Two-dimensional layered nanomaterials for gas-sensing applications

Abstract

Owing to the unique thickness dependent physical and chemical properties, two-dimensional (2D) layered nanomaterials have received tremendous attention and shown great potential in the fabrication of high-performance electronic/optoelectronic devices. Notably, the implication of 2D nanomaterials in the gas-sensing field has also drawn considerable attention but few related review studies have been reported. This critical review mainly focuses on the current progress of 2D layered nanomaterials in gas-sensing applications. Firstly, we describe the basic attributes of 2D layered nanostructures and discuss the fundamentals of their gas-sensing applications. Secondly, we have numerated recent gas-sensing studies on typical 2D layered nanomaterials, including graphene, MoS₂, MoSe₂, WS₂, SnS₂, black phosphorus, and others. Particularly, the optimized strategies for improving their gas-sensing performances are also discussed here. Finally, we conclude this review with some perspectives and the outlook on future advances in this field.

---

Tianyou Zhai

Tianyou Zhai received his B.S. degree in chemistry from Zhengzhou University in 2003, and then received his Ph.D. degree in physical chemistry from the Institute of Chemistry, Chinese Academy of Sciences (ICCAS) under the supervision of Prof. Jiannian Yao in 2008. Afterwards he joined the National Institute for Materials Science (NIMS) as a JSPS postdoctoral fellow of Prof. Yoshio Bando's group and then as an ICYS-MANA researcher within NIMS. Currently, he is a Chief Professor of the School of Materials Science and Engineering, Huazhong University of Science and Technology (HUST). His research interests include the controlled synthesis and exploration of fundamental physical properties of inorganic functional nanomaterials, as well as their promising applications in energy science, electronics and optoelectronics. He has authored and co-authored above 100 peer-reviewed journal articles, 5 book chapters, co-edited 1 book on nanotechnology, and held 7 patents. His publications have been cited more than 5100 times (H-index is 41).

---

1. Introduction

Gas sensors that identify the presence and/or concentration of a specific gas in ambient atmosphere have attracted enormous attention and have found applications in many fields, such as environmental monitoring, process control, security, medical diagnostics, etc.^1,2 Among various types of gas sensors, metal oxide-based gas sensors have been extensively investigated in the past few decades due to their relatively low cost and high sensitivities.^3–5 However, to obtain a high response to targeted gases, a relatively high operating temperature would generally be adopted, which not only increases the power consumption but also causes thermal safety problems, thus hindering their practical applications. Therefore, novel sensors which can work effectively at low temperature, especially at room temperature, are highly desired. Several strategies, such as exploiting novel gas-sensing materials, optimizing device geometry, and adopting suitable fabrication techniques, have been developed to address this issue. Notably, all these methods are strongly dependent on the characteristics of the gas-sensing materials. Thus, developing novel gas-sensing materials with ultrahigh surface-to-volume ratios and strong surface activities is a strikingly primary and reliable approach for fabricating gas sensors operated at a low or room temperature. Since these samples possess an ultra-large specific surface area and high surface activities, the interaction of gas molecules with the surfaces is very strong which ultimately improves the sensitivity and kinetics of sensing.

Two-dimensional (2D) layered nanomaterials have aroused huge interest due to their excellent semiconducting performances and unique thickness dependent physical and chemical properties, including ultra-large surface-to-volume ratios, high surface activities, etc.^6–10 Furthermore, 2D layered nanomaterials can be easily fabricated as chemiresistive field effect transistors (FETs) that have low power consumption and good safety. Such advantages of 2D layered nanomaterials distinguish them from conventional metal oxides, and render them a promising future in the development of ultrahigh sensitive and low-power-consumption sensor platforms. Up to now, some intrinsic 2D nanomaterials with layered characteristics, such as graphene,^11–13 metal dichalcogenides (e.g. MoS₂,^14–20 MoSe₂,²¹ WS₂,^22–24 SnS₂,^25,26 and VS₂ ²⁷), black phosphorus (BP),^28–31 hexagonal boron nitride (h-BN)³² and graphitic carbon nitride (g-C₃N₄),³³ have been developed as gas-sensing materials and opened up new avenues for fabricating low or room temperature gas sensors. In addition, there are many studies on the preparation of hybrid 2D functional materials in which many materials are incorporated into the 2D system for gas-sensing applications, such as layer-by-layer assembly^34–36 and the Langmuir–Blodgett (LB) method.^37–39 These strategies have been elaborated in other studies,^40,41 thus herein we focus mainly on these inherent 2D layered nanomaterials for gas-sensing applications. Recent reviews have described the synthesis, properties, and potential applications of intrinsic 2D nanomaterials in various aspects, including electronic devices, optoelectronic devices, energy storage devices, as well as sensing platforms.^42–47 However, there are few review studies systematically focused on their gas-sensing applications until now, thus it is necessary to make a full summary on such a topic.

In this critical review, we mainly focus on the current developments of gas sensors based on some kinds of typical 2D layered nanomaterials, including graphene, MoS₂, MoSe₂, WS₂, SnS₂, VS₂, BP, BN, and C₃N₄. Moreover, the optimized approaches for enhancing their gas-sensing behaviours are also discussed here. Finally, we conclude this article with some perspectives and outlooks on this new trend in the gas sensing field.

2. Fundamentals of gas-sensing on 2D layered nanomaterials

2.1 Gas-sensing mechanism for 2D layered nanomaterials

Development of gas sensors based on intrinsic 2D layered nanomaterials is strictly dependent on their sensing mechanism. Unlike typical metal oxides (e.g., SnO₂, ZnO, In₂O₃), the sensing mechanism of typical 2D layered nanomaterials (especially graphene, MoS₂, and WS₂) is based on a charge transfer process.^18,19,48 When the sensors are exposed to reactive gas, the gas species are adsorbed on the surfaces of sensing channels, leading to the variation of sensor resistance. Upon exposure to air or an inert environment, in contrast, the gas molecules are desorbed from the sensing materials, and the sensor resistances go back to their original state. In detail, taking CVD-grown monolayer MoS₂ (n-type semiconductor) as a typical example, the charge transfer processes between gaseous species (e.g., NO₂, NH₃) and sensing devices can serve as a rational sensing mechanism discussed in the following.¹⁹ When NO₂ molecules are adsorbed onto MoS₂ surfaces, an unpaired electron from the nitrogen atom in NO₂ usually acts as a strong oxidizer (electron acceptor), and tends to withdraw electrons from the MoS₂ surface (Fig. 1a). As a result, a decreased electron concentration in the conduction band of the MoS₂ occurs. Such a charge transfer induces a decreased carrier concentration and thus degraded electrical conductivity. In contrast, NH₃ having a lone electron pair always serves as an electron donor, and transfers its electrons to the conduction band of MoS₂, thus this will lead to an increased electron concentration and electrical conductivity in MoS₂ (Fig. 1b). The charge transfer processes between the adsorbed gaseous species and MoS₂ are verified by an in situ photoluminescence (PL) study. When the sensor was exposed to oxidative NO₂, the intensity of the quasi-particle (A⁺) peak in the PL spectrum increased, whereas the neutral exciton (A⁰) peak decreased (Fig. 1c). This is attributed to the extraction of electrons from the MoS₂ by NO₂ and the generation of the corresponding holes in MoS₂. In contrast, an opposite trend was observed for NH₃ gas because of dissociation of the positive trions from the neutral excitons, leading to a suppressed intensity of the A⁺ peak and increased A⁰ (Fig. 1d).

Fig. 1 Schematics of the charge transfer processes for MoS₂ in the presence of NO₂ (a) and NH₃ (b) molecules. In situ PL spectra recorded from MoS₂ with NO₂ (c) and NH₃ (d) molecules (reproduced from ref. 19).

2.2 Configurations of gas-sensing devices

The chemiresistor-type gas-sensing device is widely applied in which the resistance of the sensing material changes under exposure to the targeted gas. This type of sensing device has several advantages, such as a simple fabrication process, direct measurement and easy operation.^49–51Fig. 2a exhibits a typical structure of a chemiresistive gas sensor. The dispersed solution containing sensing materials was dropped onto a ceramic plate which was previously coated with a pair of gold electrodes deposited by the screen printing technique. A RuO₂ heater was mounted on the reverse sides of the ceramic plate to control the sensing temperature. This type of sensing device has been applied to various kinds of gas species, including NO₂, NH₃, H₂S, H₂, ethanol and formaldehyde.^52–54

Fig. 2 Two typical configurations of gas-sensing devices: (a) chemiresistor (reproduced from ref. 52); (b) field effect transistor (FET) (reproduced from ref. 18).

The field effect transistor (FET) is another important type of electronic gas sensor and has been demonstrated to be a competitive configuration for 2D layered nanomaterial-based sensors.⁵⁵ In this case, the drain current (I_ds) of a FET-based sensor depends on the gate bias (V_g), and it can be effectively tuned upon the exposure of gas species. The sensing performance of FET-based devices is mainly evaluated by the on/off current ratios. A high sensitivity usually accompanies a high on/off current ratio.⁵⁶ Up to now, there are many functional nanostructures, especially 2D layered nanomaterials, suitable as channel materials for FET type gas sensors due to their excellent semiconducting properties and large lateral size.^18,57,58Fig. 2b presents the structure of a MoS₂-based FET sensor and it has been applied for the detection of NO₂ and NH₃.¹⁸ When the sensor was exposed to reactive molecules (NO₂, NH₃), these molecules were adsorbed onto the surfaces of the sensing materials, leading to a change of the local charge carrier concentration, and this signal could be monitored sensitively as current altering.

The surface acoustic wave (SAW) technique has also been developed for the detection of a variety of gas species, such as CO, H₂, methanol, chloroform.^59,60 This type of sensor is based on the frequency changes caused by the conductance or mass change of their sensing layers by exposure to gas molecules. As a typical example, Kaner et al.⁵⁹ reported a SAW graphene sensor for the detection of H₂ and CO. The sensor responses were approximately 5.8 kHz to 1% H₂ and 8.5 kHz to 1000 ppm CO at room temperature, respectively. Notably, the molecular weight of CO is 14 times heavier than that of H₂. Thus, the dominant factor for CO sensing was the mass change, while the H₂ sensing was mainly attributed to the conductance change of graphene nanosheets. In another example, Balandin et al.⁶⁰ demonstrated a SAW sensor for distinguishing various chemical vapours. The characteristic frequency (f_c) of the Lorentzian noise bulges in the sensor is distinct for different vapors with variations from f_c = 10–20 Hz to f_c = 1300–1600 Hz for tetrahydrofuran and chloroform vapours, respectively.

