Carbon is abundantly present in nature and constitutes one of the essential elements for organic organisms. It serves as the fundamental building block for various complex molecules, such as fats, steroids, hydrocarbons, and organic solvents. With four electrons in its outer valence shell, carbon can form diverse crystalline structures, ranging from the hardest diamond to the softest graphite. Over the past two decades, discoveries of novel carbon-based materials—such as fullerenes identified by American scientists in 1985 and carbon nanotubes identified by Japanese scientists in 1991—have kept carbon nanomaterials at the forefront of scientific research, sparking sustained global interest. The discovery of graphene—a two-dimensional carbon crystal—by scientists Geim and Novoselov at the University of Manchester in 2004 overturned the theory that strictly two-dimensional crystals could not exist at finite temperatures, profoundly influencing the development of condensed matter physics. The discovery of graphene not only enriched the family of nanocarbon materials, forming a complete system from zero-dimensional fullerenes, through one-dimensional carbon nanotubes, to two-dimensional graphene and three-dimensional diamond, but its unique nanostructure and outstanding mechanical, thermal, electrical, and optical properties have made the development and research of graphene materials another international research hotspot following carbon nanotubes. Owing to its exceptional properties, workability, and low cost, graphene holds significant application value across energy, materials, electronics, and biomedicine, poised to spark a new technological revolution in the international field of advanced materials.
1 Overview of Graphene
1.1 Structure of Graphene
Graphene constitutes a single-atom-thick two-dimensional crystal formed by carbon atoms bonded via sp² hybridisation. These atoms are arranged in a regular honeycomb lattice structure. Each carbon atom forms σ bonds with three neighbouring atoms, while its remaining π electrons delocalise to form extensive π bonds with other carbon atoms. This allows electrons to move freely within the lattice, conferring graphene's exceptional electrical conductivity. Simultaneously, this closely packed honeycomb structure serves as the fundamental building block for other carbon materials. As illustrated in Figure 1, single-layer graphene can encapsulate itself to form zero-dimensional fullerenes, while single- or multi-layer graphene can curl to form single-walled or multi-walled carbon nanotubes.
Owing to the thermodynamic instability of two-dimensional crystals, graphene—whether in its free state or deposited upon a substrate—is never perfectly planar. Instead, its surface exhibits intrinsic microscopic wrinkles, as demonstrated by both Monte Carlo simulations and transmission electron microscopy ( Figure 2 ) . These microscopic wrinkles exhibit lateral dimensions within the range of 8–10 nm and longitudinal dimensions of approximately 0.7–1.0 nm. This three-dimensional variation induces electrostatic forces, facilitating the aggregation of graphene monolayers. Concurrently, differing wrinkle sizes result in distinct electrical and optical properties being exhibited by graphene.
Beyond surface wrinkles, graphene in practice is not flawless but exhibits various defects, including morphological imperfections ( such as pentagonal and heptagonal rings ) , voids, edges, cracks, and impurity atoms. These defects influence graphene's intrinsic properties, such as electrical and mechanical characteristics. However, by deliberately introducing defects through engineered methods—such as high-energy radiation exposure or chemical treatments—one can intentionally modify graphene's intrinsic properties, thereby fabricating graphene devices tailored to specific performance requirements.
1.2 Properties of Graphene
The unique single-atom-layer structure of graphene endows it with numerous exceptional physical properties. As previously noted, each carbon atom in graphene possesses an unpaired π electron. These electrons form π orbitals perpendicular to the plane, allowing π electrons to move freely within these long-range π orbitals. This confers outstanding electrical conductivity upon graphene. Research indicates that at room temperature, the carrier mobility in graphene can reach 15,000 cm²/ ( V·s ) , equivalent to one-thirtieth of the speed of light. Under specific conditions, such as at liquid helium temperatures, this figure can soar to 250,000 cm²/ ( V·s ) , far surpassing other semiconductor materials like indium antimonide, gallium arsenide, and silicon semiconductors. This renders the behaviour of electrons within graphene remarkably similar to relativistic neutrinos. Furthermore, electrons traverse the lattice without encountering barriers or scattering, conferring exceptional electronic transport properties. Concurrently, graphene's unique electronic structure gives rise to numerous peculiar electrical phenomena, including the room-temperature quantum Hall effect.
Given that each carbon atom in graphene forms strong σ bonds with three neighbouring atoms, it also exhibits outstanding mechanical properties. Scientists at Columbia University directly tested the mechanical properties of single-layer graphene using atomic force microscopy, discovering a Young's modulus of approximately 1100 GPa and a fracture strength reaching 130 GPa – a hundredfold higher than the finest steel. Graphene also serves as an exceptional thermal conductor. Owing to the low carrier density in undoped graphite, heat transfer in graphene primarily occurs via phonon conduction, with electron motion contributing negligibly to thermal conductivity. Its thermal conductivity reaches 5000 W/ ( m·K ) , surpassing that of carbon nanotubes and exceeding common metals such as gold, silver, and copper by more than tenfold.
