Hydrodynamic cavitation-assisted preparation of porous carbon from garlic peels for supercapacitors

Highlights

• •HC is used to prepare porous carbon for SCs for the first time.
• •A SSA of 3272 m²/g with a capacitance of 227 F/g is maximumly achieved.
• •HC/KOH treatment doubles the performance of SCs compared with sole KOH activation.
• •HC is a potential preparation method of porous carbon with high efficiency.

Abstract

Hydrodynamic cavitation (HC), which can effectively induce sonochemical effects, is widely considered a promising process intensification technology. In the present study, HC was successfully utilized to intensify the alkali activation of GPs for SCs, for the first time. Five BDCMs were synthesized following the method reported in the literature. For comparison, four more BDCMs with HC-treated, among which a sample was further doped with nitrogen during the HC treatment, were prepared. Then all the samples were compared from microscopical characteristics to electrochemical performance as SCs materials. The morphology study demonstrated that the HC treatment had created many defects and amorphous carbon structures on the GP-based BDCMs, with the highest SSA reaching 3272 m²/g (1:6-HCGP), which 32 folded that of the Raw carbon sample’s. The HC treatment also intensified the N-doping process. XRD and XPS results manifested that the N content had been increased and consequently changed the electronic structure of the carbon atoms, leading to the increase of specific capacitance (1:6-HCGP+N-based SC, 227 F/g at 10 A/g). The cycle performance proved that the GP-based BDCMs have long-term stability, indicating that the HC-treated BDCMs were good choices for energy storage technologies. Compared with the ultrasound-assisted method, which may have a high energy density, the HC-assisted method enables high production and energy efficiency. This work is a first time attempt towards the industrial application of HC method in energy-related materials synthesis and encourages more in-depth studies in the future.

1. Introduction

In the face of global warming and energy shortage, using energy storage facilities has seen great potential in reducing CO₂ emissions [1] and storing solar/wind-generated energy [2], [3], [4], [5], [6]. Among various devices, supercapacitors (SCs), possessing several advantages of high capacitance and power density and long cycles, are drawing increasing attention [7], [8]. SCs are generally divided into pseudocapacitors and electric double-layer capacitors (EDLCs) [9]. The formers store electric energy through reversible redox reactions with electroactive substances on the electrode surface [10], [11]. For the EDLCs, the energy storage mechanism lies in forming an electric double layer on the electrode’s surface under the electrostatic force. Hence, the charging and discharging processes in EDLCs mainly depend on physical electrostatic attraction, leading to long-life features, high rate capability, and power density [12]. Moreover, due to the high cost of Ru₂O [13] and the electrochemical instabilities of most transition metal compounds in aqueous electrolytes, it is not easy to apply pseudocapacitors in commercial devices. On the contrary, EDLCs, based on porous carbon (PC) electrodes, are more economical to be used in industrial applications [14], [15].

Among various PC materials, biomass-derived carbon materials (BDCMs) are synthesized based on biomass [16], [17]. This refers to all organic matter formed directly or indirectly through the photosynthesis of green plants [18], [19], including all animals, plants [20], microorganisms [21], and excretion and metabolites produced by these organisms, which are clean, renewable, and diverse [22]. Since they naturally possess many beneficial functional groups and heteroatoms, and properties of extensive access and sustainability, these BDCMs have large potential for commercial SCs [23], [24]. To achieve high specific capacitance, a high specific surface area is needed [25]. Therefore, it is necessary to activate the source carbons in physical or chemical ways before utilizing them in SCs. However, the activation methods reported up to now, including physical, physicochemical [26], chemical, template-assisted/induced, and microwave/ultrasonic induced/assisted methods, exhibit high energy consumption and low economic efficiency [27]. This necessitates exploring novel activation methods for BDCMs with high energy efficiency and low cost and can be extended for large-scale production [28].