3. Gas sensors based on 2D layered nanomaterials

3.1 Graphene-based gas sensors

3.1.1 Pristine graphene-based gas sensors. Graphene has been considered as an attractive building block for fabrication of gas sensors, mainly due to its atom-thick two-dimensional structure and excellent physical properties, such as high specific surface area and fast electron transport kinetics.^11–13,61 Moreover, graphene has inherently low electrical noise because of its high-quality crystal lattice along with its two-dimensional structure, making it ultrasensitive to gas adsorption even down to a single molecule level.^62,63 In 2007, Novoselov et al.⁶⁴ developed a gas sensor using mechanically exfoliated graphene with a limit of detection (LOD) down to the parts per billion (ppb) level. Furthermore, the sensing device was optimized by using Hall geometry to obtain high response through changing the carrier density near the Dirac point. Fig. 3a exhibited the carrier concentration variation (Δn) in single-layer graphene exposed to various concentrations (C) of NO₂. It was found that Δn depended linearly on C under the same exposure conditions. The step-like resistivity changes upon exposure to extremely diluted NO₂ during long term measurements were carried out as shown in Fig. 3b. The sensing device exposed to 1 ppm NO₂ at a leaking rate of 10⁻³ mbar s⁻¹ is recorded in the blue curve. After the analysis of obtained data (Fig. 3c), single NO₂ molecules could be detectable by applying the optimized sensor. Such an outstanding single-molecule-detection limit is attributed to the high carrier mobility of graphene and the resulting extremely low noise.^65,66

Fig. 3 (a) Concentration variation (Δn) of charge carriers in single-layer graphene exposed to different NO₂ concentrations (C). Upper inset: scanning electron micrograph of this device. Lower inset: characterization of the graphene device by using the electric-field effect. (b) The changes of Hall resistivity (ρ_xy) observed near the neutrality point during adsorption (blue curve) and desorption (red curve) of strongly diluted NO₂. The green curve presents a reference exposed to pure He. (c) For analysis, all changes in ρ_xy were recorded larger than 0.5 Ω and quicker than 10 s as individual steps (reproduced from ref. 64).

Inspired by the work described above, many efforts have been made to exploit the gas-sensing performances of pristine graphene prepared by the mechanically exfoliated method. For instance, Balandin et al.⁶⁰ demonstrated a exfoliated graphene based transistor that can selectively detect different organic vapours. Fig. 4a exhibited the SEM image of the back-gated graphene transistor using the electron beam lithography (EBL) technique. The noise frequency characteristics induced by different chemical vapours are also shown in Fig. 4b. The characteristic frequencies of these bulges are clearly variable for different gas chemicals. Moreover, the noise frequency range and the relative resistance changes can serve as distinctive signals for specific gases, rendering the graphene device with a highly selective gas-sensing attribute.

Fig. 4 (a) SEM images of back-gated graphene devices. (b) Noise spectral density (S_I/I²) multiplied by frequency (f) versus f for the device in the presence of open air and different vapours (reproduced from ref. 60).

Although high-quality graphene can be prepared by the mechanical exfoliation method, the size of most graphene films is usually limited below 1000 μm², which is not considered as a scalable technique.⁶⁷ Large-area synthesis of high-quality pristine graphene sheets can be achieved by the chemical vapour deposition (CVD) method. In the CVD synthesis, metal substrates such as copper, nickel and cobalt, were applied to catalytically decompose the hydrocarbons at moderate temperatures (<1000 °C).^68–73 The deposited graphene sheets could be transferred to arbitrary substrates (rigid or flexible) for the fabrication of gas sensing devices.^74–76 As a typical example, very recently, Jang et al.⁶⁸ presented a self-activated transparent gas sensor based on CVD-grown graphene with endurance to humidity and mechanical bending. Fig. 5a showed the fabrication procedure of a graphene sensor. Firstly, the graphene layer was grown and directly patterned on a Cu foil. Subsequently, poly(methyl methacrylate) (PMMA) was coated on top of the patterned graphene to transfer the sample to a target substrate. Finally, the Cu foil was etched by using FeCl₃ solution, and the patterned graphene with PMMA was transferred onto a polyimide (PI) substrate. The PMMA can be removed by acetone and heat treatment. When exposed to three consecutive pulses of 5 ppm NO₂, the as-fabricated devices exhibited improved response and recovery with increasing bias voltage (Fig. 5b). Moreover, the dynamic sensing transients of the patterned graphene sensor to 5 ppm of NO₂ in dry (0% RH) and humid (50% RH) atmospheres were also presented in Fig. 5c, indicating the negligible influence of humidity on this self-activated sensor. In addition, the sensor displayed a reliable operation under mechanical bending due to the excellent flexibility of graphene (Fig. 5d). These remarkable sensor performances significantly extended the potential applications of the all-graphene sensors in the internet of things and wearable electronics. Recently, Kumar et al.⁷⁶ demonstrated for the first time a simple, low-cost chemical sensing platform based on graphene-paper (G-paper) prepared by direct transfer of CVD-grown graphene onto paper without any intermediate layers. These sensors achieved a remarkable limit of detection (LOD) of 300 ppt for NO₂, which is better than those from other paper-based sensors. Moreover, G-paper sensors were also found to be robust against minor strain, which was expected to enable a flexible and portable gas-sensing platform.

Fig. 5 (a) Fabrication procedure of an all-graphene sensor. Response curves of the patterned graphene sensors to three pulses of 5 ppm NO₂ (b) under the increased bias voltage; (c) under 0% and 50% relative humidity conditions at 60 V; (d) without and with the bending strain (reproduced from ref. 68).

Large-area uniform pristine graphene films can also be achieved by the epitaxial growth method. The SiC substrate was usually applied to produce graphene layers by heating under ultrahigh vacuum, and the graphene layer thickness can be adjusted by controlling the annealing temperature and duration.⁷⁷ The grown graphene layers on the SiC substrate can be directly fabricated into the sensing device without a transfer process.⁷⁸ Moreover, the sensing performances of graphene-based devices were strongly influenced by the chosen substrate.^79–82 For example, Nomani et al.⁸¹ reported that epitaxial-grown graphene layers on the Si- and C-faces of the semiconducting 6H-SiC displayed opposite conductance variations when exposed to NO₂ molecules. The graphene layer grown on Si- and C-faces exhibited 10% decrease and 4.5% increase in conductance toward 18 ppm of NO₂, respectively. The phenomenon was attributed to the different doped states of both graphene materials. In this case, the graphene films grown on the Si-face were p-type, whereas those grown on the C-face were n-type. Interestingly, an opposite direction of signal change was also observed for the epitaxially grown graphene-based sensor to various concentrations of NO₂, demonstrated by Pearce et al.⁸² The single layer graphene on the SiC substrate exhibited n-type behaviour upon exposure to low NO₂ concentration, whereas a p-type response toward high NO₂ concentrations. Such results demonstrated that the single layer graphene device underwent an n–p transition when exposed to increased NO₂ concentration, resulting from the tuned electronic properties of graphene induced by the SiC substrate which can act as the electron donor.

3.1.2 Functionalized graphene-based gas sensors. Although graphene materials can be prepared by the several methods mentioned above, the chemically reduced graphene oxide approach attracted the most intensive attention, mainly owing to the fact that it can be produced on a large scale with a relatively low cost.^83–85 Graphene oxide (GO) as the precursor of reduced graphene oxide (rGO), can be usually synthesized from purified graphite powder by the modified Hummers method.^86,87 Compared with pristine graphene, GO is heavily oxygenated with abundant oxygen containing functional groups on its defect sites and edges.^88,89 These functional groups of GO can apparently act as reactive sites for gas adsorption.⁹⁰ However, GO is almost an insulator because of the conjugated electronic structure disturbed by these functional groups, hindering its practical application in electronic devices. Chemical reduction of GO is an effective method to partially restore its conductivity, meanwhile the resulting rGO also has a certain amount of oxygen groups because of incomplete reduction.⁹¹ Furthermore, the reduction process can introduce some structural defects and vacancies which can also act as adsorption sites.^92–94 At present, a variety of chemical reducing agents were employed to availably reduce GO into rGO reported in many reviews.^85,94–96 Hydrazine, as a typical reducing agent, has been generally studied for reducing GO into rGO.^97–99 For example, Robinson et al.⁹⁷ reported high-performance molecular sensors based on hydrazine-reduced GO. At first, the GO film device was created by a standard photolithography technique (Fig. 6a), and the device was subsequently exposed to hydrazine hydrate vapour for 24 h at a temperature of 100 °C. The reduction degree of these GO devices can be tuned by varying the exposure time to hydrazine vapour. Moreover, these GO devices presented distinct response characteristics upon exposure to 5 s pulses of acetone (Fig. 6b). Such results indicated that the sensing performances could be tailored by adjusting the reduction process. Meanwhile, these hydrazine-reduced GO devices displayed a lower noise response compared with the single-walled carbon nanotube (SWNT)-based device. In addition, the noise levels in the rGO devices were dependent on the thickness of the sensing film, and the noise response can be tuned down to the thermal noise limit of the instrumentation (Fig. 6c). Ascorbic acid (vitamin C) as a green and mild reducing agent, was also used to synthesize rGO from exfoliated GO demonstrated by Ruoff et al.¹⁰⁰ The resulting rGO film on the PET substrate was employed to fabricate a flexible gas sensor using the inkjet printing technique (Fig. 6d). The sensing devices can achieve reversible and selective detection of chemically aggressive gases including NO₂ and Cl₂ at room temperature, with a detection range of 100 ppm to 500 ppb (Fig. 6e and f). However, due to the strong interactions of gas molecules with graphene layers, UV light illumination (254 nm) was applied to accelerate a full signal recovery in this case. Compared with hydrazine, p-phenylenediamine (PPD) as a weaker reductant, could be applied to prepare rGO with more oxygenated groups and structural defects, and the defective rGO-based sensors may exhibit a higher sensing performance. For instance, Zhang et al.¹⁰¹ developed a gas sensor based on PPD-reduced GO. The sensor response is 3.3 times that of the hydrazine-reduced GO sensor toward 10 ppm of dimethyl methylphosphonate (DMMP). Moreover, the sensor based on PPD-reduced GO displayed high repeatability, and can be considered as an excellent sensing material for potential applications. Because of the high toxicity for hydrazine, non-toxic NaBH₄ was also used as an effective reducing agent. Guha et al.¹⁰² reported chemically reduced GO using NaBH₄ and investigated its performance for ammonia detection at room temperature. The response and recovery characteristic of the rGO-based sensing device could be optimized by the reduction time of GO. The sensor exhibited excellent selectivity to NH₃, and response varied between 5.5% at 200 ppm and 23% at 2800 ppm of ammonia without any heat or ultraviolet illumination for quick resistance recovery.

Fig. 6 (a) Optical image of the GO film device. (b) Plot of the normalized conductance response to 5 s pulses of acetone for GO devices reduced with hydrazine for 3, 6, 18, and 24 h. (c) Noise density spectra for SWNT and rGO devices (reproduced from ref. 97). (d) Inkjet-printed graphene oxide film lifts off the PET surface. Inset: digital image of an inkjet-printed RGO/PET four-probe sensor. Plot for resistance versus time when the film was exposed to NO₂ (e) and Cl₂ (f) (reproduced from ref. 100).

In addition, during the chemical reduction of GO, many foreign groups or atoms are introduced to intrinsically change the surface properties of rGO, leading to the enhanced sensing performances.^48,103,104 In a typical example, Shi et al.⁴⁸ used chemiresistor-type NO₂ sensors based on sulfonated rGO (S-G). The thickness of the S-G sheets was measured to be 0.9 to 1.1 nm by atomic force microscopy (AFM) in Fig. 7a, indicating a monolayer for the S-G. Moreover, the as-fabricated S-G sensor displayed a reversible response and recovery characteristic upon exposure to various concentrations of NO₂ without the assistance of UV/IR light illumination or thermal treatment (Fig. 7b). In addition, the sensor response was about 16.4 times that of the rGO counterpart toward 50 ppm NO₂. These results indicated that the chemical modification of sulfonic acid greatly improved the performance of rGO in the NO₂-sensing application. Furthermore, the sensitivity of the S-G sensor was also strongly influenced by the thickness of the sensing layer. Fig. 7c showed the responses of the S-G sensors with various thicknesses of sensing layers exposed to 50 ppm NO₂. The response of sensor 3 with sensing layer thickness (T) around 1–6 nm is 4.8 and 2.3 times those of sensor 1 (T = 50 nm) and sensor 2 (T = 10 nm), respectively. This is because the thinner S-G layer possessed a larger surface-to-volume ratio which is favourable for the surface adsorption of NO₂. To further improve the response-recovery characteristic of the S-G device for NO₂ sensing at room temperature, Huang et al.¹⁰³ demonstrated a flexible chemiresistor-type sensor based on Ag nanoparticles decorated S-RGO (Ag-S-RGO), which was printed onto a polyimide (PI) substrate using a gravure printing technique. The FESEM image in Fig. 7d showed that large amounts of Ag NPs are uniformly assembled on the S-RGO surfaces and the Ag-S-RGO sheets have a few wrinkles. Moreover, the printed Ag-S-RGO sensor exhibited a fast response-recovery rate toward NO₂ with a concentration range of 0.5 to 50 ppm at room temperature (Fig. 7e). When exposed to 50 ppm NO₂ at room temperature, the response and recovery time of the sensor was only 12 s and 20 s, respectively. Furthermore, the Ag-S-G sensor showed almost constant resistance after 1000 bending cycles (Fig. 7f), indicating potential application in a flexible sensing device.

