Vertical nanodiamond dominated sheets possessing both high capacitance and high n-type Hall mobility

Abstract

It is very difficult to prepare vertical nanodiamond (VND) dominated sheets compared with vertical graphene (VG) sheets. Here, we describe the preparation of VNDs by phase transformation from Ta-loaded VGs via argon/oxygen plasma treatment with varied oxygen percentage. As the oxygen percentage increases to 5%, the height of VGs decreases, accompanied by high capacitance and low Hall mobility. When the oxygen percentage increases to 10%, the VGs transform to nanocrystalline diamond (NCD) grains with trans-polyacetylene (TPA) contents in grain boundaries increasing significantly. These VNDs exhibit both high capacitance (1452 μF cm⁻²) and n-type Hall mobility (846 cm²V⁻¹s⁻¹). With the oxygen percentage increasing to 20%, NCD grains grow larger and the TPA content reduces, while the capacitance decreases considerably, but Hall mobility remaining high, suggesting that the performance improvement results from the synergistic effect of NCDs and TPA. These exhibit significant applications of VNDs in energy storage, high performance sensors, high-frequency and high-power electronic devices.

The authors transformed Ta-loaded vertical graphene into diamond by Ar/O plasma treatment at ordinary pressure and obtained vertical nanodiamond dominated sheets electrode with both high capacitance and high n-type Hall mobility.

Introduction

Fundamental electronic components and capacitors play a crucial role in various applications, including electronic devices and energy storage systems. The design of electrodes therein and the selection of materials directly impact on key performance metrics of capacitors. Common electrode materials include carbon-based materials, metal oxides, and conductive polymers^1–4, each offering distinct properties such as electrical conductivity, chemical stability, and surface area, making them suitable for various application. Among them, diamond has attracted considerable attention due to its excellent thermal conductivity, chemical stability, and mechanical strength. However, diamond is inherently insulating, specific treatments are required to transform diamond into an electrically conductive material for practical applications requiring electrically conductive diamond.

Boron doping has been recognized as a promising method^5–7 to produce boron-doped diamond (BDD). However, BDD exhibits relatively low electrochemical double-layer capacitance (EDLC) performance and high interfacial impedance⁸. To overcome these drawbacks, structural modulation strategies, such as compositing and nanotechnology were applied^9–11. Many studies have focused on nanocrystalline diamond (NCD)-based composite electrode materials. For example, boron-doped graphene (BDG)/BDD cathode and BDG/nanocrystal BDD/TiO₂ nanotube array anode were used to fabricate a capacitor¹², achieving specific capacitances of 6.676 × 10⁴ and 2.386 × 10⁴ μF cm⁻², respectively. Facure et al. prepared a composite structure of NCD and MXene and obtained the capacitance of 235 F g⁻¹ in AlCl₃ electrolyte¹³. In addition to composites, nitrogen-doping has also been reported to enhance the electronic transport and electrochemical properties of NCD films^14–16. The tight-binding studies proved that when nitrogen is introduced into the grain boundaries (GBs), the nearest-neighbor hopping could occur in the GBs and result in enhanced electron transport via the new carbon states in the band gap¹⁷.

Recently, vertically aligned nanosheet structure with high specific surface area has attracted considerable attention due to its superior reaction kinetics and mass transportation capability¹⁸. It was reported that vertically aligned reduced graphene oxide electrode obtained an areal capacitance of 1.83 F cm^–2 in 6.0 M KOH electrolyte¹⁹. It was reported that the closely-packed NCD films, in which nanocrystalline diamond grains closely-packed without amorphous carbon or nano-graphite grain boundaries, exhibited high n-type mobility²⁰. If such NCD films could be fabricated into vertical sheet structures, they may offer excellent performances due to their high Hall mobility and high specific surface area, which hold great potential for high-frequency and high-power energy storage and electronic devices. However, reports on this structure are currently lacking due to the difficulty in the preparation comparative to that of vertical graphene sheets (VGs). Encouragingly, VGs have been successfully prepared and employed in three-dimensional electrodes, exhibiting high efficient oxygen reduced reaction performance^21,22. It was also observed that reducing the number of graphene layers by the plasma treatment enhanced the carrier mobility^22,23. Moreover, it was reported that single-atom transition metals let the phase transformation from graphite to diamond at ordinary pressure^20,24, and the presence of oxygen also contributed to this promotion²⁵. It was also demonstrated that VGs prepared in Ta-filament chemical vapor deposition (CVD) equipment contained single-atom Ta²¹.

