Graphene oxide dough composites: direct mixing and structural design strategies for high-performance electronics and energy applications

Abstract

Achieving multifunctionality in nanocomposites requires both precise and straightforward strategies for compositional integration and structural design. However, the scalable and tunable incorporation of diverse nanomaterials, while enabling structural versatility, remains a significant challenge. In this study, we present a deformable, binder-free graphene oxide dough (GOD) platform that enables the uniform incorporation of various nanomaterials through straightforward mixing strategies. The viscoelastic GOD matrix facilitates both direct powder mixing (dough-to-powder) and composite-to-composite (dough-to-dough) integration, allowing for precise structural tuning from binary to quaternary composites. The resulting 3D frameworks exhibit dense architectures with interconnected conductive pathways and stable interfaces. Supercapacitor electrodes incorporating RuO₂ nanoparticles and carbon nanotubes (CNTs) within the GOD matrix deliver outstanding volumetric capacitance (285 F cm⁻³ at 0.5 A g⁻¹) and excellent high-rate performance (206 F cm⁻³ at 20 A g⁻¹). Moreover, Fe₂O₃ nanoparticles and CNT-containing GOD composites achieve exceptional electromagnetic interference shielding effectiveness (81.3 dB), demonstrating mechanical robustness and functional integrity. This scalable and versatile composite platform offers fine control over composition and structure, establishing a promising direction for practical high-performance energy storage and electronic systems.

---

Introduction

Advanced composites have revolutionized materials engineering by synergistically combining multiple components to achieve superior properties. These materials overcome the limitations of individual components, providing enhanced mechanical strength, chemical stability, and design flexibility for applications across diverse industries. Their ability to be shaped into intricate forms further enables them to meet specific design requirements.^1–3 However, despite significant advancements, challenges such as reproducibility, scalability, and the complexity of manufacturing processes hinder their seamless integration of composites into mass-produced devices.^4,5

Recent research has focused on developing novel heterostructures by combining two-dimensional (2D) layered materials, such as graphene, hexagonal boron nitride, and transition-metal dichalcogenides, with various active materials.^6–8 These 2D heterostructures provide a distinctive platform where charged particles or molecules in proximity to certain 2D materials enable precise modulation and synergistic interactions between multiple components. In addition, 2D layered materials offer excellent flexibility, mechanical strength, a high surface-to-volume ratio, and versatile functionalization, allowing them to interact effectively with a wide range of materials. This approach has proven highly effective in tailoring and enhancing material properties for practical applications, including supercapacitors (SCs), batteries, sensors, and electromagnetic interference (EMI) shielding.^9–11

Among 2D materials, graphene oxide (GO), a chemically oxidized form of graphene, is considered an attractive building block for hybrid composites. It contains oxygen-containing functional groups such as hydroxyl (–OH), carbonyl (C=O), and carboxyl (–COOH) groups, which serve as reactive sites for chemical bonding and facilitate GO dispersion in solvents and matrices.^12–14 Several fabrication techniques, including hydrothermal synthesis,^15,16 freeze-casting,^17,18 chemical reduction,^19,20 sol–gel processing,^21,22 and vacuum evaporation,^23,24 have been employed to create GO-based composites. Nevertheless, the complexity and challenges associated with multi-component composite synthesis have significantly limited their industrial applications, mostly restricting them to simpler binary composites.²⁵ Recently, advancements in GO processing have resulted in a moldable, dough-like GO material, termed graphene oxide dough (GOD), enabling facile shaping and fabrication of compact, free-standing 3D structures.²⁶ Although previous studies have explored the cohesive nature and mechanical properties of GO dough, their primary focus has been on bulk graphene formation and electrocatalytic support applications.

In this study, we present an innovative and simplified approach that significantly broadens the functionality and potential applications of graphene composites. Unlike conventional composite fabrication methods, our approach enables the direct physical integration of various functional nanomaterials into the graphene oxide dough (GOD) matrix at room temperature. The inherent viscoelastic properties of GOD facilitate straightforward physical mixing, which enables precise control over the composite morphology and thickness, while also ensuring uniform dispersion of the incorporated nanomaterials. Utilizing these tunable and structurally customizable characteristics, GOD composites exhibit enhanced performance in practical applications such as symmetrical supercapacitors (SCs) and electromagnetic interference (EMI) shielding. Specifically, binder-free SC electrodes composed of densely packed GOD composites (density: 1.63 g cm⁻³) containing RuO₂ nanoparticles and CNTs (RuO₂@CNT@RGOD) achieved a high volumetric capacitance of 285 F cm⁻³ at 0.5 A g⁻¹, outstanding high-rate capability of 206 F cm⁻³ at 20 A g⁻¹, and excellent cyclic stability with 93% capacitance retention after 10 000 cycles at 10 A g⁻¹. Additionally, GOD composite films featuring a single-layer structure incorporating Fe₂O₃ nanoparticles and CNTs (Fe₂O₃@CNT@RGOD), as well as multilayered configuration, demonstrated exceptional EMI shielding performance, reaching a total EMI shielding effectiveness (EMI SE) of 81.3 dB.

Experimental

Materials

Graphene flake (−325 mesh, 99.8%), ruthenium(IV) oxide (RuO₂, anhydrous, 99.9%), vanadium(V) oxide (V₂O₅, 99.2%), and silicon powder (Si, crystalline, 98%) were purchased from Alfa Aesar. Sulfuric acid (H₂SO₄, 95%) and potassium permanganate (KMnO₄, 99.3%) were obtained from Junsei Chemical. Hydrochloric acid (HCl, 35.0–37.0%), hydrogen peroxide (H₂O₂, 30.0–35.5%), and ethanol (anhydrous, 99.9%) were purchased from Samchun Chemicals. Isopropyl alcohol (IPA, 99.5%) and acetone (99.8%) were purchased from Daejung chemicals. Iron(III) oxide (α-Fe₂O₃, 96%) and titanium(IV) oxide (anatase, TiO₂, 99.8%) were sourced from Sigma-Adrich. Multiwalled carbon nanotubes (MWCNTs, 95 wt%, diameter: ∼20 nm, length: ∼10 μm) were obtained from ACN Technology.

Synthesis of GO dispersion and GO foams

Graphene oxide (GO) was synthesized by oxidizing graphite flakes using a modified Hummers' method. First, graphite (5 g) and KMnO₄ (25 g) were added slowly to H₂SO₄ (187.5 ml) in an ice bath, and the resulting mixture was stirred for 3 h. The mixture was then transferred to a 30 °C oil bath and stirred for 5 h, forming a thick dark-green paste. The mixture was subsequently returned to the ice bath, and 500 ml of water was added slowly. And then, 30% H₂O₂ solution (10 ml) was added until the solution changed color from brown to yellow, after which the solution was stored overnight at room temperature. The next day, the clear supernatant was removed, and 500 ml of 36% HCl was poured onto the suspension, which was then stirred for 5 min. The solution was then vacuum-filtered, and the obtained GO cake was dispersed in water and sonicated for 2 h below 20 °C. Finally, the solution was centrifuged to remove excess acid, and the resultant GO was re-dispersed in water to form a GO aqueous dispersion. The 1 wt% GO dispersion (10 ml) was placed in conical tubes and immersed in liquid N₂ for 5 min to freeze the sample. The frozen GO dispersion was then freeze-dried under vacuum for 3 days to obtain a dry sponge-like GO foam.

