Ultrasonic liquid exfoliation for producing graphene materials from rice stem: Investigating cellular components and functionalities

Highlights

• •Prospective approach for biomass-derived graphene production.
• •Roles of liquid components within the solvent system.
• •Water content enhances cavitation effects induced by ultrasound.
• •Unique porous structure of GO inherited from cell structure.
• •GO with a few-layer structure and intact properties derived from biomass.

Abstract

This study investigates a prospective and straightforward method for producing graphene material derived from biomass, examining the influence of plant cell composition and functions. The experimental outcomes highlight ultrasound's crucial role in synthesizing graphene material sourced from biomass. Ultrasound, a pivotal element in the experiment, significantly affects graphene production from biomass by working synergistically with the liquid components in the solvent system. Notably, the ethanol content reduces the solution’s surface tension, facilitating the effective dispersion of biochar and graphene oxide sheets throughout the process. Simultaneously, the water content maintains the solution’s polarity, enhancing the cavitation effect induced by ultrasound. Biomass-derived graphene is exfoliated utilizing an ultrasonic bath system (134.4 W, 40 kHz, 0.5 W/cm2) from biochar. The as-synthesized graphene oxide exhibits a structure comprising a few layers while remaining intact, featuring abundant functional groups. Interestingly, the resulting product displays nanopores with an approximate diameter of 100 nm. These nanopores are attributed to preserving specific cell structures, particularly those with specialized cell wall structures or secondary metabolite deposits from biomass resources. The study's findings shed light on the impact of cellular structure on synthesizing graphene material sourced from biomass, emphasizing the potential application of ultrasound as a promising approach in graphene production.

1. Introduction

Graphene material, a two-dimensional nanomaterial, presents exceptional properties across multiple domains, encompassing electrical, chemical, and thermal properties [1], [2]. Its distinctive qualities open the door to extensive applications in diverse fields such as biomedicine, advanced catalysts, and electronics [3]. However, the synthesis of graphene continues to encounter challenges, particularly in sustainability, stability, quality, and cost. Pristine graphite has gained popularity as a raw material in conventional synthesis approaches due to its well-ordered layered structure. However, the diminishing availability of graphite resources due to gradual depletion and mining limitations has underscored the growing importance of seeking renewable and environmentally friendly alternatives for both raw materials and production methodologies [4], [5], [6]. Among various alternative options, biomass stands out with its strong competitiveness, attributed to its abundant reserves and renewable nature [7], [8], [9], [10], [11]. Given that biomass, in contrast to traditional graphite, represents an innovative and relatively unexplored reactant, delving into its inherent properties becomes crucial for understanding its influence on subsequent reactions. See Table 1..

Table 1.

Experimental conditions for liquid exfoliation in both the experimental group (with ultrasonication) and the control group (with mechanical agitation).

	Ultrasonication	Mechanical Agitation	Ethanol Concentration (v/v%)
Experimental Group	50 °C, 6 h	Not applied	0
	50 °C, 6 h	Not applied	20
	50 °C, 6 h	Not applied	40
	50 °C, 6 h	Not applied	60
	50 °C, 6 h	Not applied	80
	50 °C, 6 h	Not applied	100
Control Group	Not applied	50 °C, 6 h	0
	Not applied	50 °C, 6 h	60

Graphene materials obtained from biomass resources exhibit distinctive characteristics compared to those synthesized from pristine graphite. Graphene derived from biomass exhibits a higher degree of defects and a greater presence of impurities, which can be attributed to the diverse organic and inorganic compounds inherent in biomass [12]. Simultaneously, certain cellular structures that contribute to plants' mechanical strength and protective functions also impact biochar formation, the precursor carbon material. The thermal conversion process from biomass to biochar involves the degradation of complex molecules found within cells, such as cellulose and lignin, leading to structural changes in the resulting biochar. Consequently, graphene derived from biochar displays modified properties compared to conventional graphene materials.