Hydrodynamic cavitation (HC) has been extensively considered a promising process intensification technology for industrial-scale applications [29], [30], [31], [32]. Like ultrasound, the intensification mechanism of HC originates from the cavitation phenomenon, which is a rapid phase-change phenomenon (from liquid to gas) within a concise period in liquids [33], [34]. The bubble collapse releases enormous energy into the surrounding fluid and induces mechanical, thermal, and chemical effects, known as sonochemical effects [35], [36]. As a result, HC can create a unique environment with high pressure (∼1000 bar), temperature (∼5000 K), and strong oxidation (hydroxyl radicals)/reduction (hydrogen radicals) environment under ambient conditions [37], [38]. Such extraordinary conditions can effectively increase the reaction rates and reduce the requirements for reaction conditions (temperature, pressure, concentration, purity, etc.) [39], [40]. Since Pandit and Joshi [41] first applied HC to oil hydrolysis in 1993, HC has demonstrated great commercialization prospects in various applications, e.g., water treatment [42], [43], emulsification [44], delignification [45], food processing [46], [47], flotation [48], surface finishing [49], heat generation [50], nanomaterial synthesis [51], and even graphene exfoliation [52. Numerous studies have demonstrated that HC can efficiently induce a sonochemical effect similar to ultrasound, with high scalability and low cost [53], [54].

Regarding the preparation of BDCMs, Albanese et al. [55] recently discovered that HC can enhance the surface area of biochar by 120 % by preserving or improving the respective chemical composition in a pilot scale facility [56] (30 kg of feedstock). Moreover, the process yield of HC (92–315 kg/kWh) was significantly higher than slow pyrolysis (less than18 kg/kWh). This novel work proved that the HC-based preparation of BDCMs is a practical method and can overcome the scale-up issue in various commonly followed methods. Nevertheless, the intensification mechanism underlying the HC effect is still absent to our knowledge.

To understand the mechanism of HC effect on the preparation of BDCMs, the present study, for the first time, investigated the intensification effect of HC on the alkali-assisted synthesis of BDCM for SCs. Nine samples from garlic peel (GP) with various synthesis conditions (i.e., HC treatment employed or not, nitrogen doping or not, carbon to alkali mass ratio in the mixture before activated) were prepared, and their microscopical characteristics to electrochemical performance as SC materials were tested and compared. The results showed that HC treatment facilitated the diffusion of KOH in the raw carbon, leading to the creation of more micropores, and greatly improved the GP-based BDCMs’ specific surface area (SSA) and electrochemical performance.

2. Methodology

2.1. Materials

GPs were achieved from Shanghe County, Jinan, Shandong Province, China [57]. The chemicals used in this study (potassium hydroxide, hydrochloric acid, sulfuric acid, nitric acid, absolute ethanol, melamine, etc.) were analytically pure and purchased from Tianjin KEMO Chemical Reagent Co., Ltd. High-purity nitrogen, purchased from Jinan Deyang Special Gas Co., Ltd., with a purity of ≥99.999 %, was utilized as the inert protective gas in the tubular furnace and atmosphere furnace.

2.2. Hydrodynamic cavitation system and the SC materials preparation process

The schematic diagram of the HC system is illustrated in Fig. 1. The HC possesses a Venturi tube with a diameter of 10 mm, a throat diameter of 1.5 mm, a contraction angle of 40°, and an expansion angle of 15°. A pair of pressure and temperature sensors were installed on the upstream and downstream sides of the Venturi. A plunger pump with a power rating of 1 kW was used to recirculate the solution in the water tank with a volume of 1 L. In this experiment, the pump pressure was set at 0.2 MPa, and the solution temperature was maintained at approximately 20 ℃ using a heat exchanger.

Fig. 1.

Schematic diagram of the HC system (1 – plunger pump, 2 – flow meter, 3 – pressure gauge, 4 – Venturi tube, 5 – temperature gauge, 6 – heat exchanger, 7 – water tank).