Preparation of GODs and GOD composites

Graphene oxide doughs (GODs) were prepared by adding a polar solvent to the GO foams. Ethanol was pipetted into the GO foams until the GO mass fraction reached 30 wt%. The GO foams and solvent were then kneaded by hand or with a resonance acoustic mixer (RAM, Resodyn LabRAM) using zirconia balls (diameter: 5 mm) at 50G (G: force of gravity) for 5 min. While hand-mixing the components with ethanol as a solvent to yield GODs with a controlled solid content of 30 wt%, the mixer was also employed to improve reproducibility and control over variables. This technique offers a simple, cost-effective, high-throughput method that enhances particle anchoring and dispersibility within the GOD matrix, facilitating fabrication of high-viscosity GOD composites. For the dough–powder (DP) mixing method, nanopowders were added directly to the GO foams containing ethanol to achieve the desired weight ratios of particles and GO. The resultant mixtures were then mixed using the RAM. For the dough–dough (DD) mixing method, doughs already mixed with nanopowders were combined to achieve the desired weight ratio of particles and GO, and the mixture was blended using the RAM. When two or three different types of nanopowders were mixed with GODs, the resultant mixtures were referred to as ternary or quaternary mixtures, respectively.

Preparation of RGOD and RGOD composite films

Exploiting the easy deformability of the kneaded GODs and GOD composites, the doughs were roll-pressed into films using a rolling press machine. The films were rolled to the desired thickness based on their application by adjusting the gap between roll and substrate on which the GOD materials were placed. Next, the samples were dried 60 °C for 1 h to prevent expansion caused by rapid solvent evaporation during the annealing process. The films were then heat-treated at 250 °C for 1 h under an Ar gas atmosphere to prepare reduced GOD (RGOD) and RGOD composite films.

Characterization

The structure and morphology of the samples were observed by field-emission scanning electron microscopy (FE-SEM, SU8230, Hitachi High-Technologies) and X-ray diffraction (XRD, SmartLab, Rigaku). The distribution and composition of the samples were determined using energy-dispersive X-ray spectroscopy (EDS, MLA 650 FEG, FEI) and X-ray photoelectron spectroscopy (XPS, Axis-Supra, Kratos). Raman spectra of the samples were recorded using a Raman spectrometer (LabRAM HR Evolution, HORIBA). The electrical conductivity was measured using a four-probe station (MST 4000A, MSTECH). Thermal stability was analyzed by thermogravimetric analysis (TGA, TGA/DSC 1, Mettler Toledo) at a scan rate of 10 °C min⁻¹ under a N₂ atmosphere up to 1000 °C. To estimate the surface area and pore structure of the samples, the Brunauer–Emmett–Teller (BET) specific surface area and Barrett–Joyner–Halenda (BJH) pore size and volume were analyzed by N₂ adsorption measurements using a surface area and porosity analyzer (TriStar 3000, Micromeritics). Mechanical behavior was tested using a Nano Indenter (STeP500 NHT3 and MCT3, Anton Paar).

Electrochemical measurements

The electrochemical performance of various free-standing films, including RGOD, RGOD mixed with RuO₂ particles (RuO₂@RGOD), and RGOD mixed with both RuO₂ particles and CNTs (RuO₂@CNTs@RGOD), was investigated in a two-electrode system using 6 M KOH solution as the electrolyte. Symmetrical SC electrodes were prepared as follows: according to Section 2.3. Preparation of GODs and GOD Composites, the GOD, RuO₂@GOD (with a weight ratio of 1 : 5), and RuO₂@CNT@GOD (with a weight ratio of 1 : 2 : 10) were fabricated. The resulting doughs, in a deformable and soft state, were compressed using a roll-press machine and punched into round shapes (electrode slices with a thickness of ∼60 μm and a diameter of ∼1.2 cm). The compressed GOD and GOD composite films were then lightly dried at 60 °C and heat-treated at 250 °C for 1 hour under an Ar atmosphere. After reduction, the dough films shrank to a diameter of 1.1 cm and a thickness of 40 μm while maintaining their round shape. Before assembling into symmetrical supercapacitors, the electrodes were soaked overnight in 6 M KOH electrolyte. Finally, two symmetrical electrodes with the same mass and thickness were assembled using a separator (Waterman, GF/C) in an HS-flat cell (Hohsan Corp., Japan). A VSP instrument (Bio-Logic Science Instruments) was used to perform cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and electrochemical impedance spectroscopy (EIS) measurements of the samples. A potential window from 0 to 1 V was used for all the samples. CV measurements were conducted at scan rates of 10, 20, 50, and 100 mV s⁻¹, and GCD tests were carried out at different current densities of 0.5, 1, 2, 3, 4, 5, 10, and 20 A g⁻¹. Galvano-EIS measurements of the prepared samples were performed in the frequency range from 0.01 Hz to 100 kHz with a 0.1 mA current amplitude. The specific gravimetric capacitance (C_g, F g⁻¹) and volumetric capacitance (C_v, F cm⁻³) were calculated from the GCD results using the equations:

C_v = ρ × C_g

where I is the constant current (A), m is the mass of one electrode (g), Δt is the discharge time (s), ΔV is the operating potential (V), and ρ is the density of the electrode (g cm⁻³).

The coulombic efficiency (η_t) was determined as the ratio of the discharging time (t_D) to the charging time (t_c) as expressed by the following equation:

The volumetric energy density (E_v, W h L⁻¹) of the electrodes was calculated using the equation:

The volumetric power density (P_v, W L⁻¹) of the electrodes was calculated using the equation:

Electromagnetic interference (EMI) shielding effectiveness measurements

The EMI shielding performance of RGOD films with various thicknesses, as well as RGOD composite films containing Fe₂O₃ particles and CNTs (Fe₂O₃@CNT@RGOD, with a mixing ratio of 2 : 1 : 10 and a thickness of 500 μm) was measured. Additionally, three-layered sandwich structures of RGOD/Fe₂O₃@RGOD (weight ratios of 1 : 1, 1 : 5, and 1 : 10)/RGOD with a thickness of 400 μm were fabricated, and their EMI shielding performance was assessed. The EMI shielding performance of the samples was recorded using a vector network analyzer (E8364B, Agilent Technologies). The S-parameters (i.e., S₁₁ (reflection coefficient) and S₂₁ (transmission coefficient)) of the samples were recorded and used to calculate the EMI SE. The EMI SE of total (SE_T), reflection (SE_R), absorption (SE_A), and multiple reflection (SE_M) along with the power coefficients for absorptivity (A), reflectivity (R), and transmissivity (T) are expressed as follows:

R = |S₁₁|²

T = |S₂₁|²

1 = A + R + T

SE_T = SE_R + SE_A + SE_M = −10 × log T

SE_R = −10 × log(1 − R)

The SE_M is generally neglected when SE_T > 15 dB.²⁷ The EMI shielding performance of the samples was analyzed in the X-band frequency range (8.2–12.4 GHz), with the EMI SE values reported for measurements taken at 12.4 GHz.