Fig. 7 (a) AFM images of S–G sheets deposited on the mica surface. (b) Response variation of the S–G-based sensor as a function of NO₂ concentrations ranging from 5 to 45 ppm. (c) Responses of the S–G-based sensors with different thicknesses of sensing layers after exposure to 50 ppm NO₂: 50 nm (1), 10 nm (2), and a few layers (1–6 nm) (3) (reproduced from ref. 48). (d) FESEM image of Ag-S-RGO. Inset: photographs of Ag-S-RGO ink and Ag-S-RGO sensing layer printed onto the PI substrate with Ag interdigitated electrodes, respectively. (e) Response of the Ag-S-RGO sensor upon exposure to various NO₂ concentrations. (f) Changes of the normalized resistance during the bend test for the Ag-S-RGO sensor (reproduced from ref. 103).

3.1.3 Functional material/graphene composite-based gas sensors. To further enhance the sensing performances of the graphene, incorporating functional micro/nanomaterials (such as metals,^105–111 metal oxides,^51,112–115 carbon nanotubes,¹¹⁶ polymers^117–122) into graphene networks to form hybrid architectures have been proven to be an effective way due to the diverse functionalities and synergistic effects in the composites.

Metals, especially noble metals (e.g., Au, Pt, Pd, and Ag) decorated graphene hybrid composites are considered as a new type of sensing material with high sensitivity and selectivity beyond conventional graphene-based materials due to their specific catalysis effect for gas-sensing. For example, Ramaprabhu et al.¹⁰⁵ fabricated a hydrogen sensor based on Pt nanoparticle (Pt NP) decorated graphene nanosheets by a simple drop casting technique. The sensor based on the Pt NP/graphene composite exhibited a higher response (R_g/R_a − 1 = 16%) to 4 vol% H₂ than that of the Pt NP decorated multi-walled carbon nanotubes (R_g/R_a − 1 = 8%) at room temperature. In another case, Liu et al.¹⁰⁶ developed a highly sensitive NO sensing device. The device comprised sensing channels with palladium-decorated reduced graphene oxide (Pd-RGO) and the electrodes covered with chemical vapour deposition (CVD)-grown graphene were fabricated by the alternating current dielectrophoresis technique, and the detailed fabrication process is shown in Fig. 8a. The SEM image of the device (device A, graphene-Pd-RGO) is also shown in Fig. 8b. After a moderate current annealing, the novel device displayed a recoverable response upon exposure to NO ranging from 2 to 420 ppb at room temperature (Fig. 8c). To totally investigate the effects of Pd decoration of RGO, electrodes covered with CVD-grown graphene, and current annealing on sensing properties, the typical comparison results of devices A, B, C, and D upon exposure to continuously increasing NO gas concentrations (2, 24, 156, and 420 ppb) are depicted in Fig. 8d. This result demonstrated that the improved NO response of the graphene-Pd-RGO sensor can be attributed to the Pd decoration and CVD-grown graphene contacts of RGO. Functional Ag nanostructures were also employed to enhance the sensing properties of the rGO materials. Cuong et al.¹⁰⁹ reported NH₃ sensors based on Ag nanomaterial (Ag nanoparticles, Ag nanowires) coated rGO nanosheets fabricated with solution processes. When exposed to 100 ppm NH₃ at room temperature, the rGO/AgNW sensor exhibited a higher sensitivity, and an excellent response and recovery characteristic compared with rGO/AgNPs, and rGO sensors. Furthermore, the effect of AgNW spray volume on the rGO/AgNW device response indicated that an excess amount of AgNWs was unfavourable for enhancing the sensing performance due to the agglomeration of the AgNWs.

Fig. 8 (a) Schematics of the fabrication process for the graphene-Pd-RGO device. (b) SEM image of Pd-RGO nanosheets trapped between two graphene-covered electrodes. (c) Device response exposed to 2, 24, 156, and 420 ppb NO after current annealing. (d) Responses of the various devices versus time, and the step-shaped response corresponding to 1, 10, 60, and 120 s pulses of NO: 2, 22, 132, and 264 ppb, respectively (reproduced from ref. 106).

Semiconducting metal oxides, including SnO₂, ZnO, In₂O₃, TiO₂, WO₃, Fe₂O₃, CuO and Cu₂O, have been developed for fabrication of gas-sensing devices in many reports.^123–126 However, the low conductivity at room temperature and high operating temperature limited their potential applications in room-temperature sensing devices. Blending metal oxides with graphene sheets to form hybrid architectures can maintain improved sensing behaviour compared with individual metal oxides and graphene.^{112,114,127,128} As a typical example, Chen et al.¹¹² developed a novel and selective gas-sensing platform using SnO₂ nanocrystal (SnO₂ NC) decorated rGO (SnO₂-rGO), as schematically depicted in Fig. 9a. The hybrid nanostructures consisting of rGO nanosheets and SnO₂ NCs worked as the sensing channel in the sensor (Fig. 9b). When the sensing device was exposed to the targeted gases (e.g., NO₂, NH₃), the electron transfer between gas molecules and rGO occurred, thereby varying the device conductivity. Moreover, this SnO₂-rGO device showed improved NO₂ sensing, whereas it showed weakened NH₃ sensing compared with the bare rGO (Fig. 9c and d). The reported findings implied a promising direction for tuning the sensitivity and selectivity of the rGO-based sensor. Similarly, Sow et al.¹¹⁴ prepared the octahedral Cu₂O nanowire mesocrystal coupled rGO composite (Cu₂O/rGO) under hydrothermal conditions and developed their applications in the detection of NO₂. When exposed to NO₂ with a concentration range of 0.4–2 ppm at room temperature, the sensor based on the Cu₂O/rGO hybrid composite exhibited a higher response than that of rGO and Cu₂O alone. The theoretically calculated LODs for the composite-, rGO- and Cu₂O-based sensors were 64 ppb, 81 ppb and 82 ppb, respectively. The enhanced gas-sensing performance of the Cu₂O/rGO composite can be mainly attributed to its high specific surface area and improved conductivity, induced by the highly oriented and porous interpenetrating structured Cu₂O nanowire mesocrystals coupled on the rGO nanosheets. In addition, metal oxide quantum dot (QD) decorated rGO hybrid composites have been reported for the fabrication of gas sensors, exhibiting enhanced performance.^129,130 As a typical example, Zeng et al.¹²⁹ fabricated novel ZnO QD decorated graphene nanocomposites by a facile solution-processed method. ZnO QDs with a size about 5 nm are nucleated and anchored onto the graphene, and their distribution density can be easily controlled by precursor concentration and reaction time. Moreover, ZnO QD/graphene nanocomposites presented enhanced formaldehyde sensing by 4 times compared to the bare graphene at room temperature. Furthermore, the sensor based on the nanocomposites possessed fast response (30 s) and recovery (40 s) behavior, excellent selectivity and stability. In another example, Mishra et al.¹³⁰ demonstrated a superior sensitive, selective and reproducible performance for a H₂ and LPG sensor based on SnO₂ QDs decorated on rGO synthesized by a surfactant assisted hydrothermal method. The sensor showed a high response of 89.3% and 92.4% to 500 ppm H₂ at 200 °C and 500 ppm LPG at 250 °C, respectively. Furthermore, the sensor also exhibited excellent selectivity for H₂ and LPG in the presence of other interfering gases, such as ammonia, acetone, chloroform, toluene, benzene, n-butylacetate, acetic acid and formic acid.

Fig. 9 (a) Schematic of the novel gas-sensing platform using RGO nanosheets decorated with SnO₂ NCs. (b) SEM image of the SnO₂ NC decorated RGO bridging a pair of Au electrode fingers. Sensor responses of RGO sensors with and without SnO₂ NCs toward 100 ppm NO₂ (c) and 1% NH₃ (d) (reproduced from ref. 112).

Carbon nanotubes (CNTs), as one-dimensional structure carbon-based materials, have been reported for the detection of various gas species, including NO₂, NH₃, H₂, organic vapor.^131–135 However, the imperfect contact of CNTs with metal electrodes limited their potential application in flexible devices. Incorporating CNTs into graphene to form a hybrid composite, not only extended the application of CNTs in flexible devices, but also improved the sensing properties of graphene. For instance, Choi et al.¹¹⁶ developed a flexible device using CNTs/rGO hybrid films for detection of NO₂ gas at room temperature. To obtain excellent contact with metal electrodes, vertically aligned CNT arrays were directly grown on rGO films by the CVD method. Significantly, the as-fabricated sensor exhibited a remarkably enhanced response compared with their counterparts based on bare graphene.

Besides, some functional polymers, including polyaniline (PANI),^117–119 poly(3,4-ethylenedioxythiophene) (PEDOT),¹²⁰ electrospun polymer (e.g. PVA: poly(vinyl alcohol), PEI: poly(ether imide), and PPr: polypyrene),^121,122 polypyrrole (PPy),¹³⁶ have been extensively studied for the chemical modification of pure graphene materials to improve their sensing performances. As a typical example, Zhang et al.¹¹⁸ developed a useful ammonia (NH₃) gas sensor based on PANI–rGO hybrids. PANI nanoparticles were successfully anchored onto the rGO sheets by using rGO/MnO₂ hybrids as both the templates and subsequent oxidants for the aniline monomer simultaneously during the process of polymerization. Fig. 10a displays an enlarged SEM image of the PANI–rGO hybrids bridging the electrode gap, and the PANI nanoparticles were distributed uniformly on the surface of rGO. Moreover, the sensor based on the PANI–rGO hybrid exhibited a much higher response toward 50 ppm NH₃ than those of the pristine graphene and individual PANI nanofiber (Fig. 10b). It was revealed that the combination of rGO and PANI nanoparticles facilitated the enhancement of the sensing properties due to the synergetic behavior of the functionalities of rGO and PANI in the hybrid composites. In another example, Shi et al.¹²² described an ultrasensitive and selective NO₂ sensor based on self-assembled graphene/electrospun polymer nanofiber (rGO/P NF) composites. These rGO/P NFs bridged the gaps of the gold digits to form conductive channels, as shown in Fig. 10c. Furthermore, this sensor exhibited high sensitivity and remarkable reversibility to NO₂ gas at a concentration ranging from 150 ppb to 5 ppm at room temperature (Fig. 10d). Moreover, the LOD was experimentally measured to be as low as 150 ppb, and this value was much lower than the threshold exposure limit proposed by the American Conference of Governmental Industrial Hygienists (200 ppb).