Synthesis of GP-based BDCMs: GPs were washed with deionized water and then put in an oven at 120 °C for 12 h to dry. GP powders with particle sizes of less than 45 μm were obtained by a pulverizer (LBH-400Y) with a rotational speed of 32000 rpm for 5 min. Finally, GP powders were carbonized in a tubular furnace (600 °C, 2 h) to achieve raw carbon products. Five BDCMs (one with nitrogen doping) were synthesized following the alkali activation method reported in the literature [58]. For comparison, four more HC-treated BDCMs, among which a sample was further doped with nitrogen (melamine as the nitrogen source) during the HC treatment, were prepared. The activated samples were washed with distilled water until the supernatant liquid’s pH reached 8 at 80 °C. Then, the pH was adjusted to 2 by adding hydrochloric acid (HCl) with a concentration of 1 mol/L. All the samples were then thoroughly washed with deionized water until the pH of supernatant liquid reached 7. Lastly, the samples were dried at 120 °C in a vacuum oven for 24 h. All the samples were named based on whether HC treatment and nitrogen doping (N-doping) were employed and the carbon to alkali mass ratio in the mixture before activation, i.e., Raw carbon, 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP, and 1:8-HCGP (Table S1).

2.3. Characterizations

Scanning electron microscopy (SEM) images were obtained by a field emission SEM (TESCAN MIRA LMS). Transmission electron microscopy (TEM) images were acquired by a JEM-2100 microscope. An SSA and pore size analyzer (JW-BK132F) was utilized to conduct the nitrogen adsorption/desorption isotherms. The SSA and pore size distribution were determined by the Brunauer-Emmett-Teller (BET) method and the Barrett-Joyner-Halenda (BJH) model, respectively. A Renishaw/RM2000 (514 nm laser) was used to conduct Raman spectroscopy measurement. X-ray diffraction (XRD) was performed using an Ultima IV diffractometer with a Cu Kα radiation source. X-ray photoelectron spectroscopy (XPS) measurements were obtained on a Thermo Scientific K-Alpha X-ray photoelectron spectrometer equipped with an Mg-Kα radiation source.

2.4. Electrochemical measurements

To prepare the electrode slurry, GP-based BDCMs, conductive graphite, and polytetrafluoroethylene (PTFE) emulsion (60 wt%) were weighed in a mass ratio of 8:1:1 and mixed with a proper amount of absolute ethyl alcohol. After that, the mixture was treated by ultrasonic dispersion for 30 min, and approximately 4 mg of the as-prepared slurry was coated on a foamed nickel electrode (diameter 16.2 mm) uniformly. Then, the electrode was kept in a vacuum oven at 80 °C for 12 h and pressed with a tablet machine at 15 MPa for 1 min. The electrochemical performance tests of the BDCM-based SCs were conducted in a two-electrode system with 6.0 M KOH as the electrolyte. Two electrode sheets with a glass fiber filter paper (Waterman, GF/B) in the middle were assembled into a button capacitor. Furthermore, an electrochemical workstation (CS350H, Wuhan Corrtest, China) was used to conduct the electrochemical tests. The specific capacitance C (F/g), energy density E (Wh/kg), and power density P (W/kg) of the SCs were calculated using Eqs. (1), (2), and (3), respectively.

$$C=\frac{4I\mathrm{\Delta }t}{m\mathrm{\Delta }V}$$ (1)

$$E=\frac{1}{8}C{(\mathrm{\Delta }V)}^2$$ (2)

$$P=\frac{E}{\mathrm{\Delta }t}$$ (3)

where I is the discharge current (A), Δt is the discharge time (s) in Eq. (1) and discharge time (h) in Eq. (3), m is the GP-based BDCM mass on the two electrode sheets (mg), and ΔV is the electrical potential difference (V).

