Ionic side-chain engineering in conjugated polyelectrolytes for high-performance pseudocapacitors

Abstract

Conjugated polyelectrolytes (CPEs) are promising materials for pseudocapacitor electrodes due to their redox-active backbones and tunable ionic functionalities. Here, we investigate the impact of ionic pendant groups on electrochemical performance by comparing anionic CPE-K (sulfonate-functionalized) and cationic CPE-Br (quaternary ammonium-functionalized), both featuring identical conjugated backbones. The opposite charge polarity influences how polarons are stabilized: notably, CPE-K undergoes self-doping due to its anionic sulfonate groups, while CPE-Br remains neutral. In their pristine states, CPE-Br exhibits a higher specific capacitance (95 F g⁻¹vs. 84 F g⁻¹), more distinct redox features, and better rate capability than CPE-K. This is attributed to its cationic nature, which prevents self-doping and enhances anion penetration from the electrolyte. Electrochemical impedance spectroscopy further corroborates these findings, revealing significantly lower charge-transfer resistance and a smaller Warburg factor for CPE-Br, indicative of faster ion diffusion compared to CPE-K. Upon forming composites with single-walled carbon nanotubes (SWCNTs), both materials exhibit enhanced performance; notably, the 2 : 1 CPE-Br/SWCNT composite outperforms its CPE-K counterpart, achieving a high specific capacitance of 291 F g⁻¹ at 1 A g⁻¹ and retaining 254 F g⁻¹ at 5 A g⁻¹. Furthermore, it demonstrates excellent cycling stability over 1000 cycles with ∼99% coulombic efficiency, underscoring its robust charge storage reversibility and long-term durability. These findings highlight the importance of pendant group design in optimizing ionic transport and redox behavior in CPE-based pseudocapacitor systems.

---

1. Introduction

The growing demand for green and sustainable energy solutions has spurred extensive research into energy storage devices. Supercapacitors have attracted significant attention due to their high power density, rapid charging capabilities, and long cycle life, making them ideal for various applications, including portable electronics, electric vehicles, and grid energy storage.¹ Among these, pseudocapacitors—a subtype of supercapacitors—store energy through fast and reversible redox reactions at the electrode surface. These faradaic processes provide pseudocapacitors with higher energy density compared to standard supercapacitors.^2,3 They are often made from materials like metal oxides or conductive polymers, positioning them as promising candidates for future energy storage solutions including multifunctional pseudocapacitors.^4,5

Conjugated polymers (CPs) have also garnered significant interest due to their unique properties, such as facile tuning of redox properties, lightweight, and mechanical flexibility.⁶ These features make CPs particularly suitable for sustainable energy solutions, especially in pseudocapacitors. Their redox behavior and tunable energy storage capabilities make them ideal candidates for flexible pseudocapacitor applications. However, challenges such as poor electrolyte contact due to their hydrophobicity, which results in poor cycling stability, must be addressed to fully realize the potential of CP-based pseudocapacitors.^7,8

Conjugated polyelectrolytes (CPEs), a class of CPs with ionic side chains on the periphery, offer both ionic and electronic conductivity.^9–11 The mixed ionic-electronic behavior of CPEs is particularly advantageous for energy storage devices, as it enables simultaneous electronic and ionic conduction within a single phase.^12,13 The use of CPEs can enhance the cycling stability of pseudocapacitors compared to traditional CPs by improving electrolyte interaction, reducing electrode swelling after discharge, minimizing degradation, and enhancing ion transport.^14,15 Bazan et al. sought to improve the cycling stability of cyclopentadithiophene-co-benzothiadiazole-based anionic CPE (CPE-K) supercapacitors, by incorporating electrostatic interchain crosslinkers, such as Mg²⁺ ions.¹⁶ The resulting mechanically rigidified CPE-K hydrogels demonstrated improved capacitance retention (>40% after 1000 cycles). This improvement is attributed to the dynamic nature of hydrogels, which can accommodate volumetric changes without significant structural degradation during repeated charge–discharge cycles. To address electrical conductivity issues, Quek et al. developed a pseudocapacitive composite of graphene oxide (GO(+) and GO(−)) with CPE-K.¹⁷ This work showed that incorporating 2D electrolytes of GO into CPE hydrogels can enhance their performance for pseudocapacitive energy storage. For instance, adding positively charged GO(+) to a self-doped CPE-K hydrogel improved its mechanical strength, specific capacitance, and cycling stability due to better electrostatic interactions and ion diffusion. However, challenges remain, such as the limited solubility of GO(+), which restricts higher additive loading, and the complexity of the synthesis process, which may hinder large-scale applications. Furthermore, most research on CPE-based pseudocapacitors has focused primarily on using anionic CPE-K as redox-active materials. The ionic groups (either cationic or anionic) on the side chains of CPEs can significantly influence ion transport from the electrolyte, charge–discharge rates, and cycling stability. Despite this, studies addressing this aspect are still limited. To fully exploit the ionic-electronic dual conduction properties of CPEs, controlling the diffusion of electrolyte ions is crucial, but this challenge remains largely unexplored.