Recent research has introduced various methods for producing graphene from biomass, employing different methods. One method involves the conversion of corn stover into reduced graphene oxide through continuous stirring in a concentrated sulfuric acid solution. The resulting product serves as an effective electromagnetic wave absorber [13]. Another approach focuses on synthesizing porous graphene material from loofah using potassium ferrate activation and graphitization processes, making it a valuable component in efficient electrode materials [14]. Furthermore, graphene material is derived from miscanthus through mechanical shearing within different solvent systems, such as N-methyl-s-pyrrolidone and detergents [15]. Similarly, some methods have incorporated ultrasound as part of the synthesis pathway for graphene materials. For instance, few-layer graphene has been successfully exfoliated from pristine graphite in organic solvents by adding an alkali aqueous solution, achieved through a nozzle-type high-pressure homogenizer [16]. Another approach integrates ultrasound treatment with Hummer's modified method, using pristine graphite to synthesize larger-sized graphene materials [17]. These methods have all produced graphene products with a certain degree of stability, indicating reduced susceptibility to aggregation. However, these production methods often involve elevated temperatures, harsh chemicals, prolonged treatment durations, and the generation of pollutive waste. Furthermore, many synthesis methods involving ultrasound are still in a stage where relatively uniform pristine graphite is used. Consequently, there is a growing need to advance greener, cost-effective, and sustainable approaches to graphene production.

In this study, our primary objective is to produce a few-layered graphene material from rice stems using a sustainable methodology. Rice stems were selected as the carbon source for graphene synthesis due to their easy accessibility, abundant availability, economic viability, and sustainable nature as agricultural waste. The exclusive transformation of biochar into graphene is achieved through ultrasonication in an ethanol–water solution, a method commonly employed for dispersing graphene solutions. In contrast to conventional graphene synthesis utilizing graphite as the raw material, our study utilizes ultrasound as the primary synthesis mechanism to convert biomass into graphene material. Notably, our approach involves using fewer harsh chemicals, potentially reducing it by approximately 10.27 % compared to other methods [17]. Our study highlights the influence of biomass's cellular structure and organic composition on graphene properties and emphasizes the significant contribution of the ethanol solvent to ultrasonication. Using rice stems as a carbon precursor, and our study presents a promising and eco-friendly approach to producing graphene material. Furthermore, our findings expand the potential applications of biomass-derived graphene material in bioenergy, advanced materials, and biomedical applications.

2. Material and methods

2.1. Preparation of biochar

The rice stem samples were obtained from Lianyungang in Jiangsu Province, China. Before size reduction, the rice stems were dried at 60 °C for 48 h. Subsequently, the dried stems underwent milling to achieve particles with a diameter of less than 0.224 mm using a Retsch ZM200 mill. These milled rice stem particles were then subjected to pyrolysis using a vertical tube furnace (OFT-1200X) under ambient conditions, gradually raising the temperature to reach a final pyrolysis temperature of 1000 °C. The residence time at this pyrolysis temperature was maintained at 1 h. Throughout the entire pyrolysis process, including the cooling phase, a continuous nitrogen flow (99.99 % N₂) was supplied at 120 mL/min to ensure a non-oxidizing atmosphere.

2.2. Synthesis of graphene oxide from biochar

The biochar was subjected to ultrasonication using an ultrasound system (CR-040S). The ultrasonic reactor, configured as a bath with 4 transducers, operated at an output power of 134.4 W, a frequency of 40 kHz, an acoustic density of 0.5 W/cm², and a temperature control set at 50 °C. Instead of conventional solvents like ortho-dichlorobenzene, an alcohol solution was chosen as the reaction medium for its milder and more environmentally friendly impact. The experiment was divided into two groups: one utilizing ultrasonication and the other serving as a reference using mechanical agitation. Each experiment involved mixing 0.05 g of biochar with a varying concentration of ethanol and water. The mixture in the experimental group was sonicated at 50 °C for 6 h [18]. As a control group, an additional set mixed 0.05 g of biochar with a 60 v/v% alcohol and water solution. Under the same temperature conditions, the mixture in the control group was stirred at a rotational speed of 1000 rpm using a magnetic stirrer (LC-MSB-HD) for the same duration. The resulting liquid products were centrifugated at 7500 rpm for 45 min, with the supernatant collected as the final product. The entire experimental process, from biochar synthesis to the ultimate acquisition of the product, is illustrated in Fig. 1.

Fig. 1.

