Engineering multiwalled carbon nanotube modified titanium carbide MXene nanocomposites for flexible symmetric supercapacitors in printed electronics

Abstract

MXene-based micro-supercapacitors (MSCs) are promising power sources for wearable electronics and distributed IoT systems, yet their scalability is limited by MXene restacking and complex fabrication protocols. Here, we present the development and optimization of delaminated Ti₃C₂T_x MXene (D-MXene)/multi-walled carbon nanotube (MWCNT) composite electrodes with enhanced electrochemical performance. The incorporation of 20 wt% MWCNTs effectively suppresses MXene restacking, facilitating improved ion transport and structural integrity. The optimized D-MXene/MWCNT composite electrode (MC-20) exhibits a high specific capacitance of 645 F g⁻¹ corresponds to galvanostatic charge discharge cycles (GCD) at a current density of 1 A g⁻¹, with 88% retention after 10 000 cycles. To demonstrate the practical utility of the optimized material, a flexible printed micro-supercapacitor (MSC) was fabricated using a D-MXene/MWCNT/PEDOT:PSS composite following a previously established printing protocol. The device exhibits an areal capacitance of 53 mF cm⁻² at 0.1 mA cm⁻² and an energy density of 4.7 µWh cm⁻², maintained even at a high-power density of 80 µW cm⁻². Remarkably, the device retains 96.7% of its capacitance after 8000 bending and charge–discharge cycles, demonstrating excellent electrochemical durability. This study highlights the effectiveness of MWCNT incorporation in enhancing MXene-based electrode architectures and establishes a co-engineering strategy that integrates material design with device architecture, offering a promising pathway to translate optimized MXene composites into flexible and wearable energy storage systems for next-generation IoT applications.

---

1 Introduction

Wearable and distributed Internet of Things (IoT) systems demand compact and mechanically compliant energy storage solutions that can be developed directly on flexible substrates. Micro supercapacitors (MSCs) can meet this demand as they offer high power density, fast charge–discharge, and long cycle life in compact, yet planar formats.^1–3 Specifically, the interdigitated architecture of MSCs minimizes electron diffusion paths and reduces overall internal resistance, thereby enabling rapid electron transfer and quick response times.⁴ Adaptability of MSCs for wearable and distributed IoT systems is largely governed by the inherent mechanical properties, charge storage mechanisms, and fabrication protocols of the adopted electrode materials.^5,6 Recently, 2D transition metal carbides, known as MXenes (M_n+1X_nT_x, where M is an early transition metal, X is C and/or N, and T_x represents terminal functional groups), have emerged as promising electrode materials for supercapacitors.² Yuri Gogotsi et al. reported an MSC with MXene/CNT electrodes, with a notable areal capacitance of approximately 317 mF cm⁻² at 50 mVs⁻¹.⁷ Despite this promise, two bottlenecks limit the translation of MXene based MSCs from laboratory devices to manufacturable modules; a primary material level bottleneck and a secondary process level bottleneck. At the material level, nanoscale MXene flakes are prone to self-restacking due to strong van der Waals interactions, resulting in densely packed structures that suppress areal/gravimetric capacitance and rate capability.⁸ At the process level, many high-performing electrodes rely on fabrication routes (e.g., vacuum filtration and ion-beam lithography) that are difficult to scale, as well as expensive to adopt for wearables or distributed IoT systems. Also, many of these fabrication protocols are not suitable for substrates in wearables including textiles/low energy polymers. Hence, a pathway that co-optimizes these two bottle necks is essential for harvesting the charge storage capabilities of MXenes for MSCs.

Common strategies to mitigate restacking of MXene nanosheets include inserting 0D/1D/2D spacers to open interlayer galleries and build percolation networks, as well as delamination of multilayer MXenes using organic intercalants (e.g., DMSO and DMF) to increase accessible surface area.^9–11 Multiwalled carbon nanotubes (MWCNTs) can be used as an interlayer spacer, and are reported to enhance the mechanical as well as electrochemical characteristics of MXene electrodes in supercapacitors. For instance, Xin Shi et al. incorporated continuous, ultrathin CNT films as interlayer materials between MXenes. These layered composite films exhibited outstanding electrochemical performance, with an energy density of 7.34 Wh kg⁻¹ at a power density of 50 W kg⁻¹.⁶ At the process level, the rich surface chemistry of MXene nanosheets, their negative surface charge, and hydrophilicity facilitate the formation of stable, viscous aqueous colloidal inks that can be screen printed to fabricated electrode materials for MSCs. Screen printing can enable cost-effective and scalable fabrication of MSCs, compatible with polymer substrates and even textiles.¹² Integrating these two approaches can therefore offer a promising solution to the current bottlenecks in MXene based micro-supercapacitors. However, research on fully printed MXene devices that co-optimizes restacking for electrochemical performance, ink parameters (rheology, substrate wetting, and adhesion under tape tests), and gel-polymer electrolyte (GPE) compatibility remains comparatively under-reported. In this context, our work emphasizes the fabrication of MSCs via a straightforward, solution-based deposition and patterning strategy, providing high scalability, mechanical flexibility, and reliable electrochemical performance.

In this work, multilayered Ti₃C₂T_x MXene was synthesized via in situ HF etching, followed by DMSO-assisted delamination and sonication to obtain well-exfoliated nanosheets, enabling a direct comparison of electrochemical performance before and after exfoliation. To further suppress restacking and enhance ion-transport pathways, MWCNTs were incorporated between MXene layers, and their percentage was systematically varied to identify an optimized D-MXene/MWCNT composition. In parallel, the PEDOT:PSS formulation was engineered to ensure print compatibility, film integrity, ink flexibility and dispersion without compromising electrochemical performance. While these approaches have been explored individually, we report here, for the first time, a fully integrated, efficient, and flexible screen-printable MSC fabricated using a composite ink that integrates delaminated Ti₃C₂T_x (D-MXene) with MWCNTs as spacers and PEDOT:PSS as a conductive binder. Co-optimization studies on (i) materials architecture (MWCNT-spaced D-MXene that suppresses restacking), (ii) ink/process engineering, and (iii) device-level validation are being reported in detail. The screen-printed electrodes offer 645 F g⁻¹ at 1 A g⁻¹ and devices achieve 53 mF cm⁻² and 4.7 µWh cm⁻² at 80 µW cm⁻², retaining 96.7% capacitance after 8000 bend/charge–discharge cycles. These results can offer a promising pathway to connect the state-of-the-art MXene composites to scalable and flexible energy storage solutions for wearables and IoT applications.

2 Results and discussion

Ti₃C₂T_x MXene is synthesized through an in situ HF etching process, followed by delamination using DMSO as an intercalant. The small, polar DMSO molecules can efficiently penetrate between MXene layers, weakening the van der Waals forces that hold the layers together.¹³ By increasing the interlayer spacing, DMSO helps prevent restacking, ensuring a stable suspension of individual MXene layers.¹⁴Fig. 1 shows a schematic representation of the synthesis of delaminated Ti₃C₂T_x MXene (D-MXene), followed by the preparation of D-MXene/MWCNT composites with varying compositions. The synthesis procedures are detailed in the SI.

Fig. 1 Schematic diagram for the synthesis of delaminated Ti₃C₂T_x MXene (D-MXene), and the D- MXene/MWCNT composite.

2.1. General characterization of delaminated MXene and D-MXene/MWCNT composites

The phase and structural characteristics of the synthesized Ti₃C₂T_x MXene and D-MXene were analysed using the 1D wide angle (WAXD) and powder X-ray diffraction (PXRD) techniques. Fig. 2a–c present the XRD patterns for the Ti₃AlC₂ MAX phase, Ti₃C₂T_x MXene and D-MXene. The characteristic peak of the MAX phase at 2θ ≈ 9.61°, corresponding to the (002) plane, shifts towards a lower angle at 2θ ≈ 6.43° for D-MXene, consistent with an ∼49% increase in interlayer spacing (Cu Kα, λ = 1.5406 Å: d₀₀₂ ∼9.20 Å → 13.74 Å), confirming the increase in interlayer spacing after delamination. The XRD traces also reveal the progressive removal of the Al layer in the parent MAX phase, confirming successful etching and gallery expansion. Notably, after approximately 48 hours of treatment, the characteristic Ti₃AlC₂ peak at 2θ ≈ 39° (104) completely disappears, indicating the breakdown of the tightly packed crystalline structure of the MAX phase (Ti₃AlC₂) and the formation of new MXene layers.¹⁵ Residual Ti₃C₂(OH)₂ is detectable at 2θ ≈ 26.45° in etched samples, as expected for hydroxyl-terminated MXene.¹⁶

Fig. 2 (a) 1D WXRD (b and c) PXRD spectra of Ti₃AlC₂ (MAX phase), MXene, and D-MXene. (d) Nitrogen adsorption–desorption isotherms (e and f) SEM images of MXene and D-MXene.

X-ray photoelectron spectroscopy (XPS) is used to precisely determine the chemical states of various species present in the D-MXene sample, where the experimental data obtained have been fitted using the Gaussian method. The survey spectrum shown in Fig. S1a confirms the presence of Ti, C, O and F atoms present in Ti₃C₂T_x MXene where the T_x can have the possibilities of –F, –OH and =O functionalities. The entire XPS spectra have been deconvoluted by keeping C 1s at 285 eV as the ref. 17. The deconvoluted Ti 2p spectra presented in Fig. S1b clearly demonstrate the presence of the Ti–C bond, which can be explained with the peaks at 454.7 eV and 460.4 eV that are respectively those corresponding to the 2p_3/2 and 2p_1/2 spin–orbit coupling states of Ti 2p. The presence of TiO₂ in the sample can be observed from the peaks at 458.7 eV and 464.5 eV, which are in coherence with the same spin orbit coupling states respectively.¹⁸ Similarly, the C 1s deconvoluted spectra (Fig. S1c) show the presence of C–Ti, C–C, C–O and O–C=O groups which can be linked to the peaks at 281.9 eV, 284.9 eV, 286.8 eV and 289 eV respectively.¹⁹ Furthermore, the deconvoluted O 1s spectra in Fig. S1d shows characteristic deconvoluted peaks at 529.8 eV, 530.9 eV, 531.9 eV, and 533.4 eV, in which the peak at 529.8 eV can be attributed to the presence of the Ti–O bond which may be arising from the presence of trace amounts of TiO₂ in the sample.²⁰ The major peaks at 530.9 eV and 531.9 eV can be connected to the presence of terminal –O and –OH functional groups present in the sample, along with a peak of less intensity at 533.4 eV arising due to surface adsorbed water molecules. Fig. S1e presents deconvoluted F 1s spectra that confirms the presence of Ti–F and C–F bonds, in light of the respective peaks at 685.7 eV and 689.9 eV.