3. Results and discussion

3.1. Morphology and structure

The morphology of the as-prepared GP-based BDCMs were analyzed by SEM, as shown in Fig. 2. The SEM images showed that all the samples possessed two-dimensional and wrinkled morphology. As the carbon to alkali mass ratio decreased (from 1:4 to 1:8), more micropores were observed, these micropores linked macropores and mesopores, improving ions transportation and storage during the charge and discharge processes of the SCs. Notably, with HC treatment employed, ravines were created on the surface of the samples which could enable full contact of active sites with electrolyte and absorb more ions, thus increasing the specific capacitance [59].

Fig. 2.

SEM images of (a) Raw carbon, (b) 1:4-GP, (c) 1:4-HCGP, (d) 1:6-GP, (e) 1:6-HCGP, (f) 1:6-GP+N, (g) 1:6-HCGP+N, (h) 1:8-GP, and (i) 1:8-HCGP.

To further investigate the structure of the GP-based BDCMs, TEM images were obtained and are shown in Fig. S1 [60]. Corresponding the SEM images, the samples showed 2D and transparent morphology. The HC-treated samples had smaller particle size of 1 – 2 μm compared with the untreated ones (particles size 30 – 40 μm). This was due to the intense mechanical effect which was featured with large shear stress during the HC treatment process. The smaller-size BDCMs derive microstructure in the SCs’ electrodes led to higher specific capacitance [61]. For the nitrogen-doped samples (1:6-GP+N and 1:6-HCGP+N, Fig. S1 f and g), many nano-size particles on the GP-based BDCMs could be observed, these particles were the N-doping dots, and more N-doping dots distributed uniformly on the 1:6-HCGP+N particle than that on 1:6-GP+N, indicating the intensification effect of HC treatment on N-doping of the GP-based BDCM. In particular, the successful N-doping could change the electronic structure of carbon atoms in the GP-based BDCMs, and increase BDCMs’ electrical conductivity [62].

N₂ adsorption–desorption isotherms were conducted to depict the SSA and pore size distribution of the samples. As shown in Fig. 3 a – c and Fig. S2 a, the N₂ adsorption–desorption isotherms of the Raw carbon, 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP and 1:8-HCGP indicated these samples possessed many micropores, with corresponding SSAs of 102, 1455, 2049, 3019, 3272, 3116, 2798, 2848, and 3071 m²/g. All the GP-based BDCMs except the Raw carbon sample were categorized as type-IV sorption isotherms with obvious hysteresis loop, which revealed they all had mesoporous distributions [63]. The detailed pore size distributions of the GP-based BDCMs were displayed in Fig. 3 d – f and Fig. S2 b, all samples except for the Raw carbon sample presented dominant micropore distributions with diameters of about 2 nm. Notably, the pore size distributions of 1:8-GP and 1:8-HCGP samples showed broaden peaks in the range of 2 – 4 nm, indicating the increase of average pore size compared to other samples. The tendency of SSAs was a volcano type, with the 1:6-HCGP sample having the largest SSA. The 1:8-GP and 1:8-HCGP had more mesopores and less micropores compared with 1:6-GP and 1:6-HCGP samples, leading to the decrease of the SSAs. The nitrogen-doped samples’ SSAs were smaller than the undoped ones, this was due to the blocking effect of N-doping dots on the material, which was corresponding to the TEM results. All the HC-treated samples showed larger SSAs than the untreated ones. The HC effect forced K⁺ and OH^– into the pores on the samples and creates micropores. By the synergistic activation of HC and alkali, the SSA was additionally increased by 40.8 % (for the carbon to alkali mass ratio of 1:4), 8.4 % (for the carbon to alkali mass ratio of 1:6), and 7.8 % (for the carbon to alkali mass ratio of 1:8).

Fig. 3.

(a)-(c) N₂ adsorption–desorption isotherms and (d)-(f) pore size distributions of 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP, and 1:8-HCGP.