Synthesis process of graphene product from pyrolysis of biomass.

2.3. Characterization of graphene material

X-ray Photoelectron Spectroscopy (XPS) and Fourier Transform Infrared Spectroscopy (FTIR) assessed the carbon bonding arrangements and functional groups within the graphene products. Furthermore, Raman spectroscopy was employed to analyse the graphene products' chemical structure. The samples were placed onto a Si/SiO₂ substrate, and measurements were conducted within a range of 200 to 4000 cm⁻¹ at room temperature. The characteristic peak and intensity data were acquired using a 514 nm Ar laser excitation with a power output of 0.2 mW. Brunauer-Emmett-Teller (BET) analysis was utilised to examine the surface area and porosity of GO product synthesized from both mechanical and ultrasonication process. The samples were firstly degassed at 150 °C for 12 h, followed by adsorption and desorption of nitrogen.

The morphology and texture characteristics of the graphene material were examined using a high-resolution transmission electron microscope (HRTEM, FEI TF20-Super-X) operating at an accelerating voltage of 200 kV. After subjecting the samples to sonication for 5 min, a single drop was deposited onto a carbon-coated copper grid. Subsequently, the grid was allowed to dry at room temperature before analysis. For quantifying the morphology of the graphene materials, including the number of layers and the distribution of pores on the surface, an atomic force microscope (AFM, Bruker Dimension Icon) was employed. To this end, a water suspension (approximately 0.01 mL) was placed onto a mica sheet and dried under ambient conditions before undergoing analysis.

3. Results and discussion

3.1. Characterization of GO sheets

3.1.1. Spectroscopic characterisation of GO

In order to explore variations in ultrasonic energy parameters and their influence on the characteristics of graphene oxide products, a diverse range of characterization techniques was employed. FTIR and XPS techniques were employed to analyze the chemical bond characteristics of the GO sheets, as depicted in Fig. 2. In Fig. 2(a), the FTIR spectrum of the GO product is presented, revealing distinct peaks. A significant peak around 3454 cm⁻¹ is attributed to the O-H stretching vibrations, indicating the presence of hydroxyl groups on the GO surface [19]. Additionally, a peak at 1643 cm⁻¹ indicates the presence of conjugated and cyclic alkenes, suggesting notable polycyclic aromatic structures in the product [20]. Moreover, peaks at 1380 cm⁻¹, 1060 cm⁻¹, and 712 cm⁻¹ correspond to vibrations associated with various functional groups, including O-H bending, C-O stretching, aromatic C-H bending, and potentially Si-phenyl vibrations. Additionally, a weak yet broad peak related to silicon is observed around 2080 cm⁻¹, indicating the presence of Si-H bonds. This peak, along with the previously indicated one, suggests the retention of silicon within the silica cell through bonding with the carbon framework during the pyrolysis process.

Fig. 2.

Chemical bonding analysis of the GO product: (a) Identification of functional groups in the GO product through FTIR (b) Examination of carbon atom bonding characteristics by XPS.

XPS was utilized with FTIR to further confirm the bonding characteristics, with a specific emphasis on the carbon atoms. The XPS spectra associated with carbon atoms, post-peak positioning, and fitting are illustrated in Fig. 2(b). Five characteristic peaks are identifiable, positioned around 285.38 eV, 285.4 eV, 289.5 eV, 293.4 eV, and 296.18 eV. Among these peaks, the ones situated at 285.38 eV and 285.4 eV are ascribed to C = C and C-OR bonding structures, respectively [21]. Notably, the peak at 285.38 eV displays notable intensity, indicating a relatively higher content of C = C bonds. Additionally, the peak at 289.5 eV corresponds to C = O bonding, signifying the presence of carbonyl groups in the GO product. Collectively, these three peaks suggest that the GO product exhibits a comparatively complete conjugated structure of graphene sheets coupled with oxygen-containing functional groups at the edges. The peak at 293.4 eV signifies the presence of –OH structures on the aromatic rings of the GO product. The findings from XPS, complemented by the FTIR analysis, reveal that biochar undergoes a layer-by-layer exfoliation during the ultrasonication process, producing low-layered, structurally intact GO sheets with oxygen-containing functional groups.