N₂ adsorption–desorption tests were conducted to analyze and compare the pore size distribution and specific surface area of MXene and D-MXene samples. The BET isotherms are presented in Fig. 2d, and the corresponding pore size distribution curves are shown in Fig. S2a. The D-MXene sample exhibits a significantly higher specific surface area of 43.63 m² g⁻¹ and a pore volume of 0.14 cm³ g⁻¹, in contrast to pristine MXene, which shows a surface area of only 4.45 m² g⁻¹ and a pore volume of 0.02 cm³ g⁻¹. The exfoliation process, facilitated by DMSO, produced few-layered MXene sheets with increased interlayer separation, which can lead to enhanced electrochemical performance. To verify this, the morphology of the samples was analysed using both SEM and TEM techniques. Fig. 2e clearly illustrates that the etched MXene exhibits familiar accordion-like stacks with tightly coupled layers, whereas D-MXene reveals few-layer sheets with well-separated lamellae.²¹ The distinctive accordion-like stack structure is due to the selective etching of the ‘Al’ layer, leading to a parallel arrangement of MXene layers. However, the intercalant, DMSO could successfully exfoliate this architecture by creating a more open crumpled morphology as observed in Fig. 2f. The interaction between the solvent and the MXene flakes is influenced by their large lateral dimensions, surface energy of the flakes and the surface tension of the solvent. DMSO is a superior intercalant due to its hygroscopic nature, and its high boiling point (189 °C) makes it more practical for intercalation processes, as solvents with lower boiling points tend to evaporate quickly and are difficult to manage during processing.^9,10,22 As shown in Fig. S3a–d, the TEM images provide detailed insights into this change in morphology. The as-prepared MXene displayed tightly stacked, multi-layered structures, while the D-MXene exhibits more distinct, well-separated layers. This increased separation between the layers can enhance ion accessibility, which is crucial for improving electrochemical performance. The enhanced gallery spacing observed by XRD is corroborated by the more open, crumpled morphology in SEM and by thinner platelets in TEM, hence confirming the successful delamination.

To prevent the restacking of D-MXenes, MWCNTs were integrated as interlayer spacers between the D-MXene sheets. To determine the optimal MWCNT content for enhancing both interlayer spacing and electrochemical performance, batches of five different compositions were prepared, with varying MWCNT weight percentages (5%, 10%, 20%, 30%, and 40%), labelled as MC-5, MC-10, MC-20, MC-30, and MC-40, respectively. The structural, morphological, and electrochemical properties of these composites were systematically evaluated. The structural properties of the synthesized D-MXene/MWCNT composites were characterized using the WAXD and PXRD techniques, as shown in Fig. 3(a–c). In the XRD patterns of the D-MXene/MWCNT composites, an additional diffraction peak appears at around 2θ 26°, which corresponds to the (002) crystal plane of carbon nanotubes. Meanwhile, the characteristic (002) peak of MXene, initially observed at 2θ ∼ 6.43°, gradually shifts toward a lower angle (2θ ∼ 5.77°) with increasing MWCNT content. This shift indicates a progressive expansion of the interlayer spacing within the MXene layers, and suggests partial exfoliation and weakened interlayer interactions, possibly due to surface adsorption or edge interconnection of MWCNTs with MXene sheets. The calculated d-spacing increases from 13.74 Å to 15.31 Å, representing an ∼11% increase, which further confirms the structural expansion due to MWCNT incorporation.^23,24 Moreover, as the proportion of MWCNTs increases, the intensity of the MXene (002) peak decreases, while the intensity of the MWCNT-related (002) peak becomes more prominent, reflecting the increasing contribution of the carbonaceous phase and a reduction in the preferred orientation of the MXene layers.

Fig. 3 (a) 1D WXRD, (b and c) PXRD spectra, (d and e) nitrogen adsorption–desorption isotherms and BET surface area values of D-MXene/MWCNT composites, (f–k) SEM images of D-MXene and D-MXene/MWCNT composites, and (l and m) TEM images of optimal composite MC-20.

N₂ adsorption–desorption isotherms and BET surface area values of D-MXene and all D-MXene/MWCNT composites are presented in Fig. 3d and e. The isotherms exhibit Type IV-like behaviour with a small hysteresis loop at intermediate to high relative pressures. However, the minimal N₂ uptake at low P/P₀ and the absence of a clear saturation plateau near P/P₀ → 1 indicate that the pore system does not strictly conform to canonical Type IV mesoporosity. Instead, the adsorption behaviour is characteristic of a pore structure dominated by meso-to macroporous slit-like voids originating from interlayer spacing between MXene sheets and interparticle voids created by MWCNT incorporation.^25,26 The obtained BJH desorption average pore diameters of 10.9, 30.0, 33.2, 37.9, 28.8, and 38.5 nm for D-MXene, MC-5, MC-10, MC-20, MC-30, and MC-40, respectively, fall within the mesoporous range (2–50 nm), supporting the presence of dominant mesopores, particularly in the upper mesopore regime. The rapid increase in adsorption near P/P₀ ≈ 1 further suggests the contribution of large mesopores and macropore-like voids, which prevents the establishment of a well-defined saturation plateau. Also, among the composites, the D-MXene/MWCNT composite (MC-20) exhibited a significant increase in specific surface area (112.04 m² g⁻¹) and a larger pore volume (1.14 cm³ g⁻¹). This enhancement reflects the formation of a 3D porous framework within the MC-20 composite, attributed to the incorporation of MWCNTs, known for their inherently high surface area (96 m² g⁻¹) and the expanded layer spacing of the MXene sheets. The introduction of MWCNTs markedly improved the composite surface area, as their porous, high-aspect-ratio structure promotes better dispersion and mitigates MXene restacking. However, at higher CNT loadings (MC-30 and MC-40), the surface area decreased (35.71 and 16.25 m² g⁻¹, respectively), likely due to CNT agglomeration and partial pore blockage at elevated concentrations. The BET isotherm of MWCNTs is shown in Fig. S2c for comparison. The resulting mesoporous structure of the composite is beneficial, as it exposes more active sites and creates numerous ion and electron transport pathways at the electrode/electrolyte interface, and thereby can enhance the electrochemical properties of the electrode material. The pore size distribution curves of all the MC-composites are illustrated in Fig. S2b and the BET surface area, BJH desorption average pore diameter, and average pore volume values for all samples are summarized in Table S1.

Morphological evaluations were carried out to understand the effect of MWCNT addition on the distribution of D-MXene flakes. The SEM images of D-MXene and the various D-MXene/MWCNT composites are illustrated in Fig. 3f–k. It is evident that all five composites contain numerous fibrous structures and flakes, where the fiber-like structures are MWCNTs, and the flake-like structures are D-MXene sheets. In the MC-5 sample, the lower MWCNT content leads to a sparsely dispersed distribution across a dominant matrix of D-MXenes, while reinforcing the composite structure without significantly altering MXene's characteristics. In MC-10, an increased quantity of MWCNTs can be observed, often appearing entangled within the MXene matrix. In MC-20, the nanotubes form a well-distributed web interwoven through the MXene sheets. The fibers bridge neighbouring flakes and keep them apart, preventing re-stacking observed in MC-5 and MC-10. This architecture is exposing more of the MXene surface, which can improve both ion access (shorter diffusion paths) and electronic conduction (continuous percolation). However, as the MWCNT content increases further, the MXene layers become entirely covered by MWCNTs, leading to aggregation that diminishes the electrochemically active surface areas. This aggregation, particularly evident in MC-30 and more in MC-40, can negatively impact the electrochemical performance of the material, when used in electrode fabrication. The TEM image of MC-20 (Fig. 3l and m) corroborates the SEM observations, showing uniformly distributed nanotubes coming into intimate contact with MXene platelets without large agglomerates. Overall, the evolution from insufficient spacing (≤10 wt%) to an interwoven, percolated network (20 wt%) and then to over-coverage/agglomeration (≥30 wt%) identifies MC-20 as the most favourable composition for subsequent electrochemical evaluation.