3.2. Chemical composition

Raman spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS) (Fig. 4) were conducted to analyze the samples’ chemical compositions. According to the Raman spectra (Fig. 4. a), the GP-based BDCMs had three prominent peaks near 1330, 1580, and 2750 cm⁻¹, representing D-band, G-band, and 2D-band, respectively. D-band represented the disordered vibration peak of BDCMs (e.g., defect or sample edge). In contrast, the G-band induced by the in-plane vibration of the hybrid carbon atom demonstrated high-order characteristics [27], [64]. The 2D-band, which was excited by a double-resonant Raman process, represented the layer of the samples [65]. The ratio of G-band intensity to D-band intensity (i.e., I_G/I_D) evaluated the graphitization degree of carbon. Generally, the commercial carbons have relatively low I_G/I_D value, as they have high graphitization degree. The I_G/I_D value of commercial carbons is ∼0.50 [66], while the I_G/I_D values of Raw carbon, 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP, and 1:8-HCGP were 0.96, 1.1, 1.24, 1.04, 1.14, 1.02, 1.08, 1.08, and 1.14, respectively, indicating all the GP-based BDCMs contained more defects and amorphous carbon than the commercial carbons. These defects and amorphous carbon offered active sites and accelerated charge transfer during the charging and discharging processes. Notably, the Raman spectra of the HC-treated samples were higher I_G/I_D values than those of the untreated ones, specifying that the microbubbles collapse from HC disrupted the GP-based BDCMs and generated more defects on the samples. A similar result was obtained from the XRD spectra of the representative samples (Fig. 4. b). Accordingly, they demonstrated a broad peak of 002 at 2θ = 21° and a flat peak at 2θ = 44°, which was an evidence of the existence of amorphous carbon. The 002 peaks of the N-doping samples 1:6-GP+N and 1:6-HCGP+N showed slight movement to the right side, demonstrating the successful doping of N into the samples, as the N atoms enlarged the interplanar distances of the amorphous carbon.

Fig. 4.

(a) Raman spectra of the samples, (b) XRD spectra of the representative samples (1:6-GP, 1:6-HCGP, 1:6-GP+N, and 1:6-HCGP+N), (c)-(j) C1s spectra, and (k)-(r) O1s spectra of 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP, and 1:8-HCGP samples.

XPS spectra of the samples were shown in Fig. 4. c – r. C1s peaks were deconvoluted into four prominent peaks at around 284.6, 285.1, 286.1, and 289.1 eV, assigned to C C, C—C, C O & C—N, and O C-O, respectively. This result manifested the co-existence of graphic sp2 C and C O & C—N species, representing the disordered structure of amorphous carbon, in the GP-based BDCMs. The result was consistent with those of the SEM and Raman spectra. Though the BDCM-based SCs’ electrochemical performance mainly originated from the electric double-layer capacitance, the C O & C—N functional groups on the GP-based BDCMs’ surface introduced pseudocapacitance [67], as they were conducive to the adsorption of ions in the charging of SCs [68]. The quantitative analysis of XPS spectra demonstrated that utilizing HC treatment led to the increasement of the functional groups’ proportions (1:4-GP: 21 %, 1:4-HCGP: 27 %, 1:6-GP: 20 %, 1:6-HCGP: 29 %, 1:6-GP+N: 24 %, 1:6-HCGP+N: 31 %, 1:8-GP: 17 %, and 1:8-HCGP: 25 %), and thus improving the performance of the HCGPs as SC materials. Moreover, the contents of hydrophilic oxygen-containing functional groups (O C-O, C O, etc.) also increased from 19 % to 23 % for 1:4-HCGP, 16 % to 21 % for 1:6-HCGP, 17 % to 22 % for 1:6-HCGP+N, and 30 % to 42 % for 1:8-HCGP, respectively, corresponding to the O1s spectra result. These hydrophilic functional groups could improve the hydrophilicity of the GP-based BDCMs, increase the contact surface of GP-based BDCMs and electrolytes, reduce free macromolecules of water and oxygen, enhance the wettability, and, consequently, develop the electrochemical performance of GP-based BDCMs. The formation of these functional groups could be attributed to the unique environment with the strong oxidation (hydroxyl radicals)/reduction (hydrogen radicals) effect created by the HC treatment. To study the HC effect on N-doping, the high-resolution spectra of N for the 1:6-GP, 1:6-HCGP, 1:6-GP+N, and 1:6-HCGP+N were measured. Accordingly, the N1s characteristic peaks (Fig. S3) were assigned to three types: pyridinic-N (N-6) at around 399 eV, pyrrolic-N (N-5) at around 400 eV, and other species of N (N-X, including oxide-N and carbon-N) at around 403 eV. N-5 and N-6 were reported to be conducive to the enhancement of the carbon materials’ conductivity [69], thus improving the corresponding SC’s electrochemical performance [70]. The proportions of the N-5 and N-6 species in the 1:6-GP,1:6-HCGP, 1:6-GP+N, and 1:6-HCGP+N samples were calculated to be 0.7 %, 0.6 %, 1.0 %, and 1.5 %, respectively, demonstrating the intensification effect of HC treatment during N-doping.

