Synthesis and applications of large-area single-layer graphene

Abstract

Large-area graphene is gaining heightened attention in the fields of optoelectronics and highly selective membrane separation due to its unique physicochemical properties including exceptionally high surface, electro mobility, thermal conductivity, mechanical strength, and one atom thickness. These outstanding properties can lead to novel or improved technologies to address the pressing global challenges in transparent conducting electrodes, transistors, ultrathin electromagnetic shielding film, osmotic energy production, highly selective single molecule gas detection, desalination, DNA sequencing, etc. This review assesses the progress in two categories of methods for preparing large-area graphene, namely chemical assembly of small graphene sheets and chemical vapor deposition method. We also discuss the challenges for preparing large-area graphene with high qualities and controllable layers and for transferring graphene film to various substrates with large area, clean, and low defect surface. Applications of large-area graphene in the optoelectronics fields have been discussed by Bonaccorso et al. We mainly discuss its applications in highly selective membrane separation including single molecule gas detection, desalination, DNA sequencing and osmotic energy production. Finally, we outline the global investment and industrial production of large-area graphene.

---

Xiao-Hong Lin

Xiao-Hong Lin received her bachelor's degree in Polymer Science and Engineering from Qilu University of Technology, China. She is now studying for a master's degree in Materials Science, Sichuan University, China. Her research focuses on seawater desalination using nanoporous graphene membrane.

Jing-Gang Gai

Jing-Gang Gai is an associate professor at the State Key Laboratory of Polymer Materials Engineering, Polymer Research Institute of Sichuan University, China. He received his MS degree from Sichuan University in 2005, and his Ph.D. from Institut National Polytechnique de Lorraine, France in 2009. His research interests are in the areas of angstrom film and its applications in membrane separations and energy production by osmotic energy.

---

1. Introduction

Graphene is composed of a single layer of carbon atoms which pack into a two-dimensional (2D) honeycomb lattice.^1,2 It is the basic building block for graphitic materials of other dimensionalities including 0D fullerenes,³ 1D nanotubes^4–8 and 3D graphite. More and more scientists have engaged in researches related with graphene since Geim and Novoselov successfully isolated pristine graphene utilizing the adhesive tape.⁹

Graphene exhibited many excellent properties^10–13 including mechanical,¹⁴ thermal,¹⁵ chemical,^16,17 optical,¹⁸ electrical properties.¹⁹ The strength of defect-free single crystalline graphene is the highest among the materials which have been tested. It is up to 130 GPa which is 100 times higher than that of the steel.¹⁴ The thermal conductivity is about 5000 W m⁻¹ K⁻¹ and is 3 times higher than that of diamond.¹⁵ The intrinsic mobility is up to 2 × 10⁵ cm² V⁻¹ s⁻¹ which is 10 times higher than that of commercial silicon.¹⁹ Besides, the theoretical thickness of single layer graphene is only 0.34 nm, and it is the thinnest 2D material up till now.¹

Chemically doped graphene, especially nitrogen-doped and boron-doped graphene, also has excellent properties and it has been widely studied. Pristine single-layer graphene is a gapless material, but the creation of band gap is necessary for many applications in energy related devices. Chemical doping of graphene can produce a band gap and it can lead to useful properties as well. Rao et al. systematically reviewed the synthesis, characterization, properties and applications of graphene doped with nitrogen, boron and other elements.²⁰ For nitrogen-doped graphene, the nitrogen source is usually ammonia and the nitrogen content can be varied from 0.25 to 15 at% through different synthetic methods. For boron-doped graphene, the boron source can be B₂H₆, H₃BO₃, B₂O₃, BCl₃ or BF₃ and the boron content can be varied from 0.22 to 3 at%. These synthetic methods of doped graphene mainly include arc-discharge, CVD, ball milling, thermal treatment and microwave treatment. Among them, arc discharge is a facile, low-cost and efficient method and it's versatile for producing various carbon nanomaterials.^21–25 Compared with pure graphene, optical, magnetic and electrical properties of doped graphene have great changes. For example, the photoluminescence emission of nitrogen-doped graphene is significantly enhanced. These changes are mainly associated with the change of the electronic structure of graphene. Due to its excellent performance, chemically doped graphene has a wide range of applications including supercapacitors, adsorption of gases, lithium ion batteries, field emission, water splitting and so on. For instance, the capacitance of nitrogen-doped graphene prepared under plasma conditions is 4 times higher than that of pristine graphene.

Due to these excellent properties mentioned above, scientists always pursue the defect-free large-area graphene (LAG). In these years, the preparation technologies and the applications of LAG have developed rapidly, especially in Korea, United States, Japan and China. LAG has been used to address the pressing global challenges of transparent conducting electrodes, field effect transistors, flexible and wearable touch screen and display, ultrathin electromagnetic shielding film, highly selective molecular sieve membranes for single molecule gas detection, desalination, DNA sequencing and osmotic energy production. Since 2004, the number of academic papers related to LAG has significantly increased. Fig. 1 illustrates the number of graphene-related and LAG-related academic papers in recent years. There were nearly 16 000 papers published with the keyword “graphene” and about 600 papers published with the keyword “LAG” in 2014 (Web of Science).²⁶

Fig. 1 The number of published academic papers about graphene (left) and large-area graphene (right) in different years.

Some representative works are listed in what follows chronologically. In 2006, for the first time, Somani et al. experimentally demonstrated that few layers of planar nano-graphenes (PFLGs) could be directly synthesized by simple thermal CVD method.²⁷ They scraped the graphene grown on Ni sheets (2 × 2 cm²) off using a sharp blade and collected the powder. They observed that the graphitic structure includes about 35 layers of graphene (Fig. 2a). In 2007, Obraztsov et al. adopted CVD method to prepare large area graphitic films with 3 to 6 graphene layers.²⁸ They used silicon and nickel square plates with lateral dimensions of 10 × 10 mm² or 25 × 25 mm² as substrates. There are obvious differences between the morphologies of films deposited on Ni and Si substrates (Fig. 2b). In 2008, Eda et al. reported a solution-based method (vacuum filtration method) of preparing large-area reduced graphene oxide (RGO), using which they successfully fabricated a 10 cm² RGO membrane with 1–5 layers of graphene (Fig. 2c).²⁹ In June 2009, Li et al. successfully prepared large area, high quality and uniform graphene films on copper substrates adopting CVD method. More than 95% of the 1 × 1 cm² film was monolayer graphene, 3–4% was bilayer graphene and the rest small fraction (<1%) was trilayer or few-layer (<10 layers) graphene (Fig. 2d).³⁰ They found that the graphene grew on copper by a surface-catalyzed process. Then in August, Li et al. showed a photograph of a 4.5 × 4.5 cm² graphene on quartz substrate (Fig. 2e) fabricated by CVD method and also reported an improved method to transfer graphene to other substrates with much lower densities of cracks and tears.³¹

Fig. 2 Representative pictures about large-area graphene before 2010. (a) HR-TEM image of PFLG film which shows graphitic structure marked by two arrows (ref. 27). (b) The SEM images of graphene films grown on Si (left) and Ni (right) respectively (ref. 28). (c) Photograph of 10 cm² GO thin films on plastic substrate (ref. 29). (d) Graphene film transferred onto a glass plate, more than 95% of the 1 × 1 cm² film is monolayer graphene (ref. 30). (e) 4.5 × 4.5 cm² graphene on quartz substrate (ref. 31).

As can be seen from Fig. 1, the researches about LAG have increased rapidly since 2010. There was a major breakthrough in the preparation of LAG in 2010. Bae et al. from Sungkynkwan University of Korea first reported a roll-to-roll (R2R) method to produce LAG films.^32,33 They adopted the CVD method to prepare LAG on flexible ultra-large copper substrates. The films were 30-inch in the diagonal direction (Fig. 3a) and were predominantly monolayer. The films can be transferred to a target substrate by R2R method.

Fig. 3 Representative pictures about large-area graphene after 2010. (a) A transparent ultralarge-area graphene film transferred on a 35-inch PET sheet (ref. 32). (b) A fabricated touch panel using the graphene-based transparent conductive films (ref. 34). (c) Photograph of 20 × 47 cm² GO film deposited directly on a PET substrate by rod coating method (ref. 35). (d) A graphene film in area of 294 × 100 mm² transferred onto PET substrate (ref. 36). (e) A 100 m-long, 230 mm-wide graphene film (ref. 37). (f) Photograph of 7 × 7 cm² monolayer graphene film (ref. 38).

In 2011, Kim et al. developed a large-area surface wave plasma (SWP) CVD apparatus to fabricate LAG-based films consisting of few-layer graphene sheets at lower temperatures (300–400 °C).³⁴ The graphene film with lateral dimensions of 23 × 20 cm² exhibited excellent optical and electrical properties, and has been successfully applied in the preparations of touch panel (Fig. 3b). Then in 2012, Wang et al. reported a novel method—rod coating to produce large-scale, transparent and conductive RGO films.³⁵ This method can be carried out on arbitrary substrates at room temperature. They showed a graphene oxide (GO) film (20 × 47 cm²) deposited directly on a poly(ethylene terephthalate) (PET) substrate (Fig. 3c) and the RGO/PET film was successfully used in electrode.

Yamada et al. established a R2R microwave plasma (MWP) CVD process to prepare LAG films at low temperature.³⁶ They deposited a graphene film with an area of 294 × 100 mm² on Cu substrate and then transferred to PET film (Fig. 3d) below 400 °C. In 2013, Kobayashi et al. from Sony Corporation reported a new method—R2R CVD which applied selective Joule heating of the copper foil to achieve low pressure thermal CVD synthesis and a direct R2R transfer process, successfully producing a 100 m-long, 230 mm-wide high-quality graphene film which was predominantly composed of single-layer graphene (Fig. 3e).³⁷

In 2014, Liu et al. reported an effective way by using CVD method to synthesize high-quality and uniform monolayer graphene without bilayer regions.³⁸ They fabricated a 7 × 7 cm² monolayer graphene (Fig. 3f). Ryu et al. adopted a method of hydrogen-free rapid thermal chemical vapor deposition (RT-CVD) to produce LAG films. They got homogeneous 400 × 300 mm² graphene films,³⁹ which was successfully used to fabricate a capacitive multitouch device and to install a mobile phone (Fig. 11d). More details are shown in Table 1.

Table 1 Basic synthesis method, size, layer number, properties and applications of large-area graphene in different years as mentioned above

Year	Synthesis method	Size/cm^2	Layer number	Properties	Applications	Reference
2006	CVD	2 × 2 (Ni)	∼20–35	—	—	27
2007	CVD	2.5 × 2.5 (Ni)	3–6	—	—	28
2008	Vacuum filtration	10	1–5	Exist structural defects, lowest sheet resistance: ∼43 kΩ sq^−1, transparency: 60–95%, electron mobility: ∼0.2 cm^2 V^−1 s^−1	Thin-film transistors (TFTs)	29
2009	CVD	1 × 1 (Cu)	1 (>95%)	Monolayer >95%, electron mobility: ∼4050 cm^2 V^−1 s^−1	Field-effect transistors	30
	CVD	4.5 × 4.5		Excellent mechanical stability, sheet resistance: ∼350 Ω sq^−1, transmittance: 86–98%	Transparent conductive electrode	31
2010	CVD	30-inch (Cu)	Monolayer (mostly)	Sheet resistance: ∼125 Ω sq^−1, optical transmittance: 97.4%, half-inter quantum Hall effect	Transparent electrode	32
2011	SWP-CVD	23 × 20 (Cu)	Few-layer	Sheet resistance: 2.2–45 kΩ sq^−1, optical transmittance: 78–94% (on glass plate)	Capacitive touch panel	34
2012	Rod-coating	20 × 47 (PET)	1–4, >4 (can be tuned)	Sheet resistance: 1.68–20.1 kΩ sq^−1, optical transmittance: 64.6–92.6%, excellent foldability and uniformity	Touch screen	35
	MWP CVD	29.4 × 10 (Cu)	1–4	Optical transmittance: 95%, sheet resistance: 0.1–7 × 10^5 Ω sq^−1	Electronic devices	36
2013	R2R CVD	100 m × 23 cm (Cu)	Monolayer (mostly)	Coverage rate: 89–98%, sheet resistance: ∼150 Ω sq^−1, carrier mobility: ∼900 cm^2 V^−1 s^−1	—	37
2014	CVD	7 × 7 (Cu)	Monolayer	Hole mobility 2534.3 ± 97.6 cm^2 V^−1 s^−1, high uniformity	—	38
	RT-CVD	40 × 30 (Cu)	—	Sheet resistance: 249 ± 17 Ω sq^−1	Touch screen	39

---

This review first discusses the preparation of LAG films. There are mainly two kinds of methods including chemical assembly of small graphene sheets and CVD method. Then, we summarize the applications of graphene films in membrane separation including gas separation, water desalination and DNA sequencing. Thirdly, we introduce the global investment in the field of graphene and the industrial production of graphene films. Finally, we will highlight challenges that should be overcome in order to achieve pervasive application of LAG films.

2. Preparation

There have been many methods for the preparation of graphene, for instance, Rao et al. summarized various synthetic methods of different layers of graphene.¹³ Up to now, chemical assembly of small graphene sheets^12,33,40 and CVD method are two principal methods for the preparation of LAG.^32,37,41 These two methods are emphatically discussed in this review.

2.1. Chemical assembly

Chemical assembly is a solution-processing procedure of graphene or graphene oxide dispersed in a solvent. The resulting solution can be directly deposited on any substrates via the following methods: rod-coating,^35,42–45 spin-coating of RGO,^46–52 dip-coating of GO,^53–56 vacuum filtration of GO suspension,^29,57–68 interfacial assembly,^69–71 spray coating^72–80 and so forth.

2.1.1. Rod coating. Rod coating, also known as Meyer rod-coating, Meyer rod, Meyer bar, was invented by Charles W. Mayer. The so-called rod is a metal bar wrapped with wire, and a solution can be spread out on the substrate surface when the rod is pulled over the solution (Fig. 4a). The initial coating consists of a series of stripes, and then the surface tension of the liquid film pulls these stripes together to form a continuous and uniform film. Finally, the solvent is removed by drying process and a film can be obtained. This coating technique is mainly used to make liquid thin films. In addition to graphene or RGO films, it can also be used to fabricate carbon nanotube (CNT) films,^44,81,82 silver nanowire films^83,84 and other thin films.

Fig. 4 (a) Schematic diagram of rod coating method (ref. 45). (b) 25 × 10 cm² SWNT thin film on a PET substrate with good transparency (ref. 44). (c) RGO film coated glass showed excellent defrosting performance (ref. 43). (d) A 14 × 18 cm² AC LED device which worked perfectly under the bending state (ref. 42).

In general, the GO solution is used as a coating material followed by reduction treatment. The obtained RGO thin film is often used to fabricate conductive and transparent electrodes. In 2012, for the first time, Wang et al. combined rod coating and room-temperature reduction of GO to fabricate a large area (20 × 47 cm²) RGO film on flexible PET substrate.³⁵ In the experiment, the concentration of GO solution was 0.05–2.5 mg mL⁻¹ and the wire diameter of the rod was 0.2 mm. With the increase of GO concentration from 0.5 to 2.5 mg mL⁻¹, the transmittance and sheet resistance of RGO films decreased from 92.6% to 64.6% and 20.1 to 1.68 kΩ sq⁻¹, respectively. The room-temperature reduction of GO reduces the consumption of energy, but the use of palladium catalyst increases the cost and the pressurized hydrogen in autoclave may induce safety problems.