Results and discussion

Fabrication and characterization

The fabrication processes for the GODs, GOD composites, and their reduced films are schematically depicted in Fig. 1. Highly concentrated GODs with a solid content of 30 wt% were prepared by adding ethanol to GO foams obtained through freeze-drying GO dispersions. A high GO concentration enhances the binding attraction between graphene sheets, disrupting the equilibrium with electrostatic repulsions and inducing gelation to promote self-assembly.²⁸ The abundant oxygen-containing functional groups on GO sheets, such as hydroxyl, carboxyl, and epoxy groups, facilitate strong hydrogen bonding and van der Waals interactions, which enhance cohesion and adhesion. As shown in Fig. S1,† these properties are significantly influenced by the solvent concentration. The GODs were prepared using different ethanol concentrations, and their formation, flexibility, and structural integrity were evaluated by pressing with a glass rod. Excessive amounts of polar solvent (<30 wt% GO content) cause the GO sheets to over-expand, weakening internal bonding and leading to structural instability, making it difficult to maintain the desired shape. Conversely, insufficient solvent (>30 wt% GO content) results in inadequate bonding forces, causing brittleness and cracking. Achieving the right balance of polar solvent is crucial for maintaining strong cohesion and adhesion while ensuring the desired flexibility of the GOD material. To further investigate the effects of different polar solvents on the formation and fabrication behavior (workability and structural integrity during rolling) of GOD, comparative experiments and XRD analyses were conducted (Fig. S2†). Equal amounts of GODs (30 wt%) were prepared using water, ethanol, isopropanol (IPA), and acetone, and subsequently rolled using a glass rod. The GODs prepared with water, ethanol, and IPA exhibited continuous and smooth pressing behavior with no significant differences in workability. In contrast, acetone, due to its rapid evaporation, caused premature drying and made the dough difficult to process. The evolution of the XRD patterns of the GODs was monitored in relation to the volatilization rates of the polar solvents. Initially, the characteristic (001) peak of GO appeared broad, indicating a larger interlayer spacing due to solvent absorption. Over time, as evaporation progressed under ambient conditions, these peaks sharpened, reflecting solvent removal and the corresponding reduction in interlayer spacing.²⁹ Specifically, the (001) peak of the GOD prepared with ethanol stabilized at 10.2° within 60 minutes, while the water-based GOD required a longer time. Meanwhile, the GODs processed with IPA and acetone exhibited relatively weaker and broader (001) peaks centered around 9.4°, suggesting less ordered structures. These findings indicate that ethanol provides optimum conditions for controlling both the final interlayer structure and mechanical integrity of the GOD composites. Unlike polar solvents such as water and ethanol, nonpolar solvents like hexane and toluene interact minimally with the functional groups of GO, leading to weak bonding between the sheets and producing fragile doughs that remained brittle even after being fabricated into films (Fig. S3†).

Fig. 1 Schematic illustration of the fabrication processes for (a) graphene oxide doughs (GODs) and GOD films, and (b) GOD-based composites and their films prepared via Dough–Powder mixing (DP mixing) and Dough–Dough mixing (DD mixing) strategies.

Because of the viscoelastic behavior of the dough structure, which exhibits enhanced adhesiveness and cohesiveness, dough composites in which nanopowders are physically integrated can feasibly be prepared either by hand or using a mixer, without requiring additional binders or complex chemical processes. The consolidated GO network, in which the balance between binding attractions and electrostatic repulsions is disrupted, exhibits both shear-thinning behavior and structural recoverability. During kneading, the applied shear force temporarily breaks GO interconnections, allowing solid nanoparticles to disperse effectively within the viscoelastic matrix. Once removal of shear force, the GO sheets reassemble through strong interlayer interactions, resulting in stable nanoparticle incorporation. The solvent present during mixing acts as a lubricant, promoting uniform interaction among components. As the solvent evaporates, the GO sheets self-assemble into a mechanically robust structure, securely embedding the particles within the conductive 3D composite framework.^28,30 Single or multiple powders can be directly incorporated into the GODs and kneaded through DP mixing to achieve a uniform mixture, resulting in the formation of DP hybrid structures. Furthermore, by taking advantage of the soft dough state, the integration of nanopowder into the dough is facilitated through DD mixing, allowing for easy blending of doughs pre-mixed with nanopowders. Using both DP and DD mixing methods, a diverse range of nanomaterials (e.g., metal and metal oxide nanoparticles, polymers, carbon materials, and biomaterials) can be homogeneously dispersed within the dough matrix, creating a versatile platform for developing 3D composites with the desired properties for various applications. After obtaining the dough composite, it was mechanically roll-pressed at room temperature to produce a film with a precisely controlled thickness. The GOD composite films were then reduced by heat treatment at 250 °C for 1 h to yield electrically conductive composites with robust and densely packed structures.

In their soft state, GODs not only withstand substantial deformation but also allow the separated pieces to be easily joined by compression, similar to kneading bread dough or working with clay. As shown in Fig. 2(a–c), the GODs can be easily shaped into specific forms such as a duck, heart, moon, clover and flexible film by cutting and kneading them by hand or using baking molds. Fig. 2(d) displays GOD composites prepared through DP mixing with Fe₂O₃, V₂O₅, CNT, TiO₂, RuO₂, and Si powders. The uniform color of the resulting composites confirms the through mixing of each powder. Additionally, the GOD composites can be produced and stored on a large scale, addressing the challenges of stacking and transporting GO in its dispersed form (Fig. S4†). The DP mixing process also allows for the simultaneous mixing of two different powders (Fig. 2(e)), creating a ternary composite such as Fe₂O₃@TiO₂@GOD, where Fe₂O₃ and TiO₂ powders are homogeneously integrated with GOD. Furthermore, DD mixing of various pre-mixed doughs containing different particles can be used to create a homogeneous ternary (Si@CNT@GOD) and quaternary mixtures (Fe₂O₃@V₂O₅@TiO₂@GOD) (Fig. 2(f)). The dough structure can also be designed into a 3D orthogonal structure, such as a sandwich, without requiring complex processes. As shown in Fig. 2(g), TiO₂@GOD, TiO₂@V₂O₅@GOD, and TiO₂@V₂O₅@Fe₂O₃@GOD were fabricated and pressed into a flat form, resulting in a three-layer structure with different particles arranged according to each layer.

Fig. 2 Photographs of (a–c) graphene oxide doughs (GODs) shaped into various geometries, (d and e) GOD-based composites fabricated via DP mixing, (f) GOD-based composites prepared via DD mixing, and (g) a multi-layered GOD-based composite film with distinct composite layers.

Fig. 3(a–d) shows surface images of RGOD films with precisely controlled thicknesses of 50, 100, 200, and 400 μm. The films have smooth surfaces without noticeable distortion, regardless of thickness. In the corresponding cross-sectional SEM images (Fig. 3(e–h)), the RGOD films display finely packed layer structures, leading to uniform and continuous conductive networks at all investigated thicknesses. The density and electrical conductivity of the RGOD films reached 2.1 g cm⁻³ and 50 S cm⁻¹, respectively, at a thickness of 10 μm (Fig. 3(i)). Notably, both the density and electrical conductivity increased as the film thickness decreased. This behavior has been observed in previous studies, primarily due to the exclusion of insulated regions as the films were compressed to smaller thicknesses.^31–35 Despite the decrease in density and electrical conductivity with increasing film thickness, the RGOD films still exhibited relatively high density and electrical conductivity compared to previously reported graphene-based films thicker than 100 μm.^36–40

Fig. 3 (a–d) Photos of the surface and measured thickness, (e–h) cross-sectional SEM images (scale bar: 20 μm), (i) density and electrical conductivity of RGOD films, (j) XRD spectra and (k) Raman spectra of GOD and RGOD films with thicknesses including 50, 100, 200 and 400 μm.