2.2. Electrochemical evaluation of MXene and D-MXene electrodes

To assess the electrochemical performance, both MXene and D-MXene based electrodes were thoroughly characterized using a standard two-electrode configuration in a 6 M KOH electrolyte at room temperature. As depicted in Fig. S4a and b, cyclic voltammetry (CV) curves were recorded within a potential window of 0 to 0.8 V at different scan rates, ranging from 5 to 200 mV s⁻¹. With increasing scan rates from 5 to 200 mV s⁻¹, the CV plots for both MXenes and D-MXene exhibit and maintain almost rectangular profiles, typical for double-layer capacitors, signifying good capacitive behaviour and efficient ion response of the electrodes.²⁷ For D-MXene, the CV curves show a more pronounced rectangular shape compared to MXenes. This enhancement is attributed to the increased electrostatic charge accumulation at the electrode–electrolyte interface, resulting from the expanded spacing between MXene layers due to delamination.²⁸ At a scan rate of 20 mV s⁻¹, the specific capacitance (C_sp) values for MXenes and D-MXene were 155 F g⁻¹ and 525 F g⁻¹, respectively. The CV profile at this scan rate shows that D-MXene exhibits a significantly larger loop area and higher current response compared to MXenes, indicating a considerably higher capacitance for D-MXene (as shown in Fig. 4a). The CV graphs demonstrated a strong dependence of the material's performance on the scan rate, with high reversibility of the capacitance, highlighting the excellent electrochemical stability of the synthesized material. Fig. S4c and d present the galvanostatic charge–discharge (GCD) curves for both MXenes and D-MXene at increasing currents from 1 mA to 5 mA. The GCD curves of D-MXene display triangular shapes without any noticeable voltage drop at the start of the charge–discharge cycle, which signifies the ideal electric double-layer capacitance behaviour of the D-MXene sheets. The discharge profiles obtained at 1 A g⁻¹ were used to evaluate the specific capacitances of MXenes and D-MXene (Fig. 4b). D-MXene demonstrated a markedly higher capacitance of 516 F g⁻¹, which can be attributed to its extended charge and discharge times at the same current density, resulting from the significant increase in surface area due to delamination. In contrast, MXenes exhibited a much lower capacitance of 98 F g⁻¹, primarily due to the restacking of its layers, which restricts ion accessibility and reduces the effective surface area. The improvement in the electrochemical performance of D-MXene can be readily attributed to the enhanced surface area upon DMSO treatment, as evident from the BET results. The plot of specific capacitance versus scan rate and current density is illustrated in Fig. 4c. As the scan rate decreases, the cyclic voltammetry curves exhibit an increased current near the maximum voltage window. This occurs because, at lower scan rates, electrolyte ions have enough time to penetrate deeper into the widened interlayer gaps formed after the removal of the intermediate aluminium layer. This phenomenon is particularly evident when the MXene layers are aligned parallel to the current collector.²⁹ The MXene electrode achieves a specific capacitance of 178 F g⁻¹ at a scan rate of 5 mV s⁻¹. However, at a higher scan rate of 200 mV s⁻¹, the specific capacitance significantly decreases to 87 F g⁻¹, indicating the poor rate capability of the multi-layered MXene. In contrast, the few-layered D-MXene exhibits higher specific capacitance across all scan rates. For instance, the maximum specific capacitance of D-MXene reaches 603 F g⁻¹ at 5 mV s⁻¹. Notably, even at a scan rate of 200 mV s⁻¹, D-MXene maintains a C_sp value of 358 F g⁻¹, which is four times higher than that of the multi-layered MXene. These results clearly show that the delamination process using DMSO significantly enhances both the rate capability and specific capacitance of MXenes. Also, the trend reveals that as the current density increases, the specific capacitance (C_sp) gradually decreases. The specific capacitance for varying scan rates and current densities was calculated using eqn (S1) and (S2).

Fig. 4 Electrochemical performance studies of the fabricated MXene and D-MXene based electrodes: (a) CV graphs at 20 mV s⁻¹, (b) GCD curves at 1 Ag⁻¹, (c) variation of specific capacitance with increasing scan rates and current densities, (d) Nyquist plot, (e) Ragone plot, and (f) cycling stability over 10 000 GCD cycles at 5 A g⁻¹.

Notably, D-MXene demonstrates excellent rate capability, retaining 70% of its initial capacitance, whereas MXenes show a retention rate of 61%. This superior rate performance of D-MXene can be attributed to its larger surface area and more defined pore structure, which promote faster ion transport and enhance access to redox-active sites compared to MXenes. To investigate the ion transport behaviour of the electrodes, electrochemical impedance spectroscopy studies were conducted in the 10 mHz to 100 KHz frequency range as depicted in Fig. 4d. The impedance spectra can be divided into two distinct regions: a small semicircle in the higher frequency domain and a straight line in the lower-frequency domain. The diameter of the smaller arc corresponds to charge transfer resistance (R_ct) at the MXene electrode/electrolyte interface, while the straight line depicts mass transfer processes.³⁰ In the Nyquist plots, both MXene and D-MXene electrodes exhibit a vertical line in the low-frequency region, indicating rapid ion diffusion at the electrode–electrolyte interfaces and demonstrating ideal capacitive behaviour.³¹ The bulk solution resistance (R_s) indicated by the intercept on the real axis, accounts for the ionic resistance of the electrolyte. D-MXene shows an R_s of 1.21 Ω and a charge transfer resistance of 4.13 Ω, significantly lower than the 1.42 Ω solution resistance and 5.81 Ω charge transfer resistance observed for MXenes. These results highlight the superior electrical conductivity of D-MXene, which enhances its performance as an electrode material.

The Ragone plot, showing the variation in energy density (E_D) as a function of power density (P_D), is presented in Fig. 4e. MXenes achieve an energy density of 8.7 Wh kg⁻¹ with an impressive power output of 798.9 W kg⁻¹, while D-MXene significantly outperforms this, exhibiting an energy density of 45.9 Wh kg⁻¹ at a similar power density of 800.1 W kg⁻¹. In addition, the MXene and D-MXene based electrodes exhibit good long-term cycling stability with 90% and 92% capacitance retention after 10 000 cycles at a current density of 5 A g⁻¹ (Fig. 4f). The superior performance of D-MXene, with high capacitance retention and energy density in aqueous electrolytes, suggests that it could serve as a cost-effective replacement for advanced carbon-based materials in future energy storage applications.

2.3. Electrochemical evaluation of D-MXene/MWCNT composites

Furthermore, the electrochemical performance of D-MXene/MWCNT composites (MC-5, MC-10, MC-20, MC-30, and MC-40) was compared with that of the D-MXene. Cyclic voltammetry (CV) curves for both D-MXene and the D-MXene/MWCNT composites, recorded at different scan rates, are shown in Fig. S5a–f. These curves were analysed to assess how the incorporation of MWCNTs influences the electrochemical behaviour of the different compositions. The results clearly show that the D-MXene/MWCNT composites (MC-5, MC-10, and MC-20) exhibit significantly higher specific capacitance and better overall capacitive behaviour compared to D-MXene. Among them, the MC-20 electrode stands out, achieving a specific capacitance of 699 F g⁻¹, outperforming D-MXene, which exhibited 525 F g⁻¹ at 20 mV s⁻¹. This enhanced performance results from the porous microstructure of the composite, which enlarges the surface area and exposes more accessible active sites for the electrolyte. Additionally, CNTs create a conductive network that facilitates electron transport between the electrodes. It is also evident from the results that even a slight variation in MWCNT content influences the capacitive properties. The CV curves of MC-10 and MC-20 show a more rectangular shape compared to those of D-MXene and MC-5, indicating improved capacitive behaviour and reduced resistance. In particular, MC-20's CV profile exhibits a more pronounced rectangular shape than MC-10, reflecting better ion transport and higher capacitance. This difference can be attributed to the lower MWCNT content in MC-5, which leads to increased restacking of the MXene layers. When insufficient CNTs are present, the MXene layers become more compact and restacked, limiting ion transport and hindering electrolyte infiltration. Although the MXene itself possesses high conductivity, this tightly packed structure reduces the rate capability and overall performance of the composites. The introduction of MWCNTs serves as a critical factor in spacing out the MXene layers, enhancing ion accessibility and overall capacitive behaviour. However, as the CNT content increases from 20% to 30% and then to 40%, the composite begins to deviate from the typical electric double-layer capacitor (EDLC) behaviour. This is due to the excessive agglomeration of CNTs on the MXene surface, which hinders the ion pathways and reduces effective ion transport. Consequently, the capacitance decreases as the CNTs completely cover the MXene layers, impeding the overall electrochemical performance. Fig. S6a–f illustrates the galvanostatic charge–discharge curves recorded from 1 A g⁻¹ to 5 A g⁻¹. The curves for the MC-5, MC-10, MC-20, and MC-30 samples exhibit a nearly linear profile with minimal IR drop and a typical triangular shape, indicative of excellent electrochemical capacitive behaviour. However, the GCD curves for MC-40 deviate from this symmetric triangular shape, consistent with the cyclic voltammetry (CV) results. As expected, the charge–discharge duration increases with a decrease in current density, which is likely due to more complete ion insertion and better contact between the electrolyte and the electrode material.³² Among all the samples, MC-20 demonstrates the highest specific capacitance across all current densities, with a maximum value of 645 F g⁻¹ at 1 A g⁻¹. However, as the current density increases to 5 A g⁻¹, the specific capacitance of MC-20 drops to 497 F g⁻¹.

The CV curves of all samples were compared at a scan rate of 200 mV s⁻¹, as illustrated in Fig. 5a, and exhibited significantly larger CV areas for MC-20, compared to other electrodes. Variation of the specific capacitance values of the symmetric supercapacitor based on D-MXene/MWCNT composite electrodes is shown in Fig. 5b. Comparison of the GCD data of all samples showed nearly triangular symmetric charge–discharge curves, with longer charge and discharge times for MC-20 electrodes, compared to other electrodes at the same current densities (Fig. 5c). The longer charge/discharge times indicated higher capacitances. The capacitances were calculated from the GCD data and are presented in Fig. 5d at different current densities. The values of specific capacitance calculated from CV and GCD for D-MXene and all the MC-composites are tabulated in Tables S2 and S3. Notably, the maximum specific capacitance of MC-20 (645 F g⁻¹) showed a 1.25-fold increase compared to that of D-MXene (516 F g⁻¹) and a 6.5-fold increase compared to that of MXene (98 F g⁻¹) at a current density of 1 A g⁻¹.

Fig. 5 Evaluation of electrochemical performance of D-MXene and D-MXene/MWCNT composite based electrodes: (a) CV at 200 mV s⁻¹, (b) variation of C_sp with scan rates (c) GCD curves at 1 A g⁻¹, and (d) variation of C_sp with current densities. (e) Nyquist plot, (f) Ragone plot and (g) cycling stability over 10 000 GCD cycles at 5 A g⁻¹.

Furthermore, to investigate the ion transport dynamics within the composites, EIS measurements were conducted across the frequency range of 10 mHz to 100 KHz. The resulting Nyquist plot, shown in Fig. 5e, illustrates the impedance characteristics of D-MXene and D-MXene/MWCNT composite electrodes, with an enlarged view of the high-frequency region provided as an inset. In the low-frequency region, the MC-20 composite demonstrates a steeper slope and shorter Warburg line compared to the D-MXene electrode, reflecting superior capacitive behaviour. The solution resistance (R_s), obtained from the intersection point of the impedance curve with the real axis, was measured for each electrode. The R_s values for the D-MXene, MC-5, MC-10, MC-20, MC-30, and MC-40 electrodes were 1.21, 1.24, 1.22, 1.21, 1.34, and 1.37 Ω, respectively. These results indicate that R_s decreases as the CNT content increases up to a certain point, but begins to increase when the CNT content becomes excessive. The semicircular region in the high-frequency range provides insight into the charge transfer resistance (R_ct), which reflects ion transfer at the interface between the electrode material and the electrolyte.³³ The inset of Fig. 5e shows that the R_ct values for D-MXene, MC-5, MC-10, MC-20, MC-30, and MC-40 are 4.13, 1.34, 1.21, 1.17, 1.47, and 2.82 Ω, respectively. The R_ct values of MC-20 significantly reduced to approximately 1.17 Ω, which is nearly 3.5 times lower than that of D-MXene (4.13 Ω). Furthermore, the energy density and power density values are calculated using eqn (S3) and (S4). The Ragone plot in Fig. 5f shows the expected trend: as the power density increases, the energy density decreases for all samples. The maximum energy densities of 46.76, 50.13, 57.34, 40.62, and 23.29 Wh kg⁻¹ were recorded at power densities of 800.3, 801.1, 802.7, 799.1, and 798.8 W kg⁻¹ for MC-5, MC-10, MC-20, MC-30, and MC-40, respectively. These values are higher than those reported for similar electrodes, demonstrating that MC-20 has significant potential as an electrode material for energy storage devices.