3.3. Electrochemical performance

The electrochemical performance tests of the BDCM-based SCs were conducted at room temperature (25 °C) in a two-electrode system with 6.0 M KOH as the electrolyte. The results are shown in Fig. 5.

Fig. 5.

Electrochemical performance tests of SCs with 6 M KOH as the electrolyte. (a) GCD curves (at a current density of 1 A/g), (b) CV curves, (c) specific capacity at various current densities, (d) Nyquist plots, (e) Ragone plots, and (f) cyclic stability of the BDCM-based SCs.

The galvanostatic charge–discharge (GCD) tests were conducted under a current density of 1 A/g. Fig. 5 a presented the GCD curves of the SCs with various GP-based BDCMs. The GCD curves exhibited highly repeatable cyclic behavior for all the samples. And their symmetrical triangular shapes suggested the SCs’ high-rate capacitive performance. The GCD times of the HC-treated samples were 20 % longer than those of the untreated ones in average, indicating that the porous morphologies led by the HC effect benefit the ions’ fast transportation during the charging and discharging processes.

The cyclic voltammetry (CV) curves of the GPs-based SCs at the scan rate of 50 mV/s were shown in Fig. 5 b. All the curves, except for those of the 1:6-GP+N and 1:6-HCGP+N, exhibited an approximately rectangular shape, and the SCs’ currents stabilized when the external circuit was reversed, indicating that pseudocapacitance did not exist. But for the 1:6-GP+N and 1:6-HCGP+N samples, the CV curves shape changed a little, as there appealed a faint peak during the forward scan, indicating the introducing of pesudocapacitance by the N-doping dots. This was consistent with the results obtained from the GCD curves. The HC-treated samples had higher current density compared with the untreated ones at the same voltage. It was due to the reason that the HC-treated BDCMs had a great amount of micropores with highly disordered structure, creating ideal channels for ion migration.

Fig. 5 c showed the specific capacitance of the BDCM-based SCs at the current densities ranging from 0.1 to 20 A/g. The specific capacitances of the Raw carbon, 1:4-GP, 1:4-HCGP, 1:6-GP, 1:6-HCGP, 1:6-GP+N, 1:6-HCGP+N, 1:8-GP, and 1:8-HCGP based SCs at a current density of 10 A/g were 23, 141, 155, 153, 199, 212, 227, 156, 191 F/g, respectively. All the HC-treated samples-based SCs preserved higher specific capacitance than those with the untreated ones, and this result originated from the modification in the morphology and heteroatomic N-doping, which was in good agreement with the morphology and structure characterization results.

