MoS2 Nanosheets Sensitized with Quantum Dots for Room-Temperature Gas Sensors

Highlights Highly sensitive and selective room-temperature NO2 gas sensors by sensitizing MoS2 nanosheets with PbS quantum dots were demonstrated. In this device architecture, the receptor and transduction function as well as the utility factor of semiconductor gas sensors could be enhanced simultaneously. The strategy of sensitizing 2D semiconductors with quantum dots as sensitive and selective receptors for gas molecules may offer a powerful new degree of freedom to the surface and interface engineering of semiconductor gas sensors. Electronic supplementary material The online version of this article (10.1007/s40820-020-0394-6) contains supplementary material, which is available to authorized users.


Introduction
Hazardous air pollutants have become a serious problem for the ecosystem and public health [1,2]. Nitrogen dioxide (NO 2 ) primarily gets in the air from the burning of fuel. Exposure to NO 2 may potentially increase susceptibility to respiratory infections, and a 5-min emergency exposure limit of 35 ppm NO 2 exposure has been proposed by the American Industrial Hygiene Association [1,3]. The large-scale networking of gas sensors for achieving online NO 2 monitoring requires the power consumption of the sensors to be lower. While semiconductor gas sensor have been widely used in home alarm system owing to their high sensitivity, simple operation, and low cost [4][5][6], their scale-up application in environmental internet has not been achieved due to the limitation of high operating temperature (typically above 300 °C) which raises the power consumption. The high operating temperature of semiconductor gas sensors also sets a limit to their integrability with CMOS technology or flexible electronic system. Thereby, novel nanostructured materials [7][8][9][10] with the potentials for room-temperature gas sensors have become a hot research topic.
MoS 2 is a well-known 2D graphene-like transition metal dichalcogenides (TMDs). With relatively high carrier mobility, large surface-to-volume ratio, and abundant edge sites which can provide active adsorption sites for gas molecules [11][12][13][14], MoS 2 has been demonstrated as one of the promising materials candidates for room-temperature NO 2 gas sensors. Liu et al. [13] reported CVD growth of monolayer MoS 2 for room-temperature detection of NO 2 with a response time of several minutes without a full recovery to the initial state. Cho et al. [15] demonstrated a chargetransfer-based sensitive NO 2 gas sensor by CVD-synthesized atomic-layered MoS 2 , with a sensitivity of 220% and a long time (more than 30 min) to recovery. Similarly, chemical exfoliated MoS 2 prepared by Jung et al. had an incompletable recovery to NO 2 at room temperature [16]. Kumar et al. fabricated a high-performance NO 2 sensor based on MoS 2 with abundant active edge sites. When operated at 60 °C, it had a fast response (16 s) with complete recovery (172 s) with a relative response of 18.1% to 5 ppm NO 2 [17]. As an alternative strategy, UV light irradiation or gate effect was employed to improve sensitivity toward NO 2 of MoS 2 sensor [18][19][20][21]. Pham et al. [18] employed LED illumination to improve sensitivity of CVD grown single-layer MoS 2 , achieving sub-ppb limit of NO 2 gas detection. However, the comparatively high sensitivity and fast response/recovery kinetics at room temperature were not simultaneously obtained for pristine MoS 2 gas sensors. They suffer from the trade-off between receptor and transducer function. For semiconductor gas sensors, the structural defects are always necessary for gas molecule reception and, on the contrary, may decrease the electronic transduction.
Recently, MoS 2 -based nanocomposites or hybrids through surface modification with noble metals [11], architecture design of hetero-nanostructures with metal oxide nanoparticles [22,23], and functionalization with other 2D-layered materials such as graphene [24][25][26][27][28] have been demonstrated with improved sensitivity and fast response/ recovery kinetics. Motivated by this strategy, we proposed to improve the room-temperature response and recovery by sensitizing MoS 2 nanosheets with quantum dots (QDs), a highly tunable zero-dimensional (0D) nanomaterial with size-dependent bandgap and excellent solution processability [29][30][31][32][33][34][35]. The huge amount of surface dangling bonds of QDs makes them sensitive receptors for gas molecules. Herein, the PbS QDs-sensitized MoS 2 nanosheets were obtained via a two-step solution process. The sensor had an excellent response of 6.15, to 10 ppm NO 2 at room temperature, almost five times greater than that of pristine MoS 2 nanosheets. The sensing mechanism was attributed to the enhanced receptor and transducer functions as well as the utility factor which determine the performance of semiconductor gas sensors.

Preparation of MoS 2 Nanosheets
In a typical hydrothermal synthesis of MoS 2 nanosheets [36], as shown in Fig. 1a 1 3 temperature. The final product was rinsed with deionized water and absolute ethanol several times to remove any possible ions. After drying at 70 °C for 6 h, black MoS 2 nanosheet powder was obtained. Figure 1a shows the synthesis of MoS 2 nanosheets sensitized with PbS QDs. Organo-hot injection method has always been proven as an effective method for QD synthesis [37][38][39]. First, the as-prepared MoS 2 powders (20 mg) were dissolved in 4 mL of oleic acid (OA). Ultrasonic dispersion was conducted for 30 min to ensure the black powder was completely dispersed in the solution. PbO (2 mmol), OA (2 mmol), 1-octadecene (ODE) (20 mL), and as-prepared MoS 2 (OA) solution (530 μL) were all mixed in a three-neck flask and heated to 90 °C under a vacuum for 6 h. Then, the reaction temperature was raised to 120 °C and 0.33 mmol bis(trimethylsilyl) sulfide (TMS) mixed with ODE (10 mL) was rapidly injected under an inert atmosphere. The reaction lasted for 30 s, and the mixture was then transferred to cold water bath for rapid cooling to room temperature. The nucleation and growth of QDs anchoring in the surface of MoS 2 nanosheets occurred in this process. The product was precipitated by acetone and re-dispersed in toluene several times to prepare PbS-MoS 2 solution for device fabrication.