BET analysis was carried out to examine the variations in specific surface area and pore distribution of GO products synthesized through different methods. Fig. 3 illustrates the adsorption and desorption isotherms of samples synthesized via mechanical agitation and ultrasonication, along with their respective pore size distributions. As suggested by the International Union of Pure and Applied Chemistry (IUPAC) classification, the isotherms obtained for these samples belong to Type IV with nearly Type H4 hysteresis loops [24]. This characteristic is attributed to the complex internal network of GO products and the preserved cellular structure derived from the rice stem. GO synthesized through ultrasonication exhibits a more typical H4 hysteresis loop compared to that synthesized via mechanical agitation. Consequently, the GO product obtained through ultrasonication tends to possess slit-shaped narrow pores, resulting in a larger specific surface area compared to that obtained via the mechanical method, approximately 283 m²/g. Meanwhile, the specific surface area exceeds that of biomass-derived GO material synthesized via different methods with similar biomass treatment conditions [15], [25]. The larger specific surface area highlights the superiority of ultrasound over mechanical agitation under equivalent processing conditions. Benefiting from the cavitation effect generated by high-frequency vibrations, the ultrasonic environment is more conducive to breaking down the structure of biochar and facilitating the exfoliation of layers, leading to a significant enhancement in the specific surface area.

Fig. 3.

Nitrogen adsorption and desorption isotherms and pore size distribution of GO product from biochar samples: (a) isotherms of GO synthesized via mechanical agitation, (b) pore size distribution of GO synthesized via mechanical agitation, (c) isotherms of GO synthesized via ultrasonication and (d) pore size distribution of GO synthesized via ultrasonication.

Simultaneously, the GO product prepared through ultrasonication exhibits a more concentrated pore size distribution, primarily comprising mesopores, while the mechanical method-derived GO product shows a limited presence of macropores. This is attributed to the high-frequency vibrations induced by ultrasonic waves, which lead to the formation of tiny bubbles in the solution. The collapse of these bubbles generates localized high temperature and pressure, conditions conducive to the formation of finer and more even pore structures within the material. Therefore, during the ultrasonic process, the effective energy of ultrasonication, utilized for the liquid exfoliation of biochar, will significantly influence the exfoliation efficiency of biochar and the properties of the resulting GO product.

Raman spectroscopy was employed to analyze the crystal structure and chemical composition of the synthesized GO. Fig. 4 illustrates the Raman spectrum, with Fig. 4(a) representing the spectrum with the most typical graphene features, while Fig. 4(b) offers a comparative illustration of spectra for all experimental groups. In most samples, four distinct peaks are discerned: the D band around 1355 cm⁻¹, the G band at approximately 1587 cm⁻¹, the 2D band around 1670 cm⁻¹, and the D + G band at approximately 2976 cm⁻¹. These peaks align with characteristic peaks of GO, indicating the presence of typical GO characteristics in the synthesized product. The D band indicates defects within the GO material, while the G band represents the vibration of sp² hybridized carbon [22]. Both the D and G bands exhibit significant intensities and smooth profiles, indicative of substantial defects resulting from oxygen-containing functional groups and distinctive lattice characteristics of GO. The calculated I_D/I_G ratio for the GO product is 0.97, surpassing values derived from GO synthesized through conventional chemical oxidation methods involving a mixture of potassium permanganate, sulfuric acid, and sodium nitrate solution [23]. This difference suggests that ultrasound-induced liquid exfoliation in the ethanol–water mixture leads to a higher GO yield with increased oxygen-containing functional groups. Additionally, the intensity and shape of the 2D band aid in estimating the number of layers present in the products. The 2D peak points toward the formation of monolayer GO. However, as the peak appears broad and less distinct than the G band, it suggests that the product is a mixture containing graphene with varying layers.

Fig. 4.

Raman spectra of GO sheets synthesized via ultrasonication: (a) graphene oxide synthesized with 60% ethanol solution, and (b) product synthesized with various ethanol concentrations.