To assess the stability of the device, a 10 000-cycle GCD test was conducted, with the results displayed in Fig. 5g. The MC-20 electrode demonstrated excellent stability, retaining 88% of its capacitance at a current density of 5 A g⁻¹. In comparison, the lowest performing electrode, MC-40, showed a gradual decrease in capacitance retention to 84%. The energy density, power density and the cycling stability data for these samples are summarized in Table S4. For practical assessment at the device level, the energy density of the symmetric two-electrode cell was calculated by accounting for the total active mass of both electrodes using eqn (S5) and (S6). This leads to a fourfold reduction compared to the single-electrode values. The calculated energy and power densities at both the single-electrode and device levels for all composite electrodes are summarized in Tables S5–S7.

The exceptional performance of the MC-20 electrode can be attributed to the highly organized structure of multi-walled carbon nanotubes (MWCNTs) integrated within the Ti₃C₂T_x MXene sheets. The well-dispersed CNTs provide a structural backbone that prevents the MXene layers from collapsing, ensuring better ion accessibility and charge transfer. This synergy between MWCNTs and Ti₃C₂T_x MXene not only enhances the electrochemical performance but also contributes to the long-term stability of the supercapacitor, making MC-20 an ideal candidate for high-performance energy storage devices. The performance metrics of MC-20 are benchmarked against recent literature and summarized in Table 1.

Table 1 Electrochemical performance comparison of MXene-based symmetric supercapacitor electrodes

Electrode material	Electrolyte	Voltage window	C sp (F g^−1)	E D (Wh kg^−1)	P D (W kg^−1)	Cyclic stability(%)	Ref.
MXene/CNT film	1 M H2SO4	−0.35 to 0.35 V	423.4 (1 A g^−1)	—	—	90.6% (5000 cycles)	34
MXene/N-CNT	6 M KOH	−1 to 0 V	167.2 (0.5 A g^−1)	12.1	195.4	73.2% (10 000 cycles)	35
MXene/CNT film	PVA/H2SO4 gel	−0.4 to 0.4 V	318 (1 A g^−1)	7.34	50	99% (5000 cycles)	36
MnO2@ MXene-CNT	1 M Na2SO4	0 to 1V	181.8 (1 A g^−1)	—	—	95% (1000 cycles)	30
MXene-CNT/PANI	1 M H2SO4	−0.7 to 0.1 V	429.4 (1 A g^−1)	—	—	93% (10 000 cycles)	37
MXene/CNT	1 M Na2SO4	−1 to 0 V	254 (0.5 A g^−1)	14.1	13 900	85% (5000 cycles)	38
D-MXene/MWCNT	6 M KOH	0 to 0.8 V	645 (1 A g^− ^1 )	46.8	800.3	88% (10 000 cycles)	This work

---

2.4. Fabrication of screen-printed electrodes for flexible MSCs

Producing a stable, shear-thinning ink suitable for screen printing on flexible substrates requires a balance of viscosity, dispersion stability, and film adhesion. Under the constraints of this study, and without extensive binder optimization, neither as-prepared D-MXene nor bare MWCNT dispersions exhibited adequate rheological properties for reproducible printing. Therefore, for the purpose of demonstrating device applicability, we employed a printable ink composed of the optimized D-MXene/MWCNT composite blended with PEDOT:PSS as the conductive binder, following our previously reported formulation protocol.³⁹ This approach yielded a stable, conductive, and flexible ink suitable for fabricating printed MSCs.

Formulating inks for printing flexible electrodes of D-MXene/MWCNT poses numerous challenges. Strong van der Waals interactions between the materials often demand a large quantity of polymeric dispersants, and conventional screen-printing formulations rely on insulating polymeric binders and elastomers for ink formulations. These polymers are often insulating and can degrade the charge storage capabilities of electrodes or might require harsh post printing treatments. To retain the electrochemical performance, as well as to achieve binding, dispersion and flexibility, we employed PEDOT: PSS as a multifunctional additive to formulate the ink. PEDOT:PSS is a conductive polymer, and well-studied as a conductive binder and dispersant for formulating MWCNT inks due to its high electrical conductivity, film-forming properties, flexibility and dispersion stability by forming π–π interactions with MWCNTs.⁴⁰ The ink was formulated using the optimized composite MC-20 as detailed in the SI and the final composite ink is termed D-MXene/MWCNT/PEDOT:PSS (MCP) composite ink. A schematic diagram illustrating the ink formulation and MSC fabrication is presented in Fig. 6a and b.

Fig. 6 Schematic representation of the (a) formulation of D-MXene/MWCNT composite ink and (b) the MSC fabrication process.

2.4.1 Properties of the ink and printed traces. The surface morphological features of the printed structures were evaluated using scanning electron microscopy, which revealed a rough, microporous surface (Fig. 7a). The observed microporosity is inevitable in printed structures, because of the evaporation of residual solvents.⁴¹ Upon a closer evaluation at a higher magnification, a continuous dense distribution of well-exfoliated Ti₃C₂T_x MXene sheets in a network of MWCNTs is observed (Fig. 7b). The sheets appeared to exfoliate better than the optimized MC-20 composite, probably due to the prolonged sonication and stirring cycles during the ink development. The printed traces also exhibited a nearly uniform cross-sectional thickness of about 23.2 µm (Fig. 7c). Printed MSCs demand that the electrodes be exposed to GPEs, which can wet the former, eventually leading to delamination or flaking. Hence, proper adhesion of the printed traces to the substrate is crucial for realizing electrodes for prolonged applications.⁴² The ASTM-D-3359-76 and DIN 53131 standards were followed to assess the rub-off resistance of the printed patterns. In this procedure, printed structures using the optimized ink were precisely cut with a parallel spacing of 0.2 cm using an X-Acto-type knife. A strip of 3M 600 clear Scotch tape was applied to the cut area, pressed down gently, and then quickly removed at a 180° angle. To evaluate the ink's adhesion to the substrate, the surface morphology of the printed traces along the cuts was observed both before and after tape removal, and the results were compared with the standard chart. Optical micrographs taken before and after the tape removal (Fig. S7a–f) show no significant flaking or delamination of the printed traces. Based on the comparison with the standard chart, the ink's adhesion was classified as ASTM Class-4, indicating that it is suitable for use in printed, wearable, and flexible electronics.

Fig. 7 SEM images of (a) & (b) the printed surface, (c) cross-section of the printed traces, (d) surface tension of the ink, (e) contact angle with the Mylar® substrate, (f) rotational viscometry, photographs of printed traces (inset), and (g) bending analysis of the printed traces.

Rheological characteristics of the ink determine the suitability of any ink for the screen-printing process. Ideally, screen inks should be non-Newtonian, shear-thinning liquids with low surface tension in the printing environment and should form a low contact angle with the substrate.^12,40 The interfacial rheological profiling of the formulated ink revealed a low surface tension (σ) of 23.9 mN m⁻¹ at room temperature (Fig. 7d) and an average contact angle (θ_c) of 49.9° with a Mylar^® substrate (Fig. 7e). The low surface tension and contact angle reveal the ink's excellent surface wettability and levelling capability, suggesting that the ink can produce clean, well-defined structures on Mylar^®.⁴³ Rotational viscometry conducted using a parallel plate model revealed that the ink exhibits non-Newtonian behaviour, as evidenced by a sharp increase in shear stress followed by a plateau (Fig. 7f).^44,45 Also, the ink exhibited a zero-shear viscosity of 9.06 Pa s, which drastically decreased to 1 Pa s at a shear rate of 5 s⁻¹ before stabilizing, confirming its shear-thinning nature (Fig. 7f).^44–46 This behaviour ensures that the ink remains on the screen during the initial flooding and flows smoothly through the screen under shear from the squeegee when printed. Hence, the rheological profiling qualifies the formulated ink for screen printing, and the ink exhibited an apparent viscosity of 1 Pa s at 5 s⁻¹, low surface tension, and contact angle. As mentioned earlier, the optimally formulated ink was then screen printed on Mylar® using a semi-automatic screen printer and dried at 100 °C. The ink generated neatly printed traces of even alphabets without losing structural fidelity, despite the marginally high contact angle (Fig. 7f, inset). The printed traces exhibited liner current (I)–voltage (V) dependence, identifying it as an ideal ohmic conductor (Fig. 7g). The traces also maintained this linear I–V relationship even after the 10 000th bending cycle to a radius of 5 mm, revealing high flexibility. The sheet resistance of the printed structure was evaluated by printing a 3 × 3 square patch, and zone-wise evaluation was carried out. At 1 mA source current, the printed traces displayed an average sheet resistance of 52.30 ± 0.02 Ω □⁻¹. Profilometry measurements determined the print thickness to be 23.31 ± 0.14 µm. The conductivity of the prints was calculated using eqn (1) (ref. 12 and 40) and found to be (8.30 ± 0.01) × 10³ S m⁻¹.