Beyond its exceptional thermal and mechanical properties, graphene exhibits several novel characteristics. The presence of lone pair electrons at its edges and defects confers magnetic properties, including ferromagnetism. Its unique single-atom-layer structure yields a theoretical specific surface area of 2630 m²/g. Graphene also possesses distinctive optical properties, with single-layer graphene exhibiting over 97% transmittance in the visible spectrum. These characteristics confer extensive application prospects for graphene in critical fields such as nanodevices, sensors, hydrogen storage materials, composite materials, and field emission materials.
2 Preparation of Graphene
A key factor constraining the widespread application and industrial production of graphene is the large-scale fabrication of single-layer or few-layer graphene materials with processable properties. Currently, the most widely applied graphene preparation methods can be broadly categorised as follows: mechanical exfoliation, epitaxial growth, chemical vapour deposition, chemical synthesis, reduction of graphene oxide, and vertical cutting of carbon nanotubes.
2.1 Mechanical Exfoliation
As early as 1999, Rouff et al. attempted to isolate graphene from graphite using micro-mechanical exfoliation. Although they failed to obtain monolayer graphene, their work laid a foundation for subsequent research. In 2004, Geim and Novoselov employed this method, repeatedly peeling high-orientation pyrolytic graphite with adhesive tape. Subsequently, the tape bearing graphene was sonicated in acetone, and graphene dispersed in acetone was then lifted out using a silicon wafer. While this method could produce large-area, high-quality graphene sheets, it was time-consuming and labour-intensive, with extremely low yields and monolayer graphene content in the product. Consequently, it was unsuitable for large-scale graphene production and remained confined to fundamental graphene research.
Ultrasonic exfoliation of graphite in liquid phase represents another commonly employed method. When selecting solvents with surface energies matching graphene, such as N-methyl-2-pyrrolidone, N, N,N-dimethylformamide, or dichlorobenzene as the medium, or in water containing surfactants, the mechanical forces generated by ultrasonication can separate graphene from the graphite matrix. Simultaneously, the interactions between graphene and solvent molecules enable the separated graphene to remain stably suspended in the solvent. Graphene obtained via this method is predominantly monolayered or few-layered. Raman spectroscopy and XPS analysis indicate that the resulting graphene contains minimal oxidised groups and exhibits fewer defects. Compared to the conventional ‘tape-off’ mechanical exfoliation method, this approach is simpler and more conducive to subsequent graphene processing and the preparation of graphene-based materials. However, this approach similarly faces the challenge of low yield, typically yielding graphene solutions with concentrations ranging from 0.01 to 0.03 mg/mL.
2.2 Epitaxial Growth Method
The first method proven capable of large-scale graphene film production involves removing silicon atoms via high-temperature ( 1200–1500°C ) sublimation from the surface of single-crystal silicon carbide ( SiC ) wafers, yielding epitaxially grown graphene. In this approach, silicon atoms are removed, allowing residual carbon atoms on the surface to rearrange into graphene structures. These structures grow continuously on the flat, suitable hexagonal silicon carbide substrate surface. Furthermore, adjusting annealing temperature and duration enables control over graphene thickness and layer number. Experiments demonstrate that graphene can form on both silicon-rich ( 0001 ) and carbon-rich ( 000-1 ) surfaces. Graphene grown on the 0001 plane exhibits a well-ordered structure, whereas that on the 000-1 plane forms a stacked disordered structure. Nevertheless, these multilayer graphene films display excellent electrical properties, enabling top-gate transistors fabricated from them to achieve electron mobility values of up to 5000 cm²/ ( V·s ) . . The advantage of the epitaxial growth method lies in its ability to fabricate multiple devices ( such as transistors ) on a single substrate. However, the number of graphene layers produced by this method varies, and substrate doping readily affects the electrical properties of the graphene, thus failing to meet the requirements for electronic device applications.
2.3 Chemical Vapour Deposition
Chemical vapour deposition involves pyrolysing carbon sources ( such as hydrocarbons ) at elevated temperatures and depositing them onto solid substrates, typically transition metals like Ni or Ru. Although this method yields higher-quality graphene with improved yield compared to mechanical exfoliation and exhibits high electron mobility, the resulting graphene films only possess practical value when transferred from the metal substrate to other substrates. Recent reports have documented the successful synthesis of single-layer or few-layer graphene on polycrystalline Ni substrates via chemical vapour deposition of carbon sources like methanol. Subsequent etching techniques have enabled transfer of these graphene films onto alternative substrates including PMMA, PDMS, Si/SiO₂, and glass. Nevertheless, the resulting films are invariably multi-layer graphene or exhibit non-uniform single-layer graphene structures.