Fig. 10 (a) An enlarged SEM image of PANI-rGO hybrid bridging electrodes. (b) The response curves of PANI-rGO hybrids, bare PANI nanofibers and bare rGO to 50 ppm NH₃ at room temperature (reproduced from ref. 118). (c) SEM image of the surface of an rGO/P NF bridged interdigital electrodes. (d) Plot of response versus time for an rGO/P NF-based sensor upon exposure to NO₂ with concentrations ranging from 150 ppb to 5 ppm (reproduced from ref. 122).

In addition, combining ionic liquids with graphene layers is also an attractive strategy to form a highly selective gas-sensing platform. In a typical example, Ariga et al.¹³⁷ prepared graphene/ionic liquid (G-IL) films by the reduction of graphene oxides in the presence of ionic liquids and subsequent layer-by-layer (LbL) assembly. The layered graphene/ionic liquid composites on quartz crystal microbalances (QCM) further demonstrated an ideal molecular discrimination for toxic aromatic hydrocarbons. This result implied the great potential of the G-IL films for practical applications in gas-sensing.

3.1.4 The others. Apart from the modification or hybridization of graphene materials to achieve enhanced sensing performance, some other methods have been developed. As a typical example, Lee et al.¹³⁸ demonstrated a high-performance three-dimensional (3D) chemical sensor using rGO formed on high aspect-ratio 3D SU8 micro-pillar arrays by designing the device geometry. To study the sensing enhancement by geometrical engineering of the sensing area, both 2D and 3D devices with identical device areas, and cross-sections of 3D devices were designed, as schematically illustrated in Fig. 11a. In addition, the as-synthesized 3D SU-8 micro-pillars with heights of 40 μm (3D40) and 70 μm (3D70) were also exhibited in the FESEM images of Fig. 11b. Moreover, upon exposure to 5 ppm NO₂, the response of the 2D, 3D40, and 3D70 devices were 19%, 21%, and 28%, respectively (Fig. 11c); while they were 20%, 75%, and 88% when exposed to 40 ppm NH₃, respectively (Fig. 11d). The highly responsive 3D rGO sensor can be attributed to the interaction cross-section of rGO on the same device area and the increasing sensing area.

Fig. 11 (a) Schematic structure of both 2D and 3D devices (left), and cross-section of 3D device (right). (b) FESEM images of SU-8 pillar arrays with different heights of 40 (left) and 70 μm (right). Sensing response of rGO sensors towards (c) 5 ppm NO₂ (exposure time is 15 min) and (d) 40 ppm NH₃ (exposure time is 30 min) (reproduced from ref. 138).

3.2 Transition metal dichalcogenide (TMD)-based gas sensors

3.2.1 MoS₂-based gas sensors. As a typical atomic-layer material, MoS₂ possessing a high surface-to-volume ratio and excellent semiconducting properties, has accelerated the development of potential applications in gas sensors.^{14,55,139,140} For instance, Zhang et al.¹⁴ reported a FET device based on mechanically exfoliated MoS₂ films for the detection of NO gas. Moreover, the influence of the MoS₂ film layer on the sensing performance was investigated. It was found that bilayer (2L), trilayer (3L), and quadrilayer (4L) MoS₂ film-based FET sensors exhibited a high response toward NO with a detection limit of 0.8 ppm, whereas the single-layer (1L) MoS₂ device showed a rapid but unstable response. Similarly, Late et al.¹⁷ focused on the effect of MoS₂ layer thickness on the sensing performance of their FET devices. Using micromechanical exfoliation, two and five layer MoS₂ films were prepared and applied to detect gases, including NO₂, NH₃ and water vapour. It was observed that the five layered MoS₂ FET device displayed a stable response with excellent sensitivity and recovery compared with the two layered device. In another example, Perkins et al.¹⁶ fabricated a sensor device using the monolayer MoS₂ flake by exfoliating its bulk counterpart. The schematic and optical images of the MoS₂ monolayer sensor prepared using the electron beam lithography (EBL) technique are shown in Fig. 12a and b, respectively. The sensor exhibited a significant response to a sequence of triethylamine (TEA) pulses with concentrations ranging from 0.002% P₀ (1 ppm) to 0.2% P₀ (100 ppm) (Fig. 12c). More interestingly, the sensor showed a highly selective response to electron donors (triethylamine, acetone) and little response to electron acceptors (NO₂). Compared to the carbon nanotube-based sensor, the MoS₂ sensor exhibited higher selectivity towards volatile organic compounds (VOCs) at room temperature (Fig. 12d).

Fig. 12 (a) MoS₂ monolayer is supported on the SiO₂/Si substrate and contacted with Au contact pads. Transient physisorption of molecules induces temporary conductivity changes of the sensing channel. (b) Optical image of the devices, showing the monolayer MoS₂ flakes electrically contacted by multiple Au leads. (c) Device responses under a series of exposure pulses of TEA concentration increases from 0.002 to 0.2 (%P₀). Inset: a model of TEA molecule. (d) Histogram of MoS₂ and CNT-network sensor responses to various gases (reproduced from ref. 16).

Vapour phase synthesis was considered as a promising choice to obtain high quality and large scale MoS₂ films.^15,18–20 As a typical example, Duesberg et al.¹⁵ developed a high-performance sensor based on MoS₂ thin films prepared by sulfurization of sputtered Mo thin-films. The MoS₂ thin-film sensor showed a fast response toward NH₃ with experimental LOD down to 300 ppb. However, they don't immediately recover at room temperature. In another example, Zhou et al.¹⁸ performed Schottky-contacted chemical sensing using CVD-grown monolayer MoS₂ synthesized with MoO₃ and sulfur as starting materials. Observed from the optical microscopy (OM) image in Fig. 13a, the as-grown monolayer MoS₂ on SiO₂/Si substrates was a triangle-shaped material, with lateral dimensions of 5–30 μm. The MoS₂ devices are also exhibited in the OM image of Fig. 13b. Furthermore, the sensing performance of the as-fabricated MoS₂ FETs also demonstrated the ultrasensitive detection of NO₂ down to a few ppb level and NH₃ down to 1 ppm (Fig. 13c and d).

Fig. 13 Optical microscopy image of (a) the as-grown MoS₂ monolayers on SiO₂/Si substrates and (b) the MoS₂ device. Real-time conductance change of MoS₂ device after exposure to NO₂ (c) and NH₃ (d) with various concentrations. Inset in (d) is a zoom-in plot of the sensor response to low NH₃ concentrations of 1, 5, and 10 ppm (reproduced from ref. 18).

To achieve the highly selective and ultra-sensitive detection of volatile organic compounds (VOCs) in the lung cancer diagnosis of breath analysis, Jung et al.¹⁴¹ developed a tunable chemiresistor sensor by using MoS₂ films and thiolated ligand (mercaptoundecanoic acid (MUA)) conjugated MoS₂ films. Both films displayed high selectivity and sensitivity (down to 1 ppm) toward the representative VOC groups (ethanol, acetone, propanal, toluene, and hexane). In particular, the primitive and MUA-conjugated MoS₂ sensing channels exhibited a distinctly opposite direction of sensor response toward VOCs. Therefore, it will be a promising approach to construct a versatile sensor array for application in lung cancer diagnosis, by conjugating a wide variety of thiolated ligands on the MoS₂ surface. In addition, metallic nanoparticles were utilized for modifying MoS₂ to improve their selectivity and sensitivity behaviors.^142,143 As a typical example, Jin et al.¹⁴² described a highly sensitive H₂ sensor based on the MoS₂-Pd composite by a facile solvent exfoliation and drop casting method. A schematic illustration of the MoS₂-Pd composite and the optical image of the sensor device are exhibited in Fig. 14a and b, respectively. In particular, upon exposure to 500–50 000 ppm H₂, the sensor showed a fast response and recovery time of 40 and 83 s, respectively (Fig. 14c). Furthermore, the favorable selectivity of the sensor is also demonstrated in Fig. 14d.

Fig. 14 (a) Schematic illustration of the MoS₂–Pd composite. (b) MoS₂–Pd composite sensor device. (c) Electrical response of the MoS₂–Pd composite exposed to different concentrations of H₂ (500–50 000 ppm) by 40 s pulses. (d) Cross-sensitivity of the MoS₂–Pd composite to 50 000 ppm hydrogen, 50 ppm ammonia, 50 000 ppm acetone and ethanol (reproduced from ref. 142).

Moreover, to avoid the unstable response of pristine MoS₂ nanosheets in an air environment, Chen et al.¹⁴⁴ developed a novel nanohybrid of the SnO₂ nanocrystal-decorated crumpled MoS₂ nanosheet (MoS₂/SnO₂). Fig. 15a illustrated schematically the preparation process of the MoS₂/SnO₂ nanohybrids. Furthermore, the MoS₂/SnO₂ sensor exhibited a reversible, reliable response and recovery characteristic toward various NO₂ concentrations ranging from 0.5 ppm to 10 ppm (Fig. 15b), unlike other graphene- or carbon nanotube-based NO₂ sensors that require annealing treatment, or UV illumination to facilitate recovery. The sensing response of the MoS₂/SnO₂ nanohybrid to various gases at a concentration of 10 ppm was also exhibited in Fig. 15c, indicating an excellent selectivity to NO₂ at room temperature.

Fig. 15 (a) Schematic illustration of the preparation process for MoS₂/SnO₂ nanohybrids. Inset: photographs of the MoS₂ suspension in water before and after adding SnCl₄ solution. (b) Dynamic sensing response of the MoS₂/SnO₂ nanohybrids toward different NO₂ concentrations. Inset: the enlarged sensing response curve for 0.5 ppm NO₂. (c) Sensing response of the MoS₂/SnO₂ nanohybrid to various gases at a concentration of 10 ppm (reproduced from ref. 144).

3.2.2 MoSe₂-based gas sensors. Similar to layered MoS₂, atomically thin MoSe₂ consists of sandwiching two atomic layers of Se atoms and one atomic layer of Mo through covalent bonds but has a larger atomic structure due to the larger atomic radius of the Se atoms than that of S atoms.^145–148 Due to its unique 2D structure, the MoSe₂ layer can be employed as a chemical sensor for detecting traces of toxic gas molecules. For instance, Late et al.²¹ described a high-performance room temperature NH₃ sensor based on single-layer MoSe₂ prepared by the micromechanical exfoliation method. Fig. 16a showed the typical optical image of the single-layer MoSe₂ sample. The single-layer MoSe₂ sensor device fabricated on the SiO₂/Si substrate by the electron beam lithography (EBL) technique and the actual sensor device mounted on the chip carrier are also present in the SEM and optical images of Fig. 16b and c, respectively. The sensor device showed comprehensible detection of NH₃ gas at concentrations of 50, 100, 200, 300, and 500 ppm (Fig. 16d). Furthermore, this single-layer MoSe₂ based NH₃ sensor displayed a relatively short response time (2.5 min), which is lower than that reported for single-layer MoS₂, ranging between 5 and 9 min.¹⁸

Fig. 16 (a) Optical image of the single-layer MoSe₂. (b) SEM image of the single-layer MoSe₂ sensor device supported on the SiO₂/Si substrate. (c) Optical image of the as-fabricated gas sensor. (d) Sensor responses as a function of NH₃ concentration (reproduced from ref. 21).