2.4.2 Fabrication of all printed micro-supercapacitors (MSCs). Micro-supercapacitors were fabricated using the D-MXene/MWCNT/PEDOT: PSS (MCP) ink through a multistep printing process. The Mylar^® surface was washed with acetone and ethanol before printing, and then dried under vacuum for 3 hours at a temperature of 80 °C. The identified interdigitated electrode-type (IDE) bottom current collector (CC) was subsequently printed and dried at 100 °C for 3 hours (Fig. S8a). IDEs of MCP were then printed using the optimized printing process and dried at 100 °C for 30 minutes (Fig. S8b). The top CC was then printed using conductive silver ink after precisely aligning the IDE with the screen using the printer's MOPS system. The printed structures were dried at 100 °C for 30 minutes to obtain IDEs with current collectors without delamination or cracking (Fig. S8c). The fabricated GPE was then printed through laser-cut stencils after aligning using an optical microscope. GPE was printed subsequently so that it did not encounter the current collector to minimize electrolyte–current collector interaction. After printing, the GPE remained stable without oozing or dripping (Fig. S8d). Even after the GPE printing, the printed electrodes stayed firmly attached to the substrate without dissolving into the GPE, confirming the ink's strong adhesion and suitability for creating printed electrodes for micro-supercapacitors. Furthermore, the fabricated MSC was encapsulated via vacuum sealing, with PET sheets laminated at the edges, allowing the contact pads of the current collectors to extend outside (Fig. S8e).

2.4.3 Electrochemical characteristics of the D-MXene/MWCNT/PEDOT: PSS ink based MSC. The D-MXene/MWCNT/PEDOT: PSS (MCP) ink was used to print an electrode on a Mylar^® substrate to fabricate a micro-supercapacitor suitable for flexible transparent electronic devices. Before printing the interdigitated electrodes (IDEs), the optimized MCP ink-based electrodes were fabricated by coating the ink onto two 1 cm² pieces of conductive carbon cloth and vertically stacking them face-to-face, with a Celgard 3501 separator soaked in 6 M KOH electrolyte placed between them. After an infiltration period of 24 hours, the electrochemical properties were evaluated in a standard EL-cell setup using two-electrode measurements. Cyclic voltammetry (CV) profiles were recorded at scan rates ranging from 5 to 200 mV s⁻¹ within a potential window of 0–0.8 V (Fig. S9a). The CV curves deviate from the ideal rectangular shape observed for MC-20, showing a pseudo-rectangular profile, especially at higher scan rates, likely due to the pseudo-capacitive behaviour of PEDOT: PSS.⁴⁷ From the CV analysis, the MCP electrodes demonstrated an impressive specific capacitance of 530 F g⁻¹ at a scan rate of 20 mV s⁻¹, which was enhanced to a maximum of 670 F g⁻¹ at 5 mV s⁻¹ as the scan rate decreased. The excellent capacitive performance of the MCP electrodes is further confirmed by galvanostatic charge–discharge (GCD) tests, which show longer discharge times and exhibited a maximum specific capacitance of 574 F g⁻¹ at a current density of 1 A g⁻¹ (Fig. S9b). The specific capacitance values derived from CV and GCD for MCP electrodes coated on carbon cloth are summarized in Table S8. The Nyquist plot (Fig. S9c) reveals that the charge transfer resistance (R_ct) was measured to be 14.56 Ω, while the solution resistance (R_s) was 1.97 Ω. These EIS results are consistent with the CV and GCD data, highlighting the strong performance of MCP electrodes. This enhanced pseudo-capacitive behaviour is attributed to the incorporation of PEDOT: PSS, which prevents MXene nanosheet aggregation while its film-forming properties ensure good binding with MWCNTs, without compromising conductivity. To further elucidate the charge-storage mechanism, the electrochemical kinetics were analyzed using the power–law relationship:

i = aν^b (2)

where i is the total current at a fixed potential, ν is the scan rate, and a and b are adjustable kinetic parameters. A b value of 0.5 indicates diffusion-controlled behaviour, whereas b = 1 corresponds to a purely capacitive process. The value of b may be derived from the slope of the log(j) against log(ν) plot, where j denotes the current density (Fig. S10a). The extracted b values of D-MXene, MC-20 and MCP electrodes are summarized in Table S9. D-MXene exhibits predominantly capacitive behaviour (b = 0.938), MC-20 shows a near-unity b value (b = 0.986), indicating almost entirely surface-controlled charge storage enabled by MWCNTs, while MCP shows a reduced b value (b = 0.869), reflecting the charge-storage behaviour that remains primarily electric double-layer capacitance (EDLC)-dominated, with an additional diffusion-influenced pseudocapacitive contribution arising from the incorporation of PEDOT:PSS.

Dunn's method was further applied to quantitatively separate surface-controlled and diffusion-controlled contributions according to:

i = k₁ν + k₂ν^1/2 (3)

where k₁ν and k₂ν^1/2 correspond to surface-controlled and diffusion-controlled processes, respectively. Fig. S10b presents the linear fitting of versus from which the kinetic parameters k₁ (slope) and k₂ (y-intercept) are extracted and summarized in Table S9.

As shown in Fig. S10c, surface-controlled capacitance dominates across all electrodes, contributing approximately 92.9%, 98.4%, and 85.6% at 20 mV s⁻¹ for D-MXene, MC-20, and MCP, respectively. These results indicate that the addition of CNTs increases the capacitive contribution from 92.9% to 98.4%, while the subsequent incorporation of PEDOT:PSS introduces a modest diffusion-influenced behaviour without altering the predominantly capacitive nature of the system.

The optimally formulated ink was then utilized to print interdigitated electrodes (IDEs) on a Mylar^® substrate, enabling the fabrication of all-printed micro-supercapacitors (MSCs) in a symmetric, two-electrode configuration. Initially, IDE-patterned bottom current collectors were printed on the Mylar^® substrate using conductive silver ink, followed by a layer of MCP ink, which was then subjected to drying. Once dried, these electrodes were meticulously aligned, and the top current collectors were printed using the same silver ink. For electrochemical characterization, a gel polymer electrolyte (GPE) composed of 1 M LiClO₄-PC-PMMA-FS was prepared and carefully coated onto the interdigitated electrodes of the printed device.

Fig. 8a illustrates the cyclic voltammograms (CVs) of printed IDEs recorded at scan rates of 5, 10, 20, 50, 100, and 200 mVs⁻¹. At lower scan rates, the CVs exhibit a quasi-rectangular shape within a potential window of 0.8 V, indicating stable capacitive behaviour. However, at higher scan rates, the curves shift towards a more linear profile, likely due to the influence of the silver paste used as the current collector. At a scan rate of 20 mV s⁻¹ the device exhibited an areal capacitance of 72 mF cm⁻². As anticipated, the areal capacitance for all devices decreases as the scan rate increases. The current output from the printed IDEs is relatively low at rapid scan rates, leading to non-linear CV profiles. This behaviour is expected due to the horizontal alignment of MXene flakes, which restricts out-of-plane electrical conductivity and hampers electrochemical performance, especially at higher charge/discharge rates.

Fig. 8 Electrochemical characteristics of the D-MXene/MWCNT/PEDOT: PSS ink based MSC. (a) CV studies conducted at scan rates ranging from 5 mV s⁻¹ to 200 mV s⁻¹, (b) GCD curves at current densities ranging from 0.1 mA cm⁻² to 1 mA cm⁻², (c) variation of areal capacitance with increasing scan rates and current densities, (d) Ragone plot (e) Nyquist plot, and (f) the corresponding equivalent circuit.

A similar trend is observed in the GCD curves (Fig. 8b), where each curve shows an IR drop of approximately 0.2 V, attributed to the influence of the silver paste current collector. Despite this voltage drop, the GCD curves maintain a predominantly triangular shape, indicative of the supercapacitive behaviour of the printed IDEs and their extended discharge time. The areal capacitance (C_areal) at different scan rates and current densities was calculated using eqn (S7) and (S8), and the results are presented in Fig. 8c. The areal capacitance decreases from 97 mF cm⁻² at 5 mV s⁻¹ to 26 mF cm⁻² at 200 mV s⁻¹, indicating the typical rate-dependent behaviour of ion transport within the electrode. Similarly, with increasing current density, the areal capacitance decreases from 53 mF cm⁻² at 0.1 mA cm⁻² to 17 mF cm⁻² at 1 mA cm⁻², attributable to limited electrolyte ion diffusion at higher charge–discharge rates. For practical comparison, the gravimetric (specific) capacitance values of the printed electrodes were also calculated and systematically compared with the corresponding areal capacitances; the summarized values are provided in Table S10.

Furthermore, the Ragone plots (Fig. 8d) reveal a notable areal energy density (E_Areal) of 4.7 µWh cm⁻² at an areal power density (P_Areal) of 80 µW cm⁻², as calculated from eqn (S9) and (S10). For comparison, the corresponding volumetric energy and power densities (E_Vol and P_Vol) were also calculated from eqn (S11) and (S12), and their Ragone plot is presented in Fig. S11a. Electrochemical impedance spectroscopy (EIS) was then employed to investigate the charge transport characteristics of the screen-printed MXene-based IDEs. The Nyquist plots were recorded over a wide frequency range, and the experimental data were fitted using an equivalent circuit model implemented in ZSimpWIN3.21 software. The fitted equivalent–circuit parameters for the printed MSC are summarized in Table S11. As shown in Fig. 8e, the Nyquist plot exhibits a semicircular pattern in the high-frequency region, followed by a nearly linear response in the low-frequency region. The measured equivalent series resistance (R_s) and charge-transfer resistance (R_ct) values were 65.5 Ω and 81.5 Ω, respectively. The relatively higher R_ct can be attributed to the use of GPE, which generally exhibits higher ionic resistance than liquid electrolytes due to limited ion mobility within the polymer matrix.³⁹ Furthermore, contact resistance at the electrode-current collector interface and the interdigitated device geometry contribute to the overall impedance. Collectively, these resistive components result in a noticeable IR drop of ∼0.2 V during galvanostatic charge–discharge, indicating voltage losses associated with both ionic and electronic transport limitations.

The equivalent circuit used to fit the impedance data is illustrated in Fig. 8f. In this model, the series resistance (R_s), obtained from the high-frequency intercept, represents the combined contributions from the intrinsic resistance of the printed electrode, the electrolyte, and the current collector. The interfacial response is modeled by a parallel combination of charge-transfer resistance (R_ct) and a constant phase element (CPE, Q), where R_ct corresponds to the resistance at the electrode–electrolyte interface, and the CPE accounts for non-ideal double-layer capacitive behaviour of the device. The Warburg element (W) incorporated in the low-frequency region represents ion diffusion within the electrode. Overall, the electrode and electrochemical performance parameters of the MCP-based MSC are summarized in Table 2.