A comparative analysis was conducted using Raman spectra to further analyze the influence of varying ethanol concentrations and the application of ultrasonication on the quality of graphene oxide products. All samples, across different concentrations, exhibit distinctive D and G bands within the characteristic peak region of graphene oxide. The observed D and G bands display prominent peak intensities and smooth profiles, indicating the successful ultrasonic exfoliation of graphene oxide with structurally intact features and oxygenated functional groups across all concentration levels. However, a noticeable phenomenon is observed in the 2D band and D + G band of groups with 40 % and 100 % ethanol concentrations, where a merging effect is evident. The convergence of these two peaks suggests a higher graphene layer count or the presence of structural defects. Changes in the profile and positioning of the 2D peak are attributed to an increased thickness of the GO sheet, potentially leading to overlap or concealment within the D + G peak. These disparities are related to the energy distribution of the ultrasonic system used in the reaction. When adjusting the ethanol concentration in the liquid medium of the reaction system, the energy allocated for the liquid exfoliation reaction varies. Depending on the reactor parameters and transducer power, the power used for the reaction and heating is illustrated in Fig. 5. As the ethanol concentration gradually increases, the energy available for exfoliation progressively rises, with a more pronounced enhancement observed when the concentration is below 80 %. However, concurrently, as the ethanol concentration gradually increases, the dielectric polarization capability of the solution weakens. Consequently, even though more energy is input into the system for liquid exfoliation, with the majority utilized to initiate cavitation effects, there is an ineffective separation of biochar into lower-layer graphene oxide products. Considering the comparative results in Fig. 4(b), the optimal concentration for ethanol appears to be 60 %, given both effective energy input and the dielectric polarization capability of the solution.

Fig. 5.

The energy distribution of the ultrasonication system during liquid exfoliation.

Correspondingly, the products derived from the control group experiment, comprising four distinct sets, were subjected to Raman analysis. This comparative analysis aimed to discern the influence of introducing ultrasonication on synthesizing graphene materials. The results are illustrated in Fig. 6. When maintaining consistent alcohol concentrations, incorporating ultrasonication in liquid exfoliation proves more effective than mechanical agitation. This is evident in the spectra, where the outcomes lean towards a more pronounced differentiation of sharper 2D and D + G bands, as demonstrated in the comparison between Ultra 0 % and ME 0 %. However, the spectra of these control group products exhibit varying degrees of merging in both the 2D band and the D + G band. As a result, all these products display more layers, indicating that liquid exfoliation may not be highly effective. To better understand the number of graphene layers in the synthesized material, AFM combined with TEM was employed for microscopic characterization.

Fig. 6.

Raman spectra of products synthesized through mechanical agitation and ultrasonication (ME: mechanical agitation, and Ultra: ultrasonication)Microscopic Characterisation of GO.

The HRTEM image of the graphene material obtained through ultrasonication from biochar reveals a highly-ordered carbon structure exhibiting relatively high transparency. The transparency suggests a low number of layers within the graphene flakes, as shown in Fig. 7 (a). Notably, the edges of these flakes display sharp and curled features, characteristic of the synthesized GO material's low thickness and exceptional flexibility. Additionally, the presence of wrinkle fringes in the image allows for the estimation of the interlayer spacing of GO [26]. The calculated interlayer spacing of the material ranges from 0.34 to 0.41 nm, aligning well with the anticipated range for graphene materials [27]. In contrast to the graphene oxide synthesized through ultrasonication, the product generated solely by mechanical agitation exhibited lower transparency. It lacked the characteristic curled morphology at the edges, as depicted in Fig. 7 (c). The edge morphology was sharper, revealing a multi-layer stacking structure. This observation indicates that the exfoliation effect of simple mechanical agitation is less efficient compared to liquid-phase exfoliation via ultrasonication. The resulting product tends to stack, resembling a multi-layer biochar rather than forming a few-layer graphene material.

Fig. 7.

HRTEM analysis of morphological characteristics and interlayer spacing in graphene material: (a) graphene oxide obtained with a 60% ethanol solution via ultrasonication, (b) interlayer spacing in graphene oxide, and (c) product obtained via mechanical agitation.