Fig. 5 d showed the Nyquist curve of the BDCM-based SCs. The Nyquist curve of an SC under ideal conditions was supposed to be a vertical straight line. In the real world, there must be an internal resistance between the electrode and the electrolyte, the semi-circular curve in the high-frequency region represents the equivalent series resistance (ESR) of the SC materials. Accordingly, the relatively low ESRs were achieved by the HC-treated samples (1:4-HCGP, 1:6-HCGP, 1:6-HCGP+N, and 1:8-HCGP) based SCs, i.e., 0.35, 0.81, 0.58, 0.48 Ω, respectively. All the HC-treated samples’ curves in the low-frequency region showed a linear trend, indicating good electric double-layer properties. On the contrary, the curves of the untreated samples-based SCs presented an obvious bending trend with a high radius, representing a large diffusion resistance and low ions transportation characteristic.

Further, the ratio of energy density and power density was analyzed to evaluate the BDCM-based SCs’ energy-storage performance. Fig. 5 e showed the Ragone plots for the BDCM-based SCs. The 1:6-HCGP+N-based SC had the best energy-storage performance, as it ranked the highest energy density, i.e., 10.19 Wh/kg, achieving at the power density of 250 W/kg (1: 6-HCGP+N-based SC). Finally, the long-term cycling performance of the GPs-based SCs was tested with 5000 charging-discharging cycles (Fig. 5 f). The specific capacitances of all the BDCM-based SCs were retained at over 96 %. It was evident that the energy density and power density values of all the HC-treated samples-based SCs were higher and comparable to those of the untreated ones, combining their long-term stability, the HC-treated samples-based SCs are very good candidates for potential applications.

3.4. Mechanism discussions

The HC-treatment during the impregnation process before activation is a key step to improve the GP-based BDCMs’ electrochemical performance, and its intensification effect mainly reflected in the following aspects: 1) the impregnation duration is greatly shortened. 2)The bubble collapse caused by the HC effect leads to extremely high temperature and pressure, which are beneficial to create defects in the GPs. 3)The defects on the GPs can absorb more KOH molecules. 4) The impurities in the GPs are washed away, leading to the exposure of complete materials. Fig. 6 shows the HC-intensified GP-based BDCMs synthesis process on a micro-scale. The mechanism underlying the reaction can be attributed to two aspects: on one hand, the extraordinary radicals (i.e., hydroxyl and hydrogen free radicals) cause significant damage to the GPs and tear them into disordered amorphous carbon with more micropores that possess higher charge-transfer ability. On the other hand, the energy wave transmitted through the HC effect can accelerate the K⁺ and OH^– ions transportation and intensify the impregnation process so that KOH can play a better role in the high-temperature activation process.

Fig. 6.

Hydrodynamic cavitation intensifies the GP-based BDCMs synthesis on a micro-scale.

4. Conclusions and outlooks

In the present study, HC was successfully utilized to intensify the alkali activation of GPs for SCs, for the first time. Six GP-based BDCMs were synthesized under different HC treatment conditions and then compared from microscopical characteristics to electrochemical performance as SCs materials. The morphology study demonstrated that the HC treatment had created many defects and amorphous carbon structures on the GP-based BDCMs, with the highest SSA reaching 3272 m²/g (1:6-HCGP), which 32 folded that of the Raw carbon sample’s. The HC treatment also intensified the N-doping process. XRD and XPS results manifested that the N content had been increased and consequently changed the electronic structure of the carbon atoms, leading to the increase of specific capacitance (1:6-HCGP+N-based SC, 227 F/g at 10 A/g). This study suggests that the activation of BDCMs in a hydrodynamic cavitation reactor is a highly effective technology for fabricating SCs materials, which can be potentially integrated into SCs and other energy-related materials synthesis strategies.

CRediT authorship contribution statement

Xiaoxu Xuan: Investigation, Resources, Writing – original draft, Writing – review & editing, Conceptualization, Supervision. Mengjie Wang: Resources. Weibin You: Resources. Sivakumar Manickam: Writing – review & editing. Yang Tao: Writing – review & editing. Joon Yong Yoon: Writing – review & editing. Xun Sun: Writing – original draft, Writing – review & editing, Conceptualization, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

The data that has been used is confidential.