In the same year, Xie et al. invented a wormlike micelle assisted rod coating method to prepare transparent and conductive thin films on flexible plastics.⁴⁴ They selected a cetyltrimethylammonium hydroxide/p-toluenesulfonic acid wormlike micelle system for the dispersion of different nanomaterials. The system can also adjust the viscosity of nanomaterial solution. They chose single-walled carbon nanotubes (SWNTs) as experimental model, successfully preparing a 25 × 10 cm² SWNT thin film on PET substrate with good transparency (Fig. 4b). After doping in thionyl chloride solution, the transmittance of SWNT thin films was about 78% and the sheet resistance was as low as 480 Ω sq⁻¹. This method can be applied to various nanomaterials and their hybrids, such as, CNTs, RGO and silver nanowires. However, the complete removal of residual surfactant which will influence the sheet resistance is difficult and complex.

In 2014, Wang et al. adopted Meyer-rod coating method followed by thermal reduction process to prepare RGO thin films on flexible glass.⁴³ They got a conductive RGO film on 10 × 20 cm² flexible glass substrate with transmittance over 40% and sheet resistance of ∼5 × 10³ Ω sq⁻¹. The RGO thin films can be used as heaters due to its high electrical conductivity. The conductive RGO thin films can be used to make defrost glasses for vehicle windows. Wang et al. added an electrode pair (Ag) on RGO film/flexible glass and observed that frost on the flexible glass was eliminated when 10 volts was applied to it (Fig. 4c). Yang et al. employed rod coating method to fabricate light-emitting diodes (LED) with double functional GO-based flexible alternating current (AC).⁴² In the experiment, GO film was used as dielectric layer and RGO film was used as conductive electrode. The sheet resistance of RGO film can be as low as 2.6 kΩ sq⁻¹ at the transmittance of 69%. They successfully fabricated a 14 × 18 cm² AC LED device (Fig. 4d) and the luminance of the device increased steadily with the rise of frequency due to low dielectric loss property of GO.

The rod coating method for the preparation of LAG has many advantages including low cost, easy to operate, scalable, and precise in thickness control (by adjusting the solution concentration or the diameter of rod wire). It can also operate directly on any substrates at room temperature. But the viscosity of the solution should not be too high, or it will reduce the flowability of the fluid between the wire windings. For this method, the optimum viscosity range of coating solution is 0.02–1.00 Pa s.⁴⁴ From what has been discussed above, we found that LAG films prepared by rod-coating method have good transparency and conductivity. Therefore, the industrial production of transparent and conductive thin films may be achieved through the combination of rod coating and R2R technology.

2.1.2. Spin coating. Spin coating method for preparing LAG mainly includes three steps: preparation of the solution, high-speed rotation and solvent evaporation (Fig. 5a). GO suspension is firstly added dropwise to the center of substrate, and then the substrate is accelerated by adjusting the rotation speed of the spin coating apparatus to spread GO suspension. Finally a thin GO film can be obtained on the substrate. The thickness of thin films can be controlled by adjusting the GO concentration, the spin speed or the number of spin-coating cycles. The concentration of GO should be appropriate, typically 0.4–5.0 mg mL⁻¹ which can generate uniform and continuous thin films.^48–50 A too high concentration will lead to aggregation of GO sheets, and too low concentration will result in nonuniform and discontinuous coverage of substrate surface. Thicker films can be obtained by reducing the spin coating speed or increasing the number of spin coating cycles. Yamaguchi et al. reported that single to bilayer films were obtained at 8000 rpm, 4–5 layers at 6000 rpm and 7–8 layers at 4000 rpm.⁵⁰

Fig. 5 (a) Schematic diagram of spin coating method (ref. 47). (b) A 300 mm, transparent and flexible graphene film transferred onto PET film (ref. 50). (c) Graphene composite electrode on PET (ref. 52).

In 2008, Robinson et al. reported a modified spin coating method—blowing dry N₂ in the spin-coating process (Fig. 5a).⁴⁷ It can accelerate the evaporation of solvent, thereby increasing the interaction between GO sheets and the substrate to obtain a continuous film. LAG film can also be obtained by spin coating method. Yamaguchi et al. prepared 300 mm wafer-scale, highly uniform and transparent GO films with controllable number of layers (Fig. 5b) by modified spin coating method.⁵⁰ The sheet resistance and transmittance of 15 nm thin film were ∼1 kΩ sq⁻¹ and ∼70%. For the first time, Chang et al. fabricated a new type of graphene composite electrode in 2010.⁵² The GO solution was prepared by surfactant-assisted exfoliation of graphite oxide followed by in situ chemical reduction. The graphene composite electrode was got by spin coating of the mixture of RGO solution and poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) without further processing (Fig. 5c). This graphene composite electrode exhibited high mechanical and electrical stability in bending tests of at least 1000 cycles. The films prepared by spin coating method are generally uniform and continuous, while the reduction of GO is usually incomplete and the reduction process is complex. The reduction of GO often requires high temperature annealing or the use of toxic hydrazine as reductant. In addition, spin motion can cause stress and a large amount of solution will be wasted in the spin-coating process.

2.1.3. Dip coating. The general process of preparing graphene films by dip coating method may be summarized as follows: firstly, the target substrate is immersed into GO suspension for a period of time; secondly, the substrate is raised out from GO suspension and dried in flowing nitrogen; finally, the obtained GO films are thermal reduced to get RGO films. In general, the concentration of GO suspension is in the range of 0.1–1.0 mg mL⁻¹.^53,56,85 Lou et al. found that the suitable concentration to form robust GO/ceramic composite membranes was 1.0 mg mL⁻¹.⁸⁵ Sun et al. obtained an ideal RGO film when the concentration of GO suspension was 0.3 mg mL⁻¹ with five coating cycles.⁵³ The sheet resistance of the RGO film was less than 60 kΩ sq⁻¹ and the transmittance was 81%. Jeon et al. reported a novel horizontal-dip-coating method to prepare thin and homogeneous GO films on ITO substrate for organic semiconducting devices.⁵⁵ As a result, they got homogeneous GO thin films with high surface coverage and low surface roughness. And Cao et al. reported a new method which can synchronously reduce and assemble GO sheets on metal substrate to form LAG films.⁵⁶ The thickness of film can be adjusted by controlling the immersion time of substrate in GO suspension. They got a paper-like film with the area of 20 × 23 cm² and a transparent few-layer film with the area of 10 × 10 cm². The transmittance of film composed of 6 layers of RGO was about 89% at 550 nm, and it decreased to 42% when the number of layers increased to 20. The sheet resistance of few-layered films was in the range of 1–10 MΩ sq⁻¹, which indicated that the conductivities of these films are not very well. Graphene films fabricated by dip coating method can be applied to many application domains, for instance, transparent and conductive electrode,^{54–56,86,87} sensor,^88,89 water treatment,⁹⁰ bone repair,⁹¹ small molecular separation,⁸⁵ photoelectrochemical water splitting⁹² and composite interface layer.⁹³

Dip coating is a direct and effective method, and it has the following advantages: low cost, easy to operate, continuous and large area production can be achieved through mechanization or automation. The thickness of GO films can be adjusted by changing the concentration of GO suspension, immersion time of substrates in GO solution and the number of coating cycles, but the resulting films are generally nonuniform due to aggregation of GO sheets. After thermal reduction, nonuniformity of the film thickness can be improved.

2.1.4. Vacuum filtration. The vacuum filtration method for the preparation of graphene films mainly includes a preparation of GO suspension, a filtration of the GO suspension through a filter membrane, transferring GO flakes onto target substrate followed by reduction.^29,57 The filter membrane can be cellulose ester membrane, anodic aluminium oxide membrane or other membranes, and the pore size of filter membrane can be tens of nanometers or even tens of millimeters in different experiments.^29,59–61 In the filtration process, the liquid can pass through the pores while GO sheets will be stuck to form film. In general, GO films on filter membrane can be transferred by placing the membrane with the film side down onto target substrate and dissolving the membrane with acetone or other solvent. The reduction of GO films is often operated by thermal annealing, exposure to hydrazine vapor or combination of hydrazine vapor exposure and low-temperature annealing. The last method is usually more efficient. Park et al. reported a simple and environmentally friendly method to reduce GO films. They exposed GO films to flash light irradiation for a few milliseconds under ambient conditions (Fig. 6a).⁵⁹ A completely reduced GO thin films can be obtained under optimal conditions (1 pulse number, 20 ms pulse width and 40 J cm⁻² energy). The sheet resistance decreased dramatically from 0.8 MΩ sq⁻¹ to 17.55 Ω sq⁻¹ after reduction which indicated the high efficiency of this reduction method. Eda et al. adopted vacuum filtration method to deposite 1–5 nm graphene films on various substrates including filtration membrane, glass and plastic (Fig. 2c). After chemical reduction and thermal annealing treatment, the transparencies of thin films varied from 60 to 95% and the lowest sheet resistance was 43 kΩ sq⁻¹. They also reported that Cl-doped RGO thin films can be used as hole-collecting electrode in organic photovoltaic devices.^29,57 Oh et al. fabricated oxide-free graphene suspension from graphite by liquid phase exfoliation using N-methyl-2-pyrrolidone (NMP) followed by thermal treatment to enhance the exfoliation (Fig. 6b), and then graphene sheets were deposited on polymer substrate by vacuum filtration method without the need for reduction treatment.⁵⁸ The sheet resistance and transmittance of graphene films treated at 800 °C were 2.8 × 10³ kΩ sq⁻¹ and 80%, respectively.

Fig. 6 (a) Schematic diagram of the flash light irradiation process for the reduction of GO thin films (ref. 59). (b) Schematic diagram of the oxide-free graphene suspension preparation process (ref. 58). (c) Schematic layout of the preparation of MoS₂/graphene composite paper (ref. 60).

Graphene films or composite materials prepared by vacuum filtration method can be used as electrode material, such as, anode for lithium-ion batteries,^60,68 self-powered electronics⁶² and field emission material.⁶³ Yang et al. fabricated a MoS₂/graphene composite film that can be used as an anode for lithium-ion batteries (Fig. 6c).⁶⁰ The capacity retention of the electrode was 65.8% at a high current rate of 1000 mA g⁻¹ and 91.1% after 100 cycles respectively. Du et al. fabricated graphene-based supercapacitors with specific capacitance of 245 F g⁻¹ and good cycling stability of 10 000 cycles. They got self-powered electronics which can be used to store solar energy by integration of graphene-based supercapacitors with high performance perovskite hybrid solar cells.⁶² Hybrid films of CNTs and graphene have excellent mechanical and electrical properties. These films may replace CNT films or ITO and they can be used as electrodes for supercapacitors, lithium-ion batteries and solar cells.^64,66 To sum up, graphene films or composite materials fabricated by vacuum filtration method have excellent electrical performance.

The vacuum filtration method can be carried out at room temperature, and the equipment is simple and easy to operate. The thickness of graphene films can be tuned by changing the concentration and filtration volume of GO suspension, but the resulting films are generally inhomogeneous. It is difficult to achieve complete transfer of LAG films on filter membrane, and small area graphene films can be transferred relatively intact, but not uniform. There is still much work need to be done in order to get high-quality LAG films by vacuum filtration method.

2.1.5. Interfacial assembly. Self-assembly of GO sheets can be carried out at various interfaces thanks to the surface chemical activity and amphiphilicity. The nature of interfacial self-assembly of GO is a self-concentration process of GO sheets.⁶⁹ There are mainly three kinds of interfaces for the self-assembly of GO sheets: liquid–air interface,^94–98 liquid–liquid interface^99–103 and liquid–solid interface.¹⁰⁴ In general, aqueous GO suspension is used as the liquid phase to form assembly interface with another phase including air, liquid or solid. People can get GO-based or graphene-based materials with diverse microstructures and morphologies (thin film, hydrogel, spherical structures, etc.) through changing and controlling these interfaces. The structure of GO sheets is anisotropic. How they assembly will affect the properties of the final material. Surface activity and assembly behavior of GO sheets have been studied. Controlled assembly of these 2D materials is important for both basic research and technical applications.⁷⁰ We mainly introduce the self-assembly of GO sheets at liquid–air and liquid–liquid interface because most of investigations have focused on these two interfaces.
2.1.5.1. Liquid–air interfacial assembly. For the interfacial assembly of GO, liquid–air interface is the most readily available. The ambient atmosphere and GO suspension are usually the gas phase and the liquid phase, respectively. Amphiphilic nature and surface chemistry activity are key factors for the assembly of GO sheets at liquid–air interface.⁶⁹ Due to the amphiphilicity, GO sheets tend to move toward the interface and the movement of GO sheets can reduce the surface tension of the liquid phase, thus, it can achieve self-assembly of GO sheets at liquid–air interface.

Zhao et al. invented a modified chemical exfoliation technique to fabricate large area GO sheets.⁹⁷ The maximum area of the GO sheets obtained was ∼40 000 μm². Thin GO films were prepared by self-assembly on a liquid/air interface, and then reduced by HI acid for the application of transparent conductive film (Fig. 7a). They found that the sheet resistance and transmittance of RGO films composed of GO sheets in area of ∼7000 μm² was 840 Ω sq⁻¹ and 78%, respectively. Evaporation-induced self-assembly of GO sheets at a liquid/air interface on glass substrate was reported by Niu et al. in 2012 (Fig. 7b), and they adopted low temperature thermal reduction at 200 °C.⁹⁴ They obtained centimeter scale RGO films which exhibited high elastic modulus of 76.18 GPa and hardness of 6.89 GPa. Liquid–air interface assembly method can be applied to both flexible and non-flexible substrates and can be used to prepare hybrid films.^96,98 Shao et al. presented the first example of GO/CNTs hybrid membrane which is uniform and flexible (Fig. 7c).⁹⁶ In the experiment, SWCNT were dispersed into GO to form homogeneous suspension, and GO sheets serve as the surfactants owing to the amphiphilic.

Fig. 7 (a) A photograph of RGO film on PET reduced by HI acid (ref. 97). (b) Schematic diagram of GO film self-assembly at liquid–air interface on glass substrate (ref. 94). (c) Schematic diagram of the self-assembly of a GO/CNT hybrid membrane at the liquid/air interface (ref. 96).