The XRD patterns of the GOD and RGOD films with different thicknesses are presented in Fig. 3(j). The GOD film (50 μm) exhibited a diffraction peak at 2θ = 9.7° (001) with a d-spacing of 0.91 nm. After reduction, the RGOD film (50 μm) displayed a broad diffraction peak at 2θ = 24.9° (d = 0.36 nm), corresponding to the (002) diffraction peak of graphite (JCPDS No. 65-6212), indicating a decrease in the interlayer spacing due to the removal of oxygen functional groups. As the thickness of the RGOD films increased, the peak position shifted slightly to smaller angles. Specifically, at thicknesses of 100, 200, and 400 μm, the peak appeared at 2θ values of 24.5° (d = 0.36 nm), 24.4° (d = 0.36 nm), and 24.1° (d = 0.37 nm), respectively. This shift suggests a subtle increase in interlayer spacing with increasing film thickness. This increase can be attributed to the mechanical roll-pressing strategy used during the fabrication of free-standing GOD films. In this process, GODs of equal weight are placed between the roll and substrate, with precise gap control to achieve the desired film thickness. As the gap increases, the pressure applied to the films decreases, resulting in less compression and a looser packing of graphene layers. This reduced pressure allows for slightly greater spacing between the graphene sheets, which explains the observed increase in interlayer spacing with increased film thickness. In contrast, thinner films, which are subjected to higher pressure, experience more compact packing, reducing interlayer spacing.⁴¹ The Raman spectra of the GOD and RGOD films of various thicknesses are shown in Fig. 3(k). Two dominant peaks corresponding to the D and G bands were observed at 1360 and 1607 cm⁻¹, respectively. The D band represents the vibration of sp³ carbon atoms in disordered GO nanosheets, while the G band represents the vibrations of sp² carbon atoms in the planar graphene structure. The ratio between the intensities of the D and G bands (I_D/I_G) is a key parameter for characterizing the degree of disorder or defects in carbon materials. As the GOD film (50 μm) was reduced to the RGOD film (50 μm), the I_D/I_G ratio increased from 0.91 to 1.10, indicating the removal of oxygen functional groups. The thickness of the RGOD films did not significantly affect the I_D/I_G ratio. Additionally, all RGOD, regardless of thickness, exhibited characteristic peaks representing the second-order overtone (2D band) and structural defects (D + G band) of graphene materials at around 2704 and 2942 cm⁻¹, respectively. The prominent 2D band in the densest 50 μm RGOD film indicates a compact stacking structure, demonstrating that the high-density RGOD films retain densification and improved electrical conductivity.⁴² Even after reduction, which removes oxygen-containing groups, the RGOD films with thicknesses below 100 μm exhibited bendable properties, with minimal changes in resistance across the bending radii (Fig. S5†). This retained flexibility makes the RGOD composite films suitable for practical applications, as they can be fabricated and handled without breaking.

Fig. 4(a) shows the XRD patterns of six RGOD composite films, each prepared by DP mixing of the GODs with six different nanomaterials (CNTs and V₂O₅, TiO₂, Fe₂O₃, RuO₂, and Si nanoparticles). The patterns of all samples display distinct peaks corresponding to both RGOD and the added particles, indicating that the particles remained well-incorporated in the doughs even after the reduction process, making them suitable for practical applications. The XPS analysis of the dough composite mixed with RuO₂ particles (Fig. 4(b) and S6†), which is investigated as an SC electrode material in this study reveals the presence of C, O, N, and Ru. The C 1s peak at 281.0 eV overlaps partially with the Ru 3d peak (280.4 eV), including contributions from C–C, C=C, and C–OH bonds associated with RGOD. The Ru 3d peak, showing binding energy corresponding to Ru⁴⁺, confirms the presence of RuO₂ in the composite. The Ru 3p spectrum presents two peaks at 462.3 and 484.4 eV, assigned to Ru 3p_3/2 and Ru 3p_1/2, respectively, further confirming the presence of RuO₂. The O 1s peak, located at 528.7 eV, is resolved into three components corresponding to Ru–O–Ru, Ru–OH, and C–OH at 529.3, 530.8, and 532.7 eV, respectively.^43,44 These results indicate that chemical bonding between the RuO₂ particles and RGOD has been fully established. The surface of the RGOD composites was further analyzed using SEM and EDS to examine the particle distribution within the dough. The RuO₂ particles were observed to be evenly dispersed across the RGOD surface (Fig. 4(c)). This uniform distribution was also evident in the composites prepared by DP mixing with several particles, as shown in the SEM cross-sectional images and EDS mapping data in Fig. S7 and S8.†Fig. 4(d) compares XRD results for ternary and quaternary composites (V₂O₅@TiO₂@RGOD and V₂O₅@TiO₂@Si@RGOD, respectively) prepared by DP mixing and DD mixing. The XRD peaks of the ternary mixtures produced by DP mixing, where V₂O₅ and TiO₂ powders were added to the dough, and DD mixing, where V₂O₅@GOD and TiO₂@GOD were blended, were nearly identical. This approach also yielded consistent results for quaternary mixtures containing V₂O₅, TiO₂, and Si particles. The EDS results in Fig. 4(e and f) confirm the uniform distribution of the three types of particles in quaternary composites prepared using both mixing methods, providing strong evidence that graphene dough is a promising composite platform for designing intricate functional electrodes. For uniform mixing, an acoustic mixer and zirconia balls were employed, and the mixing efficiency of the ternary composites (TiO₂@V₂O₅@GOD) was compared based on the mixing time using each method (Fig. S9†). The consistent color of the GOD composites highlights the effectiveness of our direct mixing process in achieving uniform particle blending within a short processing time of 3–5 minutes. To further demonstrate the practicality of these composite materials, their thermal and mechanical properties were analyzed. The TGA profiles (Fig. S10†) show that RGOD exhibits greater thermal stability compared to GOD, attributed to the reduction of oxygen-containing groups. The composites exhibited particle-dependent stability, with enhanced stability arising from complex interactions and synergistic effects. Fe₂O₃@CNT@RGOD demonstrated the highest stability, retaining 84.5 wt% at 800 °C and remaining stable up to 1000 °C, which is crucial for EMI shielding applications.³¹ The indentation tests (Table S1†) reveal increased elastic modulus (GPa), hardness (MPa), and Vickers hardness (HV) values in GOD-nanoparticle composites compared to pure GOD. These improvements are attributed to several reinforcing mechanisms. Nanoparticles provide structural support by filling matrix weaknesses and improving load distribution. Strong interfacial bonding enhances stress transfer, reducing deformation and increasing hardness. Nanoparticles also constrain matrix deformation and inhibit crack propagation, significantly boosting durability and hardness.⁴⁵ After reduction, shorter and stronger C–C bonds within GOD matrix further improve mechanical stability and durability without compromising structural integrity.⁴⁶ Additionally, cyclic nanoindentation tests were performed on RGOD composite films (RuO₂@CNT@RGOD and Fe₂O₃@CNT@RGOD) to evaluate their cyclic mechanical response during the initial 10 loading cycles (Fig. S11†). Under a constant peak load of 3 mN, both films showed consistently low maximum penetration depths with minimal deviation, indicating good mechanical robustness under cyclic loading.

Fig. 4 (a) XRD spectra of RGOD composites with various particles. (b) The wide scan of XPS spectrum, (c) SEM surface and its EDS mapping image of RuO₂@RGOD. (d) XRD spectra of ternary and quaternary RGOD composites, corresponding to V₂O₅@TiO₂@RGOD and V₂O₅@TiO₂@Si@RGOD. (e and f) SEM and EDS mapping images of quaternary RGOD composites (V₂O₅@TiO₂@Si@RGOD) fabricated by DP and DD mixing (scale bar: 2 μm).