In this study, we investigated the pseudocapacitive behavior of two types of CPEs, anionic CPE-K and cationic CPE-Br, both based on the same cyclopentadithiophene-co-benzothiadiazole backbone. We focused on how their ionic side chains influence ion transport and the resulting electrochemical and charging–discharging characteristics for pseudocapacitor applications. CPE-K exhibits self-doping without external bias, where the anionic groups compensate for the oxidized backbone. However, this limits full de-doping during discharging, restricting charge storage capacity. In contrast, CPE-Br, with its cationic pendant groups, significantly enhances anion transport from the electrolyte, leading to faster charge–discharge rates and lower iR-drop. Electrochemical impedance spectroscopy (EIS) measurements revealed that CPE-Br has lower charge-transfer resistance, resulting in more efficient redox reactions and improved cycling stability compared to CPE-K. The electrochemical and pseudocapacitive properties of both CPEs and their composites with single-walled carbon nanotubes (SWCNTs) are discussed in detail.

2. Results and discussion

2.1. Synthesis of CPEs and their electrochemical properties

To explore the effect of different ionic groups in CPEs on pseudocapacitor properties, we synthesized two types of CPEs with identical backbones but varying ionic functionalities: CPE-K (anionic) and CPE-Br (cationic) (Fig. 1a). CPE-K contains sulfonate groups on its pendant side chains, with potassium (K⁺) as the counterion, while CPE-Br features quaternary ammonium groups paired with bromide (Br⁻) as the counterion. The synthesis procedures for both CPE-K and CPE-Br are outlined in Scheme S1 (ESI†). CPE-K was synthesized following a previously reported method.¹⁸ For CPE-Br, 2,6-dibromo-4,4-bis(6-bromohexyl)-4H-cyclopenta[2,1-b:3,4-b′]dithiophene with terminal bromo groups was polymerized with 2,1,3-benzothiadiazole-4,7-bis(boronic acid pinacol ester) via Suzuki coupling, yielding the neutral precursor PCPDTBT polymer. This polymer was subsequently quaternized by treatment with an excess of trimethylamine (30 equiv. in methanol), producing CPE-Br. The chemical structures of CPE-K and CPE-Br were confirmed using ¹H and ¹³C NMR spectroscopy (Fig. S1–S5, ESI†). The number-average molecular weight (M_n) of CPE-Br was determined to be 5.4 kg mol⁻¹ with a dispersity (Đ) of 1.33, as measured by gel permeation chromatography (GPC) using the neutral polymer PCPDTBT. For CPE-K, due to the ionic nature of the polymer, direct M_n measurement was not feasible. Instead, dialysis with a cellulose membrane (7 kg mol⁻¹ molecular weight cut-off) was used to obtain a comparable M_n to that of CPE-Br for a systematic comparison. Despite having the same backbone structure, CPE-K and CPE-Br exhibited distinct solubility behaviors. CPE-K shows good solubility in both water and dimethyl sulfoxide (DMSO, ∼20 mg mL⁻¹) due to the hydrophilic nature of its sulfonate groups.¹⁹ In contrast, CPE-Br demonstrates good solubility in polar organic solvents such as methanol and DMSO, owing to the quaternary ammonium groups, which are less hydrophilic than sulfonate groups and interact more favourably with these solvents. Fig. S6 (ESI†) shows photographs of CPE-K and CPE-Br dissolved in water and methanol, respectively, highlighting their distinct solubility behaviors. The decreased water solubility of CPE-Br compared to CPE-K may enhance the cyclic stability of pseudocapacitors in aqueous electrolyte solution.

Fig. 1 (a) Molecular structures of CPE-K and CPE-Br. (b) UV-Vis-NIR spectra of CPE-K and CPE-Br in solution and in film. Cyclic voltammograms of (c) CPE-K and (d) CPE-Br films at scan rates from 10 mV s⁻¹ to 50 mV s⁻¹. (e) log i_pvs. log v plots and calculated b values of CPE-K and CPE-Br.

It has been reported that CPE-K can undergo self-doping without the addition of external dopants, due to the presence of anionic sulfonate groups in its side chains that stabilize the positive polarons generated on the conjugated backbone.²⁰ This self-doping behavior can be experimentally observed by the appearance of new absorption in the near-infrared (NIR) region in UV-vis spectra. The UV-vis spectra of CPE-K and CPE-Br in solution and film are shown in Fig. 1b. Both polymers share the same conjugated backbone and exhibit a similar π–π* transition peak around 400 nm, along with a broad intramolecular charge transfer (ICT) peak in the 600–900 nm range. Notably, CPE-K shows additional broad absorption over 900 nm in the NIR region, which is attributed to (bi)polaron formation.²¹ In contrast, CPE-Br does not exhibit any (bi)polaron absorption bands. This difference can be attributed to the distinct charge groups: the sulfonate anions in CPE-K effectively stabilize the positive polarons, while the quaternary ammonium cations in CPE-Br lack this stabilizing effect.