Langmuir–Blodgett (LB) assembly is a classical liquid–air interfacial assembly technique to prepare thin films.^105–113 In a typical LB assembly process for the preparing of GO thin films, GO sheets are firstly dissolved in a volatile organic solvent and then dropped onto water surface by using a glass syringe.^69,113 In general, methanol will be chosen as the volatile solvent, because it disperses GO well and can spread on a water surface rapidly. The optimum ratio of water/methanol mixture is 1 : 5 in most experiment. After evaporation of the volatile solvent, a monolayer film of GO sheets will form at water–air interface. Finally, the film can be transferred to a target substrate by vertical dip-coating. Fig. 8 shows a general flowchart for the synthesis of graphene films. There are mainly two interaction ways for GO sheets: edge-to-edge and face-to-face.¹¹³ Strong edge-to-edge electrostatic repulsion makes GO sheets stable and can also prevent the single layers from overlapping, while face-to-face interaction will lead to irreversible stacking. Thus, people can obtain multilayer films through layer-by-layer deposition. The morphology of GO films prepared by LB assembly can be precisely adjusted by changing the surface pressure and pH value of GO suspension.^{69,71,113,114} With the increase of surface pressure, GO sheets begin to fold at the contact points along edges so as to balance the pressure, leaving the interior flat and free of wrinkling.¹¹³ Further increasing the surface pressure, edge-overlapping can be firstly observed, then, the interior of GO films will wrinkle when there is no free space left at the edges of films. Hydrophilic edge groups (carboxyl, hydroxyl) and hydrophobic basal plane of GO sheet lead to its amphiphilic, pH can affect the degree of ionization of the edge carboxyl groups, so that amphiphilicity of GO can be tuned by pH.^71,114 The pH value of GO suspension is often modified by adding HCl or NaOH solution. GO sheets become more hydrophilic with the increase of pH which promotes the deprotonation of the carboxyl groups. After adding HCl to neutralize the base, the surface pressure can be largely recovered but it is still smaller than the initial value, which is considered that there is a loss of material at the interface under basic conditions. Under acidic conditions, parallel wrinkles can be seen in the film.

Fig. 8 Flow chart for the synthesis of graphene thin films by LB assembly technique (ref. 106).

In 2008, Li et al. reported a method to fabricate high-quality graphene sheets from graphite. About 90% of these graphene sheets were monolayer and the average size was 250 nm.¹¹¹ For the first time, high-quality graphene sheets were assembled by LB technique through layer-by-layer deposition. The sheet resistance and transmittance of three-layer films on quartz were 8 kΩ and 83%, respectively. GO films prepared by LB assembly can be used as conductive electrodes.^{105–108,110} Lin et al. prepared transparent conductive films (TCFs) composed of ultra-large GO sheets by LB assembly followed by HI acid reduction and chemical doping treatment.¹⁰⁶ The sheet resistance of the film was 1100 Ω sq⁻¹ with a transmittance of 91%. In 2011, Zheng et al. produced films from ultra-large GO sheets with the sheet resistance of ∼500 Ω sq⁻¹ with 90% transparency after thermal reduction and chemical doping treatments.¹⁰⁸ Then, in 2012, the sheet resistance and transmittance of TCFs prepared by Zheng et al. were 605 Ω sq⁻¹ and 86%, respectively.¹⁰⁷ In the experiment, they adopted a chemical method to produce ultra-large GO sheets with a diameter up to about 100 μm. Both the production of large size GO sheets with high quality and the promotion of electrical conductivity are key factors for the successful preparation of TCFs with excellent electrical properties. Films composed of GO sheets can be used as two-dimensional filler for polymer nanocomposites.¹⁰⁹ Kulkarni et al. used monolayer GO flakes as filler for polyelectrolyte, a significant increase in the elastic modulus of this composite film can be observed from 1.5 GPa of pure polyelectrolyte to about 20 GPa for only 8.0 vol% graphene oxide encapsulated membranes.

In conclusion, liquid–air interfacial assembly of GO sheets can be easily realized by using the ambient atmosphere as gas phase and GO suspension as the liquid phase. LB assembly is a classical liquid–air interface assembly technique. It is known to be the only technique that can realize layer-by-layer deposition of GO sheets and can precisely control the thickness of the film by repeated deposition.^105,107 Graphene films fabricated by LB assembly can be transferred to various substrates and it is amenable for large-scale production and industrial applications.

2.1.5.2. Liquid–liquid interfacial assembly. Liquid–liquid interface is another ideal 2D platform for the assembly of GO sheets to form films.^69,71 The self-assembly interface is usually formed between a hydrophobic organic solvent and GO suspension. The driving force for the self-assembly of GO sheets at the liquid–liquid interface is the minimization of interfacial free energy.¹⁰² After the formation of GO membrane at the interface, it can be transferred to target substrate by lift-up technique.

In 2008, Biswas et al. investigated the formation of closely packed monolayer of highly hydrophobic graphene nanosheets at a liquid–liquid interface.¹⁰² They chose chloroform as organic solvent to disperse graphene nanosheets, then, two distinct phases formed after adding water to the mixture. A large interfacial area between the two phases can be created through sonication. Graphene nanosheets preferred to be adsorbed at the interface and to form a large area closely packed monolayer driven by the minimization of interfacial free energy. These graphene films exhibit good properties. The average conductivity of 4 nm thick graphene films is more than 1000 S cm⁻¹ and the transmittance is more than 70% at a wavelength of 550 nm. Chen et al. thought that organic solvent evaporation provides little driving force which results in slow transport of GO sheets to the interface.¹⁰⁰

In 2011, Chen et al. reported an ethanol-assisted self-assembly method which allowed fast preparation of GO-based film at pentane–water interface (Fig. 9a). They studied the effect of ethanol and found that it served as a nonsolvent for GO suspension. Due to the less dispensability of GO in ethanol, the stability of GO suspension is compromised when ethanol is injected into it. Then, GO sheets are pushed to the pentane–water interface and aggregate to form a film. This approach accelerates the liquid–liquid interfacial assembly and opens up opportunities to fabricate new GO-based hybrid films. The transmittance of resulting RGO films can be tuned from 72% to 97% at 550 nm by changing the concentration of GO suspension. The surface resistivity varied from 8.3 to 464.6 kΩ sq⁻¹. In 2015, Arapov et al. adopted oil–water interfacial assembly to prepare LAG thin films.¹⁰³ Firstly, they fabricated expanded graphite via graphite intercalation method followed by thermal expansion. In the experiment, both donor and acceptor-type graphite intercalation compounds were investigated. The resulting material was dispersed in a mixture of isopropanol and propylene glycol with the optimum volume ratio of 3 : 2. Finally, graphene thin films were self-assembled at the oil–water interface by adding graphene suspension to the interface. Those films were transferred to target substrate by lift-up technique (Fig. 9b). The area of the graphene films were more than 50 cm² and potentially more than 1 m². The sheet resistance and transmittance of those films after thermal annealing varied from 10 to 0.8 kΩ sq⁻¹ and 75–45%, respectively.

Fig. 9 (a) Schematic illustration of ethanol-assisted formation of GO thin film at pentane–water interface (ref. 100). (b) Schematic diagram of lift-up transfer technique (ref. 103).

2.1.5.3. Liquid–solid interfacial assembly. The driving force for liquid–solid interfacial assembly is related to the amphiphilic nature and the chemical activity of GO sheets.⁶⁹ There are many types of interactions between GO sheets and solid surface, including covalent bonding, hydrogen bonding, π–π interaction, electrostatic effects and potential chemical reaction. For the first time, Liu et al. obtained paper-like GO films by self-assembly of GO sheets on the surface of aluminum foil.¹⁰⁴ In the experiment, GO aqueous suspension was casted on the aluminum foil followed by drying at 40–60 °C at ambient pressure, and after that a black GO/Al thin film was formed. The formation of paper-like GO thin films are affected by many factors, such as the evaporation, upward gas flow, and the increasing hydrophobicity of GO reduced by Al foil. The structure of GO films can be controlled by adjusting the concentration of GO. The thickness and the area of the film can be controlled by changing the usage amount and concentration of GO suspension.

In summary, the self-assembly of GO sheets at various interfaces is a self-concentration process. The interfacial assembly method has many advantages. It is an effectively method in controlling the homogeneity of film thickness and the layers through layer-by-layer deposition. The films of different sizes can be obtained by changing the size of the interfaces. The resulting films can be easily transferred to various substrates by lift-up technique or dip-coating method. Besides, it is possible to fabricate composite films by dispersing bi- or multi-components in GO suspension. But, there are also many challenges. (1) A deeper understanding of interfacial assembly and the precise control of self-assembly of GO sheets at various interfaces remain to further investigation. (2) It is still very difficult to precisely control the microstructure of self-assembly of GO sheets in its large scale preparation and real applications. (3) More attention should be paid to low cost and to improve scalable assembly techniques as well as effective and nondestructive reduction techniques.

2.1.6. Spray coating. Spray coating method can also be used to prepare thin graphene films on various substrates, but it often results in nonuniform deposition due to aggregation of GO sheets. Spray coating technique (Fig. 10a) mainly includes two methods: electrostatic spray deposition^73,75 and thermal spraying.⁸⁰ Electrostatic spray deposition (ESD) has mass scale capacity, less material loss and high precision for properties control.⁷³ In a typical ESD process, a liquid precursor solution is forced to flow out of a capillary nozzle and atomized at the tip of the nozzle by a high electric potential which is applied between the nozzle and the substrate.⁷⁵ After landing atomized solution on the substrate, a solid or porous film can be formed after the evaporation of solvent. The thickness of resulting films can be controlled by adjusting solution concentration and the time of deposition. The morphology can be controlled by adjusting the solution concentration, flow rate and applied potential. In a typical thermal spraying, molten or semi-molten coating precursor is sprayed onto a substrate and is often accelerated by a heat source towards substrates.⁶⁹ Jeon et al. adopted thermal spraying of GO to prepare thin hole transport layers for polymer solar cells on a hot substrate at various temperatures.⁸⁰

Fig. 10 (a) Schematic diagram of electrostatic spray deposition (ref. 75). (b) A CCG film on 4-inch quartz wafer (ref. 77). (c) Photograph of a 9 × 6 cm² graphene paper (ref. 73).

In 2007, Gilje et al. employed spray coating method to prepare graphene films.⁷⁶ A GO aqueous dispersion was deposited onto preheated SiO₂ substrates followed by chemical reduction using hydrazine. GO sheets can be deposited onto various substrates of any size by this method and highly uniform graphene sheets can be obtained with any desired coverage density. In 2010, for the first time, Pham et al. employed spray deposition of GO–hydrazine dispersion (a mixture of GO and excessive hydrazine monohydrate) to prepare chemically-converted graphene (CCG) films.⁷⁷ The preparation of thin films and reduction of GO to CCG were carried out simultaneously after depositing the mixture on preheated substrate. They successfully prepared a 4-inch, uniform CCG film on quartz wafer (Fig. 10b). The successful preparation of a stable graphene aqueous solution and RGO dispersion are critical for both basic research and practical applications of graphene. In 2011, Pu et al. studied the dispersion of graphene in aqueous solution with different types of surfactants including cationic, anionic, non-ionic and polymer types.⁷⁸ Finally, they found that the best graphene dispersion was prepared by non-ion type surfactant—polyoxyethylene-nonylphenylether at a concentration of 200–300 ppm. The graphene suspension can be spray-coated onto various substrates to form films.

In 2012, Min et al. reported a facile way to fabricate stable RGO dispersions in aqueous or organic media by adjusting the degree of reduction and pH of RGO dispersion.⁷² The degree of reduction determined the surface properties and pH affected the surface potential of the dispersion. Uniform and transparent conductive RGO films can be obtained by spray coating of well dispersed RGO aqueous solution on various substrates. In 2014, Xin et al. adopted electrostatic spray deposition integrated with R2R process to prepare LAG papers.⁷³ The graphene paper was deposited on Al foil and was peeled off from the substrate by immersing into water. This is attributed to their differences in surface properties and wettability. They got a 9 × 6 cm² graphene paper (Fig. 10c), an 18 inch graphene paper strip and a graphene paper roll with width of 4.5 inch, respectively.

In summary, all of these solution processing methods mentioned above can be easily realized on various substrates at low temperature. But the graphene films often suffer from poor quality, for instance, agglomeration of nanosheets and nonuniformity of the film. RGO films have an important feature of low cost due to the ample supply and low price of graphite. Besides, it is easy to achieve mass production of LAG films by these methods. Furthermore, reduction methods, mainly including thermal reduction and chemical reduction, have a great impact on the quality of RGO thin films.⁵⁹

Both thermal reduction and chemical reduction have drawbacks. Thermal reduction is generally operated at high temperature in inert gas or vacuum environment which is incompatible with most device fabrication techniques, especially unsuitable for heat-sensitive substrates such as PET. Chemical reduction often takes a long time and needs to use toxic hydrazine. Recently, Park et al. reported a new reduction method—exposure GO films to flash light irradiation which can be performed in room-temperature ambient conditions and only needs a few milliseconds. There are many reactive sites in GO films including hydroxyl, carboxyl and epoxy groups which can be used for chemical functionalization, but these groups will reduce the conductivity and increase interlayer spacing of films. After reduction treatment, the content of these groups can be reduced and the conductivity of RGO films can also be improved. Among all these methods, rod-coating may be the most effective method which can get 20 × 47 cm² GO film, and it can also precisely control the thickness of GO films from a single layer to tens of layers by adjusting the concentration of the GO solution or the diameter of the rod wire.

2.2. CVD method

At present, CVD is the most popular and most practical method to synthesize LAG films. The growth mechanism of CVD is self-limiting.^115–117 For CVD method, large area and high quality graphene films are produced by the catalytic decomposition of hydrocarbons on a metal (for example, Cu, Ni, Pt) surface which has catalytic activity at high temperatures. Then the films can be transferred to a target substrate by etching away the metal substrate or by R2R transfer techniques. In this review, we mainly discussed two CVD methods: thermal CVD^27,32,37,39 and microwave plasma-enhanced (MWP) CVD.^36,118–135

2.2.1. Thermal CVD. Thermal CVD method is a kind of gas phase reaction, graphene films are prepared by catalytic decomposition of hydrocarbons on a metal surface with catalytic activity at high temperatures. LAG films can be successfully prepared by thermal CVD method. But thermal CVD is often carried out at about 1000 °C that may cause evaporation of Cu and form defect in graphene films. One of the important challenges for large-scale production of graphene films by thermal CVD method is to reduce the reaction temperature.

The first report on CVD synthesis of graphene films was published in 2006.²⁷ Somani et al. chose camphor as the starting material which is natural, environmental friendly and low-cost. They employed thermal CVD method to synthesize PFLG films from camphor pyrolysis on Ni substrate. In this experiment, camphor was firstly evaporated at 180 °C and then pyrolyzed at 700–850 °C with argon as carrier gas. This method did not require vacuum system and the reaction temperature was also relatively low. Finally, they got PFLG films with graphitic structure down to about 20 graphene layers. Since then, CVD has developed by many researchers to a scalable and reliable method for the preparation of LAG. Thermal CVD is often performed on metal substrates, among which Cu and Ni substrates are most commonly used ones.^30,136

In 2009, there was a seminal breakthrough as researchers discovered that the growth of graphene on copper was self-limited to a single layer.³⁰ Since then, the development of CVD method has been very fast. In 2010, Bae et al. successfully prepared 30-inch graphene films by combining CVD method and R2R transfer method (Fig. 11a).³² The sheet resistance of films was as low as ∼125 Ω sq⁻¹ with 97.4% transmittance, and the films exhibited half-integer quantum Hall effect. In the CVD process, firstly, Cu substrate was heated to 1000 °C followed by annealing for 30 min in order to produce high-quality graphene films. Then, in order to achieve film growth, gas mixture of CH₄ and H₂ was flowed with rates of 24 and 8 sccm respectively at 460 mTorr for 30 min. Finally, the sample was cooled to room temperature. In R2R process, thermal release tap was firstly adhered to the resulting graphene film on Cu substrate by applying pressure between two rollers. Then, Cu substrate was etched away by immersing it into Cu etchant. The tap/graphene film was rinsed with deionized water to remove residual etchant. Finally, the tap/graphene film and a target substrate were inserted between the rollers simultaneously. The graphene film was transferred to target substrate by heating the rollers at 90–120 °C.