Here, we propose an innovative method for preparing vertical nanodiamond sheets (VNDs), in which argon/oxygen (Ar/O) plasma treatments with different oxygen ratio flow of 2%, 5%, 10%, 15%, and 20% are performed on VGs prepared by CVD with Ta filament to adjust the layer of graphene and facilitate the transformation of the graphene phase into diamond. The results indicate that the as-deposited VGs consist of graphene and a small amount of NCDs grains. When the oxygen ratio increases to 5%, the graphene undergoes etching, leading to the decrease in capacitance, while the Hall mobility remains low. With oxygen ratio increasing to 10%, the vertical sheets contain almost no graphene but consist of closely packed nano-sized diamond grains and significantly increased trans-polyacetylene (TPA) content in GBs, exhibiting both high double-layer capacitance (1452 μF cm⁻²) and n-type Hall mobility (846 cm²V⁻¹s⁻¹). As the oxygen ratio increases to 15% and 20%, NCD grains become larger and TPA content in GBs decreases dramatically. The capacitance of these electrodes drops significantly while Hall mobility remains high. These suggest that with oxygen proportion increasing to 10% and above, Ta and oxygen synergistically promoted the phase transformation from graphene to diamond, forming VNDs. Moreover, the NCDs with TPA chains in GBs simultaneously enhanced the capacitance and n-type Hall mobility. These VNDs provide applications in the areas of electrodes, energy storage, and high-performance electronic devices.

Results and discussion

Characterization of surface morphology

The field emission scanning electron microscope (FESEM) images show that all samples exhibit an overall morphology of clustered cauliflower-like particles with curved vertical sheets on its surface (Fig.  1). The as-prepared sample is named VS_UN and the treated samples are labeled as VS₂, VS₅, VS₁₀, VS₁₅, and VS₂₀, respectively. The average diameter of the particles in the untreated sample is approximately 1.72  μm (Fig. 1a). After Ar/O plasma treatment, the average diameter gradually decreased to approximately 0.87 μm (Fig. 1b–f), demonstrating that the etching effect of the plasma treatment, which intensified as the oxygen ratio increased. When the oxygen proportion increases to 10% and above, the reduction of particle size was not as significant as in samples at lower ratios, indicating that the vertical sheets on these samples maintained a stable state after plasma treatment.

Fig. 1. The FESEM images of the samples acquired in secondary electron (SE) mode.

The FESEM images of samples (a) VS_UN, b VS₂, c VS₅, d VS₁₀, e VS₁₅, and f VS₂₀. Inset images (a-1)–(f-1) show the top view of the samples. Inset images (a-2)–(f-2) show the samples in cross-sectional orientation with thickness value.

Characterization of electrochemical and electrical properties

The electrochemical performance and Hall mobility of the prepared samples were tested, including CV curves (Fig. 2a–g), calculated area-specific capacitance and Hall mobility (Fig. 2h), and the charge/discharge curves at various scan rates (Fig. 2i). When the oxygen proportion is at or below 10%, the CV curves of the samples exhibit rectangular-like shapes, suggesting typical EDLC performance and the capability of efficiently and reversibly adsorbing/desorbing charge during the scanning process (Fig. 2a–d). In contrast, the curves of samples VS₁₅ and VS₂₀ have changed significantly from rectangular-like curves to symmetrical arcs, suggesting the suppression of capacitance behavior (Fig. 2e, f). Especially in sample VS₂₀, where the CV curves exhibit almost straight lines shape, indicating inefficient charge storage. The CV curves of all samples were further compared at a fixed scan rate of 0.1 V s^–1 (Fig. 2g). Samples from VS_UN to VS₁₀ exhibit relatively low response current, whereas VS₁₅ and VS₂₀ show more separated and broader anodic and cathodic curves. This difference implies changes in the electrode composition and microstructures induced by Ar/O plasma treatment and the resulted transition in electrochemical behavior.