The presence or absence of self-doping behavior in CPE-K and CPE-Br significantly impacts their electrochemical performance. To investigate this, we performed cyclic voltammetry (CV) at various scan rates in a 2 M NaCl aqueous electrolyte solution. The working electrode was fabricated by drop-casting the respective CPEs onto an indium tin oxide (ITO)-coated glass substrate. A saturated calomel electrode (SCE) and a platinum (Pt) mesh with an equivalent active area were used as the reference and counter electrodes, respectively. As shown in Fig. 1c and d, both CPEs exhibit quasi-rectangular CV profiles across all scan rates, characteristic of pseudocapacitive behavior. The key difference between CPE-K and CPE-Br is the appearance of redox peaks. CPE-K displays a single broad redox peak.²² In contrast, CPE-Br shows two distinct redox peaks at approximately ∼0.2 V and ∼0.6 V, indicating a two-electron transfer process at two different potentials (Fig. 1d). This clear difference can be attributed to the self-doping in CPE-K, which induces partial oxidation in the polymer backbone, resulting in a single redox process. In contrast, CPE-Br, lacking this self-doping ability, exhibits two-electron oxidation behavior. This suggests that CPE-Br has the potential to store more energy than CPE-K during charging, as evidenced by the additional redox process.

To gain a deeper understanding of the energy storage mechanism, we conducted CV measurements at varying scan rates (10 to 50 mV s⁻¹) and calculated the b-value using the Dunn power law relationship, defined as,²³

i_p = av^b (1)

where i_p is the current response at a given potential (vs. SCE), v represents the scan rate (mV s⁻¹), and the b-value is a parameter of the power law equation that ranges from 0.5 to 1.0. This analysis provides insight into whether the charge storage process is diffusion-controlled (b = 0.5) or surface-controlled (b = 1.0).^24,25 The calculated b-values for CPE-K and CPE-Br films are shown in Fig. 1e. The plot of log i_pvs. log v gives the slope, from which the b-value can be extracted. Compared to CPE-K (with a b-value close to 0.87), CPE-Br has a b-value of 0.81, indicating a more significant diffusion-controlled contribution to charge storage. Overall, both materials exhibit b-values between 0.5 and 1.0, indicating their potential as pseudocapacitive materials.

Next, we conducted galvanostatic charge–discharge (GCD) measurements using the same three-electrode cell configuration as in the CV measurements, within a potential window of 0.0 to 1.0 V (vs. SCE), to evaluate the differences in energy storage performance between CPE-K and CPE-Br. The charging process involves the oxidation of the CPE backbones, resulting in the formation of positive polarons, which are balanced by anions from the electrolytes.^26,27 The GCD curves for CPE-K and CPE-Br at various current densities (1–5 A g⁻¹) are shown in Fig. S7 (ESI†), with representative curves at 1 and 5 A g⁻¹ illustrated in Fig. 2a. Both CPE-K and CPE-Br exhibit a triangular-shaped profile across all current densities (1–5 A g⁻¹), confirming their pseudocapacitive behavior. The main difference between CPE-K and CPE-Br lies in their iR drop: CPE-Br demonstrates a significantly lower iR-drop compared to CPE-K. For CPE-K, the iR-drop is 0.04 V at 1 A g⁻¹, increasing sharply to 0.2 V at 5 A g⁻¹ (Fig. 2b). In contrast, CPE-Br shows an iR-drop of only 0.01 V at 1 A g⁻¹, and even at higher current densities (5 A g⁻¹), the drop remains minimal at 0.05 V. This stark difference between CPE-K and CPE-Br can be attributed to their different ionic side chains, which will be discussed in more detail later.

Fig. 2 GCD profiles of (a) CPE-K and CPE-Br at different current densities at 1 A g⁻¹ and 5 A g⁻¹, and (b) zoomed-in plots at 5 A g⁻¹. (c) Specific capacitance and iR-drop as a function of current density for CPE-K (blue) and CPE-Br (red) films. Spectroelectrochemical de-doping spectra of (d) CPE-K and (e) CPE-Br films. (f) Nyquist plots from EIS measurements for CPE-K and CPE-Br (inset shows a zoomed-in spectra).

We further calculated the specific capacitance (C_sp) of CPE-K and CPE-Br from the GCD curves using the following eqn (2),²⁸

where I is the specific current density (A g⁻¹), Δt is the total discharge time (s), m is the mass of the active material (g), ΔV is the magnitude of the potential change during discharge (V). As shown in Fig. 2c, the C_sp of CPE-K, derived from the discharge time, is 84 F g⁻¹ at a low current density of 1 A g⁻¹. This value aligns with a previous report by the Bazan group.²⁹ However, as the current density increases, the C_sp drops significantly, reaching only 20 F g⁻¹ at 5 A g⁻¹. This trend correlates well with the observed iR-drop behavior. In contrast, CPE-Br exhibits a higher specific capacitance of 95 F g⁻¹ at 1 A g⁻¹. Notably, as the current density increases, the C_sp of CPE-Br remains relatively stable, with only a minor decrease. At 5 A g⁻¹, CPE-Br retains 75% (72 F g⁻¹) of its initial value, demonstrating superior performance under high-rate conditions.