Fig. 11 (a) Schematic diagram of the roll-to-roll production of graphene film (ref. 32). (b) Schematic diagram of R2R CVD system using selective Joule heating (ref. 37). (c) Schematic illustration of RT-CVD synthesis setup (ref. 39). (d) Photograph of the touch screen fabricated by graphene film (left) and ITO (right) installed in a mobile phone, respectively (ref. 39).

In 2013, Sony Corporation successfully prepared a 100 m-long, 230 mm-wide graphene film by employing R2R CVD method (Fig. 11b).³⁷ The sheet resistance of graphene films was about 500 Ω sq⁻¹ before doping and 200 Ω sq⁻¹ after doping with AuCl₃. It was a big breakthrough for the preparation of LAG films and brought new hope to researchers and enterprisers. They used selective Joule heating of the Cu foil to achieve R2R production of high quality graphene. The Cu foil was wound on the R2R system for 100 m and it can be heated to 1000 °C owing to small emissivity and heat conduction. The first 52 m graphene film was grown at 950 °C by applying a constant direct current J = 82 A mm⁻², the remaining 48 m graphene film was grown at 980 °C (J = 83 A mm⁻²). Ryu et al. adopted hydrogen-free rapid thermal CVD (RT-CVD) method (Fig. 11c) combined with R2R transfer technique to produce homogeneous graphene films.³⁹ Cu foil was mounted vertically in the heating system composed of 24 halogen lamps which can radiate light ranging from visible to infrared wavelengths. A graphite susceptor which can efficiently transform near-infrared light to thermal radiation was adopted because visible wavelength was mostly reflected by the Cu surface. Sufficient energy can be obtained by this method, so that the Cu foil can be heated to about 1000 °C. In the experiment, the optimum temperature was ∼970 °C, CH₄ was used as carbon source and N₂ was used to cool the foils. It only took 40 min throughout the experiment without using hydrogen. The sheet resistance of graphene/PET film was 249 ± 17 Ω sq⁻¹ which indicates excellent electrical performance. They successfully fabricated a capacitive multitouch device which was installed in a mobile phone (Fig. 11d) by using the graphene film.

2.2.2. Microwave plasma-enhanced CVD. It has been more than a decade since the development of microwave plasma-enhanced (MWP) CVD. At the beginning, this method was used to prepare CNTs, carbon nanowalls and diamond films. Later it was optimized for preparation of high-quality graphene films. Compared with thermal CVD method, MWP CVD can be carried out at lower temperatures and it takes only few tens of seconds for the deposition of graphene films. And MWP CVD can be performed on any substrates as long as these substrates can tolerate the processing temperature without the need for catalytic activity. Thus, MWP CVD method has great significance for large-scale industrial production and practical application of graphene films. However, the quality of graphene film prepared by MWP CVD is not good enough. This is due to the poor catalytic decomposition of carbon source and the inefficient carbon dissolution in catalyst layer at low growth temperature.¹²⁴

In the initial stage, few-layer graphene^118,120 or graphene sheets¹¹⁹ can be prepared by using MWP CVD method. Malesevic et al. reported the synthesis of few layer graphene by MWP CVD method in 2008 and further investigated the growth mechanism in 2009.^118,120 Those graphene flakes composed of 4–6 atomic layers were deposited at 700 °C on various substrates. They reported that few-layer graphene was aligned vertically to the surface of substrates and proposed a three-step growth mechanism. In 2015, Park et al. also employed microwave plasma method to prepare vertically standing multi-layer graphene films on various substrates at a temperature ranging from 432 to 225 °C and investigated the growth mechanism.¹³⁴ In 2011, Kim et al. adopted MWP CVD method to prepare centimeter scale monolayer graphene on nickel foil at temperatures ranging from 450 to 750 °C.¹²¹

Kalita et al. has made great contributions in the preparation of graphene films by using surface wave plasma CVD method.^122–124 They employed microwave assisted surface wave plasma CVD (MW-SWP-CVD) method (Fig. 12a) to prepare graphene films on Cu foil at a temperature as low as 240 °C.¹²² The resulting films were continuous and uniform, but there still existed defects. The lowest sheet resistance of the graphene film was 10.03 kΩ sq⁻¹ with 73% transmittance at 550 nm. They thought that this technique can be scaled up for metre-scale deposition of graphene films. Then, in 2012, they used camphor as a solid carbon source to prepare graphene films below 600 °C.¹²⁴ They found that camphor molecules could decompose effectively under the influence of the plasma by dehydrogenation, and then carbon radicals segregated on substrate surface to form film. They fabricated a transparent electrode by transferring graphene film on plastic substrate. The sheet resistance and transmittance of the electrode was 8.23 kΩ sq⁻¹ and 85.4% at 550 nm, respectively. Polydimethylsiloxane can also be used as a solid carbon source to prepare graphene films.¹²⁵ Hiramatsu et al. thought that excess ion bombardment generated by the plasma would damage the deposits on substrates, so they invented a new method which can simply control the position of the plasma ball and reduce ion bombardment on the surface of substrates by installing a grounded molybdenum mesh over the substrate plate.¹²⁶

Fig. 12 (a) Schematic diagram of the MW-SWP-CVD system used for the preparation of graphene films at low temperature (ref. 122). (b) Schematic of R2R CVD apparatus for graphene film deposition (ref. 36).

Yamada et al. has also made many contributions in the preparation of LAG films.^{36,129–133,135} In February 2012, they invented a R2R MWP CVD apparatus (Fig. 12b) equipped with a pair of R2R winder and unwinder system.³⁶ The deposition area was 294 × 480 mm². They successfully prepared few-layer graphene films in width of 294 mm on Cu copper bellow 400 °C. A few months later, they successfully prepared graphene films on Al and Cu substrates at around 300 °C by surface wave plasma CVD (SWP CVD) method.¹²⁹ In 2013, they used poly(methyl methacrylate) as solid carbon source to prepare graphene films on Cu surface at 280 °C.¹³⁵ They obtained graphene films which were consisted of single- or double-layer graphene flakes. The average optical transmittance was about 96% between 400 and 800 nm and the sheet resistances were ranging from 600 to 5000 Ω sq⁻¹. Both electrical properties and domain sizes of graphene films fabricated by MWP CVD were investigated.^131,133 The Hall mobility was up to 100–1000 cm² V⁻¹ s⁻¹ under a low carbon concentration and average domain size was around 30–100 nm. And the crystallinity of the synthesized graphene films can be improved by plasma pretreatment of substrate surface using He/H₂.¹³²

In conclusion, although we can gain LAG films by using CVD method even at lower temperatures, it is difficult to precisely control the number of layers and the quality of graphene films. Besides, the performance of graphene films will be destroyed more or less after transferring.¹³⁷ There are still many challenges for large-scale production of large-area and high-quality graphene films. More detailed information can be obtained in Tables 2 and 3.

Table 2 Comparison of graphene production methods mentioned above

Method	Classification	Representative work	Reference	Concentration/temperature	Merit/demerit
Chemical assembly	Rod-coating	In 2012, Wang et al. prepared a RGO film—20 × 47 cm^2	35	0.05–2.5 mg mL^−1 (GO suspension)	Merits: growth on various substrates; low temperature; simple fabrication procedure. Demerits: agglomerate after solvent evaporation; having functional groups after reduction; difficult to mass production
	Spin-coating	In 2010, Yamaguchi et al. prepared a 300 mm water-scale GO film	50	0.4–5 mg mL^−1
	Dip-coating	In 2013, Cao et al. prepared a paper-like film—20 × 23 cm^2	56	0.1–1 mg mL^−1
	Vacuum-filtration	In 2008, Eda et al. prepared a GO thin film—10 cm^2	29	—
	Interfacial assembly	In 2015, Arapov et al. reported that they can get films more than 50 cm^2	103	—
	Spray-coating	In 2014, Xin et al. prepared a 18 inch graphene paper strip	73	—
CVD	Thermal CVD	In 2010, Samsung—30 inch in diagonal	32	Higher than 1000 °C	Merit: self-limiting growth; gain LAG film; large single-crystal growth; easy to transfer to arbitrary substrates. Demerit: transfer can damage the quality and performance of graphene; somewhat expensive; high growth temperature; difficult to precisely control layers
		In 2013, Sony Corporation—100 m-long, 230 mm-wide	37
	MWP-CVD	In 2011, Kim et al. fabricated LAG-based films—23 × 20 cm^2	34	Lower than 1000 °C
		In 2012, Yamada et al. prepared LAG films—294 × 480 mm^2	36	(200–800 °C)

---

Table 3 Comparison of different methods for the preparation of LAG films

Method	Properties of graphene film	Relative merits
Rod-coating	Continuous, uniform, flexible, good transparency and conductivity	Low-cost, easy to operate, scalable, precise in thickness control, applicable to various substrates
Spin-coating	Continuous, uniform, thin, good transparency and conductivity	Low-cost, easy to operate, thickness of film can be adjusted by changing the concentration of GO suspension and cycles of coating
Dip-coating	Aggregation of GO sheets, non-uniform, few layer, high surface coverage, tunable transmittance	Low-cost, easy to operate, continuous, thickness can be adjusted, large area production can be achieved through mechanization
Vacuum filtration	Inhomogeneous, incomplete transfer, good conductivity and mechanical properties, heterogeneity of the layer number	Simple, cheap, easy to operate, thickness can be tuned
Interfacial assembly	Having wrinkles, good transparency and conductivity, high elastic modulus, diverse microstructure and morphologies	Up-scalable, self-concentration process of GO sheets, films can be easily transferred to various substrates by lift-up technique, control layers by layer-by-layer deposition
Spray coating	Aggregation of GO sheets, non-uniform	Thickness of films can be adjusted by changing solution concentration and the time of deposition
CVD	Large area, high surface coverage, high quality, excellent optical and electrical performance	Self-limiting growth, large-scale industrial production, gain LAG film, large single-crystal growth, easy to transfer to arbitrary substrates

---

3. Applications in membrane separation

Pristine graphene without any defects is impermeable to molecules as small as hydrogen and helium, because the electron density of its aromatic rings is enough to repel atoms to pass through this membrane. However, graphene films with nanopores can be used as molecular separation membrane.¹³⁸ Due to its negligible thickness, chemical stability and high mechanical strength, graphene has become an ideal membrane material that can be used for molecular separation. In this review, we mainly discuss applications of graphene films in water desalination,^139–161 DNA sequencing^162–182 and gas separation.^183–202

3.1. Desalination and power generation

With the increase of population and factories as well as increased water pollution, freshwater resources are becoming increasingly less and desalination is expected to be an important way to get clean water. In recent years, both reverse osmosis (RO) and forward osmosis (FO) technologies have attracted growing attention in desalination. The improvements of the water flux, salt rejection, mechanical strength and anti-pollution of the membranes are critical to the two osmosis technologies. A lot of simulation^{142,152–155,158–161} and experimental researches.^{140,141,143,144,147,150,151,156} indicated that large area, porous graphene films exhibited high water flux and salt rejection for desalination even under low pressure conditions. Gai et al. reported a FO system using fluorinated porous graphene as the FO membrane which has an excellent performance for salt rejection, especially the water flux is about 1.8 × 10⁴ times higher than that of a typical cellulose triacetate membrane.¹⁴² Besides, due to the superior strength (1 TPa) and thickness (one carbon diameter), nanoporous graphene is a potential semi-permeable membrane where a fabric support layer is not necessary.

For the first time, Sint et al. studied ion selective permeability through functionalized graphene nanopores, they found that negatively charged graphene nanopores allowed passage of cations, and vice versa.¹⁵⁸ Experimental results showed that graphene films with nanopore can serve as ionic sieves of high selectivity. In 2010, Suk et al. investigated the transportation of water through a porous graphene film, and compared it with CNT membrane.¹⁵⁵ The pore diameter is critical to whether the graphene film can achieve a higher transport flux than CNT membrane. They found that graphene film can be used as a highly efficient water transporter when the pore diameter is larger than 0.8 nm.

Cohen-Tanugi et al. has done a lot of work in this area.^{152,154,160,161} In 2010, they investigated the effect of pore size, chemical functionalization and applied pressure on film performance.¹⁵² They found that the maximum pore diameter was around 5.5 Å. Hydrophilic functional groups at the edge of the pore could increase the water flux, but the rejection performance was destroyed as the pressure increases. The water permeability is ranging from 39 to 66 L cm⁻² d⁻¹ MPa⁻¹ (with 100% salt rejection) which is 2 orders of magnitude higher than that of the general RO membranes. They also investigated the water permeability of nanoporous graphene at external pressure employed in RO desalination.¹⁵⁴ Ultrahigh water flux of 1041 ± 20 L m⁻² h⁻¹ bar⁻¹ was got at a pressure of less than 100 bars. The mechanical strength of nanoporous graphene was also studied.¹⁶⁰ Both the size and spacing of graphene nanopores and the radius of the pores in support layer can affect the maximum pressure that nanoporous graphene film can withstand. When pores in supporting layer are smaller than 1 μm, nanoporous graphene film can withstand hydraulic pressures of 57 MPa which is nearly 10 times higher than typical RO membrane.

Konatham et al. investigated the effect of pore size in pristine graphene and functionalization of pores on the performance of the membrane in 2013.¹⁵⁹ Results showed that graphene with pristine pores in size of 7.5 Å can be used in water desalination effectively and the functionalized pores can generate large free-energy barriers for salt ions due to electrostatic and steric effects. Nicolaï et al. found that water permeability in graphene oxide framework membranes can be tuned from 5 to 400 L cm⁻² d⁻¹ MPa⁻¹.¹⁵³

Many experimental studies on LAG membrane were carried out and results showed that LAG membrane has excellent desalination performance which is consistent with the simulation results.^{140,141,143,144,147,150,151,156} In 2013, Hu et al. adopted layer-by-layer deposition of GO nanosheets to prepare GO membrane which was cross-linked by 1,3,5-benzenetricarbonyl trichloride on a polydopamine-coated polysulfone support (Fig. 13a).¹⁴⁰ After coating with 5–50 layers of GO, water flux of the composite membrane ranges from 80 to 276 L m⁻² h⁻¹ MPa⁻¹ which is 4–10 times higher than that of most commercial nanofiltration membranes. The rejection of monovalent and divalent salts was only 6–46%, but the rejection rates of Methylene blue and Rhodamine-WT were 46–66% and 93–95%, respectively. Choi et al. found that GO sheets on polyamide membrane can improve the durability of the RO membrane.¹⁴¹ The GO layer can resist degradation of polyamide membrane induced by fouling and chlorination, but will not reduce the water flux.

Fig. 13 (a) Schematic illustration of water transportation through GO/polysulfone composite membrane (ref. 140). (b) Schematic diagram for the preparation of the PSf-rGO/PANI composite membrane (ref. 156). (c) Schematic representation of the mechanism for water molecule transport through GO/PAN composite membrane (ref. 144).