Fig. 2. Electrical and electrochemical properties of the samples.

Cyclic voltammogram (CV) curve of the samples (a) VS_UN, b VS₂, c VS₅, d VS₁₀, e VS₁₅ and f VS₂₀ at different scan rates from 0.1 to 0.5 V s^–1. g CV curves of the different samples at 0.1 V s^–1. h The green column graph shows the calculated area-specific capacitance of the samples, and the red line shows the Hall mobility of the samples. i Charge/discharge curve of sample VS₂₀. Source data are provided as a Source Data file.

Although the Hall mobility of the VS_UN electrode is relatively low at 5 cm²V⁻¹s⁻¹, its capacitance reaches the highest value of 1626 μF cm⁻² (Fig. 2h), which exceeds part of the previously reported typical carbon-based electrode materials^2,26 and is higher than that of graphene electrodes with the similar vertical nanosheet structure (265 μF cm⁻²)²⁷. After plasma treatment with oxygen ratio of 2% and 5%, the capacitance of the VS₂ and VS₅ electrodes decreases to 1324 μF cm⁻² and 1068 μF cm⁻², with the Hall mobility of 103 cm²V⁻¹s⁻¹ and 0.91 cm²V⁻¹s⁻¹, respectively. Interestingly, when the oxygen ratio is further increased to 10%, the VS₁₀ electrode exhibits both high capacitance (1452 μF cm⁻²) and significantly enhanced Hall mobility (846 cm²V⁻¹s⁻¹). The capacitance of the VS₁₀ electrode is higher than that of other similar hybrid carbon materials, such as the BDD-carbon nanowall electrode reported²⁸. However, further increasing the oxygen ratio to 15% and 20% results in a sharp decline in capacitance to 370 and 138 μF cm⁻². Despite this, their Hall mobilities remain relatively high at 497 cm²V⁻¹s⁻¹ and 668 cm²V⁻¹s⁻¹. These results reveal that sample VS₁₀ possesses both high capacitance and high n-type Hall mobility. To the best of our knowledge, although the capacitance of some vertically aligned graphene and polymers composites electrode can reach as high as approximately 2300–5200 μF cm^−2 29,30, such a combination of high charge storage capacity and superior n-type mobility is rarely observed in conventional carbon-based electrode systems.

To further evaluate the stability and practicality of the VS₁₀ electrode, GCD measurements were conducted on sample VS_UN, VS₅, and VS₁₀ (Supplementary Fig. S1 and Fig. 2i). The curves exhibit a nearly linear and symmetrical relationship between voltage and current (Fig. 2i), indicating good reversibility and capacitive characteristics of the VS₁₀ electrode. As current density increases, the duration of the charging/discharging decreases accordingly, consistent with typical capacitive behavior. The above results indicate the pronounced impact of oxygen plasma treatment on both the surface morphology and electrical properties of these samples.