3.2.3 WS₂-based gas sensors. Similar to graphene and MoS₂, tungsten disulfide (WS₂) is an atomic-layer nanomaterial. The monolayer WS₂ possessed a hexagonal configuration, each three pairs of S atoms anchored on one W atom, and formed alternating corners (S–W–S) in a honeycomb network.^149,150 Due to its excellent electronic properties, favourable band structure and high surface-to-volume ratio, WS₂ has been reported in photoelectronic applications, such as field effect transistors (FETs),^151,152 photodetectors,^22,153 photocatalysts,^154,155 gas sensors.^22–24

For example, Li et al.²² developed a transistor based on multilayer WS₂ nanoflakes using a conventional mechanical exfoliation technique. The thickness of WS₂ nanoflakes is 42 nm, and the width (W) and length (L) of the sensing channel in the device is 15 μm and 20 μm, respectively (Fig. 17a and b). The charge transfer processes between adsorbed gas molecules (O₂, ethanol, and NH₃) and the multilayer WS₂ device are schematically illustrated in Fig. 17c. When exposed to various gas atmospheres (NH₃, ethanol, air, and O₂), the WS₂ device exhibited obviously distinct electrical and photosensitive responses (Fig. 17d). Furthermore, the gas sensitivities of the device under different conditions were also investigated in Fig. 17e. The gas sensitivity was defined as the following: S = (I_gas − I_vacuum)/I_vacuum × 100%, where I_gas is the current of the device in the targeted gas, and I_vacuum is the current under vacuum, V_DS was fixed at 1 V. In this case, ethanol and NH₃ molecules acted as electron donors to increase the electron concentration in WS₂, thus leading to positive response. In particular, the device exhibited a higher sensitivity toward NH₃ than that of the other gases, meaning the WS₂ flakes were more sensitive to NH₃ molecules. In contrast, O₂ was the electron acceptor and can extract electrons from the WS₂ flakes, displaying a negative response. In addition, the device exhibited higher sensitivity in O₂ than that in air, due to the higher concentration of O₂ acting as an electron acceptor to deplete electrons.

Fig. 17 AFM (a) and SEM image (b) of the actual transistor based on multilayer WS₂ nanoflakes. (c) Schematic diagram of the charge transfer process between adsorbed gas molecules and the multilayer WS₂ nanoflakes transistor. (d) Time-dependent photocurrent response under various gas atmospheres. (e) The gas sensitivities (A) and current changes (B) under different conditions (reproduced from ref. 22).

Very recently, Pumera et al.²³ presented an impedance phase spectrum of the interdigitated gold electrode modified with chemically exfoliated 1T-WS₂ for selective vapour sensing (Fig. 18a). The 1T-WS₂ impedimetric device demonstrated distinct resonant frequencies with a maximum around 1 Hz in the presence of methanol vapour and around 1 kHz in the presence of water vapour (Fig. 18b). Such a well-distinguished signal allowed their selective detection even though they are present in a mixture. Moreover, the impedance device displayed highly selective detection of methanol and water vapour in the presence of excess concentrations of interference species (Fig. 18c). This selective impedimetric vapour sensing based on 1T-WS₂ opens a new avenue for spectra applications in gas-sensing.

Fig. 18 (a) Schematic representation of an interdigitated gold electrode modified with 1T-WS₂ nanosheets. (b) Impedance phase spectra of 1T-WS₂ device bare and toward 506 ppm methanol vapour, 26.6 g kg⁻¹ water vapour, and in the mixture of both vapours. (c) Selectivity studies of the 1T-WS₂ platform toward toluene (1394 ppm), acetone (1268 ppm), ethanol (1264 ppm), isopropyl alcohol (1259 ppm), methanol (380 ppm), and water vapour (27 g kg⁻¹) (reproduced from ref. 23).

3.3 SnS₂-based gas sensors

Similar to the aspect of transition metal dichalcogenides (TMDs), tin disulfide (SnS₂) exists in a layered crystal phase, which is composed of tin atoms sandwiched between two layers of hexagonally close-packed sulfur atoms.¹⁵⁶ Compared to 2D layered MoS₂, SnS₂ possessed a larger electronegativity, potentially enhancing gas adsorption sites, which is demonstrated for the gas-sensing application.^25,26,157 For instance, Zhang et al.²⁵ developed a room-temperature NH₃ sensor based on flower-shaped SnS₂ consisting of interconnected nanosheets with thicknesses of about 40 nm prepared by a mild hydrothermal method. Fig. 19a shows a schematic diagram of a typical gas sensor. The SnS₂ exhibited densely packed layers of well-edged hexagonal nanosheets with an average size of 1 μm (Fig. 19b). Upon exposure to the NH₃ concentrations in the range 5 to 150 ppm, the SnS₂ sensor presents a rapid response and recovery rate at room temperature (Fig. 19c), suggesting the promise of applying flower-shaped SnS₂ nanostructures in the detection of NH₃ molecules at room temperature. Very recently, Ou et al.²⁶ presented a high sensitivity, excellent selective and reversible NO₂ sensor based on two-dimensional (2D) SnS₂ flakes at low operating temperatures. The as-obtained hexagonal 2D SnS₂ flakes were prepared by a solution-based method and their lateral dimensions mainly ranged from 80 to 200 nm according to transmission electron microscopy (TEM) in Fig. 19d. Upon exposure to NO₂ with concentrations ranging from 0.6 to 10 ppm, the sensor based on 2D SnS₂ flakes displayed a quick response and recovery kinetics at an operating temperature of 120 °C (Fig. 19e). An excellent reversibility of the sensor is demonstrated, which has rarely been observed in other 2D material counterparts. Moreover, the 2D SnS₂ flake-based sensor exhibited very strong selectivity to NO₂ among other industrially meaningful concentrations of gases, including H₂ (1%), CH₄ (10%), CO₂ (10%), and H₂S (56 ppm) (Fig. 19f). The 2D SnS₂-based sensor provided a feasible approach to achieve high-performance NO₂ sensing for a wide range of applications in environmental monitoring, industry, and healthcare.

Fig. 19 (a) Schematic diagram of a typical gas sensor. (b) SEM image of the as-synthesized SnS₂ nanoflowers. (c) Response curve on cycling between increasing concentration of NH₃ gas (5–150 ppm) (reproduced from ref. 25). (d) TEM image of 2D SnS₂ flakes. (e) Dynamic sensing performance of the sensor based on 2D SnS₂ flakes to NO₂ at concentrations ranging from 0.6 to 10 ppm under the operation temperature of 120 °C. (f) Sensor responses of 2D SnS₂ flakes toward H₂ (1%), CH₄ (10%), CO₂ (10%), H₂S (56 ppm), and NO₂ (10 ppm) (reproduced from ref. 26).

3.4 Black phosphorus (BP)-based gas sensors

Recently, the rediscovery of black phosphorus (BP) as a new single-element 2D layered material has sparked a great deal of interest in the gas-sensing field.^28–31 As a typical example, Zhou et al.²⁹ described NO₂ sensing using field-effect transistors based on multilayer BP. Fig. 20a revealed an optical image showing the multilayer BP FET used for NO₂ gas detection. When exposed to NO₂ gas, this BP sensor exhibited increased conductivity and excellent sensitivity for the detection of NO₂ down to 5 ppb (Fig. 20b). Moreover, when this BP field-effect transistor was exposed to NO₂ concentrations of 5, 10, 20, and 40 ppb, its relative conductivity change obeyed the Langmuir isotherm for molecules adsorbed on a surface (Fig. 20c). Very recently, Pumera et al.³⁰ demonstrated a highly selective methanol sensor based on layered black phosphorus (BP) using the electrochemical impedance spectroscopy transduction method. Fig. 20d presents the device based on an interdigitated gold electrode modified with layered BP. A strong signal of the impedance phase spectra was clearly observed in the presence of methanol vapour at a resonance frequency of approximately 1 kHz. In addition, this peak enhanced with the increasing concentration of methanol vapour (Fig. 20e). The excellent selectivity of the BP sensor toward methanol was also demonstrated by recording the impedance phase at a frequency of 1 kHz after successive additions of toluene (2790 ppm), acetone (2535 ppm), chloroform (4775 ppm), dichloromethane (4262 ppm), ethanol (2519 ppm), isopropanol (2519 ppm), and water (1602 ppm) (Fig. 20f). These results implied that BP was a promising candidate for gas-sensing application. Besides, Salehi-Khojin et al.³¹ demonstrated an ultrasensitive and selective humidity sensor based on stacked BP flakes. The drain current of the FET-based BP sensor ranked the highest ever reported values for humidity detection with about 4 orders of magnitude as the relative humidity (RH) varied from 10% to 85%. Moreover, the BP sensor remained nearly unchanged in response after prolonged exposure (up to 3 months) to ambient conditions.

Fig. 20 (a) Optical image of the multilayer BP flake between two Ti/Au electrodes. (b) Relative conductance change (ΔG/G₀) of the BP sensor toward NO₂ with concentration ranging from 5 to 40 ppb. Inset: a zoomed image of exposure response to 5 ppb NO₂. (c) ΔG/G₀ plotted vs. NO₂ concentration applied to the BP FET showing an agreement between the measured values (red squares) and the fitted Langmuir isotherm (reproduced from ref. 29). (d) Vapour sensor based on BP. (e) Impedance phase spectra of the BP device exposed to different concentrations of methanol vapour. (f) Impedance phase at 1 kHz in the presence of either methanol (1140 ppm), toluene (2790 ppm), acetone (2535 ppm), chloroform (4775 ppm), dichloromethane (4262 ppm), ethanol (2519 ppm), isopropanol (2519 ppm), or water (1602 ppm) (reproduced from ref. 30).

3.5 The others

Some other layered 2D materials, such as VS₂, BN, and C₃N₄, have been developed for gas-sensing application. Xie et al.²⁷ presented a highly responsive humidity sensor using liquid-exfoliated VS₂ nanosheets fabricated on a flexible PET substrate. Feng et al.³² reported a gas sensor based on atomic-layered BN nanosheets synthesized using the CO₂-PLD technique. The as-fabricated sensor displayed a high sensitivity toward methane due to the possible reaction of CH₄ molecules with B and N atoms in honeycomb-like BN nanosheets. In another example, Chen et al.³³ performed a novel C-rich C₃N₄ for the detection of NO₂ gas. The C₃N₄ was synthesized by a two-step thermal treatment of glucose and urea. Moreover, the C₃N₄ sensor possessed good sensitivity and excellent selectivity toward NO₂ at room temperature, and LOD was low to 140 ppb.

4. Conclusion and outlook

In this critical review, we have summarized the recent developments in the reported typical 2D layered nanomaterials for gas-sensing applications. The results demonstrate that 2D layered nanomaterials are competitive candidates for fabrication of high-performance gas sensors operated at low or room temperature due to their particular characteristics. In contrast, among these 2D layered nanomaterials, graphene has a zero band gap, and its semiconductivity fails to be modulated upon exposure to gas molecules. However, semiconductivity is an important property for a gas-sensing material, which mainly decides the potential of materials in gas sensing applications. Thus, pristine graphene materials may not be ideal for fabricating sensing devices. Fortunately, doped or modified graphene with foreign atoms and groups and functional micro/nanomaterials have been proved to be beneficial for enhancement of gas-sensing performance. Unlike graphene, metal dichalcogenides (MoS₂, MoSe₂, WS₂, SnS₂ and VS₂) possess tunable band gaps dependent on their layer thickness, and their excellent semiconductivities make them attractive candidates for fabrication of gas sensors. However, up to now, there has been no reliable strategy to obtain scalable and controlled synthesis of certain layers, which hampers the development of integrated sensor research. Tables 1 and 2 summarize the important studies on gas sensors based on some typical 2D layered nanomaterials. Certainly, it is not possible for us to cover all progress in this review and there are definitely many interesting and exciting research topics waiting exploration. Some of the remaining challenges are listed as the following:

(1) Improvement of response and recovery characteristics: 2D layered nanomaterial based gas sensors always exhibit extremely sluggish response and incomplete recovery to the initial state during operation at room temperature. This phenomenon is attributed to their charge transfer sensing mechanism. To improve these issues, the following strategies are worthy of attention: developing novel gas-sensing nanomaterials, surface functionalization and doping, accurate fabrication techniques, unique measurement techniques, and optimizing device geometry.