In 2014, Akin et al. prepared a polysulfone-reduced graphene oxide/polyaniline (PSf-rGO/PANI) composite material (Fig. 13b) which was used as desalination membrane.¹⁵⁶ The rGO/PANI composite material was first produced by an enzymatic-reaction-based approach, and then it was dispersed in PSf. After phase inversion polymerization, PSf-rGO/PANI composite membrane was got finally. The maximum salt rejection of the membrane with 0.5% (w/w) rGO/PANI loading is 82% for NaCl at 10 bar pressure, and water flux ranges from 30 to 180 L m⁻² h⁻¹ at pressure of 4–20 bars. Fluorine-terminated graphene membrane can also be used to remove copper and mercury from aqueous solution.¹⁴³ In 2015, Liang et al. prepared a GO/polyacrylonitrile (GO/PAN) composite membrane (Fig. 13c) as pervaporation membrane for desalination.¹⁴⁴ The water flux is up to 65.1 L m⁻² h⁻¹ with 99.8% of salt rejection at 90 °C. There are nanometer-sized intrinsic defects on graphene film prepared by CVD method and large tears may form after transferring. Thus, O'Hern et al. adopted a multiscale sealing procedure to seal these defects and then used high-energy gallium ion bombardment to produce subnanometer pores in the graphene film.¹⁵¹ The water flux of the membrane is 1.41 ± 0.23 L m⁻² h⁻¹ bar⁻¹ and rejection of MgSO₄, allura red and dextran are ∼70%, ∼90% and ∼83%, respectively.

Surwade et al. employed oxygen plasma etching process to create nanometer-sized pores in monolayer graphene film and investigated water desalination performance of the nanoporous film.¹⁵⁰ The salt rejection is nearly 100% and the water flux is up to 10⁶ g m⁻² s⁻¹ at 40 °C under the condition of pressure difference as a driving force. It is 5 orders of magnitude higher than water flux (70 g m⁻² s⁻¹ atm⁻¹) measured using osmotic pressure as driving force. He et al. prepared a GO/polyamide composite membrane for water desalination and investigated antibacterial activity of the membrane.¹⁴⁷ Water flux of the composite membrane with 0.12 wt% of GO increased to 0.219 L μm m⁻² h⁻¹ bar⁻¹ and salt rejection is nearly 90% at 20 bar. The membrane exhibits a strong antibacterial activity, thus it can be used for the purification of biologically active waste slurries. Studies show that nanoporous graphene film is an ideal material for water desalination.^{139,145,146,148,149,157} The thickness of graphene film is only one atomic layer and the surface is smooth and frictionless, so high water permeability and low energy consumption can be achieved. The durability of the composite desalination membrane can also be improved due to high mechanical strength and chemical inertness of graphene.

In fact, there are many difficulties in applying LAG films to desalination field. For example, the cost of graphene film is still very high; water flux of the membrane is limited by the concentration polarization phenomenon, fouling effects and inherent thermodynamic limit of the separation process; precisely controlling the pore sizes and the distribution and creating nanopores in LAG film are difficult; reducing energy consumption and infrastructure costs of existing desalination technologies remains a challenge. Although the development of LAG film in seawater desalination is still immature now, but it has great potential application in the field of desalination and may lead the development of the next generation water treatment membrane.

To date, FO has been one of the most promising approaches for power generation from the mixing of fresh water and seawater in estuaries.^203–207 However, existing FO membranes suffer from low power densities, which seriously hinder its practical applications in power generation. The enhancement of the water flux of a FO membrane is therefore greatly beneficial for power density.²⁰⁸ Internal concentration polarization (ICP) caused by bulky support layers of membranes has been one of the most common and unavoidable phenomena in osmotically driven membrane processes, which is the most important factor in reducing the water flux. Fortunately, porous graphene with its incomparable strength can be used as a FO membrane without the requirement of a support layer, and consequently, exhibits zero ICP. Gai et al. first reported that the water flux of fluorinated porous graphene membrane is about 1.8 × 10⁴ times higher than that of a typical cellulose triacetate membrane.¹⁴² Such high water flux will certainly bring about a very high power density in pressure retarded osmosis power generation. This work may generate potential opportunities for functionalized graphene in FO power generation.

3.2. DNA sequencing

Graphene film with single-atom-thick has excellent mechanical strength and electrical sensitivity which makes it an ideal material for high-resolution, nanopore-based molecule detectors compared with traditional solid-state nanopores made from silicon nitride, aluminum oxide or silicon oxide. Recent researches have found that nanoporous graphene membranes can be used for DNA sequencing. The experimental operation of DNA sequencing is summarized as follows. Firstly, nanometer-scale pores are formed in graphene films using electron beam. Then, the membrane system composed of graphene film, silicon nitride, silicon and other components is placed in a measurement cell which is separated into two compartments and each compartment is filled with ionic solutions electrically contacted with electrodes. Finally, DNA molecules are added into a compartment and a bias voltage is applied to the other compartment, which will cause changes in the current when DNA molecule translocates through the nanopore. A typical graphene nanopore device for DNA sequencing is shown in Fig. 14a.

Fig. 14 (a) A typical graphene nanopore device for DNA sequencing (ref. 164). (b) A graphene membrane with nanopore used as trans-electrode membrane (ionic solutions are not shown) (ref. 163). (c) Schematic of a nanochannel device functionalized by a graphene nanoribbon for DNA sequencing (ref. 177). (d) Schematic diagram of a graphene nanopore with four additional graphene sheets in order to retard the translocation of DNA (ref. 171).

It has been nearly 40 years since the first DNA sequencing by Sanger technique.^209,210 Sanger sequencing is considered to be the first-generation DNA sequencing technique. After 30 years of development, it can achieve read-lengths of up to ∼1000 base pairs (bp) and per-base ‘raw’ accuracies as high as 99.999%.^170,181 The initial draft of the first human genome sequence spent about $300 million and the second genome sequence of nonhuman primate (the rhesus macaque) cost $22 million.¹⁷³ The demand for reducing the cost stimulated the development of second-generation sequencing methods including microelectrophoretic methods, sequencing by hybridization, real-time observation of single molecules and cyclic-array sequencing, etc.^170,181 In order to reduce the cost of genome sequencing to $1000, researchers are trying to find more efficient sequencing technology. Third-generation sequencing methods will take advantage of the characteristic electrical conductance of the four different nucleotides.¹⁷⁸

Due to one-atom thickness and excellent electrical sensitivity, graphene has aroused the interest of researchers in nanopore sequencing. At the beginning, the electronic properties of DNA bases have been investigated by researchers.^168,169 Zwolak et al. investigated charge transport in single-stranded DNA in the direction perpendicular to the backbone axis in 2005.¹⁶⁸ They found that each nucleotide carried a different electrical signature which was independent of the nearest-neighbor nucleotides, and the electric currents of different nucleotides were adenine > guanine ∼ cytosine > thymine (A > G ∼ C > T). Xu et al. also found that different DNA bases showed base-specific electronic signatures.¹⁶⁹ This makes it possible for electronic sequencing of DNA by exploiting different electrical signals of four nucleotides.

There have been many literatures related with DNA sequencing using porous graphene films since 2010.^{163–167,171,172,174–180} Postma chose graphene nanogaps as the electrode as well as the membrane material for rapid sequencing of a single DNA molecule.¹⁷⁵ The translocation speed can be affected by pore size, aspect ratio, DNA–pore interaction strength and the unfolding speed of DNA into the pore. For the first time, Merchant et al. experimentally realized the translocation of DNA through graphene nanopores in diameter of 5 to 10 nm.¹⁶⁴ The blocked currents of DNA translocation through graphene nanopores were higher than that through traditional solid-state nanopores, but ion current noise levels were higher than that of SiN nanopores. They found that current noise can be reduced by coating the graphene film with thin titanium dioxide layer. Schneider et al. also investigated DNA translocation through graphene nanopores with diameter of 10–25 nm drilled by electron beam.¹⁶⁵ The average translocation time is 2.7 ± 0.8 ms for the nonfolded DNA which is higher than that for SiN nanopores (1.2 ± 0.3 ms). Nelson et al. adopted density functional theory for the calculation of electrical current and conductance perpendicular to the DNA backbone resulting from translocation of single-stranded DNA through graphene nanopore.¹⁷⁶ They found that the proposed device was able to discriminate different nucleotides due to significant differences in the conductance spectra. Garaj et al. immersed graphene membrane in an ionic solution, therefore producing a new electrochemical structure of graphene which was called trans-electrode (Fig. 14b).¹⁶³ A nanometer-sized pore in the graphene trans-electrode membrane can increase its ionic conductance by orders of magnitude.

When a DNA molecule passes through the nanopore, the ionic current can be transiently reduced or blocked and changes in ion current can reflect both polymer size and conformation. Then, in 2011, Min et al. designed a novel device composed of a fluidic nanochannel functionalized with a graphene nanoribbon for fast DNA sequencing (Fig. 14c).¹⁷⁷ The π–π interaction between graphene and nucleobases was strong to sustain the stacking structures for current measurements, but it was weak compared with external force, so as not to affect the translocation of DNA through the nanochannel. In 2012, Wells et al. assessed the applicability of graphene nanopores for DNA sequencing by simulation.¹⁶⁷ They found that different DNA nucleotides showed different ionic current blockades which represented the nucleotide type.

In 2013, Lv et al. investigated the impact of the number of layers of a graphene nanopore on DNA translocation by molecular dynamics simulations and potential of mean force (PMF) calculations.¹⁷² They found that the translocation velocity of DNA can be slowed down by reducing the pore size and increasing the thickness of graphene, and the translocation velocity was more sensitive to the number of graphene layers when the size of the nanopores is 2.4 nm. Liang et al. also theoretically studied the translocation of DNA through multilayer graphene nanopores in 2014.¹⁷⁹ They found that multilayer graphene nanopores within 7 layers were superior to monolayer graphene in DNA sequencing and the resolution of DNA sequencing can be improved by modifying the nanopore surface. Kumar et al. investigated the impact of substrate materials and membrane thicknesses on noise characteristics of graphene nanopore devices.¹⁷⁴ Results showed that fewer layer graphene transferred on to SiN_x/quartz substrate possess lower noise, higher blockade current and signal to noise ratio. Avdoshenko et al. designed a new DNA sequencing setup in which four additional graphene plates were deposited on top of the main graphene platform (Fig. 14d).¹⁷¹ This device could retard the DNA translocation and the translocation velocity was estimated to be 17 bp ns⁻¹ for an electric potential intensity of 2 V. Graphene nanopore geometries can also affect DNA sequencing.¹⁶⁶ Zhang et al. found that an axisymmetric nanopore was more suitable for DNA sequencing than an asymmetric nanopore. These studies in DNA sequencing will deepen people's understanding of genome and can enhance the control and treatment of diseases.

Both theoretical and experimental studies have shown that graphene membrane with nanopores is a good platform for DNA sequencing due to its one-atom thickness, high conductivity and mechanical stability. The preparation of sample is simple and cheap because it only needs a small amount of chemical substances.¹⁸² For an ideal nanopore sequencing method, both purified fluorescent reagents and amplified genomic DNA are not required, thereby eliminating enzymes, cloning and amplification steps. Graphene nanopore device has the potential for rapid and inexpensive detection of DNA. However, DNA sequencing by graphene nanopores is currently only in its infancy, there are still a lot of work needs to be done. One of the challenges is lowering the cost of whole-genome sequencing to a few hundred dollars. Reducing the current noise and improving signal-to-noise ratios of nanopore-based graphene films is the second important challenge. Thirdly, how to make graphene nanopores with particular size and shape is a critical issue for DNA sequencing which needs to be resolved as soon as possible. Last but not the least, the mechanism of nanopore sequencing and stochastic motion of the DNA molecule in transit should be systematically studied. It is expected that rapid, long reads and inexpensive DNA sequencing can be achieved by graphene nanopores in the near future.

3.3. Gas separation

Many kinds of materials have been studied in this field of gas separation, for instance, polymers, zeolites, carbon, metal–organic frameworks, etc. Compared with traditional systems, membrane-based system has attracted widespread interest due to low energy cost. The outstanding properties including one-atom thickness, high specific surface area, excellent mechanical strength and chemical inertness make graphene become an ideal membrane material for gas separation. Many researchers have demonstrated that nanoporous graphene films have excellent gas separation performance.^183–202

The application of nanoporous graphene membranes in gas separation is mainly in the stage of theoretical research by molecular dynamics (MD) simulations. In 2009, for the first time, Jiang et al. investigated the selectivity and permeability of porous graphene functionalized with nitrogen and hydrogen for H₂/CH₄ separation by using first principles density functional theory calculations.¹⁸³ The selectivity of nitrogen-functionalized porous graphene for H₂/CH₄ is on the order of 10⁸, while it is 10²³ for all-hydrogen pores. The permeability of H₂ is higher for all-hydrogen pores than that of nitrogen-functionalized pores. In 2010, Blankenburg et al. fabricated a porous graphene membrane by self-assembly of the hexaiodo-substituted macrocycle cyclohexa-m-phenylene (CHP) and investigated its separation performance of atmospheric gases including H₂, O₂, N₂, He, CO₂, etc.¹⁸⁵ They found that the separation performance of this membrane was superior to traditional filters including polymer and silica. For example, the selectivity for H₂/CO₂ is 10¹⁷ and it is 10²³ for H₂/ammonia. And the selectivity for He or H₂ is higher than that of other gases due to the much smaller diffusion barriers of He or H₂. In 2011, Du et al. studied the impact of the pore sizes and shapes of nanoporous graphene on H₂/N₂ separation.¹⁸⁶ They found that the permeation mechanism of H₂ and N₂ through porous graphene was different. The permeation of H₂ was affected by pore size and it showed a linear relationship to pore area, while the permeation of N₂ showed no correlation to the pore area. When the pore size is big enough to allow N₂ to permeate without any restriction, the flow of N₂ is higher than that of H₂. The reason is that N₂ molecules accumulate on the surface of graphene due to the van der Waals interactions between them.

Nanoporous graphene membrane can also be used for separating gas isotopes.^187,188 Hauser et al. adopted nitrogen-functionalized nanoporous graphene membrane to separate ³He from ⁴He.¹⁸⁷ The separation performance can be affect by the choice of tunneling barrier height and gas temperature. Hankel et al. adopted asymmetrically decorated, doped porous graphene membrane for hydrogen isotope separation.¹⁸⁸ One-side decoration of porous graphene membrane with lithium or titanium can provide asymmetric low-energy pathway for gas molecules. They found that high selectivity for deuterium from hydrogen can be achieved by nitrogen-functionalized porous graphene and the necessary flux can be achieved simultaneously at low temperatures.