Table 2 Physical parameters and electrochemical performance of the D-MXene/MWCNT/PEDOT:PSS based printed electrodes

Electrode parameters	Values
Active electrode area	0.5 cm^2
Active electrode mass	0.7 mg
Areal mass density	1.5 mg cm^−2
Electrode thickness	23.3 µm
Specific capacitance, Csp	34.3 F g^−1 at 0.1 A g^−1
Areal capacitance, Careal	53 mF cm^−2 at 0.1 mA cm^−2
Volumetric capacitance, CVol	22.7 F cm^−3 at 0.1 mA cm^−2
Solution resistance, Rs	65.5 Ω
Charge transfer resistance, Rct	81.5 Ω
Cyclic stability	96.7% (8000 GCD cycles)
Energy density, (Esp, EAreal, and EVol)	3.36 Wh kg^−1, 4.704 µWh cm^−2 and 2.019 mWh cm^−3
Power density, (Psp, PAreal and PVol)	53.32 W kg^−1, 79.99 mW cm^−2 and 34.33 W cm^−3

---

2.4.4 Flexibility, bending studies and series-parallel evaluation. To assess the long-term cycling stability of the device, GCD analysis was performed over 8000 continuous cycles at a current density of 0.5 mA cm⁻², demonstrating a 96.7% retention of capacitance, highlighting the excellent durability of the device (Fig. 9a). The MSC was subjected to bending deformations of different diameters (1.5, 1, 0.5, and 0.2 cm) using a custom-built bending deformation tester (Fig. S12a and b) to evaluate the reliability for wearable applications. To assess the impact of bending on electrochemical performance, cyclic voltammetry (CV) tests were conducted at a scan rate of 200 mV s⁻¹ across these bending diameters and are presented in Fig. 9b. The results show that the CV curves remained unchanged, indicating that the supercapacitor maintained its capacitive characteristics even at a minimal bending diameter of 0.2 cm (Video S1). Additionally, the MSC was subjected to repeated extreme bending (bending diameter of 1.5 cm) at a frequency of 4 Hz using a modified triboelectric measurement system (Holmarc, India) (Video S2) (Fig. 9c). Remarkably, the device demonstrated excellent stability, retaining its performance with no obvious change in the shapes of CV profiles even after 8000 bending cycles demonstrating exceptional mechanical robustness of a customised MSC.

Fig. 9 Reliability analysis of an MCP based MSC. (a) Cycling stability performance over extended cycles. Flexibility studies with CV measurements under (b) various bending radii and (c) repeated cycles under extreme bending. Images showing 3 MSCs in (d) series connection and (e) parallel connection. CV profiles of MSCs in (f) series and (g) parallel configurations.

The leakage current and self-discharge characteristics are important parameters to validate the practical applicability of the device. The self-discharge test of the MSC was performed by charging the device from 0 to 0.8 V for 1 h, followed by monitoring the open-circuit potential (OCP) over time. The OCP profile (Fig. S13a) shows a rapid voltage drop during the first hour, followed by a gradual decay. Even after 4.5 h, the MSC retains an output voltage of approximately 0.25 V, which indicates good charge retention. Furthermore, to evaluate the leakage current, the device was charged to 0.8 V and held at this voltage for 1 h while recording the current response (Fig. S13b). After the initial decay, the current becomes stable at ∼0.04 mA. This steady-state current is taken as the leakage current, confirming the good electrochemical stability of the device. All electrochemical and mechanical stability measurements were conducted under ambient laboratory conditions at room temperature (∼25 °C). In addition, the device preserved its structural integrity throughout testing, with no observable electrode delamination or gel polymer electrolyte (GPE) leakage, confirming the high flexibility and durability of the printed electrodes and current collector, making it highly suitable for wearable applications. To enhance the power and energy densities, multiple MSCs can be printed on the same substrate and connected in series or parallel configurations, as illustrated in Fig. 9d and e. The CV profiles of a single MSC compared with those of two and three MSCs interconnected in series and parallel arrangements are shown in Fig. 9f and g. As anticipated, connecting MSCs in series extends the operational voltage window, while parallel connections lead to an increase in current output. The CV curves for the series and parallel combinations of 1, 2, and 3 MSCs demonstrate the practical viability of these configurations, effectively showcasing their potential for scaling energy storage and power delivery in integrated devices. The ink, composed of a stable mixture of D-MXene/MWCNT/PEDOT: PSS, displayed remarkable areal capacitance and mechanical stability, maintaining performance under various bending conditions. The composite ink was optimized to maximize conductivity and long-term stability. The resulting interdigitated pattern displayed excellent flexibility and moderate sheet resistance. Moreover, the device showed robust electrochemical performance even after multiple bending cycles, with minimal changes in areal capacitance, making it suitable for use in flexible electronics. While the composite design effectively minimized the restacking tendency of MXenes, a complete suppression was not achieved, which may have a minor influence on ion diffusion and charge transport at higher mass loadings. At the device level, the internal resistance and associated IR drop are mainly governed by the resistance of the interdigitated current collectors and the limited ionic conductivity of the gel polymer electrolyte (GPE), in agreement with the observed R_s and R_ct values. In order to achieve consistent printability, the ink formulation required the addition of several processing additives, which can lead to small variations in the electrochemical response of the printed devices. Moreover, oxidation of MXenes under ambient processing conditions and during prolonged cycling remains a potential concern for long term device stability. Future work will therefore focus on low resistance current collectors including Ag-carbon hybrid collectors and graphene/carbon-based current collectors. In parallel, efforts will be directed toward developing higher-conductivity and better-infiltrating GPE formulations and adopting effective strategies to suppress MXene oxidation. Further optimization of electrode thickness and interdigitated architecture will be pursued to better balance areal capacitance, ion transport, mechanical flexibility, and resistive losses.

Despite these limitations, the developed ink formulation and device architecture demonstrate strong potential for scalable fabrication of high-performance, flexible micro-supercapacitors. Overall, the obtained performance is on par with many reported supercapacitors and batteries, underscoring the device's exceptional energy storage capabilities. To clearly position the performance of the present printed MXene-based micro-supercapacitor relative to the state of the art, an overlaid Ragone plot comparing the areal energy and power densities of our device with recently reported printed MXene-based MSCs is presented in Fig. S11b, while a detailed comparison of the electrochemical performance with recent literature is summarized in Table 3.

Table 3 Electrochemical performance comparison of MXene-based micro-supercapacitors

Electrode material	Electrolyte	Voltage window	C areal	E D	P D	Cycling stability (%)	Ref.
MXene/Ag NWs/PEDOT:PSS	H3PO4/PVA polymer gel	0–1 V	8.34 mF cm^−2	1.5 µWh cm^−2	0.6 mW cm^−2	81% (10 000 cycles)	48
MXene-CNT MSC	PVA/H2SO4 hydrogel	0–0.8 V	270.80 at 1 mA cm^−2	111 µWh cm^−2	0.4 mW cm^−2	80% (5000 cycles)	49
Ni/MXene MSC	1.0 M NaOH	0–1.4 V	756 mFcm^−2 at 1.5 mAcm^−2	206 µWh cm^−2	1.94 mW cm^−2	89.3% (7000 cycles)	50
MWCNTs-MXene@CC	1 M H2SO4	0–0.4 V	114.58 mF cm^−2 at 1 mA cm^−2	22.11 mWh cm^−3	2.99 W cm^−3	95.4% (5000 cycles)	51
MXene-CNT MSC	PVA/H2SO4 gel	0–0.8 V	61.38 mF cm^−2 at 0.5 mA cm^−2	5.46 mWh cm^−3	0.20 W cm^−3	85% (5000 cycles)	52
MWCNT/PEDOT:PSS MSC	PVA/H2SO4 gel	0–1 V	20.6 mF cm^−2 at 0.1 mA cm^−2	2.02 µWh cm^−2	2.82 µW cm^−2	99.9% (20 000 cycles)	53
3D MXene-rGO MSC	Ionic liquid	0-3V	66.69 mF cm^2 at 0.9 mA cm^2	83.4 µWh cm^−2	350 µW cm^−2	95% (5000 cycles)	54
MXene-PANI@MWCNTs MSC	PVA/H2SO4 gel	0–1 V	30.2 mF cm^−2 at 0.1 mA cm^−2	—	—	70.2% (10 000 cycles)	55
3D MXene-rGO aerogel	PVA/H2SO4 gel	0–0.6 V	34.6 mF cm^−2 at 1 mV s^−1	2.18 µWh cm^−2	60 µWcm^−2	91% (15 000 cycles)	56
MWCNT/PEDOT: PSS MSC	PMMA-PC-LiClO4	0–1.2	9.82 mF cm^−2 at 0.06 mA cm^−2	1.964 µWh cm^−2	63 µW cm^−2	99.7% (6000 cycles)	39
D-MXene/MWCNT/PEDOT:PSS (MCP) MSC	1 M PMMA-PC-LiClO 4	0-0.8 V	53 mF cm^− ^2 at 0.1 mA cm^− ^2	4.7 µWh cm^− ^2	80 µW cm^− ^2	96.7% (8000 cycles)	This work

---

3 Conclusions

This study presents an effective strategy for developing high-performance MXene-based composite electrodes through the systematic optimization of material architecture and device design. Delaminated Ti₃C₂T_x (D-MXene) was integrated with an optimized amount of MWCNTs to suppress restacking, enhance ion accessibility, and improve charge transport. The composite containing 20 wt% MWCNT (MC-20) exhibited the most favorable characteristics, with an increased surface area and pore volume, forming a conductive network that effectively prevents MXene re-stacking. The optimized MC-20 electrode delivered a high specific capacitance of 645 F g⁻¹ at 1 A g⁻¹ and retained 88% of its capacitance after 10 000 cycles, confirming excellent electrochemical durability. To demonstrate practical applicability, a flexible, screen-printed micro-supercapacitor (MSC) was fabricated by co-optimizing the material architecture and device design. The printed MSC achieved a high areal capacitance of 53 mF cm⁻² at 0.1 mA cm⁻², along with excellent long-term cycling stability (retaining 96.7% of its capacitance after 8000 cycles). Furthermore, the device exhibited outstanding mechanical robustness, maintaining stable performance at a bending radius of 0.2 cm and after 10 000 bending cycles. Overall, this study demonstrates a scalable route for integrating MXene-based composites into printable, high-performance MSCs. By synergistically combining material design with device architecture optimization, the study addresses key challenges in translating laboratory-scale MXene electrodes into flexible energy-storage modules. The presented approach offers promising potential for future integration into wearable electronics, distributed IoT nodes, and smart textile platforms, where lightweight, low-profile, and reliable energy storage is essential.