(2) Fabrication of gas sensors: most of the gas sensors are fabricated by a drop casting or microfabrication process. However, these processes not only are complicated and time-consuming, but also fail to display the intrinsic properties of the channel materials. Thus, it still remains a huge challenge to develop a facile, effective and reliable strategy to fabricate gas sensors.

(3) Practical application of gas sensors: after development for several decades, sensitivity, has shown great improvement, and is no longer the main concern of a gas sensor, while the selectivity and stability are still great challenges in practical applications. For gas sensors based on 2D layered nanomaterials, they always exhibit a cross-response to several kinds of reactive gases, leading to poor selectivity. Furthermore, most of them possess unstable response signals while being operated at low or room temperature. To address these issues, several strategies may be adopted, including surface modification, functional doping, and optimizing test conditions.

Table 1 Graphene materials for gas-sensing applications (G: graphene, DNT: 2,4-dinitrotoluene, DMMP: dimethyl methylphosphonate)

No.	Graphene materials	Sensor type	Detected gas	Sensitivity	LOD	Ref.
1	Mechanically exfoliated G	Hall geometry	NO2	ΔR = 2.5 Ω (1 electron)	1 molecule	64
2	CVD-grown G	FET	NO2	ΔR/R = 12.49% (5 ppm)	6.87 ppb	68
3	CVD-grown G	FET	NO2	ΔI/I0 = 1.67 ppm^−1	300 ppt	76
4	Epitaxially grown G	Chemiresistor	NO2	ΔG/G = −9% (18 ppm, Si-face)		81
				ΔG/G = 4% (18 ppm, C-face)
5	Hydrazine-reduced GO	Chemiresistor	DNT	ΔR/R = −0.028% (52 ppb)		98
6	Vc-reduced GO	Chemiresistor	NO2	ΔR/R = 6% (10 ppm)	400 ppt	100
7	PPD-reduced GO	Chemiresistor	DMMP	ΔR/R = 8% (20 ppm)		101
8	NaBH4-reduced GO	Chemiresistor	NH3	ΔR/R = 5.5% (200 ppm)		102
9	Sulfonated rGO	Chemiresistor	NO2	ΔG/G = 0.443 ppm^−1	3.6 ppm	48
10	Ag-S-RGO	Chemiresistor	NO2	ΔR/R = −45% (50 ppm)		103
11	Pt–G	Chemiresistor	H2	ΔR/R = 16% (4%)		105
12	Pd-RGO	Chemiresistor	NO	ΔG/G = 4% (2 ppb)		106
13	AgNW coated rGO	Chemiresistor	NH3	ΔR/R = 3% (15 ppm)		107
14	SnO2-RGO	FET	NO2	I g/Ia = 1.11 (1 ppm)		112
15	Cu2O-rGO	Chemiresistor	NO2	ΔI/I0 = 67.8% (2 ppm)	64 ppb	114
16	CuO–ZnO/G	Chemiresistor	Acetone	R a/Rg = 9.4 (10 ppm)		115
17	ZnO QDs/G	Chemiresistor	Formaldehyde	ΔG/G = 1.1 (100 ppm)		129
18	Pt–SnO2/G	Chemiresistor	H2	ΔR/R = 2 (1%)		43
19	CNT/rGO	FET	NO2	ΔR/R = 20% (20 ppm)		116
20	PANI/G	Chemiresistor	NH3	ΔR/R = 32.5% (10 ppm)		118
21	PSS-PANI/G	Chemiresistor	H2S	ΔR/R = 60% (20 ppm)	1 ppm	119
22	PEDOT/G	FET	NO2	ΔR/R = 20% (2 ppm)	500 ppb	120
23	P NFs/G	Chemiresistor	NO2	ΔR/R = 1.03 ppm^−1	150 ppb	122

---

Table 2 2D layered inorganic nanomaterials for gas-sensing applications

No.	Materials	Sensor type	Detected gas	LOD	Ref.
1	2 layer MoS2	FET	NO	0.3 ppm	14
2	MoS2 thin film	FET	NH3	300 ppb	15
3	MoS2 monolayer	Chemiresistor	Triethylamine	10 ppb	16
4	2, 5 layered MoS2	FET	NH3, NO2		17
5	MoS2 monolayer	FET	NH3, NO2	1 ppm, 10 ppb	18
6	Thiolated ligand MoS2	Chemiresistor	VOCs	<1 ppm	141
7	Pd–MoS2	Chemiresistor	H2	500 ppm	142
8	Pt–MoS2	FET	NO2	0.5 ppm	143
9	SnO2–MoS2	FET	NO2	0.5 ppm	144
10	Atomic layered MoS2	FET	NH3, NO2	1 ppm, 1 ppm	20
11	Monolayer MoSe2	FET	NH3	50 ppm	21
12	Multilayer WS2	FET	NH3, ethanol		22
13	Metallic 1T-WS2	Impedance	Methanol	5.6 ppm	23
14	Flower-shaped SnS2	Chemiresistor	NH3	5 ppm	25
15	2D SnS2 flakes	Chemiresistor	NO2	20–30 ppb	26
16	SnS2–SnO2 nanohybrids	Chemiresistor	NH3	10 ppm	157
17	Black phosphorus	FET	NO2	5 ppb	29
18	Black phosphorus	Impedance	Methanol	28 ppm	30
19	BN nanosheets	Chemiresistor	CH4		32
20	C-rich C3N4	Chemiresistor	NO2	140 ppb	33

---

Acknowledgements

This work was supported by the National Nature Science Foundation of China (21322106, 21501060 and 51472097), the National Basic Research Program of China (973 program, 2015CB932600), the Program for New Century Excellent Talents in University (NCET-13-0227), the Program for HUST Interdisciplinary Innovation Team and the Fundamental Research Funds for the Central University. The authors are indebted for the kind permission from the corresponding publishers/authors to reproduce their materials, especially figures, used in this article.