The HRTEM analysis determines the interlayer spacing, while AFM provides insights into the graphene material's morphology, size, and thickness. In Fig. 8(a) and (b), the morphological features of GO sheets exfoliated from biochar are presented. Notably, the surface of the synthesized GO exhibits pores that traverse through all layers, with sizes predominantly falling below 450 nm, as illustrated in Fig. 8. The presence and characteristics of these pores can be traced back to the original cellular structure of the source material and its behaviour during pyrolysis. By examining the radial section of the rice stem, starting from the outer epidermis and moving inward, distinct tissues emerge, including the epidermis, cortex, parenchymatous tissue, and vascular bundles. These components significantly form the observed pore structures within the synthesized GO material. The outermost layer of the rice stem, known as the epidermis, is a protective barrier and consists of a single layer of tightly packed cells. Beneath the epidermis lies the cortex, a tissue comprising parenchymatous cells, transfer cells, and sclereids. The diverse composition allows the cortex to fulfill structural and storage functions. Further within the rice stem, vascular bundles are dispersed within the parenchymatous tissue. These vascular bundles are crucial to conducting water, minerals, and other essential substances. Concurrently, the parenchymatous tissue itself contributes significantly to storage and transportation processes. Specific specialized cells exhibit remarkable mechanical strength within these essential tissues compared to other cell types, such as parenchymatous cells and companion cells. This enhanced strength is attributed to their distinctive cell wall structures or the deposition of secondary metabolites. In addition to these unique characteristics, the rice stem's internal components contribute to its overall mechanical integrity and functionality. Silica cells, stomata guard cells, cork cells found in the epidermis, sclereids, transfer cells in the cortex, and sclerenchyma fibers and xylogen in the vascular bundles exemplify these specialized cell types. Silica deposition and suberin in silica cells and cork cells contribute significantly to their resilience against environmental changes. Additionally, the thickened cell walls rich in lignin content play a crucial role in fortifying the stability of these cells. Compared to other cellular components such as cellulose and protein, these constituents that offer mechanical support and protective functions also exhibit higher thermal stability, even during pyrolysis. This attribute ensures the preservation of these cells following high-temperature pyrolysis of rice stems.

Fig. 8.

Surface morphology of the synthesized graphene oxide at different sizes (a) and (b); (c) Surface pore size distribution, and (d) layer distribution of GO products.

Beyond their functions in protection and support, these cells also hold a significant role in material transportation, consequently influencing their cell walls' development. Specialized pore structures have evolved on their longitudinal cell walls to facilitate the exchange and transport of substances, including structures like bordered and simple pits. The diameter of these pore structures varies depending on the plant’s developmental stage and growth environment, usually falling within the range of a few micrometers. When subjected to pyrolysis, these pores undergo structural changes alongside the cell. During thermal conversion, cells experience varying degrees of size reduction in tangential, radial, and longitudinal directions, with reduction rates spanning from 4 % to 50 % [28]. Tangential and longitudinal contractions predominantly govern the decrease in size of these pore structures due to their location on the longitudinal cell wall. The longitudinal contraction of the cell primarily results from the breaking of longitudinally arranged cellulose chains within the secondary cell wall's second layer. Meanwhile, the reduction in the tangential axis primarily stems from the release of lignin and hemicellulose through devolatilization. Based upon the inherent dimensions of biomass pore structures and the simultaneous contraction of cells both longitudinally and tangentially, it is estimated that the pores identified in this study originated from pit structures with a size under 500 nm, preferentially between 200 nm and 500 nm. The size distribution of pores on GO, as shown in Fig. 8(c), closely corresponds to the predicted value derived from cellular information. This similarity suggests that these pores are not a consequence of volatile or gas release during pyrolysis or newly formed through cavitation effects during ultrasound treatment. Instead, these pores originate from the inherent cell wall structure of cells known for their high thermal stability, including silica cells, stomata guard cells, cork cells in the epidermis, sclereids and transfer cells in the cortex, as well as sclerenchyma fibers and xylogen in the vascular bundles.