The superior performance of CPE-Br compared to CPE-K can be attributed to the distinct self-doping behavior and the higher Cl⁻ ion transport resistance of CPE-K. As discussed above, the self-doping in CPE-K limits further oxidation compared to CPE-Br, resulting in a relatively lower charge storage capacity. This characteristic is clearly observed in the spectro-electrochemical analysis (Fig. 2d). The in situ UV-Vis-NIR absorption was measured using the same 3-electrode electrochemical setup. A detailed description of the experimental setup and procedure is provided in the ESI.† As the applied voltage increases (from 0 V to 0.8 V) for both CPEs, the bandgap absorption in the visible region (300–900 nm) gradually decreases, while there is a concomitant increase in the midgap transition in the NIR region (over 900 nm), indicating (bi)polaron generation via electrochemical p-doping. For CPE-K, the NIR absorption peaks gradually decrease as the voltage is reduced from 0.8 to 0 V but does not reach complete disappearance. When the voltage is further reduced to −0.5 V, the absorption intensity continues to decrease, suggesting that a more negative bias is required for full de-doping. This implies that CPE-K remains partially doped at zero bias and requires additional potential to return to a fully neutral state.³⁰ In contrast, CPE-Br, which contains cationic end groups, exhibits different behavior. As shown in Fig. 2e, the NIR peaks for CPE-Br gradually decrease as the voltage is reduced from 0.8 to 0 V, and no further decrease is observed when additional negative bias is applied. This indicates that CPE-Br does not exhibit self-doping and completes its de-doping process without requiring an extended negative bias. A similar pattern can be observed during the re-doping process (Fig. S8, ESI†): while the NIR peaks for CPE-K begin to increase from an initial potential of −0.5 V, the peaks for CPE-Br start to rise at 0 V. This difference highlights that CPE-K's self-doping nature requires a broader potential range for full de-doping. The retention of NIR absorption at 0 V suggests that CPE-K remains partially doped even at zero bias, which reduces discharge capacitance and negatively impacts charge storage efficiency. Additionally, the negatively charged sulfonate terminal groups in CPE-K stabilize the positive polarons in the self-doped polymer. However, during charging, when Cl⁻ ions approach the polymer chain to balance the charge of the oxidized polymer backbone, electrostatic repulsion interferes, hindering this process.

Next, EIS measurements were conducted for CPE-K and CPE-Br films in 3-electrode setup over a frequency range from 5 MHz to 0.1 Hz with an applied AC amplitude of 10 mV at a potential of 0.5 V. The resulting Nyquist plot is shown in Fig. 2f and the data were fitted using an equivalent EIS circuit model (Fig. S9a, ESI†).^31,32 In this model, R_s represents the equivalent series resistance, while R_ct denotes the charge transfer resistance. The corresponding capacitance is represented as a constant phase element (C_ct) in an R_ct–C_ct parallel circuit. Additionally, C_dl accounts for the double-layer capacitance formed during doping (charging) of CPE-K and CPE-Br. For both CPE-K and CPE-Br, the model exhibited excellent fits, with χ² (goodness of fit) values consistently below 0.1. Both CPE-K and CPE-Br exhibit similar R_s values (∼2 Ω) at high frequencies, while a semicircle representing R_ct can be observed at medium frequencies. Notably, CPE-Br shows a lower R_ct (1.48 Ω) compared to CPE-K (5.95 Ω), indicating superior charge transfer characteristics and reduced internal resistance in CPE-Br. This difference can be attributed to the nature of their ionic side chains. In CPE-K, the sulfonate anions may hinder the efficient approach of Cl⁻ anions from the electrolyte to the conjugated backbone, likely due to electrostatic repulsion. In contrast, CPE-Br benefits from quaternary ammonium cations, which enhance interaction with Cl⁻ anions, facilitating their penetration into the conjugated backbone from the electrolyte. The improved anion diffusion lowers R_ct, resulting in better overall electrochemical performance. The observed differences in R_ct between CPE-K and CPE-Br impact the iR-drop, with lower R_ct in CPE-Br leading to a smaller iR-drop and improved electrochemical performance compared to CPE-K. Furthermore, the Warburg factor (σ), which is inversely proportional to the ion diffusion coefficient (D), was calculated from the slope of Z′ (Ω) versus ω^−1/2 (Hz^−1/2) in the 45–90° Warburg region.³³ A comparison reveals that CPE-K exhibits a higher Warburg factor (σ = 1.40 Ω s^−0.5) than CPE-Br (σ = 0.71 Ω s^−0.5), indicating slower ion diffusion in CPE-K (Fig. S9b, ESI†). This trend is further supported by the calculated D, where CPE-K exhibits a D value of ∼4.51 × 10⁻¹⁵ cm² s⁻¹, while CPE-Br shows a significantly higher D of ∼1.75 × 10⁻¹⁴ cm² s⁻¹. These results corroborate the enhanced ion mobility and accelerated diffusion of Cl⁻ ions in CPE-Br, highlighting the important role of pendant ionic groups in ion transport dynamics.