In 2012, Shan et al. investigated the effect of chemical functionalization of the graphene sheet and pore rim on the separation performance of CO₂/N₂.¹⁸⁹ They found that the functionalization of the graphene sheet can increase the active surface area, thereby improving the adsorption ability of CO₂. The functionalization of the pore rim can increase the selectivity of CO₂. They also found that all-nitrogen functionalized pore showed a higher CO₂ selectivity over N₂ due to an enhanced electrostatic interaction. Koenig et al. adopted ultraviolet-induced oxidative etching technique to drill pores in bilayer graphene film and experimentally studied the separation performance of H₂, CO₂, Ar, N₂, CH₄ and SF₆.¹⁹⁰ The results indicated that the leak rate of various gases varies in the order of H₂ > CO₂ > Ar > N₂ > CH₄. In 2013, Qin et al. introduced octagonal defect structure passivated by hydrogen in graphene membrane whose properties were studied via first principles calculations.¹⁹¹ The membrane showed high selectivity, for example, the selectivity for H₂/CH₄ was on the order of 10²² and the selectivities of CO/CO₂/O₂/N₂ from CH₄ were on the order of 10¹¹ to 10¹⁴. Jungthawan et al. theoretically studied the impact of tensile stress on gas separation properties of porous graphene by using the first-principles density functional theory.¹⁹² They found that diffusion rate can be significantly increased by applying tensile stress. For instance, the diffusion rates of H₂, O₂ and CO₂ in porous graphene were increased up to 7, 13, and 20 orders of magnitude by applying symmetrical stretch strains of 10%. Water-solubility of gases can also affect the separation properties. Lee et al. introduced a water slab between graphene membrane and gas mixture by molecular dynamics simulations, and studied the separation of CO₂ from CO₂/O₂, CO₂/N₂ and CO₂/CH₄ mixtures.¹⁹³ Results showed that CO₂ molecules were selectively transport through the membrane due to high solubility, while the transportation of other gas molecules was limited.

GO membrane can also be used for gas separation because the existance of nanochannels between GO sheets.^194,195 For example, Li et al. found that the separation selectivities of GO membrane for H₂/CO₂ and H₂/N₂ mixtures were 3400 and 900, respectively.¹⁹⁵ In 2014, Lei et al. investigated the effect of charges around the pore on the separation of H₂S/CH₄ mixture.¹⁹⁶ Results showed that the selectivity for H₂S was higher than that for CH₄ due to stronger electrostatic interaction between H₂S and the graphene sheet, while the permeability of H₂S was lower. Liu et al. found that the flux sequence of different gases through nitrogen-functionalized nanoporous graphene membrane was H₂ > CO₂N₂ > Ar > CH₄ by using MD simulations.¹⁹⁷ The results agreed well with the work of Koenig et al. Koolivand et al. fabricated nanocomposite membranes by dispersing different amounts of functionalized GO sheets into polyetherimide (PEI) for the separation of CO₂/CH₄.²⁰⁰ The PEI composite membrane composed of 0.5 wt% aminated GO sheets showed 18% reduction in CO₂ permeation, but the selectivity of CO₂/CH₄ increased 5.14 times compared with pure PEI. They attributed the excellent selectivity to the combination of multiple permselectivity including diffusivity–selectivity, solubility–selectivity and reactivity selectivity.

Researchers found that GO membrane exhibited excellent CO₂ separation performance.^201,202 Kim et al. found that GO membrane under wet-conditions was CO₂-philic and was suitable for the separation of CO₂ from other gases.²⁰¹ Relative humidity (RH) can affect gas permeance in GO membranes. The permeance varied in the order CO₂ > H₂ > CH₄ > O₂ > N₂ at low RH (∼33%), while it was CO₂ > H₂ > O₂ > CH₄ > N₂ at high RH (>68%). Shen et al. reported a method of constructing hydrogen bonds between GO and polymer, which can enable the well stacking of GO sheets. Therefore, molecular-sieving interlayer spacing and straight diffusion pathways can be formed.²⁰² The membrane showed fastest transportation for CO₂ and low relative permeation of other gases. The permeance order was CO₂ > H₂ > CH₄ > N₂. The permeability of CO₂ reached 100 Barrer and the selectivity of CO₂/N₂ was 91.

It can be seen that many studies focus on the separation of H₂ and CO₂ from other gases. The reason is that environmental pollution is getting worse with industrial development and population growth. Environmental management and looking for clean energy have become more and more important. H₂ is a kind of clean energy and only water can be produced after combustion. The main environmental effect of CO₂ is global warming. Therefore, the separation of H₂ and CO₂ from other gases is critical for the development of a new clean energy technique and environmental management. Studies show that nanoporous graphene functionalized with hydrogen or nitrogen has exhibited superior separation performance to traditional membrane system. But most of the researches are realized by MD simulation. For experimental studies, the preparation of nanoporous graphene with uniform pore sizes and distribution remains as a significant challenge.

4. Global investment and industrial production of LAG

Since Geim and Novoselov won the Nobel Prize due to their great contributions to the field of graphene, many countries have made substantial investments (Fig. 15) and have upgraded graphene research to strategic height.²¹¹

Fig. 15 Investments in the field of graphene in different countries (ref. 211–218).

European Commission ranked graphene as one of the “Future and Emerging Technologies Flagship Projects” in January 2013 and will provide €1 billion in funding in the next 10 years.^212,213 More than 140 institutions from 23 countries had participated in the Graphene Flagship program at 2014. Graphene Flagship program will be divided into two parts: initial stage and stable stage. There will be €54 million of funding at the initial stage which is from October 1, 2013 to March 31, 2016 and the budget will be distributed among 126 academic and industrial research groups. In the stable stage (from Apr. 2016), it is expected to finance €50 million per year.

British identified graphene as one of the four emerging technologies in 2011. George Osborne, the UK's Chancellor of the Exchequer, announced that Britain would invest £50 million for graphene research and would take graphene from the British laboratory to the British factory floor.^214,215 Among them, £38 million was used to set up a global hub for the research on the material at the University of Manchester. Many universities have also received funding, for instance, Cambridge University received more than £12 million for research in advanced flexible and light-based electronics. Imperial College London received about £4.5 million to research the use of graphene to get lightning-resistant and tougher aircraft components. In 2012, Britain appended £21.5 million to explore the commercialization of graphene. British announced that they will focus on the development of graphene and the British National Laboratory of Graphene will be completed in 2015.

South Korea funded $18.7 million for graphene project from 2007 to 2009. In 2013, Ministry of Trade, Industry and Energy of Korea announced that it will integrate the strength of Korean domestic research institutions and enterprises to promote commercial development of graphene and form graphene league in domestic. The graphene league included 41 research institutes and six companies. The government invested $42.3 million. South Korea planned to invest $200 million in graphene research from 2012 to 2018. Besides, many private companies like Samsung have made a significant contribution to it.²¹¹

Singapore is also working on graphene.^216,217 In August of 2010, the Graphene Research Centre (GRC) at the National University of Singapore (NUS) was established with a start-up fund of $40 million from NUS. This centre aims to become a world leader in the field of 2D materials and plans to be a fully self-contained institute where experiment, theory and applications can be carried out simultaneously. GRC has also been supported by the National Research Foundation (NRF) of Singapore. It can obtain financing through competitive grant applications. For example, Prof. Castro Neto was awarded a $10 million Competitive Research Program (CRP) grant to study 2D crystals beyond graphene. In 2014, NRF of Singapore awarded NUS with a $50 million grant over the next 10 years in order to support the operational costs based on 2D materials. GRC also organizes a very important graphene conference—Recent Progress in Graphene Research which can bring together the most important Asian researchers in research of 2D materials and they can discuss the latest research progress.

The rise of the graphene has aroused great attention of Chinese government. Although China started late in the industrial production of LAG films, its development is very rapid. From 2007 to 2013, National Natural Science Foundation of China funded 1096 projects on graphene. In 2013, China invested 210 million to promote the development of graphene industry. China has successfully achieved the preparation of large single crystals,^218,219 pure monolayers,²²⁰ three-dimensional graphene networks,²²¹ and has invented a non-destructive electrochemical bubbling transfer method.²¹⁸ Many other countries (America, Japan) have also invested in the graphene industry which will greatly promote the development of graphene industry.

Many enterprises have devoted to the industrial production of LAG films (Table 4).³³ Bluestone Global Tech (in America) was founded in New York in 2011. This company is committed to the development of graphene technology and produces graphene films on various substrates (for example, Cu, PET or SiO₂/Si). The largest films prepared on PET and Cu are 8 × 10 inch² and 24 × 300 inch², respectively. Sony (in Japan) developed a new method in 2012 using selective Joule heating of the copper foil to achieve the production of graphene films with a area of 100 m × 230 mm. Graphene Square (in Korea) which was established in January 2012 has developed a device for the fabrication of LAG films in size of A4. The largest graphene film they prepared and transferred to PET substrate is 30 inch (700 × 400 mm²).

Table 4 Synthesis method, main product, basic properties and applications of several graphene companies

Company	Synthesis method	Main product	Basic properties	Applications
Bluestone Global Tech	CVD	Graphene films on Cu, SiO2/Si, PET	Maximum size: 24 × 300 inch^2 on Cu, 8 × 10 inch^2 on PET; sheet resistance: <150 Ω sq^−1 at 95% transmittance	Field-effect transistors; touch panels
Sony Corporation	CVD	Graphene films on Cu, SiO2/Si, PET	Maximum size: 100 m × 230 mm, sheet resistance: <150 Ω sq^−1	Transparent conductive film; touch panels
Graphene Square	CVD	Films on Ni, Cu, PET, glass, SiO2	Maximum size: 30 inch, sheet resistance: 125 Ω sq^−1, transmittance: >97%	Touch screens
2D Carbon Tech	CVD	On Cu, SiO2/Si, PET, glass	Maximum size: 45 × 55 cm^2 on PET, monolayer >90%, sheet resistance: 200–400 Ω sq^−1, transmittance: >85%	Touch panels
Power Booster	CVD	Films on Cu, PET	Maximum size: 7.5 m^2 on Cu, carrier mobility: >3500 cm^2 V^−1 s^−1, sheet resistance: <100 Ω sq^−1	Touch panels
Samsung	CVD	Films on Cu, PET, SiO2/Si	Maximum size: 30 inch, sheet resistance: 125 Ω sq^−1, half-integer quantum Hall effect, transmittance: 97.4%	Transparent electrodes; touch screens

---

Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences successfully prepared their first piece of monolayer graphene film in domestic in size of 15-inch on Cu substrate. The film can be transferred entirely to the flexible PET substrate and other substrates. 2D Carbon Tech was established in December 2011. The largest graphene film they fabricated on PET is 450 × 550 mm². In January 2012, the company announced the first capacitive touch panel based on LAG. Power Booster was co-founded by China-US scientific research team in 2009. The company realized the production of 60-inch (1 m²) graphene films in 2012, 7.5 m² films on copper and 3–20 inch touch panels in 2013.

5. Challenges and outlook

LAG has grown considerably in both area of applications and preparations in recent years. Many enterprises can produce high-quality LAG films for now, and these films can be used to fabricate flexible and wearable touch screens which can be applied in smart phones or tablets. Recently, ‘graphene phone’ has appeared on the market. Besides, LAG has exhibited great potentialities in the applications of transparent conducting electrodes, transistors, ultrathin electromagnetic shielding film, highly selective single molecule gas detection, desalination, DNA sequencing and osmotic energy production, etc. LAG is likely to be the next generation of “miracle material” in the future, but it may take a long time to be widely applied owing to many challenges, such as, the cost, the production volume, quality and transfer of LAG.^33,222 The cost of graphene is a critical factor which can speed up or slow down the development of LAG. The price of per square centimeter graphene has fallen significantly compared with its initial price in the past few years, but it's still significantly higher than the price of traditional materials, for example, in 2008, the price of mechanically exfoliated graphene crystallite estimated by Geim was $1000 (ref. 223) and the price of graphene films in area of 10 × 10 cm² produced by 2D Carbon Tech in 2014 is ¥1000, while the price of ITO conductive film is about 300–450 ¥ m⁻². So the cost of graphene is still a big challenge and the development of graphene films will be resisted by the related manufacturers in some degree. Another challenge is how to achieve controllability of mass-production techniques in producing LAG films which are uniform, continuous, defect-free and the number of layers can be controlled. At the same time, it should realize the growth of LAG films on arbitrary substrates at low temperatures and realize intact, clean transfer of graphene films from metal substrates without destroying the substrates. How to achieve the application of LAG films in daily life as soon as possible is another challenge. Although the development of LAG faces many challenges, its development prospect is promising.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (51473097), (51003067), the Opening Project of State Key Laboratory of Polymer Materials Engineering (Sichuan University) (sklpme2014-3-14), and the Fundamental Research Funds for the Central Universities (2012SCU04A03).