Electrochemical performance

To explore the practical applications of the dough-based composites, their electrochemical performance was evaluated using a two-electrode symmetrical cell. Leveraging the ease of processing these dough structures, we synthesized RuO₂, CNT, and RGOD composites (RuO₂@CNT@RGOD) for SC electrodes (Fig. 5(a)). Extensive research has been conducted on integrating graphene with functional materials to develop high-capacitance SC electrodes. One effective approach to enhancing electrochemical performance is incorporating transition metal oxides, which combine conductive carbon materials with redox-active nanoparticles. Among these, RuO₂ stands out as a promising pseudocapacitive material due to its high specific capacitance, superior charge transport properties, and excellent chemical stability. The introduction of CNTs into the electrode not only serves as conductive bridges between RGO layers but also prevents the aggregation of RuO₂ nanoparticles and graphene sheets. This strategy promotes the formation of expanded interfacial structures and efficient ion diffusion pathways, synergistically enhancing the capacitance of RGOD through a combination of faradaic and non-faradaic charge storage processes.^43,44 Moreover, the binder-free design reduced unnecessary ion transport distances and increases the accessible interfacial area for the electrolyte. The porous structure of the RGOD, RuO₂@RGOD, and RuO₂@CNT@RGOD were analyzed using N₂ physisorption measurements, with corresponding adsorption isotherms presented in Fig. S12(a).† The calculated specific surface area (SSA) and pore volume of RuO₂@RGOD were 55 m² g⁻¹ and 0.29 cm² g⁻¹, respectively, slightly higher than those of RGOD (46 m² g⁻¹ and 0.23 cm² g⁻¹). Notably, the RuO₂@CNT@RGOD electrode exhibited the highest SSA and pore volume (107.3 m² g⁻¹ and 0.53 cm² g⁻¹) among the three materials. The pore size distributions (Fig. S12(b)†) further confirm that RuO₂@CNT@RGOD contains the largest proportion of mesopores among the samples, facilitating electrolyte penetration and maintain performance even at high current densities.

Fig. 5 (a) Schematic illustration of the structure of RuO₂@CNT@RGOD. (b) Comparison of CV curves at a scan rate of 20 mV s⁻¹, (c) GCD curves at a current density of 5 A g⁻¹, (d) gravimetric capacitance, (e) volumetric capacitance at various current densities, and (f) Nyquist plots of RGOD, RuO₂@RGOD and RuO₂@CNT@RGOD. (g) CV curves at different scan rates of 10, 20, 50 and 100 mV s⁻¹, (h) cycling stability and coulombic efficiency at 10 A g⁻¹ of RuO₂@CNT@RGOD (inset: GCD curves of the last 10 cycles). (i) Ragone plot of the RuO₂@CNT@RGOD compared with previously reported papers.

To assess the effectiveness of RuO₂ and CNT incorporation, the supercapacitance performance of RGOD, RuO₂@RGOD, and RuO₂@CNT@RGOD electrodes was evaluated using CV, GCD, and EIS measurements. Fig. 5(b) presents the cyclic voltammograms of the different electrodes within a potential range from 0 to 1 V at a scan rate of 20 mV s⁻¹. Among the three, the RuO₂@CNT@RGOD electrode exhibited the largest CV curve area, indicating the highest specific capacitance. Given the slight increase in the specific surface area of RuO₂@RGOD in compared RGOD, the substantially enhanced capacitance of the RuO₂@CNT@RGOD electrode can be attributed to the effective role of RuO₂ particles in boosting charge storage. As shown in Fig. 5(c), the GCD profiles at a current density of 5 A g⁻¹ display nearly symmetrical isosceles triangles for all electrodes. The coulombic efficiencies of the RuO₂@RGOD and RuO₂@CNT@RGOD electrodes were 99.3% and 99.7%, respectively, approaching 100%. These results demonstrate excellent electrochemical reversibility, facilitated by enhanced ion accessibility and efficient charge transport, which result from the homogeneous integration of RuO₂ nanoparticles and CNTs into the dense RGOD matrix.

The RuO₂@CNT@RGOD electrode exhibited the longest discharge time among the three, consistent with the CV results, further confirming its superior specific capacitance. The gravimetric and volumetric capacitance of the electrodes at different current densities are compared in Fig. 5(d) and (e). The RuO₂@RGOD and RuO₂@CNT@RGOD electrodes exhibited slightly lower gravimetric capacitances of 178 and 175 F g⁻¹ at 0.5 A g⁻¹, respectively, compared with RGOD (206 F g⁻¹). However, as the current density increased, the capacitance of the RGOD electrode dropped sharply due to its dense and bulk-like matrix, which restricted ion diffusion at high current densities. In contrast, the RuO₂@RGOD and RuO₂@CNT@RGOD electrodes maintained capacitance retention rates of 55% (97 F g⁻¹) and 72% (126 F g⁻¹), respectively, even at 20 A g⁻¹. The dense dough structure contributed to the high volumetric capacitance values, with the RuO₂@CNT@RGOD, RuO₂@RGOD, and RGOD electrodes exhibiting 285, 304, and 360 F cm⁻³, respectively, at 0.5 A g⁻¹. The slightly lower volumetric capacitance of RuO₂@CNT@RGOD compared with RuO₂@RGOD at low current densities is attributed to the incorporation of CNTs, which led to the formation of some voids, reducing electrochemical activity. Nevertheless, the RuO₂@CNT@RGOD electrode exhibited the highest capacitance retention (72%, 206 F cm⁻³), highlighting the advantages of its 3D composite structure in enabling with efficient charge transport through a short diffusion pathway and high electrical conductivity. To further investigate ion transport kinetics, EIS measurements were carried out. As shown in Fig. 5(f), the Nyquist plots reveal semicircles in the high-frequency region, representing charge transfer resistance (R_CT), and nearly vertical lines in the low-frequency region, indicative of capacitive behavior. The RuO₂@CNT@RGOD electrode exhibited the smallest diameter semicircle and the steepest straight line demonstrating lower internal resistance and enhanced capacitive performance compared with the RuO₂@RGOD and RGOD. The RuO₂@RGOD electrode also showed a smaller semicircle than RGOD, suggesting that RuO₂ particles effectively reduce graphene stacking within the dough structure, leading to a lower R_CT.

To electrochemical feasibility of the RuO₂@CNT@RGOD electrode was further examined. Fig. 5(g) shows CV curves recorded at various scan rates, where the rectangular shape was well maintained even at 100 mV s⁻¹, indicating excellent capacitive behavior and rate performance. The GCD curves at different current densities ranging from 0.5 to 20 A g⁻¹ retained their symmetrical shapes, demonstrating stable charge–discharge behavior across a wide current range (Fig. S13†). The superior electrochemical kinetics of this dough composite structure establish the RuO₂@CNT@RGOD as a high-performance SC electrode capable of sustaining high discharge rates and prolonged cycling. As shown in Fig. 5(h), long-term cycling stability was evaluated by galvanostatic charge–discharge (GCD) measurements at 10 A g⁻¹. After 10 000 cycles, the RuO₂@CNT@RGOD electrode retained 93% of its initial capacitance with a coulombic efficiency of 96%, indicating excellent electrochemical stability and minimal structural degradation. Additionally, the RuO₂@CNT@RGOD electrode exhibited a volumetric energy density of 9.9 W h L⁻¹ at a power density of 204 W L⁻¹, and retained 7.13 W h L⁻¹ at 8119 W L⁻¹. As depicted in the Ragone plot (Fig. 5(i)), this performance surpasses that of state-of-the-art carbon-based SCs with high packing densities.^47–57 Further details, including volumetric capacitance, volumetric energy and power densities, and cycling stability for each reference, are summarized in Table S2† for comparative analysis. The RuO₂@CNT@RGOD electrode achieves energy and power densities within the upper range of reported high-density carbon-based materials, including composite systems. These improvements arise from the unique RGOD-based composite framework, which maintains high packing density while allowing structural flexibility for incorporating nanomaterials tailored to specific application needs. This architecture improves pore accessibility and ion mobility, while harnessing the high electrochemical activity of RuO₂ and the excellent electrical conductivity and mechanical support provided by CNTs.