Apart from the morphological characteristics of GO sheets, determining the number of layers in GO was accomplished by combining overall thickness measurements from AFM with the interlayer spacing of the graphene material assessed via HRTEM. The results are depicted in Fig. 8(d). At an ethanol concentration of 20 %, the total yield of GO product is relatively lower compared to other concentrations under the same conditions. Specifically, the yield of GO, encompassing all products with a GO structure, is approximately 66.67 % lower in relative terms. As the ethanol concentration gradually increases, the layer distribution of the produced graphene shows a more pronounced trend towards concentration. Notably, most GO sheets obtained with a 60 % ethanol concentration exhibit a thickness of less than three layers, highlighting the advantageous use of ultrasound for liquid exfoliation in an ethanol–water solution. Compared to groups with different ethanol concentrations, this group yields GO products with less than three layers at the highest rate, reaching 90.24 %. This can be attributed to the introduction of ethanol and the consequent development of a relatively uniform layered structure of biochar at elevated pyrolysis temperatures. Incorporating ethanol into water effectively reduces the surface tension within the liquid system, promoting optimal biochar precursor dispersion during the reaction's initial stages and preventing re-agglomeration of GO sheets following the exfoliation process. Notably, with an increase in alcohol concentration, the exfoliation effect is gradually weakening. This is evidenced by a decrease in the yield of products with three layers or fewer, reaching 84.21 %, of which 56.25 % represent products with precisely three layers. This weakening effect is attributed to the water-based solvent system's role in preserving the solution's polarity, creating an environment conducive to cavitation. This effect enhances the generation of temperature and pressure differentials, thus enhancing the energy utilization efficiency inherent in ultrasound during exfoliation.

3.2. Effect of ultrasonication and ethanol-water system in liquid exfoliation of biochar

Upon comparing the outcomes of different alcohol concentrations and various liquid exfoliation techniques, it becomes evident that the introduction of ultrasound and the choice of the solvent system play a crucial role in the liquid exfoliation process. Employing common ultrasonic frequency conditions typically used in treating graphene materials, introducing ultrasonic vibration into the biochar-solvent system initiates the cavitation effect due to the introduction of strong energy [18]. This effect leads to gas release within the naturally occurring pores of the biochar, causing the fragmentation and subsequent re-dispersion of biochar particles within the liquid phase. Consequently, the surface area conducive to exfoliation is increased, enhancing reaction efficiency and the overall quality of the produced graphene material. Moreover, ultrasonic vibration generates numerous vacuoles and small bubbles [29], [30], [31], [32]. As these cavitation bubbles collapse, they create high-pressure and high-temperature conditions, increasing the contact area with the liquid phase due to the abrupt pressure change. This promotes the exfoliation of biochar. The alteration in energy and the accompanying rise in temperature induced by ultrasonic vibration provide sufficient energy to break the bond connections and inter-layer interactions between biochar sheets. This phenomenon assists in exfoliating graphene material from the biochar structure.

In addition to the energy introduced by ultrasonic vibration, the choice of solvent plays a crucial role in the exfoliation of biochar within the liquid medium. Centripetal intermolecular forces govern molecular interactions at the liquid's surface, causing surface molecules to experience more force than interior ones. In a pure water system, this leads to the clustering of molecules on the surface. Ethanol, in comparison to water, exhibits lower molecular polarity due to differences in their molecular structures [33]. The electron surrounding the oxygen atom in an ethanol molecule is directed towards the carbonyl group, leading to its molecular polarity. However, compared to water, the hydroxyl group in the ethanol molecule has notably lower electronegativity than the oxygen atom in the water molecule, resulting in a lower molecular polarity for ethanol than for water. When ethanol is dissolved in water, a layer of ethanol molecules forms among water molecules due to the weaker interaction forces within the ethanol–water system than those in a pure water solution. This arrangement of dispersing ethanol molecules imparts resistance and dispersion effects on biochar particles. As a result, electrostatic repulsion between particles within the liquid is established, rendering them relatively stable within the dispersing system during and after the exfoliation process. Consequently, no additional energy is required during exfoliation to re-disperse the aggregated biochar particles.