References

1. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
2. C. N. R. Rao, A. K. Sood, K. S. Subrahmanyam and A. Govindaraj, Angew. Chem., Int. Ed., 2009, 48, 7752–7777.
3. H. W. Kroto, J. R. Heath, S. C. O'Brien, R. F. Curl and R. E. Smalley, Nature, 1985, 318, 162–163.
4. H. Shioyama and T. Akita, Carbon, 2003, 41, 179–181.
5. H. E. Lim, Y. Miyata, R. Kitaura, Y. Nishimura, Y. Nishimoto, S. Irle, J. H. Warner, H. Kataura and H. Shinohara, Nat. Commun., 2013, 4, 2548.
6. S. Kruss, A. J. Hilmer, J. Zhang, N. F. Reuel, B. Mu and M. S. Strano, Adv. Drug Delivery Rev., 2013, 65, 1933–1950.
7. D. Yu and F. Liu, Nano Lett., 2007, 7, 3046–3050.
8. M. Zhang and J. Li, Mater. Today, 2009, 12, 12–18.
9. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
10. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson, I. V. Grigorieva, S. V. Dubonos and A. A. Firsov, Nature, 2005, 438, 197–200.
11. Y. Zhang, Y. W. Tan, H. L. Stormer and P. Kim, Nature, 2005, 438, 201–204.
12. Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.
13. C. Rao, K. Subrahmanyam, H. R. Matte, B. Abdulhakeem, A. Govindaraj, B. Das, P. Kumar, A. Ghosh and D. J. Late, Sci. Technol. Adv. Mater., 2010, 11, 054502.
14. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385–388.
15. A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao and C. N. Lau, Nano Lett., 2008, 8, 902–907.
16. F. Schedin, A. K. Geim, S. V. Morozov, E. W. Hill, P. Blake, M. I. Katsnelson and K. S. Novoselov, Nat. Mater., 2007, 6, 652–655.
17. D. C. Elias, R. R. Nair, T. M. G. Mohiuddin, S. V. Morozov, P. Blake, M. P. Halsall, A. C. Ferrari, D. W. Boukhvalov, M. I. Katsnelson, A. K. Geim and K. S. Novoselov, Science, 2009, 323, 610–613.
18. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. R. Peres and A. K. Geim, Science, 2008, 320, 1308.
19. J. H. Chen, C. Jang, S. Xiao, M. Ishigami and M. S. Fuhrer, Nat. Nanotechnol., 2008, 3, 206–209.
20. C. Rao, K. Gopalakrishnan and A. Govindaraj, Nano Today, 2014, 9, 324–343.
21. K. S. Subrahmanyam, L. S. Panchakarla, A. Govindaraj and C. N. R. Rao, J. Phys. Chem. C, 2009, 113, 4257–4259.
22. U. A. Palnitkar, R. V. Kashid, M. A. More, D. S. Joag, L. S. Panchakarla and C. N. R. Rao, Appl. Phys. Lett., 2010, 97, 063102.
23. D. J. Late, A. Ghosh, K. S. Subrahmanyam, L. S. Panchakarla, S. B. Krupanidhi and C. N. R. Rao, Solid State Commun., 2010, 150, 734–738.
24. C. N. R. Rao, K. S. Subrahmanyam, H. S. S. R. Matte, B. Abdulhakeem, A. Govindaraj, D. Barun, K. Prashant, G. Anupama and J. L. Dattatray, Sci. Technol. Adv. Mater., 2010, 11, 054502.
25. C. N. R. Rao, K. S. Subrahmanyam, H. S. S. Ramakrishna Matte and A. Govindaraj, Mod. Phys. Lett. B, 2011, 25, 427–451.
26. E. P. Randviir, D. A. C. Brownson and C. E. Banks, Mater. Today, 2014, 17, 426–432.
27. P. R. Somani, S. P. Somani and M. Umeno, Chem. Phys. Lett., 2006, 430, 56–59.
28. A. N. Obraztsov, E. A. Obraztsova, A. V. Tyurnina and A. A. Zolotukhin, Carbon, 2007, 45, 2017–2021.
29. G. Eda, G. Fanchini and M. Chhowalla, Nat. Nanotechnol., 2008, 3, 270–274.
30. X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo and R. S. Ruoff, Science, 2009, 324, 1312–1314.
31. X. Li, Y. Zhu, W. Cai, M. Borysiak, B. Han, D. Chen, R. D. Piner, L. Colombo and R. S. Ruoff, Nano Lett., 2009, 9, 4359–4363.
32. S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. Ri Kim, Y. I. Song, Y.-J. Kim, K. S. Kim, B. Ozyilmaz, J.-H. Ahn, B. H. Hong and S. Iijima, in Nature Nanotechnology, Nature Publishing Group, 2010, vol. 5, pp. 574–578.
33. W. Ren and H. M. Cheng, Nat. Nanotechnol., 2014, 9, 726–730.
34. J. Kim, M. Ishihara, Y. Koga, K. Tsugawa, M. Hasegawa and S. Iijima, Appl. Phys. Lett., 2011, 98, 091502.
35. J. Wang, M. Liang, Y. Fang, T. Qiu, J. Zhang and L. Zhi, Adv. Mater., 2012, 24, 2874–2878.
36. T. Yamada, M. Ishihara, J. Kim, M. Hasegawa and S. Iijima, Carbon, 2012, 50, 2615–2619.
37. T. Kobayashi, M. Bando, N. Kimura, K. Shimizu, K. Kadono, N. Umezu, K. Miyahara, S. Hayazaki, S. Nagai, Y. Mizuguchi, Y. Murakami and D. Hobara, Appl. Phys. Lett., 2013, 102, 023112.
38. J. Liu, P. Li, Y. Chen, Z. Wang, J. He, H. Tian, F. Qi, B. Zheng, J. Zhou, W. Lin and W. Zhang, J. Alloys Compd., 2014, 615, 415–418.
39. J. Ryu, Y. Kim, D. Won, N. Kim, J. S. Park, E.-K. Lee, D. Cho, S.-P. Cho, S. J. Kim, G. H. Ryu, H.-A. S. Shin, Z. Lee, B. H. Hong and S. Cho, ACS Nano, 2013, 8, 950–956.
40. G. Eda and M. Chhowalla, Adv. Mater., 2010, 22, 2392–2415.
41. R. Muñoz and C. Gómez-Aleixandre, Chem. Vap. Deposition, 2013, 19, 297–322.
42. Y. Yang, Z. Liu, Z. Yin, Z. Du, L. Xie, M. Yi, J. Liu and W. Huang, RSC Adv., 2014, 4, 55671–55676.
43. J. Wang, Z. Fang, H. Zhu, B. Gao, S. Garner, P. Cimo, Z. Barcikowski, A. Mignerey and L. Hu, Thin Solid Films, 2014, 556, 13–17.
44. J. Xie, H. Wang, H. Bai, P. Yang, M. Shi, P. Guo, C. Wang, W. Yang and H. Song, ACS Appl. Mater. Interfaces, 2012, 4, 2891–2896.
45. S. K. Kalpathy, L. F. Francis and S. Kumar, J. Colloid Interface Sci., 2012, 383, 155–166.
46. H. A. Becerril, J. Mao, Z. Liu, R. M. Stoltenberg, Z. Bao and Y. Chen, ACS Nano, 2008, 2, 463–470.
47. J. T. Robinson, M. Zalalutdinov, J. W. Baldwin, E. S. Snow, Z. Wei, P. Sheehan and B. H. Houston, Nano Lett., 2008, 8, 3441–3445.
48. J. T. Robinson, F. K. Perkins, E. S. Snow, Z. Wei and P. E. Sheehan, Nano Lett., 2008, 8, 3137–3140.
49. S. Pang, H. N. Tsao, X. Feng and K. Müllen, Adv. Mater., 2009, 21, 3488–3491.
50. H. Yamaguchi, G. Eda, C. Mattevi, H. Kim and M. Chhowalla, ACS Nano, 2010, 4, 524–528.
51. J. Wu, M. Agrawal, H. A. Becerril, Z. Bao, Z. Liu, Y. Chen and P. Peumans, ACS Nano, 2010, 4, 43–48.
52. H. Chang, G. Wang, A. Yang, X. Tao, X. Liu, Y. Shen and Z. Zheng, Adv. Funct. Mater., 2010, 20, 2893–2902.
53. H. B. Sun, J. Yang, Y. Z. Zhou, N. Zhao and D. Li, Mater. Technol., 2014, 29, 14–20.
54. X. Wang, L. Zhi and K. Müllen, Nano Lett., 2008, 8, 323–327.
55. H. G. Jeon, Y. H. Huh, S. H. Yun, K. W. Kim, S. S. Lee, J. Lim, K.-S. An and B. Park, J. Mater. Chem. C, 2014, 2, 2622–2634.
56. X. Cao, D. Qi, S. Yin, J. Bu, F. Li, C. F. Goh, S. Zhang and X. Chen, Adv. Mater., 2013, 25, 2957–2962.
57. G. Eda, Y.-Y. Lin, S. Miller, C.-W. Chen, W.-F. Su and M. Chhowalla, Appl. Phys. Lett., 2008, 92, 233305.
58. S. Y. Oh, S. H. Kim, Y. S. Chi and T. J. Kang, Appl. Surf. Sci., 2012, 258, 8837–8844.
59. S. H. Park and H. S. Kim, Nanotechnology, 2015, 26, 205601.
60. M. Yang, S. Ko, J. S. Im and B. G. Choi, J. Power Sources, 2015, 288, 76–81.
61. Y.-K. Kim and D.-H. Min, Chem.–Eur. J., 2015, 21, 7217–7223.
62. P. Du, X. Hu, C. Yi, H. C. Liu, P. Liu, H.-L. Zhang and X. Gong, Adv. Funct. Mater., 2015, 25, 2420–2427.
63. D. D. Nguyen, Y.-T. Lai and N.-H. Tai, Diamond Relat. Mater., 2014, 47, 1–6.
64. H. X. Kong, Curr. Opin. Solid State Mater. Sci., 2013, 17, 31–37.
65. L. Huang, Y. Wang, J. Tang, H. Wang, H. Wang, J. Qiu, Y. Wang, J. Liu and J. Liu, Int. J. Electrochem. Sci., 2012, 7, 11068–11075.
66. U. Khan, I. O'Connor, Y. K. Gun'ko and J. N. Coleman, Carbon, 2010, 48, 2825–2830.
67. J. Geng and H.-T. Jung, J. Phys. Chem. C, 2010, 114, 8227–8234.
68. M. Kim, W. J. Hyun, S. C. Mun and O. O. Park, J. Phys.: Conf. Ser., 2015, 602, 012027.
69. J.-J. Shao, W. Lv and Q.-H. Yang, Adv. Mater., 2014, 26, 5586–5612.
70. F. Kim, L. J. Cote and J. Huang, Adv. Mater., 2010, 22, 1954–1958.
71. J. Kim, L. J. Cote, F. Kim, W. Yuan, K. R. Shull and J. Huang, J. Am. Chem. Soc., 2010, 132, 8180–8186.
72. K. Min, T. H. Han, J. Kim, J. Jung, C. Jung, S. M. Hong and C. M. Koo, J. Colloid Interface Sci., 2012, 383, 36–42.
73. G. Xin, H. Sun, T. Hu, H. R. Fard, X. Sun, N. Koratkar, T. Borca-Tasciuc and J. Lian, Adv. Mater., 2014, 26, 4521–4526.
74. R. K. Layek, A. K. Das, M. U. Park, N. H. Kim and J. H. Lee, J. Mater. Chem. A, 2014, 2, 12158–12161.
75. M. Beidaghi, Z. Wang, L. Gu and C. Wang, J. Solid State Electrochem., 2012, 16, 3341–3348.
76. S. Gilje, S. Han, M. Wang, K. L. Wang and R. B. Kaner, Nano Lett., 2007, 7, 3394–3398.
77. V. H. Pham, T. V. Cuong, S. H. Hur, E. W. Shin, J. S. Kim, J. S. Chung and E. J. Kim, Carbon, 2010, 48, 1945–1951.
78. N.-W. Pu, C.-A. Wang, Y.-M. Liu, Y. Sung, D.-S. Wang and M.-D. Ger, J. Taiwan Inst. Chem. Eng., 2012, 43, 140–146.
79. X. Wu, F. Li, W. Wu and T. Guo, Appl. Surf. Sci., 2014, 295, 214–218.
80. Y.-J. Jeon, J.-M. Yun, D.-Y. Kim, S.-I. Na and S.-S. Kim, Appl. Surf. Sci., 2014, 296, 140–146.
81. Y. Meng, X.-B. Xu, H. Li, Y. Wang, E.-X. Ding, Z.-C. Zhang and H.-Z. Geng, Carbon, 2014, 70, 103–110.
82. B. Dan, G. C. Irvin and M. Pasquali, ACS Nano, 2009, 3, 835–843.
83. L. Hu, H. S. Kim, J.-Y. Lee, P. Peumans and Y. Cui, ACS Nano, 2010, 4, 2955–2963.
84. C.-H. Liu and X. Yu, Nanoscale Res. Lett., 2011, 6, 75.
85. Y. Lou, G. Liu, S. Liu, J. Shen and W. Jin, Appl. Surf. Sci., 2014, 307, 631–637.
86. H. Qiu, L. Zeng, T. Lan, X. Ding and M. Wei, J. Mater. Chem. A, 2015, 3, 1619–1623.
87. P. Meenakshi, R. Karthick, M. Selvaraj and S. Ramu, Sol. Energy Mater. Sol. Cells, 2014, 128, 264–269.
88. Y. A. Samad, Y. Li, S. M. Alhassan and K. Liao, ACS Appl. Mater. Interfaces, 2015, 7, 9195–9202.
89. A. A. Shabaneh, S. H. Girei, P. T. Arasu, W. B. W. A. Rahman, A. A. A. Bakar, A. Z. Sadek, H. N. Lim, N. M. Huang and M. H. Yaacob, Opt. Commun., 2014, 331, 320–324.
90. K. Goh, L. Setiawan, L. Wei, R. Si, A. G. Fane, R. Wang and Y. Chen, J. Membr. Sci., 2015, 474, 244–253.
91. C. Zhao, X. Lu, C. Zanden and J. Liu, Biomed. Mater., 2015, 10, 015019.
92. F. A. Jumeri, H. N. Lim, Z. Zainal, N. M. Huang and A. Pandikumar, Ceram. Int., 2014, 40, 15159–15165.
93. Y. Niu, X. Zhang, W. Pan, J. Zhao and Y. Li, RSC Adv., 2014, 4, 7511–7515.
94. Y. Niu, J. Zhao, X. Zhang, X. Wang, J. Wu, Y. Li and Y. Li, Appl. Phys. Lett., 2012, 101, 181903.
95. Y. F. Bai, Y. F. Zhang, A. W. Zhou, H. W. Li, Y. Zhang, J. H. Luong and H. F. Cui, Nanotechnology, 2014, 25, 405601.
96. J.-J. Shao, W. Lv, Q. Guo, C. Zhang, Q. Xu, Q.-H. Yang and F. Kang, Chem. Commun., 2012, 48, 3706–3708.
97. J. Zhao, S. Pei, W. Ren, L. Gao and H.-M. Cheng, ACS Nano, 2010, 4, 5245–5252.
98. P. Zhang, X. Zhang, S. Zhang, X. Lu, Q. Li, Z. Su and G. Wei, J. Mater. Chem. B, 2013, 1, 6525–6531.
99. S. Z. Moghaddam, S. Sabury and F. Sharif, RSC Adv., 2014, 4, 8711–8719.
100. F. Chen, S. Liu, J. Shen, L. Wei, A. Liu, M. B. Chan-Park and Y. Chen, Langmuir, 2011, 27, 9174–9181.
101. J. Xin, B. Xie, Y. Li, J. Shang, Y. Qiu, L. Liu, S. Zhao, L. Fan and R. Zhang, Compos. Interfaces, 2014, 21, 623–630.
102. S. Biswas and L. T. Drzal, Nano Lett., 2009, 9, 167–172.
103. K. Arapov, A. Goryachev, G. d. With and H. Friedrich, Synth. Met., 2015, 201, 67–75.
104. Q. Liu, M. He, X. Xu, L. Zhang and J. Yu, New J. Chem., 2013, 37, 181–187.
105. T. Yang, J. Yang, L. Shi, E. Mader and Q. Zheng, RSC Adv., 2015, 5, 23650–23657.
106. X. Lin, J. Jia, N. Yousefi, X. Shen and J.-K. Kim, J. Mater. Chem. C, 2013, 1, 6869–6877.
107. Q.-b. Zheng, L.-f. Shi and J.-h. Yang, Trans. Nonferrous Met. Soc. China, 2012, 22, 2504–2511.
108. Q. Zheng, W. H. Ip, X. Lin, N. Yousefi, K. K. Yeung, Z. Li and J.-K. Kim, ACS Nano, 2011, 5, 6039–6051.