EMI shielding performance

The exceptional formability, high electrical conductivity, and abundant interfaces of the graphene dough structures critically influence their EMI shielding performance, leading to effective fabrication methods for efficient EMI shielding materials. Fig. 6(a) illustrates the EM dissipation in RGOD film, where its dense structure and continuous conductive networks with good interlayer contact induced consecutive internal reflection and absorption. Given that the thickness of RGOD films can be readily adjusted, the influence of thickness on the EMI SE was investigated within the X-band frequency range (8.2–12.4 GHz). As shown in Fig. 6(b), the total EMI SE (SE_T) of 25.2 dB was achieved at a thickness of 50 μm, satisfying the commercial requirement for EMI shielding (>20 dB). With increasing film thickness, the EMI SE exhibited a positive trend, reaching a maximum value of 71.1 dB at a thickness of 500 μm. To further examine the contributions to SE_T, the SE attributed to reflection (SE_R) and absorption (SE_A) were analyzed as a function of film thickness (Fig. 6(c)). Notably, the SE_R value remained nearly unchanged as the film thickness increased, while the SE_A value gradually rose, attributable to the formation of more interfacial structures that facilitate multiple absorption of radiation between the interlayers.

Fig. 6 (a) Schematic illustration of EM wave dissipation in RGOD structure. (b) SE_T in the frequency range of 8.2–12.4 GHz (X-band), and (c) SE_T, SE_A and SE_R at 12.4 GHz for RGODs according to thickness. (d) Comparison of electrical conductivity, (e) SE_T, SE_A, and SE_R, and (f) proportion of reflection (R), absorption (A), and transmission (T) at 12.4 GHz for RGOD, Fe₂O₃@RGOD, CNT@RGOD and Fe₂O₃@CNT@RGOD with thickness of 500 μm. (g) Schematic illustration of EM wave dissipation in multilayer composite structure. (h) SE_T in X-band and (i) SE_T, SE_A, SE_R, R, A and T coefficients at 12.4 GHz for RGOD/Fe₂O₃@RGOD/RGOD multilayer composites, according to mixing ratio of Fe₂O₃ : RGOD in Fe₂O₃@RGOD, which is the middle layer.

The EMI SE can be further improved by incorporating particles beneficial for EMI shielding into the doughs. Among such particles, CNTs contribute to conduction loss due to their excellent electrical conductivity, whereas α-Fe₂O₃ functions as a magnetic loss material, effectively absorbing the magnetic components of EM waves. By uniformly blending these components within the RGOD matrix, an integrated nanomaterial is created, exhibiting a synergistic effect that enhances EMI SE.^58,59 Differences in electrical conductivity, which critically impact EMI shielding performance, were observed depending on the mixing of Fe₂O₃ and CNT particles in 500 μm-thick films. As depicted in Fig. 6(d), the addition of CNTs markedly enhanced the conductivity of the composites, while the inclusion of Fe₂O₃ caused a slight reduction. Nevertheless, the Fe₂O₃@CNT@RGOD composite, containing both Fe₂O₃ and CNTs, exhibited higher conductivity than bare RGOD, a trend also reflected in EMI SE values (Fig. 6(e)). The incorporation of CNTs contributed to improvements in both SE_R and SE_A, thereby enhancing overall EMI shielding performance. While the addition of Fe₂O₃ alone to RGOD slightly decreased SE_T due to reduced electrical conductivity, the combined addition of both particles (Fe₂O₃@CNT@RGOD) resulted in a noticeable improvement in SE_T. To further evaluate the EMI shielding mechanism based on composition, the proportion of reflection (R), transmission (T), and adsorption (A) in Fig. 6(f) were calculated from the measured scattering parameters. The T values for all samples were nearly zero, indicating that most incident EM waves were effectively blocked. Although Fe₂O₃@RGOD exhibited lower SE_T value compared to other samples, it displayed the highest A value, demonstrating its contribution to enhancing EM wave absorption. The addition of CNTs primarily increased R value, driven by their high electrical conductivity. Remarkably, the simultaneous addition of Fe₂O₃ and CNT improved both conductivity and the A value compared with RGOD, achieving the highest SE_T value of 81.3 dB. High conductivity increases conduction loss, converting incident EM waves into heat and dissipating them, thereby enhancing EMI shielding performance. However, it can also cause impedance mismatch at the material surface, hindering overall absorption, making it essential to balance the combination of CNTs and Fe₂O₃.⁶⁰ The co-addition of CNTs and Fe₂O₃ proved more effective than using either component alone, as their seamless integration within the RGOD matrix generated a pronounced synergistic effect that optimized both conduction and magnetic loss mechanisms for superior EMI shielding. Without polymer additives or supports, an excellent EMI SE value of 81.3 dB was achieved relative to thickness, surpassing previously reported graphene-based composites and commercial EMI shielding materials (Fig. S14†). The direct physical mixing approach employed for hybrid fabricating Fe₂O₃@CNT@RGOD offers clear advantages in simplicity, room-temperature processing, and rapid fabrication, compared to other particle incorporation techniques for graphene composites without polymer fillers or matrices (Table S3†). Compared to conventional EMI shielding materials such as metal foils and carbon-based materials, composite RGODs demonstrate favorable shielding performance relative to thickness, while also enabling a simplified fabrication process under mild conditions with reduced processing time. These combined advantages highlight the strong potential of composite RGODs for practical applications by offering not only effective EMI shielding but also economic benefits and enhanced manufacturing feasibility.

Beyond single-layered composites, designing novel architectures for effective EMI shielding is a critical aspect.⁶¹ The formable GODs can be readily adapted for various structural designs advantageous for EMI shielding. As reported in previous studies, a structure featuring two outer layers with high electrical conductivity and a magnetic loss layer between them effectively improves EMI shielding properties.^62,63 As illustrated in Fig. 6(g), a sandwich structure was fabricated by inserting a magnetic Fe₂O₃@RGOD film between two conductive RGOD films, inducing multiple reflections and absorption losses between the layers. The multilayer composite featuring Fe₂O₃ in the middle layer facilitates multiple reabsorptions as EM waves pass through the sandwich layers. Cross-sectional SEM and EDS mapping images of the multilayered composite (RGOD/Fe₂O₃@RGOD/RGOD) are presented in Fig. S15.† In the EDS mapping, the blue middle layer, representing iron (Fe), is clearly distinguishable from the two red RGOD layers, which indicate carbon (C). This confirms that the accurate positioning of the desired layers, successfully forming the sandwich-type multilayer structure. Each layer measured approximately 135 μm (top layer), 132 μm (middle layer), and 137 μm (bottom layer), aligning uniformly to achieve a targeted total thickness of 400 μm, demonstrating stable assembly without collapse. Consequently, the sample with a Fe₂O₃ : RGOD ratio of 1 : 5 was identified as the optimal material, achieving an SE_T value of 70.8 dB, which exceeded that of the bare RGOD film (61.0 dB) (Fig. 6(h)). By comparing the reflection and absorption rates for samples with different mixing ratios, it was verified that the A value increased as more Fe₂O₃ was incorporated into the multilayer structure. However, excessive loading of Fe₂O₃ particles reduced conductivity and diminished the material's ability to effectively reflect and absorb EM waves, negatively affecting overall SE (Fig. 6(i)), with the Fe₂O₃ : RGOD ratio of 1 : 1 resulted in a decrease in SE_T to 55.7 dB. On the other hand, samples with Fe₂O₃ : RGOD ratios of 1 : 10 and 1 : 5 exhibited improved shielding performance, reaching 66.1 and 70.8 dB, respectively. This enhancement can be attributed to the improved EMI shielding performance that complements the insufficient magnetic properties of the graphene sheet without compromising conductivity.