References

1. X. Zhou, S. Lee, Z. C. Xu and J. Yoon, Chem. Rev., 2015, 115, 7944–8000.
2. A. Tricoli, M. Righettoni and A. Teleki, Angew. Chem., Int. Ed., 2010, 49, 7632–7659.
3. S. Su, W. H. Wu, J. M. Gao, J. X. Lu and C. H. Fan, J. Mater. Chem., 2012, 22, 18101–18110.
4. Y. F. Sun, S. B. Liu, F. L. Meng, J. Y. Liu, Z. Jin, L. T. Kong and J. H. Liu, Sensors, 2012, 12, 2610–2631.
5. G. F. Fine, L. M. Cavanagh, A. Afonja and R. Binions, Sensors, 2010, 10, 5469–5502.
6. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699–712.
7. G. Fiori, F. Bonaccorso, G. Iannaccone, T. Palacios, D. Neumaier, A. Seabaugh, S. K. Banerjee and L. Colombo, Nat. Nanotechnol., 2014, 9, 768–779.
8. C. N. R. Rao, H. S. S. Ramakrishna Matte and U. Maitra, Angew. Chem., Int. Ed., 2013, 52, 13162–13185.
9. X. Zhou, L. Gan, W. M. Tian, Q. Zhang, S. Y. Jin, H. Q. Li, Y. Bando, D. Golberg and T. Y. Zhai, Adv. Mater., 2015, 27, 8035–8041.
10. F. Schwierz, J. Pezoldt and R. Granzner, Nanoscale, 2015, 7, 8261–8283.
11. Y. X. Liu, X. C. Dong and P. Chen, Chem. Soc. Rev., 2012, 41, 2283–2307.
12. Q. Y. He, S. X. Wu, Z. Y. Yin and H. Zhang, Chem. Sci., 2012, 3, 1764–1772.
13. W. J. Yuan and G. Q. Shi, J. Mater. Chem. A, 2013, 1, 10078–10091.
14. H. Li, Z. Y. Yin, Q. Y. He, X. Huang, G. Lu, D. W. H. Fam, A. I. Y. Tok, Q. Zhang and H. Zhang, Small, 2012, 8, 63–67.
15. K. Lee, R. Gatensby, N. McEvoy, T. Hallam and G. S. Duesberg, Adv. Mater., 2013, 25, 6699–6702.
16. F. K. Perkins, A. L. Friedman, E. Cobas, P. M. Campbell, G. G. Jernigan and B. T. Jonker, Nano Lett., 2013, 13, 668–673.
17. D. J. Late, Y. K. Huang, B. Liu, J. Acharya, S. N. Shirodkar, J. J. Luo, A. M. Yan, D. Charles, U. V. Waghmare, V. P. Dravid and C. N. R. Rao, ACS Nano, 2013, 7, 4879–4891.
18. B. L. Liu, L. Chen, G. Liu, A. N. Abbas, M. Fathi and C. W. Zhou, ACS Nano, 2014, 8, 5304–5314.
19. B. Cho, M. G. Hahm, M. Choi, J. Yoon, A. R. Kim, Y. J. Lee, S. G. Park, J. D. Kwon, C. S. Kim, M. Song, Y. Jeong, K. S. Nam, S. Lee, T. J. Yoo, C. G. Kang, B. H. Lee, H. C. Ko, P. M. Ajayan and D. H. Kim, Sci. Rep., 2015, 5, 8052.
20. B. Cho, A. R. Kim, Y. Park, J. Yoon, Y. J. Lee, S. Lee, T. J. Yoo, C. G. Kang, B. H. Lee, H. C. Ko, D. H. Kim and M. G. Hahm, ACS Appl. Mater. Interfaces, 2015, 7, 2952–2959.
21. D. J. Late, T. Doneux and M. Bougouma, Appl. Phys. Lett., 2014, 105, 233103.
22. N. J. Huo, S. X. Yang, Z. M. Wei, S. S. Li, J. B. Xia and J. B. Li, Sci. Rep., 2014, 4, 5209.
23. C. C. Mayorga-Martinez, A. Ambrosi, A. Y. S. Eng, Z. Sofer and M. Pumera, Adv. Funct. Mater., 2015, 25, 5611–5616.
24. M. O'Brien, K. Lee, R. Morrish, N. C. Berner, N. McEvoy, C. A. Wolden and G. S. Duesberg, Chem. Phys. Lett., 2014, 615, 6–10.
25. W. D. Shi, L. H. Huo, H. S. Wang, H. J. Zhang, J. H. Yang and P. H. Wei, Nanotechnology, 2006, 17, 2918–2924.
26. J. Z. Ou, W. Y. Ge, B. Carey, T. Daeneke, A. Rotbart, W. Shan, Y. C. Wang, Z. Q. Fu, A. F. Chrimes, W. Wlodarski, S. P. Russo, Y. X. Li and K. Kalantar-zadeh, ACS Nano, 2015, 9, 10313–10323.
27. J. Feng, L. L. Peng, C. Z. Wu, X. Sun, S. L. Hu, C. W. Lin, J. Dai, J. L. Yang and Y. Xie, Adv. Mater., 2012, 24, 1969–1974.
28. L. Z. Kou, T. Frauenheim and C. F. Chen, J. Phys. Chem. Lett., 2014, 5, 2675–2681.
29. A. N. Abbas, B. L. Liu, L. Chen, Y. Q. Ma, S. Cong, N. Aroonyadet, M. Kopf, T. Nilges and C. W. Zhou, ACS Nano, 2015, 9, 5618–5624.
30. C. C. Mayorga-Martinez, Z. Sofer and M. Pumera, Angew. Chem., Int. Ed., 2015, 54, 14317–14320.
31. P. Yasaei, A. Behranginia, T. Foroozan, M. Asadi, K. Kim, F. Khalili-Araghi and A. Salehi-Khojin, ACS Nano, 2015, 9, 9898–9905.
32. M. Sajjad and P. Feng, Mater. Res. Bull., 2014, 49, 35–38.
33. D. H. Wang, W. Gu, Y. W. Zhang, Y. Hu, T. Zhang, X. M. Tao and W. Chen, RSC Adv., 2014, 4, 18003–18006.
34. Q. M. Ji, S. B. Yoon, J. P. Hill, A. Vinu, J. S. Yu and K. Ariga, J. Am. Chem. Soc., 2009, 131, 4220–4221.
35. K. Ariga, A. Vinu, Q. M. Ji, O. Ohmori, J. P. Hill, S. Acharya, J. Koike and S. Shiratori, Angew. Chem., Int. Ed., 2008, 47, 7254–7257.
36. Z. Li, J. Lu, S. D. Li, S. H. Qin and Y. M. Qin, Adv. Mater., 2012, 24, 6053–6057.
37. B. R. Takulapalli, G. M. Laws, P. A. Liddell, J. Andreasson, Z. Erno, D. Gust and T. J. Thornton, J. Am. Chem. Soc., 2008, 130, 2226–2233.
38. F. Naso, F. Babudri, D. Colangiuli, G. M. Farinola, F. Quaranta, R. Rella, R. Tafuro and L. Valli, J. Am. Chem. Soc., 2003, 125, 9055–9061.
39. M. C. Tanese, G. M. Farinola, B. Pignataro, L. Valli, L. Giotta, S. Conoci, P. Lang, D. Colangiuli, F. Babudri, F. Naso, L. Sabbatini, P. G. Zambonin and L. Torsi, Chem. Mater., 2006, 18, 778–784.
40. K. Ariga, Y. Yamauchi, G. Rydzek, Q. M. Ji, Y. Yonamine, K. C. W. Wu and J. P. Hill, Chem. Lett., 2014, 43, 36–68.
41. K. Ariga, Y. Yamauchi, T. Mori and J. P. Hill, Adv. Mater., 2013, 25, 6477–6512.
42. X. Huang, X. Y. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012, 41, 666–686.
43. Y. W. Zhu, S. Murali, W. W. Cai, X. S. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.
44. X. Huang, Z. Y. Zeng and H. Zhang, Chem. Soc. Rev., 2013, 42, 1934–1946.
45. Y. F. Sun, S. Gao, F. C. Lei and Y. Xie, Chem. Soc. Rev., 2015, 44, 623–636.
46. P. K. Kannan, D. J. Late, H. Morgan and C. S. Rout, Nanoscale, 2015, 7, 13293–13312.
47. G. H. Yang, C. Z. Zhu, D. Du, J. J. Zhu and Y. H. Lin, Nanoscale, 2015, 7, 14217–14231.
48. W. J. Yuan, A. R. Liu, L. Huang, C. Li and G. Q. Shi, Adv. Mater., 2013, 25, 766–771.
49. P. A. Russo, N. Donato, S. G. Leonardi, S. Baek, D. E. Conte, G. Neri and N. Pinna, Angew. Chem., Int. Ed., 2012, 51, 11053–11057.
50. W. Yang, P. Wan, X. D. Zhou, J. M. Hu, Y. F. Guan and L. Feng, ACS Appl. Mater. Interfaces, 2014, 6, 21093–23100.
51. L. Li, S. J. He, M. M. Liu, C. M. Zhang and W. Chen, Anal. Chem., 2015, 87, 1638–1645.
52. H. Zhang, J. C. Feng, T. Fei, S. Liu and T. Zhang, Sens. Actuators, B, 2014, 190, 472–478.
53. H. Meng, W. Yang, K. Ding, L. Feng and Y. F. Guan, J. Mater. Chem. A, 2015, 3, 1174–1181.
54. K. Anand, O. Singh, M. P. Singh, J. Kaur and R. C. Singh, Sens. Actuators, B, 2014, 195, 409–415.
55. D. Jariwala, V. K. Sangwan, L. J. Lauhon, T. J. Marks and M. C. Hersam, ACS Nano, 2014, 8, 1102–1120.
56. Z. Y. Zhang, X. M. Zou, L. Xu, L. Liao, W. Liu, J. Ho, X. H. Xiao, C. Z. Jiang and J. C. Li, Nanoscale, 2015, 7, 10078–10084.
57. Y. P. Dan, Y. Lu, N. J. Kybert, Z. T. Luo and A. T. Johnson, Nano Lett., 2009, 9, 1472–1475.
58. B. Kumar, K. Min, M. Bashirzadeh, A. B. Farimani, M. H. Bae, D. Estrada, Y. D. Kim, P. Yasaei, Y. D. Park, E. Pop, N. R. Aluru and A. Salehi-Khojin, Nano Lett., 2013, 13, 1962–1968.
59. R. Arsat, M. Breedon, M. Shafiei, P. G. Spizziri, S. Gilje, R. B. Kaner, K. Kalantar-zadeh and W. Wlodarski, Chem. Phys. Lett., 2009, 467, 344–347.
60. S. Rumyantsev, G. X. Liu, M. S. Shur, R. A. Potyrailo and A. A. Balandin, Nano Lett., 2012, 12, 2294–2298.
61. X. Huang, Z. Y. Yin, S. X. Wu, X. Y. Qi, Q. Y. He, Q. C. Zhang, Q. Y. Yan, F. Boey and H. Zhang, Small, 2011, 7, 1876–1902.
62. Y. M. Lin and P. Avouris, Nano Lett., 2008, 8, 2119–2125.
63. Z. G. Cheng, Q. Li, Z. J. Li, Q. Y. Zhou and Y. Fang, Nano Lett., 2010, 10, 1864–1868.
64. 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.
65. 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.
66. Y. B. Zhang, Y. W. Tan, H. L. Stormer and P. Kim, Nature, 2005, 438, 201–204.
67. X. S. Li, W. W. Cai, J. An, S. Kim, J. Nah, D. X. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo and R. S. Ruoff, Science, 2009, 324, 1312–1314.
68. Y. H. Kim, S. J. Kim, Y. J. Kim, Y. S. Shim, S. Y. Kim, B. H. Hong and H. W. Jang, ACS Nano, 2015, 9, 10453–10460.
69. Y. J. Ren, C. F. Zhu, W. W. Cai, H. F. Li, H. X. Ji, I. Kholmanov, Y. P. Wu, R. D. Piner and R. S. Ruoff, Appl. Phys. Lett., 2012, 100, 163114.
70. F. Yavari, E. Castillo, H. Gullapalli, P. M. Ajayan and N. Koratkar, Appl. Phys. Lett., 2012, 100, 203120.
71. R. K. Paul, S. Badhulika, N. M. Saucedo and A. Mulchandani, Anal. Chem., 2012, 84, 8171–8178.
72. Y. W. Yu, L. Gan, X. F. Wan and T. Y. Zhai, RSC Adv., 2015, 5, 56293–56298.
73. F. Yavari, Z. Chen, A. V. Thomas, W. Ren, H. M. Cheng and N. Koratkar, Sci. Rep., 2011, 1, 166.
74. J. W. Suk, A. Kitt, C. W. Magnuson, Y. F. Hao, S. Ahmed, J. An, A. K. Swan, B. B. Goldberg and R. S. Ruoff, ACS Nano, 2011, 5, 6916–6924.
75. L. G. P. Martins, Y. Song, T. Y. Zeng, M. S. Dresselhaus, J. Kong and P. T. Araujo, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 17762–17767.
76. S. Kumar, S. Kaushik, R. Pratap and S. Raghavan, ACS Appl. Mater. Interfaces, 2015, 7, 2189–2194.
77. C. Berger, Z. M. Song, X. B. Li, X. S. Wu, N. Brown, C. Naud, D. Mayou, T. B. Li, J. Hass, A. N. Marchenkov, E. H. Conrad, P. N. First and W. A. d. Heer, Science, 2006, 312, 1191–1196.
78. M. Qazi, M. W. K. Nomani, M. V. S. Chandrashekhar, V. B. Shields, M. G. Spencer and G. Koley, Appl. Phys. Express, 2010, 3, 075101–075103.
79. S. Y. Zhou, G.-H. Gweon, A. V. Fedorov, P. N. First, W. A. d. Heer, D.-H. Lee, F. Guinea, A. H. C. Neto and A. Lanzara, Nat. Mater., 2007, 6, 770–775.
80. J. Y. Dai and J. M. Yuan, Chem. Phys., 2012, 405, 161–166.