Characterization of composition and microstructure

To investigate the underlying mechanisms of the effects of Ar/O plasma treatment on the capacitance and Hall mobility, visible Raman spectroscopy was used to characterize the compositional evolution. Figure 3 shows the Raman spectra of the samples with their Gaussian fitting peaks. It is observed that the spectra of samples VS_UN, VS₂ and VS₅ exhibit primary peaks at 1350, 1580, 1620, 2700, and 2950 cm⁻¹, as shown in Fig. 3a, b, which is attributed to D, G, D’, 2D band, and the D and D’ combination band, respectively, indicating that the vertical sheets of these samples are predominantly graphene^31–34. With oxygen ratio increasing to 10–20%, the Raman spectra show a significant change, which exhibit the typical characteristic of NCD films. Samples VS₁₀, VS₁₅, and VS₂₀ still show the D and G peaks at 1350 and 1580 cm⁻¹. Notably, the 2D and D’ peaks disappear, indicating a substantial reduction in graphene content. A broad and weak peak at 1250 cm⁻¹ emerges, which is commonly associated with sp² carbon phase³⁵. The diamond-related peak fitted at 1332 cm⁻¹confirms the significantly increased NCD contents. Additionally, the TPA-related peaks at 1150 and 1480 cm⁻¹ attributed to TPA segments at the GBs³⁶, are also observed.

Fig. 3. Raman spectra and their fitted data of the samples.

Raman spectra for the samples during the wavenumber range of (a) 800–2000 cm⁻¹ with fitting curves and b 2200–3500 cm⁻¹. The Raman intensity is expressed in arbitrary units (a.u.). c The black line represents the integrated intensity ratio of D to G peaks (I_D/I_G) of the sample, and the blue line represents the integrated intensity ratio of 2D to G peaks (I_2D/I_G) of the sample. d The black line represents the position of the G peak and the blue line represents the full width at half maximum (FWHM) of the G peak in Raman spectra. e The black line represents the integrated intensity ratio of the diamond related peak to the sum of peak D and G of the samples, and the blue line represents the integrated intensity ratio of the TPA to the sum of peak D and G of the sample. Source data are provided as a Source Data file.

The dependence of the fitted data in Raman spectra on the oxygen ratio is compared to further analyze the correlation of the component to the capacitance and Hall mobility. An estimation of the defect content can be obtained from the integrated intensity ratio of the D and G peaks (I_D/I_G)³⁷. It is noted that the capacitance exhibits the same trend as the I_D/I_G in which more defects lead to higher capacitance, suggesting that the defects serve as additional sites for charge storage or adsorption (Fig. 3c). Besides, for graphene dominated samples VS_UN, VS₂ and VS₅, higher I_2D/I_G corresponding to higher Hall mobility. Higher I_2D/I_G indicates fewer graphene layer number^38,39, suggesting that for these samples, fewer number of graphene layers lead to higher Hall mobility. A significant shift in G peak position and an increase in full width at half maximum (FWHM) is observed from VS₅ to VS₁₀ (Fig. 3d), indicating a rise in disorder of the graphene^40,41. Figure 3e presents the integrated intensity ratio of TPA and diamond-related peaks (I_TPA and I_DIA) to the sum of D and G peaks (I_SUM). Particularly, in samples VS₁₀, VS₁₅, and VS₂₀, the Hall mobility is observed to increase with higher I_TPA/I_SUM value, implying a potential contribution of TPA to charge transport in these NCD-dominated samples. It is also observed that where a decrease in I_TPA/I_SUM corresponds to an increase in I_DIA/I_SUM, further indicating a correlation between TPA and the diamond phase.

Raman results demonstrate the microstructure transition from graphene-dominated to diamond-dominated sheets when the oxygen ratio increases to 10%, which accompanied by the rebound of capacitance (1452 μF cm⁻²) and a rapid increase in Hall mobility (846 cm²V⁻¹s⁻¹), while the capacitance of typical diamond electrodes range from 10–1200 μF cm^−2 42,43, and the Hall mobility of closely-packed NCD films is 522 cm²V⁻¹s⁻¹²⁰. This structure possessing both high n-type Hall mobility and enhanced capacitance performance represents a significant breakthrough for diamond materials in electrode applications.