2.2. Electrochemical characteristics of CPE/SWCNT composites

Fabricating composites of polymers with single-wall carbon nanotubes (SWCNTs) can be an effective strategy to enhance the electrochemical performance of pseudocapacitors.^25,34,35 The exceptional electrical conductivity of SWCNTs contributes to more efficient energy storage and improved cycling stability.³⁶ However, SWCNTs also exhibit hydrophobic properties and lack ionic conductivity and redox activity, leading to poor electrochemical performance, including low capacitance, reduced rate performance, and significant iR-drop (see Fig. S10, ESI†). By compositing SWCNTs with CPEs, a synergistic effect can be achieved. The π-conjugated polymer backbone in CPEs can form cofacial π–π stacking interactions with SWCNTs, forming a homogeneous mixture with increased surface area of the composite, providing more active sites for charge storage. In addition, the ionic side chains in CPEs enhance ion migration from the electrolyte, promoting better ion transport. This combination of properties results in enhanced electrochemical performance, including higher capacitance, improved rate performance, and better cycling stability in the final composite. To verify this strategy, we prepared CPE/SWCNT composites with varying CPE-to-SWCNT weight ratios of 1 : 2, 1 : 1, and 2 : 1. These composites take advantage of the synergistic effect between SWCNTs, which offer high electrical conductivity, and CPEs, which contribute redox-active charge storage (Fig. 3a). The detailed fabrication procedure is outlined in the ESI.†

Fig. 3 (a) Schematic of CPE/CNT composites formation. (b) SEM images of 2 : 1 CPE-K/SWCNT and 2 : 1 CPE-Br/SWCNT composite films. (c) Contact angle measurements of SWCNT, 2 : 1 CPE-K/SWCNT and 2 : 1 CPE-Br/SWCNT films using aqueous 2 M NaCl solution. Cyclic voltammograms of (d) CPE-K/SWCNT and (e) CPE-Br/SWCNT composites at a scan rate of 50 mV s⁻¹. (f) log i_pvs. log v plots of various weight ratios of CPE/SWCNT composites.

We first investigated the surface morphology of the composite films by scanning electron microscopy (SEM) measurements. SEM images of pristine SWCNTs, CPE-K, and CPE-Br are shown in Fig. S11 (ESI†) while the corresponding composite films are displayed in Fig. 3b. Both CPE-K/SWCNT and CPE-Br/SWCNT composites exhibit well-integrated and homogeneously intermixed morphology. The CPEs are observed to fill the interstitial voids between the nanotubes, indicating strong interfacial interactions, likely arising from π–π stacking between the conjugated polymer backbones and the graphitic surface of SWCNTs.^17,37 Notably, the arrangement of SWCNTs differs between the two composite types. The CPE-Br/SWCNT film displays more aligned and ordered SWCNT bundles compared to the relatively entangled morphology observed in CPE-K/SWCNT. This enhanced alignment in CPE-Br/SWCNT may result from differences in processing solvents. Specifically, the use of methanol for CPE-Br leads to lower solution viscosity and faster solvent evaporation compared to the water used for CPE-K, promoting directional solvent flow that aids in nanotube reorganization during film formation.

The well-constructed composite structure was further validated through contact angle measurements. Initially, we measured the water contact angles of pristine SWCNT, CPE-K, and CPE-Br films (Fig. S12, ESI†). The SWCNT film exhibits a high contact angle of 95.0°, indicating its hydrophobic nature. In contrast, CPE-K and CPE-Br films show significantly lower contact angles of 39.0° and 65.6°, respectively, reflecting their more hydrophilic character. Together with the solubility behavior, the contact angle measurements provide further evidence that CPE-K is intrinsically more hydrophilic than CPE-Br. To more accurately simulate actual device operation conditions, contact angle measurements were also conducted using a 2 M NaCl electrolyte on SWCNT, CPE-K/SWCNT, and CPE-Br/SWCNT films (Fig. 3c). The contact angle of the SWCNT film decreases from 95.0° to 74.0°. It is 37.0° for the CPE-K/SWCNT composite and 43.4° for the CPE-Br/SWCNT composite. This substantial reduction in contact angle suggests enhanced wettability of the composites, likely due to the hydrophobic surface of the SWCNT being encapsulated by the CPE, decreasing its exposure, and the hydrophilic ionic groups of the CPE being exposed on the surface. These interactions promote better electrolyte spreading across the electrode surface, which is beneficial for pseudocapacitor operation by improving electrolyte accessibility and interfacial charge transfer. In addition, the slightly lower hydrophilicity of CPE-Br compared to CPE-K promotes more favorable interfacial interactions with SWCNTs, leading to a more uniform composite morphology and improved charge transport, which ultimately influences overall device performance.

Next, we performed CV measurements for the CPE/SWCNT composites. In Fig. 3d and e, the CV curves for different CPE-to-SWCNT ratios are presented for CPE-K/SWCNT and CPE-Br/SWCNT composites, respectively, with additional data at various scan rates provided in Fig. S13 (ESI†). For the CPE-K/SWCNT composite, at the lowest CPE-K content (1 : 2 wt ratio), the CV curve exhibits an almost rectangular shape with no distinct redox peaks, indicating low faradaic activity of the composite at this concentration and capacitor-like behavior dominated by the SWCNTs. As the CPE-K content increased to 1 : 1 and 2 : 1 ratios, broad redox peaks appear around ∼0.6 V, reflecting enhanced faradaic contributions. Notably, the 2 : 1 composite exhibits the highest current density, surpassing that of pristine CPE-K or SWCNT alone. This improvement is mainly attributed to the high electrical conductivity and large surface area of SWCNTs, which facilitate efficient charge transport and enhances electrochemical activity.^38,39