Although a high ethanol concentration is beneficial for reducing surface tension and improving dispersion, its low dielectric polarisation capacity weakens its exfoliation performance. The lower dielectric constant of ethanol indicates its reduced potential for polarisation under an external electric field. Ethanol cannot store a substantial charge, making it challenging to develop an intense polarisation field [34]. However, under the influence of ultrasound, the dielectric polarisation of liquids becomes significant. Ultrasound induces various interactions within the medium, such as intermolecular interactions, friction, compression, deformation, and the creation of repulsion and densely populated regions. This process accelerates the generation of cavitation. Dielectric polarisation plays a crucial role in initiating cavitation effects. When ultrasound propagates through the liquid medium, the energy from the sound waves transforms into vibration and heat, instigating polarisation among liquid molecules and facilitating interactions within the medium. A higher dielectric polarisation capacity indicates a higher concentration of polarisable charges within the liquid, evenly distributed. Consequently, a more potent cavitation effect is generated. In contrast, weak dielectric polarisation results lead to comparatively weaker cavitation effects in the liquid, thereby complicating the efficient exfoliation of particles within the liquid phase. Most of the products exhibited a thickness of less than five layers, signifying a highly successful exfoliation process facilitated by ultrasound. This can be attributed to adding an ethanol solution and developing a uniform layered biochar structure at elevated pyrolysis temperatures [35]. The precisely introduction of ethanol into the aqueous medium reduces the interfacial tension within the solution system, promoting optimal dispersion of the biochar precursor throughout the solution during the initial reaction stages. This measure efficiently prevents subsequent re-agglomeration of the resulting GO sheets. Furthermore, the controlled addition of ethanol maintains the polarity of the solution system, creating an effective and favorable environment for the cavitation effect. This results in a more prominent generation of temperature and pressure variations, thereby enhancing the efficiency of harnessing energy from ultrasound during the exfoliation process.

4. Conclusion

This study demonstrates the fabrication of graphene material by liquid exfoliation of rice stem-derived biochar in an ethanol–water solvent system using ultrasonication. The results of this study clearly demonstrate that the final product exhibits a relatively intact graphene structure with a limited number of layers. Through the incorporation of ethanol, the surface tension within the liquid system can be reduced, thus enhancing the efficacy of the exfoliation process. It is through this inclusion that the biochar precursor is dispersed optimally during the reaction, and the re-agglomeration of GO sheets is effectively prevented. In addition, water in the solvent maintains the system's dielectric polarisation capacity, which is essential for the conversion of energy during the application of ultrasound. Simultaneously, the ultrasound-induced cavitation effect plays a crucial role in the production process by generating variations in temperature and pressure. Significantly, the group with a 60 % ethanol concentration demonstrated superior performance. More precisely, the yield of products comprising fewer than three layers reached an impressive 90.24 %, predominantly consisting of single-layer graphene. In contrast, mechanical agitation and lower ethanol concentrations, such as 20 %, proved ineffectual or less effective, with a 66.67 % lower yield compared to the 60 % ethanol concentration group. Conversely, higher ethanol concentrations were found to compromise the efficacy of the exfoliation process, leading to an augmented proportion of higher-layer graphene within the GO product. These GO sheets exhibit higher oxygen-containing functional groups than those produced using conventional chemical methods. These functional groups offer active sites on the graphene structure, which expands the possibilities for GO sheet functionalization. Furthermore, the presence of nanopores with an approximate diameter of 100 nm is observable. The formation of these pores can be attributed to the adaptable pit structures in cells that are endowed with robust mechanical strength and resistance considering the original cell sizes in plants and the cell shrinkage during pyrolysis. These include silica cells, stomata guard cells, cork cells within the epidermis, sclereids and transfer cells in the cortex, and sclerenchyma fibers and xylogen in the vascular bundles. This investigation unveils the potential of employing renewable biomass resources as precursors to produce high-quality graphene material through an environmentally friendly and efficient approach. Further, the study reveals the influence of plant cell components and functionalities on the characteristics of biomass-derived graphene, opening up new possibilities for the creation of graphene materials for various applications.

CRediT authorship contribution statement

Xinyun Wu: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation. Manickam Sivakumar: Writing – review & editing, Validation, Methodology, Conceptualization. Siew Shee Lim: Visualization, Supervision, Project administration. Tao Wu: Supervision, Resources, Funding acquisition, Conceptualization. Pang Cheng Heng: Conceptualization, Methodology, Validation, Resources, Writing – original draft, Writing – review & editing, Supervision, Project administration, Funding acquisition.

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.