To understand the microstructural evolution responsible for the improved performance, especially for sample VS₁₀, high-resolution transmission electron microscope (HRTEM) analysis was conducted (Figs. 4 and 5). The height of the vertical sheets in sample VS_UN is about 500 nm (Fig. 4a). As shown in the inset selected area electron diffraction (SAED) image (SAED-a), the clear diffraction rings of graphite (002) and diamond (111), (220) confirm a composite phase of graphene and diamond. At higher magnification, distinct darker circular domains can be observed within the layers (Figs. 4a-1 and 4a-2). Inset FFT-a-1 and FFT-a-2 patterns show the fast Fourier transformation (FFT) images of these two images, respectively, demonstrating that both regions contain graphite and diamond phases. The representative areas from these two regions as selected by the square frames were enlarged (Fig. 4a-ⅰ and 4a-ⅱ). Lattice fringes are observed with spacing of 0.251 nm and 0.337 nm, attributing to the plane of diamond (110) and graphite (002), respectively (Fig. 4a-ⅰ), showing the structure of NCDs wrapped in graphene. Massive interspersed ribbon structures are observed (Figs. 4a-2). The selected area shows a lattice fringe spacing of 0.337 nm (Fig. 4a-ⅱ), indicating that this ribbon structure consists of multi-layered graphene, which are available to create defects providing active sites for charge storage, corresponding to the high I_D/I_G value in Raman results. These results reveal that the untreated sample is composited of nanodiamond and graphene, which is different from the previously reported pure graphene vertical sheets. It was reported that using electrochemically active nanodiamond particles as intercalation materials for graphene oxide nanosheets holds promise for producing high capacitance composite materials⁴⁴. Therefore, the higher capacitance of the VS_UN sample compared to the vertical graphene sheets reported previously is attributed to the effects of trace nanodiamond particles between the graphene layers.

Fig. 4. HRTEM images of the samples VS_UN, VS₂, and VS₅.

HRTEM images and their inset SAED patterns for samples (a) VS_UN, b VS₂, and c VS₅ at lower magnification. The yellow lines with arrows denote the length of the vertical sheets, and the dashed curves denote crystal planes of the same color; Dia (111), Dia (220), Dia (311), Dia (110) and G (002) denote the indexed crystal planes of diamond and graphite, respectively; (a-1)–(c-2) HRTEM images of the samples at higher magnification as selected by white square frame and their insert FFT patterns; (a-ⅰ)–(c-ⅱ) Magnified images of the regions selected by the green and blue square frames, respectively, and the highlighted lines mark pairs of adjacent lattice fringes and the corresponding spacing between them.

Fig. 5. HRTEM images of the samples VS₁₀, VS_15, and VS₂₀.

HRTEM images and their inset SAED patterns for (a) VS₁₀, b VS₁₅, and c VS₂₀ at lower magnification. The dashed curves indicate crystal planes corresponding to the same color; Dia (111), Dia (220) and Dia (311) denote the indexed crystal planes of diamond, respectively; (a-1)–(c-2) HRTEM images of the samples at higher magnification as selected by white square frame and their insert FFT patterns; (a-ⅰ)–(c-ⅱ) Magnified images of the regions selected by the green and blue square frames, respectively, and the highlighted lines mark pairs of adjacent lattice fringes and the corresponding spacing between them.

The height of vertical graphene sheets was reduced to about 300 nm in the sample VS₂ (Fig. 4b), revealing the etch effect of oxygen plasma. The inset SAED pattern displays clear diffraction rings of both graphene and diamond. Notably, whisker structures region consisting of fine grains is observed in the upper right area (Fig. 4b), which is absent in sample VS_UN, suggesting that these whisker structures formed during Ar/O plasma treatment. To investigate the composition of this structure, region 1 is selected and enlarged (Figs. 4b-1), where the FFT image reveals diffraction information of diamond (111) plane. A selected grain exhibits the lattice fringe spacing of 0.204 nm and an angle of 70° (Fig. 4b-ⅰ), suggesting that these whisker structures consist of nanodiamonds. Moreover, the needle-like structures exhibit more alignment compared to those in sample VS_UN (Figs. 4b-2), which are confirmed as graphene by the lattice fringe spacing (Fig. 4b-ⅱ). Furthermore, the diffraction rings of the FFT images are more distinct compared to the VS_UN, suggesting that the plasma treatment not only etched the graphene but also affected the crystal arrangement, potentially due to atomic rearrangement induced by the plasma.