In contrast to the CPE-K/SWCNT composites, the CPE-Br/SWCNT systems exhibited more prominent pseudocapacitive behavior with significantly enhanced faradaic activity (Fig. 3e). As the CPE-Br content increased, the current densities rise accordingly, with the 2 : 1 composite showing the highest response. Importantly, two distinct redox peaks, consistent with the behavior of pristine CPE-Br. This stands in sharp contrast to CPE-K, where the self-doped nature suppressed the first redox transition. These results highlight that the absence of intrinsic self-doping in CPE-Br allows for well-defined electrochemical oxidation, resulting in more efficient charge storage when integrated with SWCNTs. To further validate the superior pseudocapacitive behavior of CPE-Br, we measured the b-values using the Dunn power law relationship. The b-values for CPE-K/SWCNT are 0.98, 0.80, and 0.74 for the 1 : 2, 1 : 1, and 2 : 1 ratios, respectively, while the corresponding values for CPE-Br/SWCNT are slightly lower at 0.82, 0.70, and 0.62 (Fig. 3f). These results indicate that CPE-Br/SWCNT composites exhibit stronger redox-controlled (diffusion-limited) behavior compared to CPE-K/SWCNT composites. This trend is consistent with previous observations in CPEs: the sulfonate anions in CPE-K hinder the diffusion of Cl⁻ ions toward the conjugated backbone, thereby limiting redox reactions—a behavior that persists even in the composite form.

To further evaluate electrochemical performance, GCD measurements were conducted on the optimized 2 : 1 CPE/SWCNT composites at various current densities (1–10 A g⁻¹). Full datasets are presented in Fig. S14 (ESI†), with representative profiles at 1 and 5 A g⁻¹ shown in Fig. 4a. The CPE-Br/SWCNT composite consistently outperforms its CPE-K counterpart, exhibiting higher C_sp and better rate performance. At 1 A g⁻¹, the CPE-Br/SWCNT composite achieved 291 F g⁻¹, compared to 255 F g⁻¹ for CPE-K/SWCNT. Even at 5 A g⁻¹, CPE-Br/SWCNT retained 254 F g⁻¹, while CPE-K/SWCNT dropped to 239 F g⁻¹, demonstrating the superior rate capability of the CPE-Br/SWCNT system (Fig. 4b). The iR-drop is similar for both composites at 1 A g⁻¹ (∼0.005 V), but diverges at 5 A g⁻¹, where CPE-K/SWCNT shows a larger drop (∼0.035 V) compared to CPE-Br/SWCNT (∼0.026 V). It is noteworthy that the CPE/SWCNT composites exhibit a distinct reduction in iR-drop compared to the CPE alone for both CPE-K and CPE-Br.

Fig. 4 (a) GCD profiles at 1 A g⁻¹ and 5 A g⁻¹, (b) specific capacitance and iR-drop as a function of current density, (c) Nyquist plots from EIS measurements, and (d) capacitance retention and CE plots during 1000 charge–discharge cycles at a current density of 5 A g⁻¹ for 2 : 1 CPE-K/SWCNT and 2 : 1 CPE-Br/SWCNT composite films.

Furthermore, EIS analysis was performed (Fig. 4c). The CPE-Br/SWCNT composite exhibits a markedly lower charge transfer resistance (R_ct = 0.69 Ω) compared to CPE-K/SWCNT (3.71 Ω), indicating enhanced interfacial electron transfer (Fig. S15a and c, ESI†). More notably, the Warburg factor (σ), which reflects ion diffusion resistance, decreases significantly in the CPE-Br/SWCNT composite (σ = 0.2 Ω s^−0.5), even lower than that of pristine CPE-Br (σ = 0.71 Ω s^−0.5). This suggests that the composite structure—with its porous SWCNT network and favorable ion–ion interactions between electrolytes and the ionic side chains in CPEs—not only retains but further amplifies ion transport efficiency. The synergistic combination of highly conductive SWCNTs and ion-accessible CPE-Br facilitates rapid Cl⁻ diffusion and lowers internal resistance, contributing to its excellent electrochemical performance. For the 2 : 1 CPE-Br/SWCNT composite, the calculated diffusion coefficient (D) is approximately 2.21 × 10⁻¹³ cm² s⁻¹, which is significantly higher than that of CPE-Br itself (D ≈ 1.75 × 10⁻¹⁴ cm² s⁻¹).

Further, the cycling stability tests at 5 A g⁻¹ over 1000 cycles further highlight the advantages of CPE-Br. While the CPE-K/SWCNT composite shows rapid capacitance fading (∼50% retention) (Fig. 4d and Fig. S15b, ESI†), the CPE-Br/SWCNT composite retains over 95% of its initial capacitance and exhibits nearly 99% coulombic efficiency (CE) (Fig. 4d and Fig. S12d, ESI†). The deterioration observed in the 2 : 1 CPE-K/SWCNT composite can be attributed to the weaker adhesion between the SWCNT and the doped conjugated backbone of CPE-K. The p-doped (positively charged) backbone of CPE-K is hydrophilic, in contrast to CPE-Br, which leads to gradual polymer leaching and results in swelling and microstructural degradation of the SWCNT framework. This exceptional durability emphasizes the mechanical integrity and aqueous stability of the CPE-Br composite, highlighting its suitability for long-term pseudocapacitor applications. The resulting electrochemical properties of the cationic and anionic CPEs, along with their composites, are summarized in Table 1.

