPy₅₀-CNT₅₀/NaCl interface, indicating the good influence of PPy-CNT and the NaCl electrolyte. The mass of the PPy-CNT/NaCl electrode (excluding membrane weight) ranged from 1.4 mg to 1.6 mg, depending on the percentage of PPy vs. CNT. For PPy-CNT/DMSO, it was 10.8 mg. The thickness range of the PPy-CNT powder deposited on the separator was estimated to be between 20 and 40 µm.

When comparing the PPy₅₀-CNT₅₀/NaCl interface with other SCs based on PPy-CNT electrodes [20,48,49,50], the values are consistent with those reported in this work. In comparison with KCl (29.7 F/g) [48], the PPy₅₀-CNT₅₀/NaCl interface (58.3 F/g) shows that the use of NaCl as an electrolyte significantly influences the $C_{sg}$ value. In contrast, the PPy₅₀-CNT₅₀/NaCl interface shows lower gravimetric capacitance than those reported in [49,50] (586 F/g and 510 F/g, respectively) where H₂SO₄ was used as the electrolyte. Although the acid electrolyte provides higher conductivity, electrochemical stability, and adsorption efficiency and lower internal resistance, it is toxic and environmentally harmful. In comparison, NaCl is a nontoxic and eco-friendly alternative that facilitates handling and disposal during the manufacture of supercapacitors.

In addition, $C_G$ and $C_{Gsg}$ measurements for the CNT/NaCl, PPy₂₅-CNT₇₅/NaCl, PPy₅₀-CNT₅₀/NaCl, PPy₇₅-CNT₂₅/NaCl, and PPy/NaCl assemblies were calculated using GCD at a scan rate of 50 µA/s. Galvanostat/potentiostat equipment supplied an $I$ of 5 mA. $\Delta t$ was calculated by measuring the time from the peak voltage to the minimum voltage, obtaining 24 s for CNT/NaCl, 46 s for PPy₂₅-CNT₇₅/NaCl, 62 s for PPy₅₀-CNT₅₀/NaCl, 84 s for PPy₇₅-CNT₂₅/NaCl, and 104 s for PPy/NaCl. The $\Delta V$ applied was 1 V, and the $m$ used was around 1.5 mg. $C_{GsA}$ was calculated with an electrode of 1 cm². The electrochemical parameters obtained using GCD measurements are shown in

Table 2. Electrochemical characteristics of SC at 50 µA/s.

Electrode	Electrolyte	${\mathit{C}}_{\mathit{G}}$ (F)	${\mathit{C}}_{\mathit{G}\mathit{s}\mathit{g}}$ (F/g)	${\mathit{C}}_{\mathit{G}\mathit{s}\mathit{A}}$ (F/m2)	${\mathit{E}}_{\mathit{G}}$ (W·h/kg)	$\mathit{P}$ (W/kg)
CNT	NaCl	0.120	86	1200	60	357.1
PPy25-CNT75	NaCl	0.230	159	2300	115	229.9
PPy50-CNT50	NaCl	0.310	207	3100	155	166.7
PPy75-CNT25	NaCl	0.420	271	4200	210	129
PPy	NaCl	0.515	322	5150	258	78.1

. An increase in the $C_{sg}$, $C_{Gsg}$, $C_{GsA}$, and $E_G$ values was observed as the PPy ratio increased.

The power and energy density relationship of the SC assemblies is shown in the Ragone diagram (Figure 5a). The PPy₅₀-CNT₅₀/DMSO electrode shows a significant increase in P relative to the PPy/DMSO electrode due to the addition of carbon nanostructures. Likewise, when comparing the CNT electrode and PPy-CNT electrodes at the 25:75, 50:50, and 75:25 ratios, all dissolved in NaCl as an electrolyte, and a decrease in $P$ is observed when the ratio of PPy increases. However, at the PPy₅₀-CNT₅₀/NaCl interface, a significant increase in $E$ and $P$ is obtained, due to an equilibrium interaction between the active materials and the electrolyte. These values are higher than those of PPy₅₀-CNT₅₀/DMSO due to the use of an inorganic electrolyte. NaCl provides improved power and energy densities due to its lower internal resistance and higher ionic conductivity, facilitating rapid charge transfer.

The energy and power densities obtained for CNT, PPy, and PPy-CNT electrodes in the NaCl electrolyte are consistent with previously reported literature values [38,51,52]. The CNT/NaCl interface (60 W·h/kg and 357 W/kg) has higher E and P values than the AC-MWCNT/NaSO₄ interface (activated carbon–multiwalled carbon nanotubes deposited in stainless steel/sodium sulfate) (17.64 W·h/kg and 124.99 W/kg) and AC-MWCNT_NiF/NaSO₄ interface (activated carbon–multiwalled carbon nanotubes deposited in Nickel foam/sodium sulfate) (4.9 W·h/kg and 199.18 W/kg [51]), indicative of the influence of the inorganic electrolyte on both electrical parameters. The PPy₇₅-CNT₂₅/NaCl (155 W·h/kg and 166.7 W/kg), PPy₅₀-CNT₅₀/NaCl (210 W·h/kg and 129 W/kg), and PPy/NaCl (258 W·h/kg and 78 W/kg) interfaces have a higher E value and lower P value than PPy-MWCNT/H₂SO₄ (PPy–multiwalled carbon nanotubes/sulfuric acid) (12.5 W·h/kg and 200 W/kg) [38] and PPy-GO-MWCNT/ZnCl₂ (117 W h/kg and 340 W/kg) [52], reflecting the effect of electrolyte type on supercapacitor performance.

From the graphs of the galvanostatic charge and discharge cycles of the electrodes, the one that obtained the best result (longer life cycle time) is PPy₅₀-CNT₅₀/NaCl, which is why it is shown in Figure 5b. GCD measurements of the PPy₅₀-CNT₅₀/NaCl interface were performed at 50 µA/s. Cycling stability is relatively good, over 1050 cycles, as can be observed in Figure 5b, and it is compared with references working with PPy [53]. However, it is important to modify the molarity or type of salt of the electrolyte because its stability can be affected by the oxidation of cupper electrodes. The GCD curve exhibits an ideal triangular shape, representing the pseudocapacitance behavior that suggests low electrical resistance and fast charge/discharge kinetics. However, the cycle stability attainment is lower than that obtained in SCs with other PPy-CNT/electrolyte interfaces reported, over three hundred cycles [8,10,21,51,54,55].

4. Conclusions

Cyclic voltammetry and galvanostatic charge–discharge measurements show that the PPy₅₀-CNT₅₀/NaCl interface exhibits the optimal combination of energy and power density, resulting from the fast response of PPy and the storage capacity of CNT. The NaCl electrolyte exhibits superior performance compared to DMSO, proving its low electrical resistance, rapid charge/discharge kinetics, and enhanced pseudocapacitive behavior. Additionally, increasing the PPy content in PPy-CNT/NaCl interfaces enhances energy density, highlighting the importance of electrode composition for high-performance supercapacitors. The findings demonstrate the advantages of using NaCl as a nontoxic and high-conductivity electrolyte and the potential of the PPy₅₀-CNT₅₀ composition for efficient energy storage. Further studies should address the optimization of electrolyte type, molarity, and salt composition, as well as improving cycle stability and long-term pseudocapacitive performance to further enhance device reliability and performance.