Author contributions

Sophy Mariam Varghese: conceptualization, data curation, formal analysis, investigation, methodology, and writing – original draft. Adarsh Sivan Pillai: data curation, formal analysis, investigation, methodology, writing – review and editing. Raghavan Baby Rakhi: conceptualization, project administration, resources, supervision, validation, funding acquisition, review and editing of the manuscript. Surendran Kuzhichalil Peethambharan: conceptualization, supervision, validation, review, and editing of the manuscript.

Conflicts of interest

The authors declare that they have no known competing financial or personal conflicts of interest that could have influenced the work reported in this paper.

Data availability

The data supporting this article have been included as part of the supplementary information (SI) and will be made available upon request. Supplementary information is available. See DOI: https://doi.org/10.1039/d5ta09649a.

Acknowledgements

The research was supported through the IC MAP project (DST/TMD/IC-MAP/2K20/01), the Advanced Manufacturing Technology project (DST/TDT/AMT/2021/001 G) provided by Department of Science & Technology (DST), Ministry of Science & Technology, India, the Core Research Grant project (CRG/2021/0495) from Science & Engineering Research Board (SERB), New Delhi, and the Mission Mode Project (MMP 035201) from the Council of Scientific and Industrial Research (CSIR) New Delhi. The authors are also thankful for Department of Science & Technology (DST)- India INSPIRE scheme and to University Grant Commission (UGC) for a research fellowship. The authors also acknowledge Dr Subrata Das and Ms. Anjali Santhosh for XRD analysis, Mr Harish Raj for SEM, Mr Kiran Mohan for TEM, Dr Hareesh U. S. and Mr Peer Mohammed for BET surface area and viscosity measurements, and Dr E. Bhoje Gowd, Ms. Lisha V S, Ms. Sruthi Suresh and Mr Amal Raj R. B. for their assistance with the synthesis procedure, WAXD, and contact angle and surface tension measurements.