2.4 Reduction of Graphene Oxide
Although the three methods described above can all produce high-quality graphene, they all encounter issues such as low yield and poor workability, severely limiting graphene's application in many fields. Currently, the most widely used and promising method for achieving large-scale industrialisation is to utilise graphene oxide as a precursor. Through thermal or chemical reduction, the oxygen-containing groups on the surface of graphene oxide are removed. Although this approach does not yield perfect graphene, it significantly restores its intrinsic properties. Furthermore, compared to other graphene preparation methods, the reduction of graphene oxide offers abundant raw materials, straightforward equipment and operational processes, and produces highly processable graphene, making it a subject of considerable interest.
2.5 Longitudinal Cutting of Carbon Nanotubes
The preparation of graphene from carbon nanotubes represents a novel approach developed in recent years. Unlike the isotropic graphene sheets derived from graphite, slicing carbon nanotubes yields anisotropic graphene ribbons. Due to their structure resembling one-dimensional nanomaterials, graphene ribbons exhibit distinct properties from two-dimensional sheets, such as higher energy bands, conferring significant potential applications in nanoelectronics.
The Tour research group has reported an oxidative cutting method for carbon nanotubes, as illustrated in Figure 3. They first uniformly disperse multi-walled carbon nanotubes in concentrated sulphuric acid, then add a measured quantity of potassium permanganate as the oxidising agent. The mixture is slowly heated until the potassium permanganate undergoes complete reaction. When the mass of potassium permanganate was five times that of the carbon nanotubes, the nanotubes were completely cleaved into graphene nanoribbons. The graphene nanoribbons obtained via this method contained abundant oxygen-containing functional groups and could be uniformly dispersed in water and various organic solvents. Subsequently, sodium dodecyl sulphate was employed as a stabiliser, with hydrazine hydrate used for reduction, yielding graphene nanoribbons with lower oxidation levels. They further reported an improved oxidation-cutting method for carbon nanotubes, replacing the single concentrated sulphuric acid system with a mixture of concentrated sulphuric acid and concentrated phosphoric acid, thereby producing high-quality graphene nanoribbons.
2.6 Alternative Preparation Methods
Beyond the aforementioned graphene synthesis approaches, scientists have developed additional methodologies to achieve high-quality graphene. Dai's research group reported a three-step ‘thermal expansion-intercalation-exfoliation’ process for graphene production. where graphite intercalated with concentrated sulphuric acid was first treated at 1000°C for one minute to yield expanded graphite. This expanded graphite was then subjected to a second intercalation with fuming sulphuric acid and tetrabutylamine to increase the interlayer spacing. Finally, ultrasonic exfoliation in DMF containing surfactants yielded graphene with high exfoliation degree and low oxidation.
Several bottom-up approaches have also been employed for graphene synthesis. Stride et al. achieved kilogram-scale graphene production via solvothermal synthesis using ethanol and metallic sodium as precursors, offering advantages such as high yield, low pollution, and straightforward operation. Recent reports also describe graphene and graphene nanoribbon synthesis from aromatic small molecules. This approach primarily involves synthesising hexabenzocoronene via Diels-Alder reactions or Pd-catalysed coupling reactions, followed by cyclisation and dehydrogenation to yield graphene. This method enables the precise synthesis of graphene with specific structures and sizes. However, the reactions are complex, yields are low, and the process is time-consuming, rendering it unsuitable for large-scale production.
3 Applications of Graphene
3.1 Nanoelectronic Devices
Due to graphene's unique electronic structure and excellent electrical conductivity, it holds great promise as the optimal material for constructing nanoelectronic devices. One of the most extensively researched and prominent topics currently involves preparing transparent conductive films based on graphene to replace costly indium tin oxide ( ITO ) electrodes. Given that graphene oxide can be produced on a large scale and exhibits excellent workability, its use as a raw material for preparing graphene transparent conductive films represents a significant fabrication approach. In this method, graphene oxide films are first formed via techniques such as spin coating, dip coating, vacuum filtration, or LB assembly. These graphene oxide films are then reduced to graphene films through chemical reduction or thermal reduction processes. Scientists have also developed alternative methods for fabricating transparent conductive films using dispersions of graphene or reduced graphene oxide. For instance, Li et al. first adjusted the system pH to 10 prior to reducing graphene oxide to obtain a stable graphene dispersion, subsequently producing transparent conductive films via spray coating. The Dai group employed graphene dispersions obtained via ‘thermal expansion-intercalation-exfoliation’ as raw material, utilising LB assembly to produce graphene transparent conductive films. These films exhibited a sheet resistance of 8 kΩ/sq and a visible light transmittance of 83%. Biswas et al. employed interfacial self-assembly at the water/chloroform binary interface to produce conductive films with a sheet resistance of 100 Ω/sq and visible light transmittance of 70%. The Coleman group directly sonicated graphene exfoliated in organic solvents, followed by vacuum filtration, yielding conductive films with a sheet resistance of approximately 3 kΩ/sq and visible light transmittance of 75%. Currently, graphene conductive films are employed in electronic devices such as dye-sensitised solar cells, liquid crystal devices, organic light-emitting diodes, and transistors.