81. M. W. K. Nomani, R. Shishir, M. Qazi, D. Diwan, V. B. Shields, M. G. Spencer, G. S. Tompa, N. M. Sbrockey and G. Koley, Sens. Actuators, B, 2010, 150, 301–307.
82. R. Pearce, T. Iakimov, M. Andersson, L. Hultman, A. L. Spetz and R. Yakimova, Sens. Actuators, B, 2011, 155, 451–455.
83. S. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217–224.
84. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228–240.
85. C. K. Chua and M. Pumera, Chem. Soc. Rev., 2014, 43, 291–312.
86. D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. Dommett, G. Evmenenko, S. T. Nguyen and R. S. Ruoff, Nature, 2007, 448, 457–460.
87. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS Nano, 2010, 4, 4806–4814.
88. J. Kim, L. J. Cote and J. X. Huang, Acc. Chem. Res., 2012, 45, 1356–1364.
89. J. Kim, L. J. Cote, F. Kim, W. Yuan, K. R. Shull and J. X. Huang, J. Am. Chem. Soc., 2010, 132, 8180–8186.
90. Q. Xu, H. Xu, J. R. Chen, Y. Z. Lv, C. B. Dong and T. S. Sreeprasad, Inorg. Chem. Front., 2015, 2, 417–424.
91. A. Bagri, C. Mattevi, M. Acik, Y. J. Chabal, M. Chhowalla and V. B. Shenoy, Nat. Chem., 2010, 2, 581–587.
92. L. Liao, H. L. Peng and Z. F. Liu, J. Am. Chem. Soc., 2014, 136, 12194–12200.
93. Y. R. Choi, Y.-G. Yoon, K. S. Choi, J. H. Kang, Y.-S. Shim, Y. H. Kim, H. J. Chang, J.-H. Lee, C. R. Park, S. Y. Kim and H. W. Jang, Carbon, 2015, 91, 178–187.
94. S. F. Pei and H. M. Cheng, Carbon, 2012, 50, 3210–3228.
95. R. L. D. Whitby, ACS Nano, 2014, 8, 9733–9754.
96. T. Kuila, A. K. Mishra, P. Khanra, N. H. Kim and J. H. Lee, Nanoscale, 2013, 5, 52–71.
97. J. T. Robinson, F. K. Perkins, E. S. Snow, Z. Q. Wei and P. E. Sheehan, Nano Lett., 2008, 8, 3137–3140.
98. J. D. Fowler, M. J. Allen, V. C. Tung, Y. Yang, R. B. Kaner and B. H. Weiller, ACS Nano, 2009, 3, 301–306.
99. G. H. Lu, S. Park, K. H. Yu, R. S. Ruoff, L. E. Ocola, D. Rosenmann and J. H. Chen, ACS Nano, 2011, 5, 1154–1164.
100. V. Dua, S. P. Surwade, S. Ammu, S. R. Agnihotra, S. Jain, K. E. Roberts, S. Park, R. S. Ruoff and S. K. Manohar, Angew. Chem., Int. Ed., 2010, 49, 2154–2157.
101. N. T. Hu, Y. Y. Wang, J. Chai, R. G. Gao, Z. Yang, E. S. Kong and Y. F. Zhang, Sens. Actuators, B, 2012, 163, 107–114.
102. R. Ghosh, A. Midya, S. Santra, S. K. Ray and P. K. Guha, ACS Appl. Mater. Interfaces, 2013, 5, 7599–7603.
103. L. Huang, Z. P. Wang, J. K. Zhang, J. L. Pu, Y. J. Lin, S. H. Xu, L. Shen, Q. Chen and W. Z. Shi, ACS Appl. Mater. Interfaces, 2014, 6, 7426–7433.
104. F. Niu, J. M. Liu, L. M. Tao, W. Wang and W. G. Song, J. Mater. Chem. A, 2013, 1, 6130–6133.
105. A. Kaniyoor, R. I. Jafri, T. Arockiadoss and S. Ramaprabhu, Nanoscale, 2009, 1, 382–386.
106. W. W. Li, X. M. Geng, Y. F. Guo, J. Z. Rong, Y. P. Gong, L. Q. Wu, X. M. Zhang, P. Li, J. B. Xu, G. S. Cheng, M. T. Sun and L. W. Liu, ACS Nano, 2011, 5, 6955–6961.
107. Y. Pak, S. M. Kim, H. Jeong, C. G. Kang, J. S. Park, H. Song, R. Lee, N. Myoung, B. H. Lee, S. Seo, J. T. Kim and G. Y. Jung, ACS Appl. Mater. Interfaces, 2014, 6, 13293–13298.
108. D. H. Shin, J. S. Lee, J. Jun, J. H. An, S. G. Kim, K. H. Cho and J. Jang, Sci. Rep., 2015, 5, 12294.
109. Q. T. Tran, H. T. M. Hoa, D.-H. Yoo, T. V. Cuong, S. H. Hur, J. S. Chung, E. J. Kim and P. A. Kohl, Sens. Actuators, B, 2014, 194, 45–50.
110. V. Tjoa, W. Jun, V. Dravid, S. Mhaisalkar and N. Mathews, J. Mater. Chem., 2011, 21, 15593–15599.
111. B. Cho, J. Yoon, M. G. Hahm, D. H. Kim, A. R. Kim, Y. H. Kahng, S. W. Park, Y. J. Lee, S. G. Park, J. D. Kwon, C. S. Kim, M. Song, Y. Jeong, K. S. Nam and H. C. Ko, J. Mater. Chem. C, 2014, 2, 5280–5285.
112. S. Mao, S. M. Cui, G. H. Lu, K. H. Yu, Z. H. Wen and J. H. Chen, J. Mater. Chem., 2012, 22, 11009–11013.
113. X. Q. An, J. C. Yu, Y. Wang, Y. M. Hu, X. L. Yu and G. J. Zhang, J. Mater. Chem., 2012, 22, 8525–8531.
114. S. Deng, V. Tjoa, H. M. Fan, H. R. Tan, D. C. Sayle, M. Olivo, S. Mhaisalkar, J. Wei and C. H. Sow, J. Am. Chem. Soc., 2012, 134, 4905–4917.
115. C. Wang, J. W. Zhu, S. M. Liang, H. P. Bi, Q. F. Han, X. H. Liu and X. Wang, J. Mater. Chem. A, 2014, 2, 18635–18643.
116. H. Y. Jeong, D. S. Lee, H. K. Choi, D. H. Lee, J. E. Kim, J. Y. Lee, W. J. Lee, S. O. Kim and S. Y. Choi, Appl. Phys. Lett., 2010, 96, 213105.
117. L. Al-Mashat, K. Shin, K. Kalantar-zadeh, J. D. Plessis, S. H. Han, R. W. Kojima, R. B. Kaner, D. Li, X. L. Gou, S. J. Ippolito and W. Wlodarski, J. Phys. Chem. C, 2010, 114, 16168–16173.
118. X. L. Huang, N. T. Hu, R. G. Gao, Y. Yu, Y. Y. Wang, Z. Yang, E. Siu-Wai Kong, H. Wei and Y. F. Zhang, J. Mater. Chem., 2012, 22, 22488–22495.
119. S. Cho, J. S. Lee, J. Jun, S. G. Kim and J. Jang, Nanoscale, 2014, 6, 15181–15195.
120. Y. J. Yang, S. B. Li, W. Y. Yang, W. T. Yuan, J. H. Xu and Y. D. Jiang, ACS Appl. Mater. Interfaces, 2014, 6, 13807–13814.
121. L. Zhang, C. Li, A. R. Liu and G. Q. Shi, J. Mater. Chem., 2012, 22, 8438–8443.
122. W. J. Yuan, L. Huang, Q. H. Zhou and G. Q. Shi, ACS Appl. Mater. Interfaces, 2014, 6, 17003–17008.
123. M. M. Arafat, B. Dinan, S. A. Akbar and A. S. M. A. Haseeb, Sensors, 2012, 12, 7207–7258.
124. K. J. Choi and H. W. Jang, Sensors, 2010, 10, 4083–4099.
125. E. Comini, C. Baratto, I. Concina, G. Faglia, M. Falasconi, M. Ferroni, V. Galstyan, E. Gobbi, A. Ponzoni, A. Vomiero, D. Zappa, V. Sberveglieri and G. Sberveglieri, Sens. Actuators, B, 2013, 179, 3–20.
126. Y. B. Hahn, R. Ahmad and N. Tripathy, Chem. Commun., 2012, 48, 10369–10385.
127. Y. Yang, C. G. Tian, L. Sun, R. J. Lü, W. Zhou, K. Y. Shi, K. Kan, J. C. Wang and H. G. Fu, J. Mater. Chem. A, 2013, 1, 12742–12749.
128. L. S. Zhou, F. P. Shen, X. K. Tian, D. H. Wang, T. Zhang and W. Chen, Nanoscale, 2013, 5, 1564–1569.
129. Q. W. Huang, D. W. Zeng, H. Y. Li and C. S. Xie, Nanoscale, 2012, 4, 5651–5658.
130. R. K. Mishra, S. B. Upadhyay, A. Kushwaha, T. H. Kim, G. Murali, R. Verma, M. Srivastava, J. Singh, P. P. Sahay and S. H. Lee, Nanoscale, 2015, 7, 11971–11979.
131. S. Chopra, K. McGuire, N. Gothard, A. M. Rao and A. Pham, Appl. Phys. Lett., 2003, 83, 2280.
132. J. Li, Y. J. Lu, Q. Ye, M. Cinke, J. Han and M. Meyyappan, Nano Lett., 2003, 3, 929–933.
133. S. Hanna Varghese, R. Nair, B. G. Nair, T. Hanajiri, T. Maekawa, Y. Yoshida and D. Sakthi Kumar, Curr. Nanosci., 2010, 6, 331–346.
134. Y. Battie, O. Ducloux, P. Thobois, N. Dorval, J. S. Lauret, B. Attal-Trétout and A. Loiseau, Carbon, 2011, 49, 3544–3552.
135. S. Dhall, N. Jaggi and R. Nathawat, Sens. Actuators, A, 2013, 201, 321–327.
136. H. Bai, K. X. Sheng, P. F. Zhang, C. Li and G. Q. Shi, J. Mater. Chem., 2011, 21, 18653–18658.
137. Q. M. Ji, I. Honma, S. M. Paek, M. Akada, J. P. Hill, A. Vinu and K. Ariga, Angew. Chem., Int. Ed., 2010, 49, 9737–9739.
138. L. T. Duy, D. J. Kim, T. Q. Trung, V. Q. Dang, B. Y. Kim, H. K. Moon and N. E. Lee, Adv. Funct. Mater., 2015, 25, 883–890.
139. R. Ganatra and Q. Zhang, ACS Nano, 2014, 8, 4074–4099.
140. Q. Y. He, Z. Y. Zeng, Z. Y. Yin, H. Li, S. X. Wu, X. Huang and H. Zhang, Small, 2012, 8, 2994–2999.
141. J. S. Kim, H. W. Yoo, H. O. Choi and H. T. Jung, Nano Lett., 2014, 14, 5941–5947.
142. C. Kuru, C. Choi, A. Kargar, D. Choi, Y. J. Kim, C. H. Liu, S. Yavuz and S. Jin, Adv. Sci., 2015, 2, 1500004.
143. D. Sarkar, X. J. Xie, J. H. Kang, H. J. Zhang, W. Liu, J. Navarrete, M. Moskovits and K. Banerjee, Nano Lett., 2015, 15, 2852–2862.
144. S. M. Cui, Z. H. Wen, X. K. Huang, J. B. Chang and J. H. Chen, Small, 2015, 11, 2305–2313.
145. D. S. Kong, H. T. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
146. Y. H. Chang, W. J. Zhang, Y. H. Zhu, Y. Han, J. Pu, J. K. Chang, W. T. Hsu, J. K. Huang, C. L. Hsu, M. H. Chiu, T. Takenobu, H. N. Li, C. I. Wu, W. H. Chang, A. T. S. Wee and L. J. Li, ACS Nano, 2014, 8, 8582–8590.
147. J. C. Shaw, H. L. Zhou, Y. Chen, N. O. Weiss, Y. Liu, Y. Huang and X. F. Duan, Nano Res., 2014, 7, 1–7.
148. Y. Yang, S. T. Wang, J. C. Zhang, H. Y. Li, Z. L. Tang and X. Wang, Inorg. Chem. Front., 2015, 2, 931–937.
149. H. S. S. R. Matte, A. Gomathi, A. K. Manna, D. J. Late, R. Datta, S. K. Pati and C. N. R. Rao, Angew. Chem., Int. Ed., 2010, 122, 4153–4156.
150. X. P. Fang, C. X. Hua, C. R. Wu, X. F. Wang, L. Y. Shen, Q. Y. Kong, J. Z. Wang, Y. S. Hu, Z. X. Wang and L. Q. Chen, Chem. – Eur. J., 2013, 19, 5694–5700.
151. D. Ovchinnikov, A. Allain, Y. S. Huang, D. Dumcenco and A. Kis, ACS Nano, 2014, 8, 8174–8181.
152. Z. Q. Xu, Y. P. Zhang, S. H. Lin, C. X. Zheng, Y. L. Zhong, X. Xia, Z. P. Li, P. J. Sophia, M. S. Fuhrer, Y. B. Cheng and Q. L. Bao, ACS Nano, 2015, 9, 6178–6187.
153. N. Perea-López, A. L. Elías, A. Berkdemir, A. Castro-Beltran, H. R. Gutiérrez, S. M. Feng, R. T. Lv, T. Hayashi, F. López-Urías, S. Ghosh, B. Muchharla, S. Talapatra, H. Terrones and M. Terrones, Adv. Funct. Mater., 2013, 23, 5511–5517.
154. B. Mahler, V. Hoepfner, K. Liao and G. A. Ozin, J. Am. Chem. Soc., 2014, 136, 14121–14127.
155. Y. H. Sang, Z. H. Zhao, M. W. Zhao, P. Hao, Y. H. Leng and H. Liu, Adv. Mater., 2015, 27, 363–369.
156. M. Chhowalla, H. S. Shin, G. Eda, L. J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263–275.
157. K. Xu, N. Li, D. W. Zeng, S. Q. Tian, S. S. Zhang, D. Hu and C. S. Xie, ACS Appl. Mater. Interfaces, 2015, 7, 11359–11368.

---