Table 1 Electrochemical performance parameters of CPEs and CPE/SWCNT composites

Material	Electrical conductivity^a [mS cm^−1]	C sp [F g^−1]	iR-drop^c [V]	R ct [Ω]	Capacitance retention^d [%]
a From two-point probe measurements. b Specific capacitance determined from GCD profiles at a current density of 1 A g^−1. c Observed iR-drop at a current density of 5 A g^−1. d Capacitance retention after 1000 charge–discharge cycles at a current density of 5 A g^−1.
SWCNTs	213.76	26	0.49	—	—
CPE-K	2.71	84	0.2	5.95	100
CPE-Br	3.47 × 10^−3	95	0.05	1.48	100
2 : 1 CPE-K/SWCNT	171.04	255	0.035	3.71	∼50
2 : 1 CPE-Br/SWCNT	47.42	291	0.026	0.69	95

---

3. Conclusion

This study demonstrates that the electrochemical performance of CPEs is strongly governed by the nature of their ionic pendant groups. By comparing anionic CPE-K and cationic CPE-Br with identical conjugated backbones, we found that CPE-Br exhibited higher specific capacitance (95 F g⁻¹vs. 84 F g⁻¹), more distinct redox features, and better rate capability. These improvements stem from its cationic pendant groups, which do not induce self-doping and instead promote efficient anion (Cl⁻) transport from the electrolyte. In contrast, CPE-K undergoes self-doping due to its sulfonate side chains, leading to incomplete de-doping and electrostatic repulsion with electrolyte anions, which limits redox reversibility and rate performance. To further enhance their pseudocapacitor performance, we fabricated composites with SWCNTs, which offer high electrical conductivity and large surface area to facilitate rapid charge transport and ion accessibility. Both CPEs showed enhanced performance with SWCNTs; notably, the 2 : 1 CPE-Br/SWCNT composite outperformed its CPE-K counterpart, achieving a specific capacitance of 291 F g⁻¹ at 1 A g⁻¹ and retaining 254 F g⁻¹ at 5 A g⁻¹. It also maintained over 95% capacitance after 1000 charge–discharge cycles with ∼99% CE, highlighting its long-term electrochemical stability. Overall, this work not only provides fundamental insights into how ionic pendant groups influence redox behavior and ion transport in CPEs, but also underscores the importance of rational molecular design strategies for optimizing pseudocapacitor performance.

Author contributions

Jinesh Lalitkumar Chouhan: investigation, methodology, formal analysis, writing – original draft. Haim Kwon: figure revising, data analysis. Jung Min Ha: data analysis, formal analysis, editing-reviewing. Wonho Lee: conceptualization, supervision, writing – review & editing, Han Young Woo: conceptualization, supervision, writing – review & editing, funding acquisition.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Research Foundation (NRF) grants funded by the Ministry of Science and ICT (MSIT), Korean government (RS-2024-00334832 and 2023K2A9A2A06059546).