References

1. Y. Wang and Y. Wang, MXene ink printing of high-performance micro-supercapacitors, Carbon Neutralization, 2024, 3, 798–817,  DOI:10.1002/cnl2.165.
2. C. John) Zhang, M. P. Kremer, A. Seral-Ascaso, S.-H. Park, N. McEvoy, B. Anasori, Y. Gogotsi and V. Nicolosi, Stamping of Flexible, Coplanar Micro-Supercapacitors Using MXene Inks, Adv. Funct. Mater., 2018, 28, 1705506,  DOI:10.1002/adfm.201705506.
3. W. Yang, J. Yang, J. J. Byun, F. P. Moissinac, J. Xu, S. J. Haigh, M. Domingos, M. A. Bissett, R. A. W. Dryfe and S. Barg, 3D Printing of Freestanding MXene Architectures for Current-Collector-Free Supercapacitors, Adv. Mater., 2019, 31, 1902725,  DOI:10.1002/adma.201902725.
4. X. Feng, M. Xia, J. Ning and D. Wang, Interface-Modified Ti3C2Tx MXene/1T-WSe2 Heterostructure for High-Capacitance Micro-Supercapacitors, ACS Appl. Energy Mater., 2023, 6, 6391–6400,  DOI:10.1021/acsaem.2c04046.
5. J. Orangi, F. Hamade, V. A. Davis and M. Beidaghi, 3D Printing of Additive-Free 2D Ti3C2Tx (MXene) Ink for Fabrication of Micro-Supercapacitors with Ultra-High Energy Densities, ACS Nano, 2020, 14, 640–650,  DOI:10.1021/acsnano.9b07325.
6. X. Shi, F. Guo, K. Hou, G. Guan, L. Lu, Y. Zhang, J. Xu and Y. Shang, Highly Flexible All-Solid-State Supercapacitors Based on MXene/CNT Composites, Energy Fuels, 2023, 37, 9704–9712,  DOI:10.1021/acs.energyfuels.3c01420.
7. E. Kim, B.-J. Lee, K. Maleski, Y. Chae, Y. Lee, Y. Gogotsi and C. W. Ahn, Microsupercapacitor with a 500 nm gap between MXene/CNT electrodes, Nano Energy, 2021, 81, 105616,  DOI:10.1016/j.nanoen.2020.105616.
8. S. M. Varghese, R. S. Srihari, M. O. Anagha and R. B. Rakhi, Transition Metal Oxide–Decorated Titanium Carbide MXene Electrodes for High-Energy Supercapacitors with an Expanded Potential Window, Batter. Supercaps, 2025, 2500304,  DOI:10.1002/batt.202500304.
9. O. Mashtalir, M. Naguib, V. N. Mochalin, Y. Dall'Agnese, M. Heon, M. W. Barsoum and Y. Gogotsi, Intercalation and delamination of layered carbides and carbonitrides, Nat. Commun., 2013, 4, 1716,  DOI:10.1038/ncomms2664.
10. S. Munir, A. Rasheed, T. Rasheed, I. Ayman, S. Ajmal, A. Rehman, I. Shakir, P. O. Agboola and M. F. Warsi, Exploring the Influence of Critical Parameters for the Effective Synthesis of High-Quality 2D MXene, ACS Omega, 2020, 5, 26845–26854,  DOI:10.1021/acsomega.0c03970.
11. L. Liu, M. Orbay, S. Luo, S. Duluard, H. Shao, J. Harmel, P. Rozier, P.-L. Taberna and P. Simon, Exfoliation and Delamination of Ti3C2Tx MXene Prepared via Molten Salt Etching Route, ACS Nano, 2022, 16, 111–118,  DOI:10.1021/acsnano.1c08498.
12. A. S. Pillai, A. Chandran and S. K. Peethambharan, MWCNT Ink with PEDOT:PSS as a multifunctional additive for energy efficient flexible heating applications, Appl. Mater. Today, 2021, 23, 100987,  DOI:10.1016/j.apmt.2021.100987.
13. X. Guan, Z. Yang, M. Zhou, L. Yang, R. Peymanfar, B. Aslibeiki and G. Ji, 2D MXene Nanomaterials: Synthesis, Mechanism, and Multifunctional Applications in Microwave Absorption, Small Struct., 2022, 3, 2200102,  DOI:10.1002/sstr.202200102.
14. W. Meng, X. Liu, H. Song, Y. Xie, X. Shi, M. Dargusch, Z. G. Chen, Z. Tang and S. Lu, Advances and challenges in 2D MXenes: From structures to energy storage and conversions, Nano Today, 2021, 40, 101273,  DOI:10.1016/j.nantod.2021.101273.
15. S. M. Varghese, V. V. Mohan, S. Suresh, E. Bhoje Gowd and R. B. Rakhi, Synergistically modified Ti3C2Tx MXene conducting polymer nanocomposites as efficient electrode materials for supercapacitors, J. Alloys Compd., 2024, 973, 695019,  DOI:10.1016/j.jallcom.2023.172923.
16. X. Li, G. Fan and C. Zeng, Synthesis of ruthenium nanoparticles deposited on graphene-like transition metal carbide as an effective catalyst for the hydrolysis of sodium borohydride, Int. J. Hydrogen Energy, 2014, 39, 14927–14934,  DOI:10.1016/j.ijhydene.2014.07.029.
17. J. W. Pinder, G. H. Major, D. R. Baer, J. Terry, J. E. Whitten, J. Čechal, J. D. Crossman, A. J. Lizarbe, S. Jafari, C. D. Easton, J. Baltrusaitis, M. A. van Spronsen and M. R. Linford, Avoiding common errors in X-ray photoelectron spectroscopy data collection and analysis, and properly reporting instrument parameters, Appl. Surf. Sci. Adv., 2024, 19, 100534,  DOI:10.1016/j.apsadv.2023.100534.
18. B. Ahmed, D. H. Anjum, M. N. Hedhili, Y. Gogotsi and H. N. Alshareef, H2O2 assisted room temperature oxidation of Ti2C MXene for Li-ion battery anodes, Nanoscale, 2016, 8, 7580–7587,  10.1039/C6NR00002A.
19. J. Halim, K. M. Cook, M. Naguib, P. Eklund, Y. Gogotsi, J. Rosen and M. W. Barsoum, X-ray photoelectron spectroscopy of select multi-layered transition metal carbides (MXenes), Appl. Surf. Sci., 2016, 362, 406–417,  DOI:10.1016/j.apsusc.2015.11.089.
20. V. Natu, M. Benchakar, C. Canaff, A. Habrioux, S. Célérier and M. W. Barsoum, A critical analysis of the X-ray photoelectron spectra of Ti3C2Tz MXenes, Matter, 2021, 4, 1224–1251,  DOI:10.1016/j.matt.2021.01.015.
21. L. Liu, M. Orbay, S. Luo, S. Duluard, H. Shao, J. Harmel, P. Rozier, P. L. Taberna and P. Simon, Exfoliation and Delamination of Ti3C2TxMXene Prepared via Molten Salt Etching Route, ACS Nano, 2022, 16, 111–118,  DOI:10.1021/acsnano.1c08498.
22. A. Thakur, N. Chandran B.S., K. Davidson, A. Bedford, H. Fang, Y. Im, V. Kanduri, B. C. Wyatt, S. K. Nemani, V. Poliukhova, R. Kumar, Z. Fakhraai and B. Anasori, Step-by-Step Guide for Synthesis and Delamination of Ti3C2Tx MXene, Small Methods, 2023, 7, 2300030,  DOI:10.1002/smtd.202300030.
23. L. Yang, W. Zheng, P. Zhang, J. Chen, W. B. Tian, Y. M. Zhang and Z. M. Sun, MXene/CNTs films prepared by electrophoretic deposition for supercapacitor electrodes, J. Electroanal. Chem., 2018, 830–831, 1–6,  DOI:10.1016/j.jelechem.2018.10.024.
24. Y. O. Yesilbag and A. J. S. Salih, Optimization of MXene/CNT free-standing composite electrodes for enhanced electrochemical performance in symmetric supercapacitors, Mater. Res. Bull., 2026, 193, 113703,  DOI:10.1016/j.materresbull.2025.113703.
25. M. Thommes, K. Kaneko, A. V Neimark, J. P. Olivier, F. Rodriguez-reinoso, J. Rouquerol and K. S. W. Sing, Physisorption of Gases , with Special Reference to the Evaluation of Surface Area and Pore Size Distribution, IUPAC Technical Report, 2015, 87, 1051–1069,  DOI:10.1515/pac-2014-1117.
26. K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Reporting Physisorption Data for Gas/Solid Systems with Special Reference to the Determination of Surface Area and Porosity, IUPAC Commission on Colloid and Surface Chemistry Including Catalysis, Pure Appl. Chem., 1985, vol. 57, pp. 603–619,  DOI:10.1351/pac198557040603.
27. S. Xu, G. Wei, J. Li, Y. Ji, N. Klyui, V. Izotov and W. Han, Binder-free Ti3C2Tx MXene electrode film for supercapacitor produced by electrophoretic deposition method, Chem. Eng. J., 2017, 317, 1026–1036,  DOI:10.1016/j.cej.2017.02.144.
28. G. Wang, L. Zhang and J. Zhang, A review of electrode materials for electrochemical supercapacitors, Chem. Soc. Rev., 2012, 41, 797–828,  10.1039/c1cs15060j.
29. R. Syamsai, A. N. Grace, G. A. Babu, K. B. Karuppanan, S. K. Eswaran and M. Navaneethan, Titanium-Tantalum Double-Ordered MXene Nanosheets as Supercapacitor Electrodes, ACS Appl. Nano Mater., 2023, 6, 5224–5232,  DOI:10.1021/acsanm.2c05081.
30. Q. Liu, J. Yang, X. Luo, Y. Miao, Y. Zhang, W. Xu, L. Yang, Y. Liang, W. Weng and M. Zhu, Fabrication of a fibrous MnO2@MXene/CNT electrode for high-performance flexible supercapacitor, Ceram. Int., 2020, 46, 11874–11881,  DOI:10.1016/j.ceramint.2020.01.222.
31. S. M. Varghese, S. R. Sarath Kumar and R. B. Rakhi, Carbide-derived carbon by room temperature chemical etching of MAX phase for supercapacitor application, Appl. Phys. Lett., 2024, 124(2), 023905,  DOI:10.1063/5.0178716.
32. L. Li, F. Wang, J. Zhu and W. Wu, The facile synthesis of layered Ti2C MXene/carbon nanotube composite paper with enhanced electrochemical properties, Dalton Trans., 2017, 46, 14880–14887,  10.1039/c7dt02688a.
33. T. S. Mathis, N. Kurra, X. Wang, D. Pinto, P. Simon and Y. Gogotsi, Energy Storage Data Reporting in Perspective—Guidelines for Interpreting the Performance of Electrochemical Energy Storage Systems, Adv. Energy Mater., 2019, 9, 1902007,  DOI:10.1002/aenm.201902007.
34. W. Xu, Z. Xu, Y. Liang, L. Liu and W. Weng, Enhanced tensile and electrochemical performance of MXene/CNT hierarchical film, Nanotechnology, 2021, 32(35), 355706,  DOI:10.1088/1361-6528/ac04cf.
35. Q. Wang, H. Yuan, M. Zhang, N. Yang, S. Cong, H. Zhao, X. Wang, S. Xiong, K. Li and A. Zhou, A Highly Conductive and Supercapacitive MXene/N-CNT Electrode Material Derived from a MXene-Co-Melamine Precursor, ACS Appl. Electron. Mater., 2023, 5, 2506–2517,  DOI:10.1021/acsaelm.2c01768.
36. X. Shi, F. Guo, K. Hou, G. Guan, L. Lu, Y. Zhang, J. Xu and Y. Shang, Highly Flexible All-Solid-State Supercapacitors Based on MXene/CNT Composites, Energy Fuels, 2023, 37, 9704–9712,  DOI:10.1021/acs.energyfuels.3c01420.
37. Y. Cai, Y. Fang, W. Cao, P. He and M. Cao, MXene-CNT/PANI ternary material with excellent supercapacitive performance driven by synergy, J. Alloys Compd., 2021, 868, 159159,  DOI:10.1016/j.jallcom.2021.159159.
38. Y. Ozkan, A. Jalal and S. Salih, Optimization of MXene/CNT free-standing composite electrodes for enhanced electrochemical performance in symmetric supercapacitors, Mater. Res. Bull., 2026, 193, 113703,  DOI:10.1016/j.materresbull.2025.113703.
39. A. S. Pillai, S. M. Varghese, R. B. Rakhi and S. K. Peethambharan, Synergistic effect of solvent addition and temperature treatment on conductivity enhancement of MWCNTs: PEDOT:PSS composite ink for electrodes in all printed solid-state micro-supercapacitors, Chem. Eng. J., 2024, 495, 153495,  DOI:10.1016/j.cej.2024.153495.
40. A. Sivan Pillai, A. Chandran and S. Kuzhichalil Peethambharan, Silver Nanoparticle-Decorated Multiwalled Carbon Nanotube Ink for Advanced Wearable Devices, ACS Appl. Mater. Interfaces, 2022, 14, 46775–46788,  DOI:10.1021/acsami.2c14482.
41. A. Pullanchiyodan and K. P. Surendran, Formulation of Sol-Gel Derived Bismuth Silicate Dielectric Ink for Flexible Electronics Applications, Ind. Eng. Chem. Res., 2016, 55, 7108–7155,  DOI:10.1021/acs.iecr.6b00871.
42. ASTM D3359-09e2, Standard Test Methods for Measuring Adhesion by Tape Test, ASTM International, West Conshohocken, PA, 2009,  DOI:10.1520/D3359-09E02.
43. T. X. Liang, W. Z. Sun, L. D. Wang, Y. H. Wang and H. D. Li, Effect of surface energies on screen printing resolution, IEEE Trans. Compon., Packag., Manuf. Technol., Part B, 1996, 19, 423–425,  DOI:10.1109/96.496047.
44. B. Blunden and J. Birkenshaw, The Printing Processes, The Printing Ink Manual, 1988, pp. 10–68,  DOI:10.1007/978-94-011-7097-0_2.
45. Z. Zołek-Tryznowska, Rheology of Printing Inks, Printing on Polymers: Fundamentals and Applications, 2015, pp. 87–99,  DOI:10.1016/B978-0-323-37468-2.00006-3.
46. A. S. Bhavya, H. Varghese, A. Chandran and K. P. Surendran, Two-dimensional Hexagonal Boron Nitride Nanosheet-Polycarbonate Composite Ink-Based Printed Flexible Triboelectric Nanogenerator for Scavenging Mechanical Energy, ACS Appl. Electron. Mater., 2023, 5, 5483–5493,  DOI:10.1021/acsaelm.3c00712.
47. Z. Su, Y. Jin, H. Wang, Z. Li, L. Huang and H. Wang, PEDOT:PSS and Its Composites for Flexible Supercapacitors, ACS Appl. Energy Mater., 2022, 5, 11915–11932,  DOI:10.1021/acsaem.2c01524.
48. N. Sharma, S. Kale, N. Thomas, S. Kale, D. Ray and P. Swaminathan, Titanium Carbide MXene and Silver Nanowire Composite Ink for Transparent Flexible Printed Micro-Supercapacitors, Batteries Supercaps, 2025, 8, e202500319,  DOI:10.1002/batt.202500319.
49. Y. Zhou, J. Li, H. Fu and N. Li, Additive - Free Ti₃C₂T_x MXene/Carbon Nanotube Aqueous Inks Enable Energy Density Enriched 3D - Printed Flexible Micro - Supercapacitors for Modular Self - Powered Systems, Carbon Energy, 2025, 7, e698,  DOI:10.1002/cey2.698.
50. N. Darmiani, A. Iraji, A. Esfandiar and P. Asen, Cauli fl ower-Like Ni/MXene-Bridged Fiber-Shaped Electrode for Flexible Microsupercapacitor, ACS Energy Fuels, 2022, 36(4), 2140–2148,  DOI:10.1021/acs.energyfuels.1c03618.
51. H. Li, R. Chen, M. Ali, H. Lee and M. J. Ko, In Situ Grown MWCNTs/MXenes Nanocomposites on Carbon Cloth for High-Performance Flexible Supercapacitors, Adv. Funct. Mater., 2020, 30, 2002739,  DOI:10.1002/adfm.202002739.
52. R. Wang, S. Luo, C. Xiao, Z. Chen, H. Li, M. Asif, V. Chan, K. Liao and Y. Sun, MXene-carbon nanotubes layer-by-layer assembly based on-chip micro-supercapacitor with improved capacitive performance, Electrochim. Acta, 2021, 386, 138420,  DOI:10.1016/j.electacta.2021.138420.
53. M. Tahir, L. He, W. A. Haider, W. Yang, X. Hong, Y. Guo, X. Pan, H. Tang, Y. Li and L. Mai, Co-Electrodeposited porous PEDOT-CNT microelectrodes for integrated micro-supercapacitors with high energy density, high rate capability, and long cycling life, Nanoscale, 2019, 11, 7761–7770,  10.1039/c9nr00765b.
54. X. Zhu, C. Ren, Y. Liang, X. Liang, N. Lu, Y. Zhang, Y. Zhao and J. Gao, Laser-assisted one-step fabrication of interlayer-spacing-regulated, Chem. Eng. J., 2024, 483, 149253,  DOI:10.1016/j.cej.2024.149253.
55. Q. Wang, Y. Fang and M. Cao, Constructing MXene-PANI@MWCNTs heterojunction with high specific capacitance towards flexible micro-supercapacitor, Nanotechnology, 2022, 33, 295401,  DOI:10.1088/1361-6528/ac6432.
56. Y. Yue, N. Liu, S. Wang, W. Liu, C. Luo, H. Zhang, F. Cheng, J. Rao, X. Hu, J. Su and Y. Gao, Highly Self-Healable 3D Microsupercapacitor with MXene – Graphene Composite Aerogel, ACS Nano, 2018, 12(5), 4224–4232,  DOI:10.1021/acsnano.7b07528.

---