3.2 Molecular Detectors
Given that graphene's conductivity varies with the concentration of adsorbed molecules on its surface, coupled with its exceptionally high specific surface area and low Johnson noise ( thermal noise ) , both experimental and theoretical evidence demonstrates that monolayer graphene serves as an excellent molecular detector for various small molecules. Charge transfer between graphene and adsorbed molecules is considered the mechanism underlying this chemical detection. When small molecules adsorb onto the graphene surface, charge transfer occurs between graphene—acting as either an electron donor or acceptor—and the adsorbed molecules. Consequently, parameters such as the Fermi level, carrier density, and resistivity undergo alteration. Recent experiments demonstrate that ammonia and carbon monoxide molecules can function as electron donors, while water molecules and nitrogen dioxide molecules can act as electron acceptors interacting with graphene. Reduced graphene has also been demonstrated as an exceptionally effective detector for certain chemicals and explosives, achieving detection limits at the parts-per-billion ( ppb ) level.
3.3 Biological Applications
Reduced graphene and modified graphene have been employed in biological fields such as drug delivery systems, live-cell imaging, and biomolecular detection. Compared to carbon nanotubes, graphene-based materials exhibit distinct advantages in biological applications. Firstly, they contain no impurities such as metallic catalysts, thereby avoiding cellular biostress. Secondly, modified graphene dispersions require no surfactants and demonstrate superior water solubility. Thirdly, graphene's exceptionally high specific surface area enables significantly enhanced drug loading capacities.
Dai et al. employed polyethylene glycol-modified graphene for bioimaging of viable cells, confirming its non-toxicity. They further successfully loaded the hydrophobic drug camptothecin derivative SN38 onto this modified graphene. SN38 adsorbed onto the hydrophilic modified graphene surface via π-π conjugation interactions, enabling controlled in vivo release. Zhang et al. employed folic acid-modified graphene to load both doxorubicin and camptothecin, achieving targeted drug delivery to human breast cancer cells ( MCF-7 ) . Yang et al. first grew magnetic Fe₃O₄ nanoparticles in situ on graphene, then modified these particles with folic acid, yielding a drug delivery system with multifunctionality, pH-controlled release, and targeted drug delivery capabilities.
Modified graphene has also been employed in various bio-devices for detecting biological cells and biomolecules. It functions as an interface for identifying individual bacteria, serves as a label-free, reversible DNA detector, or acts as a polarity-specific molecular crystal for adsorbing proteins/DNA.
3.4 Nanoparticle Composite Materials
In recent years, the loading of nanoparticles onto carbon nanotubes has garnered significant attention and research. This novel nanostructure has already achieved considerable progress in biomedicine, catalysis, and sensor applications. Compared to carbon nanotubes, graphene possesses similar stable physical properties but exhibits a higher specific surface area. Consequently, loading nanoparticles onto graphene similarly holds promise for generating new nanostructures, altering their physical characteristics, and yielding richer functionalities and applications.
Currently, the graphene/nanoparticle composites most frequently reported in research primarily encompass graphene/metal nanoparticle composites, graphene/metal oxide nanoparticle composites, and graphene/quantum dot composites. Preparation methods are mainly categorised into two approaches: one utilises the interactions between precursors of metal particles or metal oxide particles and graphene or graphene oxide to achieve in situ reaction and assembly of nanoparticles onto graphene or graphene oxide. For instance, Xu et al. reported dispersing precursors of Au, Pd, and Pt—AuCl₄²⁺, PdCl₄²⁺, and PtCl₄²⁺ respectively—into aqueous solutions of graphene oxide. Subsequently, simultaneous reduction of the noble metal salts and graphene oxide using ethylene glycol yielded graphene loaded with Au, Pt, and Pd particles.
Beyond nanoparticle composites, graphene can also be assembled with other carbon-based nanomaterials to form hybrid composites. Liu et al. prepared graphene/fullerene composites via covalent bonding, observing a significant enhancement in the nonlinear optical properties of fullerene-modified graphene. Yang et al. blended carbon nanotubes with graphene to fabricate a novel supercapacitor, achieving a specific capacitance of 326.5 F/g when graphene content reached 90%. Concurrently, numerous research groups have demonstrated the advantages of graphene/carbon nanotube composites in producing transparent conductive films, observing superior performance in films prepared from this composite mixture compared to those from single-component conductive films.