References

1. S. Fleischmann, J. B. Mitchell, R. Wang, C. Zhan, D. E. Jiang, V. Presser and V. Augustyn, Chem. Rev., 2020, 120, 6738–6782.
2. R. S. Redekar, A. T. Avatare, J. L. Chouhan, K. V. Patil, O. Y. Pawar, S. L. Patil, A. A. Bhoite, V. L. Patil, P. S. Patil and N. L. Tarwal, Chem. Eng. J., 2022, 450, 137425.
3. M. M. Baig, M. A. Khan, I. H. Gul, S. U. Rehman, M. Shahid, S. Javaid and S. M. Baig, J. Electron. Mater., 2023, 52, 5775–5794.
4. S. Y. Kim, J. Jang, Y. M. Kim, J. K. Lee and H. C. Moon, Adv. Funct. Mater., 2022, 32, 2200757.
5. Y. J. Jang, S. Y. Kim, Y. M. Kim, J. K. Lee and H. C. Moon, Energy Storage Mater., 2021, 43, 20–29.
6. Y. Yan, Y. Zhou, Y. Li and Y. Liu, Carbon Lett., 2023, 33, 1461–1483.
7. P. Siwach, L. Gaba, S. Dahiya, R. Punia, A. S. Maan, K. Singh and A. Ohlan, Appl. Surf. Sci. Adv., 2024, 19, 100555.
8. T. Liu, L. Finn, M. Yu, H. Wang, T. Zhai, X. Lu, Y. Tong and Y. Li, Nano Lett., 2014, 14, 2522–2527.
9. A. K. Harit, J. M. Ha, A. Tripathi, H. Y. Woo, E. D. Jung, J. H. Park, Y. W. Noh and M. H. Song, Small, 2022, 18, 2104933.
10. G. Pace, O. Nordness, P. H. Nguyen, Y. J. Choi, C. Tran, R. J. Clément and R. A. Segalman, Macromolecules, 2023, 56, 6078–6085.
11. J. H. Park, Y. W. Noh, J. M. Ha, A. K. Harit, A. Tripathi, J. Lee, B. R. Lee, M. H. Song and H. Y. Woo, ACS Appl. Mater. Interfaces, 2024, 16(16), 19937–19946.
12. W. Lee, J. L. Chouhan, A. K. Harit, C. Park, D. Kim, S. R. McCuskey, G. C. Bazan and H. Y. Woo, ACS Nano, 2025, 19, 5965.
13. K. A. Larasati, C. Park, L. Ayuningtias, H. Cha and Y.-H. Kim, Macromol. Res., 2025, 1–8.
14. Y. Song, T.-Y. Liu, X.-X. Xu, D.-Y. Feng, Y. Li, X.-X. Liu, Y. Song, X. Xu, D. Feng, X. Liu, T. Liu and Y. Li, Adv. Funct. Mater., 2015, 25, 4626–4632.
15. L. Shen, Y. Ahn, Y. Kim, S. Kim, S. Choi, T. D. Kim and D. Lee, Macromol. Res., 2024, 32, 767–776.
16. Y. Su, S. R. McCuskey, D. Leifert, A. S. Moreland, L. Zhou, L. C. Llanes, R. J. Vazquez, L. Sepunaru, G. C. Bazan, Y. Su, A. S. Moreland, L. Zhou, L. C. Llanes, L. Sepunaru, S. R. McCuskey, D. Leifert, R. J. Vazquez and G. C. Bazan, Adv. Funct. Mater., 2021, 31, 2007351.
17. G. Quek, Y. Su, R. K. Donato, R. J. Vázquez, V. S. Marangoni, P. R. Ng, M. C. F. Costa, B. Kundukad, K. S. Novoselov, A. H. C. Neto and G. C. Bazan, Adv. Electron. Mater., 2022, 8, 2100942.
18. C. K. Mai, H. Zhou, Y. Zhang, Z. B. Henson, T. Q. Nguyen, A. J. Heeger and G. C. Bazan, Angew. Chem., Int. Ed., 2013, 52, 12874–12878.
19. P. Jing, P. J. Rodgers and S. Amemiya, J. Am. Chem. Soc., 2009, 131, 2290–2296.
20. D. X. Cao, D. Leifert, V. V. Brus, M. S. Wong, H. Phan, B. Yurash, N. Koch, G. C. Bazan and T. Q. Nguyen, Mater. Chem. Front., 2020, 4, 3556–3566.
21. D. Tsokkou, L. Peterhans, N. Banerji, D. X. Cao, C. Mai, G. C. Bazan and T. Nguyen, Adv. Funct. Mater., 2020, 30, 1906148.
22. R. J. Vázquez, G. Quek, Y. Jiang, B. Y. Rui Peng, S. R. McCuskey, D. Ohayon, B. Kundukad, X. Wang and G. C. Bazan, J. Mater. Chem. A, 2023, 11, 18843–18852.
23. W. Pholauyphon, P. Charoen-amornkitt, T. Suzuki and S. Tsushima, Electrochem. Commun., 2024, 159, 107654.
24. S. Pervez and M. Z. Iqbal, Small, 2023, 19, 2305059.
25. Y. Yang, P. Zhang, L. Hao, P. Cheng, Y. Chen and Z. Zhang, Angew. Chem., Int. Ed., 2021, 60, 21838–21845.
26. X. Liu, M. Zheng, Y. Xiao, Y. Yang, L. Yang, Y. Liu, B. Lei, H. Dong, H. Zhang and H. Fu, ACS Appl. Mater. Interfaces, 2013, 5, 4667–4677.
27. P. Simon and Y. Gogotsi, Nat. Mater., 2008, 7, 845–854.
28. A. T. Avatare, R. S. Redekar, J. L. Chouhan, S. Yadav, I. H. Lee, S. B. Patil and N. L. Tarwal, J. Electroanal. Chem., 2025, 976, 118809.
29. B. R. P. Yip, R. Javier Vázquez, Y. Jiang, S. R. McCuskey, G. Quek, D. Ohayon, X. Wang and G. C. Bazan, Adv. Mater., 2024, 36, 2308631.
30. D. Ohayon, G. Quek, B. R. P. Yip, F. Lopez-Garcia, P. R. Ng, R. J. Vázquez, D. V. Andreeva, X. Wang and G. C. Bazan, Adv. Mater., 2024, 36, 2410512.
31. J. Bobacka, A. Lewenstam and A. Ivaska, J. Electroanal. Chem., 2000, 489, 17–27.
32. M. Bianchi, S. Carli, M. Di Lauro, M. Prato, M. Murgia, L. Fadiga and F. Biscarini, J. Mater. Chem. C, 2020, 8, 11252–11262.
33. G. Quek, D. Ohayon, P. R. Ng and G. C. Bazan, Small, 2024, 20, 2401395.
34. C. Peng, S. Zhang, D. Jewell and G. Z. Chen, Prog. Nat. Sci., 2008, 18, 777–788.
35. K. Lee, J. Hwang, J. H. Park, J. Park, K. Lee and J. M. Ko, Macromol. Res., 2023, 31, 171–179.
36. H. Pan, J. Li and Y. P. Feng, Nanoscale Res. Lett., 2010, 5, 654–668.
37. B. Kim, H. Park and W. M. Sigmund, Langmuir, 2003, 19, 2525–2527.
38. K. A. Sree Raj, K. Pramoda and C. S. Rout, J. Mater. Chem. C, 2023, 11, 2565–2573.
39. S. Hema, G. U. Chandran, M. Sajith, K. Rashid Sulthan and S. Sambhudevan, Polym. Blend Nanocompos. Energy Storage Appl., 2023, 241–270.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tc01491c

---