Conclusions

This study introduces a technique for fabricating a versatile and scalable 3D graphene composite framework. A compact graphene composite platform with functionalities was developed by physically mixing highly concentrated GO dough with nanoparticles, without the use of binders or complex processing steps. The intrinsic characteristics of the dough, including its remarkable cohesion and adhesion, enable the formation of a dense structure capable of various shape transformations. These properties also facilitate the uniform incorporation of nanoparticles into the GOD matrix. The continuous conductive networks within the dense structure allow the platform to function as both electrodes for SCs and EMI shielding materials, demonstrating impressive volumetric capacitance of 285 F cm⁻³ at 0.5 A g⁻¹, a high-rate performance of 206 F cm⁻³ at 20 A g⁻¹, and an outstanding SE_T of 81.3 dB with an absorption-dominated mechanism. This work presents an efficient and innovative strategy for developing advanced composite structures for future electronic applications, achieved through the integration of functional components and the design of controllable interfacial structures.

Data availability

All data supporting this study are available in the ESI.†

Author contributions

Seoyeon Park: investigation, formal analysis, validation, writing – original draft. Jeonghyun Yoo, Kyeong Ja Kim, Soong Ju Oh: methodology and writing – review & editing. Byung-Su Kim & Ji-Hyuk Choi: conceptualization, investigation, data analysis, project administration, writing – review & editing, supervision.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The Basic Research Projects (Grant Number: 25-3815 and 25-3225) of the Korea Institute of Geoscience and Mineral Resources funded by the Ministry of Science and ICT of Korea. This work was also supported by the National Research Foundation of Korea [grant number 2022R1A2C100829211(25-6806)].