In sample VS₅, the height of the vertical sheets further decreases to 250 nm and a morphology of closely packed grains appears at the top of these sheets (Fig. 4c). Numerous circular structures appear in the needle-like structures (Figs. 4c-1). The enlarged image of selected circle reveals the structure of diamond particles wrapped within graphene (Fig. 4c-ⅰ), indicating that the continuous graphene structure is disrupted by these nanodiamond particles, thereby impeding the charge migration and resulting in a significant decrease of Hall mobility compared to sample VS₂. The diffraction spots of graphene appear more focused in the FFT image (Figs. 4c-2) and the number of graphene layers is significantly reduced compared to the untreated sample (Fig. 4c-ⅱ), which is also reflected in the gradual decrease of the I_D/I_G from the untreated sample to VS₅ in the Raman results.

For sample VS₁₀, different from the previous samples, the needle-like vertical graphene sheets have been almost completely etched and replaced by short whisker structures at the edges (Fig. 5a). These vertical whiskers consist of fine grains with an average diameter of approximately 5 nm (Figs. 5a-1 and a-2). The selected regions a-ⅰ and a-ⅱ both exhibit lattice fringes with a spacing of 0.207 nm. Combined with the diffraction ring corresponding to the diamond (111) plane, confirming these fine grains are NCDs. Different from sample VS₅, sample VS₁₀ contains few graphene, while remaining vertical sheet structure, indicating the phase transformation from graphene to diamond and the subsequently prepared nanodiamond dominated vertical sheets, which are defined as VNDs. According to the increase of TPA contents confirmed by the Raman results, the enhancement of electrical performance is correlated with TPA chains in GBs, which provides a conductive network for charge transportation⁴⁵.

For the sample VS₁₅, NCDs with size increasing to around 10 nm are packed to form VNDs which are slightly shorter than VS₁₀ (Fig. 5b). The enlarged images of the selected regions b-ⅰ and b-ⅱ at the sample boundary in Figs. 5b-1 and 5b-2 show crystalline fringes of about 0.206 nm and 0.207 nm, respectively (Fig. 5b-ⅰ and 5b-ⅱ), corresponding to the diamond (111) plane. The TPA segments are present in the GBs of nanodiamonds⁴⁶, hence the TPA content in VS₁₅ decreases as the diamond grain size increases, suggesting that TPA transit to diamond via the help of Ta single atoms and oxygen plasma. The capacitance of this sample shows an associated significant decrease from 1452.48 to 369.96 μF cm⁻², indicating that TPA structures provide active sites for charge loading. Moreover, the Hall mobility also decreased from 846 to 497 cm²V⁻¹s⁻¹, further confirming that TPA supplies sites for both charge loading and conductivity channel for the samples.

Sample VS₂₀ no longer exhibits whisker structures and displaced by less curved edges (Fig. 5c). The selected two regions in this sample exhibit the diffraction information of diamond (111) (Figs. 5c-1 and 5c-2). The selected grains on both regions reveal lattice fringe spacing of 0.206 nm, which corresponds to the diffraction pattern of the diamond (111) plane (Fig. 5c–ⅰ and 5c-ⅱ). Moreover, the nanodiamond particles on VNDs in the VS₂₀ sample are larger overall compared to those in samples VS₁₀ and VS₁₅. The selected area electron diffraction graph also exhibits clearer diffraction rings and spots as shown by inset SAED-c, indicating more consistent crystal orientation in this sample. The change in crystal size and orientation reduces charge scattering at GBs, resulting in a slight enhancement of the Hall mobility in the sample VS₂₀ compared to VS₁₅. Furthermore, the Raman results show a decrease in I_TPA/I_SUM and an increase in I_DIA/I_SUM when the oxygen content is beyond 10%, inferring the transformation from TPA chains to diamond with the aid of both oxygen and single-atom Ta in VGs, which is also confirmed by theoretical calculations²⁵.