109. D. D. Kulkarni, I. Choi, S. S. Singamaneni and V. V. Tsukruk, ACS Nano, 2010, 4, 4667–4676.
110. V. Divakar Botcha, G. Singh, P. K. Narayanam, D. S. Sutar, S. S. Talwar, R. S. Srinivasa and S. S. Major, AIP Conf. Proc., 2012, 1447, 327–328.
111. X. Li, G. Zhang, X. Bai, X. Sun, X. Wang, E. Wang and H. Dai, Nat. Nanotechnol., 2008, 3, 538–542.
112. T. H. Han, J. Nanosci. Nanotechnol., 2015, 15, 1191–1194.
113. L. J. Cote, F. Kim and J. Huang, J. Am. Chem. Soc., 2008, 131, 1043–1049.
114. L. J. Cote, J. Kim, Z. Zhang, C. Sun and J. Huang, Soft Matter, 2010, 6, 6096–6101.
115. X. Zhang, H. Li and F. Ding, Adv. Mater., 2014, 26, 5488–5495.
116. B.-J. Lee and G.-H. Jeong, Appl. Phys. A: Mater. Sci. Process., 2014, 116, 15–24.
117. K. Yan, L. Fu, H. Peng and Z. Liu, Acc. Chem. Res., 2013, 46, 2263–2274.
118. A. Malesevic, R. Vitchev, K. Schouteden, A. Volodin, L. Zhang, G. V. Tendeloo, A. Vanhulsel and C. V. Haesendonck, Nanotechnology, 2008, 19, 305604.
119. A. Dato, V. Radmilovic, Z. Lee, J. Phillips and M. Frenklach, Nano Lett., 2008, 8, 2012–2016.
120. R. Vitchev, A. Malesevic, R. H. Petrov, R. Kemps, M. Mertens, A. Vanhulsel and C. Van Haesendonck, Nanotechnology, 2010, 21, 095602.
121. Y. Kim, W. Song, S. Y. Lee, C. Jeon, W. Jung, M. Kim and C. Y. Park, Appl. Phys. Lett., 2011, 98, 263106–2631063.
122. G. Kalita, K. Wakita and M. Umeno, RSC Adv., 2012, 2, 2815–2820.
123. G. Kalita, M. S. Kayastha, H. Uchida, K. Wakita and M. Umeno, RSC Adv., 2012, 2, 3225–3230.
124. G. Kalita, S. Sharma, K. Wakita, M. Umeno, Y. Hayashi and M. Tanemura, Phys. Status Solidi A, 2012, 209, 2510–2513.
125. C. Wang, Y. Zhou, L. He, T.-W. Ng, G. Hong, Q.-H. Wu, F. Gao, C.-S. Lee and W. Zhang, Nanoscale, 2013, 5, 600–605.
126. M. Hiramatsu, M. Naito, H. Kondo and M. Hori, Jpn. J. Appl. Phys., 2013, 52, 01AK04.
127. A. Kumar, A. A. Voevodin, D. Zemlyanov, D. N. Zakharov and T. S. Fisher, Carbon, 2012, 50, 1546–1553.
128. A. Kumar, A. A. Voevodin, R. Paul, I. Altfeder, D. Zemlyanov, D. N. Zakharov and T. S. Fisher, Thin Solid Films, 2013, 528, 269–273.
129. T. Yamada, J. Kim, M. Ishihara and M. Hasegawa, J. Phys. D: Appl. Phys., 2013, 46, 063001.
130. T. Yamada, M. Ishihara and M. Hasegawa, Thin Solid Films, 2013, 532, 89–93.
131. Y. Okigawa, K. Tsugawa, T. Yamada, M. Ishihara and M. Hasegawa, Appl. Phys. Lett., 2013, 103, 153106.
132. R. Kato, K. Tsugawa, T. Yamada, M. Ishihara and M. Hasegawa, Jpn. J. Appl. Phys., 2014, 53, 015505.
133. Y. Okigawa, R. Kato, T. Yamada, M. Ishihara and M. Hasegawa, Carbon, 2015, 82, 60–66.
134. H. J. Park, B. W. Ahn, T. Y. Kim, J. W. Lee, Y. H. Jung, Y. S. Choi, Y. I. Song and S. J. Suh, Thin Solid Films, 2015, 587, 8–13.
135. T. Yamada, M. Ishihara and M. Hasegawa, Appl. Phys. Express, 2013, 6, 115102.
136. Q. Yu, J. Lian, S. Siriponglert, H. Li, Y. P. Chen and S.-S. Pei, Appl. Phys. Lett., 2008, 93, 113103.
137. K. E. Whitener and P. E. Sheehan, Diamond Relat. Mater., 2014, 46, 25–34.
138. H. Huang, Y. Ying and X. Peng, J. Mater. Chem. A, 2014, 2, 13772–13782.
139. T. Humplik, J. Lee, S. C. O'Hern, B. A. Fellman, M. A. Baig, S. F. Hassan, M. A. Atieh, F. Rahman, T. Laoui, R. Karnik and E. N. Wang, Nanotechnology, 2011, 22, 292001.
140. M. Hu and B. Mi, Environ. Sci. Technol., 2013, 47, 3715–3723.
141. W. Choi, J. Choi, J. Bang and J. H. Lee, ACS Appl. Mater. Interfaces, 2013, 5, 12510–12519.
142. J.-G. Gai, X.-L. Gong, W.-W. Wang, X. Zhang and W.-L. Kang, J. Mater. Chem. A, 2014, 2, 4023.
143. J. Azamat, A. Khataee and S. W. Joo, J. Mol. Graphics Modell., 2014, 53, 112–117.
144. B. Liang, W. Zhan, G. Qi, S. Lin, Q. Nan, Y. Liu, B. Cao and K. Pan, J. Mater. Chem. A, 2015, 3, 5140–5147.
145. K. A. Mahmoud, B. Mansoor, A. Mansour and M. Khraisheh, Desalination, 2015, 356, 208–225.
146. P. S. Goh and A. F. Ismail, Desalination, 2015, 356, 115–128.
147. L. He, L. F. Dumée, C. Feng, L. Velleman, R. Reis, F. She, W. Gao and L. Kong, Desalination, 2015, 365, 126–135.
148. A. Subramani and J. G. Jacangelo, Water Res., 2015, 75, 164–187.
149. S. Daer, J. Kharraz, A. Giwa and S. W. Hasan, Desalination, 2015, 367, 37–48.
150. S. P. Surwade, S. N. Smirnov, I. V. Vlassiouk, R. R. Unocic, G. M. Veith, S. Dai and S. M. Mahurin, Nat. Nanotechnol., 2015, 10, 459–464.
151. S. C. O'Hern, D. Jang, S. Bose, J. C. Idrobo, Y. Song, T. Laoui, J. Kong and R. Karnik, Nano Lett., 2015, 15, 3254–3260.
152. D. Cohen-Tanugi and J. C. Grossman, Nano Lett., 2012, 12, 3602–3608.
153. A. Nicolaï, B. G. Sumpter and V. Meunier, Phys. Chem. Chem. Phys., 2014, 16, 8646–8654.
154. D. Cohen-Tanugi and J. C. Grossman, J. Chem. Phys., 2014, 141, 074704.
155. M. E. Suk and N. R. Aluru, J. Phys. Chem. Lett., 2010, 1, 1590–1594.
156. I. Akin, E. Zor, H. Bingol and M. Ersoz, J. Phys. Chem. B, 2014, 118, 5707–5716.
157. F. Perreault, A. F. de Faria and M. Elimelech, Chem. Soc. Rev., 2015, 44, 5861–5896.
158. K. Sint, B. Wang and P. Král, J. Am. Chem. Soc., 2008, 130, 16448–16449.
159. D. Konatham, J. Yu, T. A. Ho and A. Striolo, Langmuir, 2013, 29, 11884–11897.
160. D. Cohen-Tanugi and J. C. Grossman, Nano Lett., 2014, 14, 6171–6178.
161. D. Cohen-Tanugi and J. C. Grossman, Desalination, 2015, 366, 59–70.
162. M. Drndić, Nat. Nanotechnol., 2014, 9, 737.
163. S. Garaj, W. Hubbard, A. Reina, J. Kong, D. Branton and J. A. Golovchenko, Nature, 2010, 467, 190–193.
164. C. A. Merchant, K. Healy, M. Wanunu, V. Ray, N. Peterman, J. Bartel, M. D. Fischbein, K. Venta, Z. Luo, A. T. Johnson and M. Drndic, Nano Lett., 2010, 10, 2915–2921.
165. G. F. Schneider, S. W. Kowalczyk, V. E. Calado, G. Pandraud, H. W. Zandbergen, L. M. Vandersypen and C. Dekker, Nano Lett., 2010, 10, 3163–3167.
166. Z. Zhang, J. Shen, H. Wang, Q. Wang, J. Zhang, L. Liang, H. Ågren and Y. Tu, J. Phys. Chem. Lett., 2014, 5, 1602–1607.
167. D. B. Wells, M. Belkin, J. Comer and A. Aksimentiev, Nano Lett., 2012, 12, 4117–4123.
168. M. Zwolak and M. Di Ventra, Nano Lett., 2005, 5, 421–424.
169. M. Xu, R. G. Endres and Y. Arakawa, Small, 2007, 3, 1539–1543.
170. M. Xu, D. Fujita and N. Hanagata, Small, 2009, 5, 2638–2649.
171. S. M. Avdoshenko, D. Nozaki, C. Gomes da Rocha, J. W. González, M. H. Lee, R. Gutierrez and G. Cuniberti, Nano Lett., 2013, 13, 1969–1976.
172. W. Lv, M. Chen and R. a. Wu, Soft Matter, 2013, 9, 960–966.
173. R. F. Service, Science, 2006, 311, 1544–1546.
174. A. Kumar, K. B. Park, H. M. Kim and K. B. Kim, Nanotechnology, 2013, 24, 495503.
175. H. W. C. Postma, Nano Lett., 2010, 10, 420–425.
176. T. Nelson, B. Zhang and O. V. Prezhdo, Nano Lett., 2010, 10, 3237–3242.
177. S. K. Min, W. Y. Kim, Y. Cho and K. S. Kim, Nat. Nanotechnol., 2011, 6, 162–165.
178. G. T. Feliciano, C. Sanz-Navarro, M. D. Coutinho-Neto, P. Ordejón, R. H. Scheicher and A. R. Rocha, Phys. Rev. Appl., 2015, 3, 034003.
179. L. Liang, Z. Zhang, J. Shen, K. Zhe, Q. Wang, T. Wu, H. Agren and Y. Tu, RSC Adv., 2014, 4, 50494–50502.
180. M. Xu, Y. Gao, X. Yang and H. Chen, Chin. Sci. Bull., 2012, 57, 3000–3009.
181. J. Shendure and H. Ji, Nat. Biotechnol., 2008, 26, 1135–1145.
182. D. Branton, D. W. Deamer, A. Marziali, H. Bayley, S. A. Benner, T. Butler, M. Di Ventra, S. Garaj, A. Hibbs, X. Huang, S. B. Jovanovich, P. S. Krstic, S. Lindsay, X. S. Ling, C. H. Mastrangelo, A. Meller, J. S. Oliver, Y. V. Pershin, J. M. Ramsey, R. Riehn, G. V. Soni, V. Tabard-Cossa, M. Wanunu, M. Wiggin and J. A. Schloss, Nat. Biotechnol., 2008, 26, 1146–1153.
183. D.-e. Jiang, V. R. Cooper and S. Dai, Nano Lett., 2009, 9, 4019–4024.
184. J. Schrier, J. Phys. Chem. Lett., 2010, 1, 2284–2287.
185. S. Blankenburg, M. Bieri, R. Fasel, K. Müllen, C. A. Pignedoli and D. Passerone, Small, 2010, 6, 2266–2271.
186. H. Du, J. Li, J. Zhang, G. Su, X. Li and Y. Zhao, J. Phys. Chem. C, 2011, 115, 23261–23266.
187. A. W. Hauser and P. Schwerdtfeger, J. Phys. Chem. Lett., 2012, 3, 209–213.
188. M. Hankel, Y. Jiao, A. Du, S. K. Gray and S. C. Smith, J. Phys. Chem. C, 2012, 116, 6672–6676.
189. M. Shan, Q. Xue, N. Jing, C. Ling, T. Zhang, Z. Yan and J. Zheng, Nanoscale, 2012, 4, 5477–5482.
190. S. P. Koenig, L. Wang, J. Pellegrino and J. S. Bunch, Nat. Nanotechnol., 2012, 7, 728–732.
191. X. Qin, Q. Meng, Y. Feng and Y. Gao, Surf. Sci., 2013, 607, 153–158.
192. S. Jungthawan, P. Reunchan and S. Limpijumnong, Carbon, 2013, 54, 359–364.
193. J. Lee and N. R. Aluru, J. Membr. Sci., 2013, 428, 546–553.
194. H. W. Kim, H. W. Yoon, S.-M. Yoon, B. M. Yoo, B. K. Ahn, Y. H. Cho, H. J. Shin, H. Yang, U. Paik, S. Kwon, J.-Y. Choi and H. B. Park, Science, 2013, 342, 91–95.
195. H. Li, Z. Song, X. Zhang, Y. Huang, S. Li, Y. Mao, H. J. Ploehn, Y. Bao and M. Yu, Science, 2013, 342, 95–98.
196. G. Lei, C. Liu, H. Xie and F. Song, Chem. Phys. Lett., 2014, 599, 127–132.
197. H. Liu, Z. Chen, S. Dai and D.-e. Jiang, J. Solid State Chem., 2015, 224, 2–6.
198. H. Liu, S. Dai and D.-e. Jiang, Solid State Commun., 2013, 175–176, 101–105.
199. I. S. Chae, J. H. Lee, J. Hong, Y. S. Kang and S. W. Kang, Chem. Eng. J., 2014, 251, 343–347.
200. H. Koolivand, A. Sharif, M. Kashani, M. Karimi, M. Salooki and M. Semsarzadeh, J. Polym. Res., 2014, 21, 1–12.
201. H. W. Kim, H. W. Yoon, B. M. Yoo, J. S. Park, K. L. Gleason, B. D. Freeman and H. B. Park, Chem. Commun., 2014, 50, 13563–13566.
202. J. Shen, G. Liu, K. Huang, W. Jin, K.-R. Lee and N. Xu, Angew. Chem., Int. Ed., 2015, 54, 578–582.
203. R. S. Norman, Science, 1974, 186, 350–352.
204. R. L. McGinnis and M. Elimelech, Environ. Sci. Technol., 2008, 42, 8625–8629.
205. R. D. Cusick, Y. Kim and B. E. Logan, Science, 2012, 335, 1474–1477.
206. B. E. Logan and M. Elimelech, Nature, 2012, 488, 313–319.
207. R. E. Pattle, Nature, 1954, 174, 660.
208. S. Zhao, L. Zou, C. Y. Tang and D. Mulcahy, J. Membr. Sci., 2012, 396, 1–21.
209. F. Sanger, G. M. Air, B. G. Barrell, N. L. Brown, A. R. Coulson, J. C. Fiddes, C. A. Hutchison, P. M. Slocombe and M. Smith, Nature, 1977, 265, 687–695.
210. F. Sanger, S. Nicklen and A. R. Coulson, Proc. Natl. Acad. Sci. U. S. A., 1977, 74, 5463–5467.
211. J. Ribeiro-Soares and M. Dresselhaus, Braz. J. Phys., 2014, 44, 278–282.
212. http://www.observatorioplastico.com/detalle_noticia.php?no_id=360448.
213. Editorial, Nat. Nanotechnol., 2014, 9, 725.
214. Q. Tannock, Nat. Mater., 2011, 11, 2.
215. http://www.theguardian.com/science/2012/dec/27/super-material-graphene-george-osborne.
216. A. H. Castro Neto, Asia Pacific Physics Newsletter, 2013, 02, 60–65.
217. http://www.graphene.nus.edu.sg/content/graphene-centre.
218. L. Gao, W. Ren, H. Xu, L. Jin, Z. Wang, T. Ma, L.-P. Ma, Z. Zhang, Q. Fu, L.-M. Peng, X. Bao and H.-M. Cheng, Nat. Commun., 2012, 3, 699.
219. T. Ma, W. Ren, Z. Liu, L. Huang, L.-P. Ma, X. Ma, Z. Zhang, L.-M. Peng and H.-M. Cheng, ACS Nano, 2014, 8, 12806–12813.
220. B. Y. Dai, L. Fu, Z. Y. Zou, M. Wang, H. T. Xu, S. Wang and Z. F. Liu, Nat. Commun., 2011, 2, 522.
221. Z. Chen, W. Ren, L. Gao, B. Liu, S. Pei and H.-M. Cheng, Nat. Mater., 2011, 10, 424–428.
222. A. Zurutuza and C. Marinelli, Nat. Nanotechnol., 2014, 9, 730–734.
223. A. K. Geim and P. Kim, Sci. Am., 2008, 298, 90–97.

---