References

1. T. Ramanathan, A. A. Abdala, S. Stankovich, D. A. Dikin, M. Herrera-Alonso, R. D. Piner, D. H. Adamson, H. C. Schniepp, X. Chen, R. S. Ruoff, S. T. Nguyen, I. A. Aksay, R. K. Prud'Homme and L. C. Brinson, Nat. Nanotechnol., 2008, 3, 327–331.
2. H. Gu, C. Liu, J. Zhu, J. Gu, E. K. Wujcik, L. Shao, N. Wang, H. Wei, R. Scaffaro, J. Zhang and Z. Guo, Adv. Compos. Hybrid Mater., 2018, 1, 1–5.
3. Z. Wang, H. Gao, Q. Zhang, Y. Liu, J. Chen and Z. Guo, Small, 2019, 15, 1803858.
4. L. Lin, H. Ning, S. Song, C. Xu and N. Hu, Compos. Sci. Technol., 2020, 192, 108102.
5. M. R. Islam, S. Afroj and N. Karim, ACS Nano, 2023, 17, 18481–18493.
6. P. Solís-Fernández, M. Bissett and H. Ago, Chem. Soc. Rev., 2017, 46, 4572.
7. S. Roy, X. Zhang, A. B. Puthirath, A. Meiyazhagan, S. Bhattacharyya, M. M. Rahman, G. Babu, S. Susarla, S. K. Saju, M. K. Tran, L. M. Sassi, M. A. S. R. Saadi, J. Lai, O. Sahin, S. M. Sajadi, B. Dharmarajan, D. Salpekar, N. Chakingal, A. Baburaj, X. Shuai, A. Adumbumkulath, K. A. Miller, J. M. Gayle, A. Ajnsztajn, T. Prasankumar, V. V. J. Harikrishnan, V. Ojha, H. Kannan, A. Z. Khater, Z. Zhu, S. A. Iyengar, P. A. d. S. Autreto, E. F. Oliveira, G. Gao, A. G. Birdwell, M. R. Neupane, T. G. Ivanov, J. Taha-Tijerina, R. M. Yadav, S. Arepalli, R. Vajtai and P. M. Ajayan, Adv. Mater., 2021, 33, 2101589.
8. Q. Lv and R. Lv, Carbon, 2019, 145, 240–250.
9. L. Liang, C. Yao, X. Yan, Y. Feng, X. Hao, B. Zhou, Y. Wang, J. Ma, C. Liu and C. Shen, J. Mater. Chem. A, 2021, 9, 24560–24570.
10. D. X. Yan, H. Pang, B. Li, R. Vajtai, L. Xu, P. G. Ren, J. H. Wang and Z. M. Li, Adv. Funct. Mater., 2015, 25, 559–566.
11. N. Choudhary, M. A. Islam, J. H. Kim, T. J. Ko, A. Schropp, L. Hurtado, D. Weitzman, L. Zhai and Y. Jung, Nano Today, 2018, 19, 16–40.
12. W. Lv, C. Zhang, Z. Li and Q. H. Yang, J. Phys. Chem. Lett., 2015, 6, 658–668.
13. J. H. Choi, C. Lee, S. Cho, G. D. Moon, B. su kim, H. Chang and H. D. Jang, Carbon, 2018, 132, 16–24.
14. J. H. Choi, Y. Kim and B. su Kim, J. Power Sources, 2020, 454, 227917.
15. K. Hu, X. Xie, M. Cerruti and T. Szkopek, Langmuir, 2015, 31, 5545–5549.
16. Y. Xu, K. Sheng, C. Li and G. Shi, ACS Nano, 2010, 4, 4324–4330.
17. Y. Shao, M. F. El-Kady, C. W. Lin, G. Zhu, K. L. Marsh, J. Y. Hwang, Q. Zhang, Y. Li, H. Wang and R. B. Kaner, Adv. Mater., 2016, 28, 6719–6726.
18. C. Wang, X. Chen, B. Wang, M. Huang, B. Wang, Y. Jiang and R. S. Ruoff, ACS Nano, 2018, 12, 5816–5825.
19. W. Chen and L. Yan, Nanoscale, 2011, 3, 3132–3137.
20. C. Hu, X. Zhai, L. Liu, Y. Zhao, L. Jiang and L. Qu, Sci. Rep., 2013, 3, 2065.
21. M. B. Lim, M. Hu, S. Manandhar, A. Sakshaug, A. Strong, L. Riley and P. J. Pauzauskie, Carbon, 2015, 95, 616–624.
22. M. A. Worsley, S. O. Kucheyev, H. E. Mason, M. D. Merrill, B. P. Mayer, J. Lewicki, C. A. Valdez, M. E. Suss, M. Stadermann, P. J. Pauzauskie, J. H. Satcher, J. Biener and T. F. Baumann, Chem. Commun., 2012, 48, 8428–8430.
23. F. Liu and T. K. Seo, Adv. Funct. Mater., 2010, 20, 1930–1936.
24. F. Liu, S. Chung, G. Oh and T. S. Seo, ACS Appl. Mater. Interfaces, 2012, 4, 922–927.
25. M. Zhong, M. Zhang and X. Li, Carbon Energy, 2022, 4, 950–985.
26. C. N. Yeh, H. Huang, A. T. O. Lim, R. H. Jhang, C. H. Chen and J. Huang, Nat. Commun., 2019, 10, 422.
27. S. Park, J. Bang, B. S. Kim, S. J. Oh and J. H. Choi, Mater. Today Adv., 2021, 12, 6.
28. Z. Sun, S. Fang and Y. H. Hu, Chem. Rev., 2020, 120, 10336–10453.
29. A. Klechikov, J. Yu, D. Thomas, T. Sharifi and A. V. Talyzin, Nanoscale, 2015, 7, 15374–15384.
30. X. Yun, B. Lu, Z. Xiong, B. Jia, B. Tang, H. Mao, T. Zhang and X. Wang, RSC Adv., 2019, 9, 29384–29395.
31. J. Yu, Z. Yuan, L. Wang, S. Chen, W. Wei and F. Zhang, Mater. Chem. Front., 2022, 6, 2201–2210.
32. J. Xi, Y. Li, E. Zhou, Y. Liu, W. Gao, Y. Guo, J. Ying, Z. Chen, G. Chen and C. Gao, Carbon, 2018, 135, 44–51.
33. X. Zhang, Y. Guo, Y. Liu, Z. Li, W. Fang, L. Peng, J. Zhou, Z. Xu and C. Gao, Carbon, 2020, 167, 249–255.
34. B. Hao, Z. Tao, Z. Liu, D. Zhang, Z. He, Q. Guo, L. Liu and D. Jiang, J. Mater. Sci.: Mater. Electron., 2019, 30, 6734–6744.
35. V. B. Mohan, K. Jayaraman, M. Stamm and D. Bhattacharyya, Thin Solid Films, 2016, 616, 172–182.
36. Z. Liu, J. Zhu, D. Lu, W. Wang, L. Yue and Y. Li, J. Power Sources, 2022, 523, 231046.
37. L. Zhang, D. Han, Y. Tao, C. Cui, Y. Deng, X. Dong, W. Lv, Z. Lin, S. Wu, Z. Weng and Q. H. Yang, J. Mater. Chem. A, 2020, 8, 461–469.
38. D. Wang, L. Sheng, M. Jiang, X. Jin, X. Lin, S. Y. Lee, J. Shi and W. Chen, Battery Energy, 2022, 1, 20220017.
39. S. Han, N. Li, Y. Song, L. Chen, C. Liu, M. Xi, X. Yu, Y. Qin, T. Xu, C. Ma, S. Zhang and Z. Wang, Carbon, 2021, 182, 393–403.
40. H. Banda, D. Aradilla, A. Benayad, Y. Chenavier, B. Daffos, L. Dubois and F. Duclairoir, J. Power Sources, 2017, 360, 538–547.
41. P. Huang, Y. Li, G. Yang, Z. X. Li, Y. Q. Li, N. Hu, S. Y. Fu and K. S. Novoselov, Nano Mater. Sci., 2021, 3, 1–16.
42. S. Zhu, F. Zhang, H. G. Lu, J. Sheng, L. Wang, S. D. Li, G. Han and Y. Li, ACS Mater. Lett., 2022, 4, 1863–1871.
43. S. Kong, K. Cheng, T. Ouyang, Y. Gao, K. Ye, G. Wang and D. Cao, Electrochim. Acta, 2017, 246, 433–442.
44. W. Wang, S. Guo, I. Lee, K. Ahmed, J. Zhong, Z. Favors, F. Zaera, M. Ozkan and C. S. Ozkan, Sci. Rep., 2014, 4, 4452.
45. K. Chen, X. Tang, B. Jia, C. Chao, Y. Wei, J. Hou, L. Dong, X. Deng, T. Xiao, K. Goda and L. Guo, Nat. Mater., 2022, 21, 1121–1130.
46. A. Kamal, B. Li, A. Solayman, S. Luo, I. Kinloch, L. Zheng and K. Liao, Nanoscale Horiz., 2025, 10, 512–536.
47. Y. Zhang, K. Liu, X. Liu, W. Ma, S. Li, Q. Zhou, H. Pan and S. Fan, Energy Fuels, 2022, 36, 8506–8514.
48. C. Huang, A. Hu, Y. Li, H. Zhou, Y. Xu, Y. Zhang, S. Zhou, Q. Tang, C. Chen and X. Chen, Nanoscale, 2019, 11, 16515–16522.
49. Y. Zhang, S. Fan, S. Li, Y. Song and G. Wen, J. Mater. Sci., 2020, 55, 12214–12231.
50. J. Chen, Q. Zhou, W. Ma, C. Wang, S. Fan and Y. Zhang, J. Mater. Sci.: Mater. Electron., 2024, 35, 1–13.
51. S. Fan, L. Wei, X. Liu, W. Ma, C. Lou, J. Wang and Y. Zhang, Int. J. Hydrogen Energy, 2021, 46, 39969–39982.
52. H. Gao, D. Zhang, H. Zhou, J. Wu, G. Xu, Z. Huang, M. Liu, J. Yang and D. Chen, Appl. Surf. Sci., 2020, 534, 147613.
53. X. Che, J. Yang, S. Liu, M. Wang, S. He and J. Qiu, Ind. Eng. Chem. Res., 2022, 61, 8908–8917.
54. S. Feng, Z. Liu, Q. Yu, Z. Zhuang, Q. Chen, S. Fu, L. Zhou and L. Mai, ACS Appl. Mater. Interfaces, 2019, 11, 4011–4016.
55. M. Moussa, M. F. El-Kady, S. Abdel-Azeim, R. B. Kaner, P. Majewski and J. Ma, Compos. Sci. Technol., 2018, 160, 50–59.
56. Y. Zhang, L. Wei, X. Liu, W. Ma, C. Lou, J. Wang and S. Fan, Int. J. Hydrogen Energy, 2022, 47, 33827–33838.
57. N. Díez, C. Botas, R. Mysyk, E. Goikolea, T. Rojo and D. Carriazo, J. Mater. Chem. A, 2018, 6, 3667–3673.
58. S. H. Lee, D. Kang and I. K. Oh, Carbon, 2017, 111, 248–257.
59. Z. H. Zhou, M. Z. Li, H. D. Huang, L. Li, B. Yang, D. X. Yan and Z. M. Li, ACS Appl. Mater. Interfaces, 2020, 12, 18840–18849.
60. K. Y. Chen, S. Gupta and N. H. Tai, Compos. Commun., 2021, 23, 100572.
61. Y. Xu, Z. Lin, Y. Yang, H. Duan, G. Zhao, Y. Liu, Y. Hu, R. Sun and C. P. Wong, Mater. Horiz., 2022, 9, 708–719.
62. M. Wang, X. H. Tang, J. H. Cai, H. Wu, J. Bin Shen and S. Y. Guo, Carbon, 2021, 177, 377–402.
63. Q. Qi, L. Ma, B. Zhao, S. Wang, X. Liu, Y. Lei and C. B. Park, ACS Appl. Mater. Interfaces, 2020, 12, 36568–36577.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta01292a

---