The evolution of microstructure and performance of the samples caused by Ar/O plasma treatment with different oxygen proportions is illustrated in Fig. 6. As the oxygen proportion increases gradually, the samples exhibit three typical states with different structures and compositions. When the oxygen ratio is below 10%, the vertical sheets consist of multi-layer graphene and minor nanodiamond particles (Fig. 6a). These sheets provide numerous active sites for charge storage due to their large specific surface area and multi-layer graphene, resulting in high capacitance. However, the high number of graphene layers leads to significant phonon scattering (as indicated by the red arrow line in the figure), hindering carrier movement (Supplementary Table S1) and resulting in low Hall mobility. When the oxygen ratio is at 10%, the height of the vertically aligned sheets decreases (Fig. 6b) and the graphene transits to diamond. A representative region selected by the orange box indicates that these VNDs, are composed of nanodiamond particles, with continuous TPA segments at the GBs, which supplies active sites for carrier loading and conductive pathways for carrier transport. Therefore, this structure exhibits both the highest Hall mobility (846 cm²V⁻¹s⁻¹) and relatively high capacitance (1452 μF cm⁻²). When the oxygen ratio is above 10%, the height of the VNDs exhibits minimal change because diamonds are more difficult to etch than graphene (Fig. 6c). The nanodiamond particles on VNDs become larger and TPA segments at GBs decrease, leading to the decrease of the capacitance, which also obstacles the continuous structures, impeding carrier transportation (as shown by the green arrows) and thereby resulting in a decreased Hall mobility.

Fig. 6. Schematic illustration of the process of Ar/O plasma treatment on VS samples.

a Cauliflower-like spherical particles with VGs, colored blue, for oxygen ratio below 10%; b Cauliflower-like spherical particles with VNDs containing large amount of TPA, colored green, for oxygen ratio at 10%; c Cauliflower-like spherical particles with VNDs with small amount of TPA, colored green, for oxygen ratio above 10%. (a-1)–(c-1) show the detail structure of vertical sheets on spherical particles as selected by the black square frames. (a-2)–(c-2) show the state and direction of carriers movement and scattering in the sheets.

In summary, Ar/O plasma treatment was performed on the Ta-loaded VGs with varying oxygen proportions to synthesize the VNDs. The results show that when the oxygen ratio is below 10%, the vertical sheets dominated by graphene exhibit limited Hall mobility, and the electrochemical double-layer capacitance gradually decreases with increasing oxygen content. Particularly, at an oxygen proportion of 10%, VNDs with both enhanced double-layer capacitance (1452 μF cm⁻²) and n-type Hall mobility (846 cm²V⁻¹s⁻¹) were successfully prepared. This sample primarily consists of NCDs with significantly increased TPA content, indicating a pronounced synergistic interaction between the nanodiamond phase and TPA. However, at higher oxygen proportions (15% and 20%), the formation of larger-sized nanodiamond grains and a substantial reduction in TPA lead to a notable decline in capacitance, while n-type Hall mobility remains relatively high, suggesting an optimized microstructure that promotes both efficient charge storage and carrier transport at an oxygen proportion of 10%. This work transforms Ta-loaded graphene into diamond through Ar/O plasma treatment at ordinary pressure, providing an approach for the synthesis of VNDs and further expanding the application of diamond electrodes in energy storage, high-performance sensors, high-frequency and high-power electronic devices.

Methods

Preparation of the vertical sheets

Vertical graphene sheets (VGs) were grown by using hot filament chemical vapor deposition (HFCVD) (JUHFCVD001). Silicon wafers undergo surface seeding treatment by polishing with diamond pastes for 20 min. After that the polished silicon wafers were washed with deionized water and dried with nitrogen gas, then placed into the cavity of the HFCVD equipment. The gas source is composed of hydrogen and acetone. Here, the acetone as carbon source was brought into cavity by hydrogen bubbling with 90 sccm, and hydrogen was 200 sccm. A two-step method to construct the VGs was according to our previous reports^21,22. In the first step, the growth power was 2200 W, and the growth time was 20 min. In the second step, the carbon source was increased to 100 sccm and hydrogen was stopped. The growth power was reduced to 1800 W and growth time was 3 min.

After finishing the growth process, an as-prepared VGs sample was divided into several smaller pieces and then treated with microwave plasma by using a microwave plasma vapor deposition equipment (MP-RCVD, Tam kang University, Taiwan) in atmosphere containing a mixture of argon and oxygen. The power of plasma was 600 W, and the pressure was 15 Torr. The processing time was 4 min. The total flow of argon and oxygen was 100 sccm with the oxygen ratio flow controlled to be 2%, 5%, 10%, 15% and 20% for the respective pieces.

Characterization of the vertical sheets

The surface morphology of the samples was characterized using field emission scanning electron microscope (FESEM, FEI Nano Nova 450) acquired in secondary electron (SE) mode at a low voltage of 5 kV. Their phase and composition characterization were investigated by the visible Raman spectroscopy (Raman, Horiba Jobin-Yvon) under room temperature (25 °C) with a 532 nm excitation wavelength and the spectra were collected in the range of 500–3500 cm⁻¹. The spectra within the range of 800–2000 cm⁻¹ had been fitted with Gaussian lines. To protect the microstructure of the films from damage during TEM sample preparation, small fragments of film were scraped down from the silicon substrate and loaded on copper mesh. The microstructures were characterized by the high-resolution transmission electron microscope (HRTEM, FEI Talos) at 200 kV accelerator voltage. The electrical properties of these samples were measured by Hall effects tester (LakeShore 8400 Series) under room temperature (25 °C). Hall effects measurements were performed on the samples using the Van der Pauw geometry. Ohmic contacts were made by applying silver paste at the four corners of the square-shaped samples.

Electrochemical characterization

The capacitance characteristics of the samples were measured by cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD) techniques in an electrochemical workstation (CHI760D, Shanghai Chen Hua) at room temperature (25 °C). Electrochemical tests were systematically conducted in 1 M KOH electrolyte with a standard three-electrode system, consisting of a Pt (99.99%) counter electrode and the saturated calomel electrode (SCE) conference electrode. The apparent area of a working electrode that was in contact with electrolyte was 0.1257 cm². The cyclic voltammograms of different films were recorded within a potential window of −0.3 to 0.3 V, of which scan rates were varied from 0.1 to 0.5 V s⁻¹. The GCD tests were conducted at a current density in the range of 0.2–0.5 mA cm^–2. The area-specific capacitance (C_A) of these electrodes was respectively calculated from the integrated charges of these cyclic voltammograms according to the equations of:

$$C_A=\frac{\int I\bullet \mathrm{d}V}{v\times A\times △E}$$ 1

where I is the current density, V is the electrode potential, v is scan rate, ∆E is the potential window, A is the apparent area of the electrode, respectively.

Author contributions

X.H. and Z.Z. conceived this research project and designed the experiments. Z.Z. prepared the samples. Y.G. and Z.Z. completed the sample characterization and analysis. M.J. and C.C. provided support for the technical issues during the experiments. X.H. and S.L. provided support for the analysis of the experimental results. Y.G. and X.H. wrote the manuscript with the essential input of other co-authors. All authors have given approval of the final manuscript.

Peer review

Peer review information

Nature Communications thanks Tsuyoshi Yoshitake and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Source data are provided with this paper. Should any raw data files be needed in another format, they are available from the corresponding author upon request. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-70089-9.