Sensor Fabrication
The layer-by-layer spin-coating deposition technique of the sensitized MoS 2 -based thin film was carried out in ambient air at room temperature (a schematic illustration can be seen in Fig. 1b). Alumina ceramic substrates (15 × 15 × 0.8 mm 3 ) prepatterned with a pair of interdigital Ag electrode (the spacing and width are 5 mm) were prepared via screen printing. Then 70 μL of PbS-MoS 2 solution was dropped onto the substrate, which was then spun at 2350 rpm for 30 s. Next, four drops of NaNO 2 diluted in methanol (10 mg mL −1 ) were added dropwise to the film for ligand exchange, with a wait time of 45 s, and spun dry at 2500 rpm for 30 s, followed by repeating the NaNO 2 treatment twice. Finally, the film was washed by methanol flush and then spun dry three times to obtain 3-layers thin-film device. The film deposition process was  repeated three times. For comparison, the pristine MoS 2 nanosheet device was prepared according to the following steps. First, the prepared substrates were placed in a hotplate with a heating temperature of 135 °C. Next, a drop of MoS 2 ethanol solution was deposited dropwise onto the thermal substrate and naturally dried for a few seconds, followed by repeating the process twice. Finally, the fabricated MoS 2 sensor was maintained under the thermal treatment for 20 min.

Characterization and Measurements
A field emission scanning electron microscope (FE-SEM, GeminiSEM 300, Zeiss, Oberkochen, Germany) equipped with an energy-dispersive X-ray spectrometer (EDS, X-MAX, Oxford, UK) was used to obtain SEM images and elemental mapping data. Transmission electron microscopy (TEM) images were recorded with a Tecnai G2 20 microscope operating at an accelerating voltage of 200 kV. X-ray diffraction (XRD) measurements were obtained using a diffractometer (Empyrean, PANalytical B. V., Netherlands) with Cu Kα radiation in the 2θ range of 10-70 °C. An energy-dispersive X-ray spectrometer (EDS) was performed on a XL 30 ESEM FEG. X-ray photoelectron spectroscope (XPS) measurements were using by an AXIS-ULTRA DLD-600 W with an Al source, and C 1s peak at 284.5 eV is used as reference. Similarly, ultraviolet photoelectron spectroscopy (UPS) measurement was also performed by using the same system with a He-Iα 21.22 eV UV light. Work functions were measured by a KP 020 K probe (KP Technology, Wick, Scotland). UV-Vis-NIR absorption spectra were measured using a PerkinElmer Lambda 950 UV-Vis-NIR spectrophotometer.
The NO 2 sensing measurements were carried out by a computer-connected source meter system (Model Keithley 2450/6487, Keithley Instruments, USA) under static conditions controlled with the relative humidity (RH) being 19-85% at room temperature (sensor setup details as shown in Fig. S1). The sensor response was defined as the ratio of R a to R g , where R a is the baseline resistance in the ambient atmosphere and R g is the resistance of the sensor device in the presence of NO 2 gas. The response time (T 90 ) and the recovery time (T 10 ) were defined as the time taken by the sensor response to reach 90% of its maximum value upon exposure to NO 2 gas and drop to within 10% of its original baseline value after removal of gas.

Structural Properties of MoS 2 Nanosheets and QD-Sensitized MoS 2 Nanosheets
The morphology of the MoS 2 and QD-sensitized MoS 2 nanosheets was characterized with SEM and TEM, respectively. Figure 2a displays a low-magnification TEM image of MoS 2 nanosheets revealing the ultrathin nanosheet morphology with slightly assembly character. Further, more lattice fringes were clearly indicated from high-magnification TEM image ( Fig. 2b), revealing the labeled lattice spacing of 0.625 nm, which was in a good agreement with the (002) lattice plane with MoS 2 nanosheets. The abundant MoS 2 nanosheets layers provide large quantities of edge sites, which may beneficial for gas molecules absorption. Moreover, Fig. S2 shows an SEM image of the as-prepared MoS 2 nanosheets distributed on the alumina ceramic substrate, and the observable flowerlike MoS 2 nanosheets were uniformly assembled by a mass of bent flakes. Similarly, TEM images of different magnifications in Fig. 2c, d used to observe more detailed microstructure information of the QD-sensitized MoS 2 nanosheets. A large amount of QDs formed on the edge sites of the MoS 2 nanosheets as demonstrated in Fig. 2c. This could be attributed to the edge area defects, which provide more active sites for the nucleation of PbS QDs. Pb atoms can fill the vacancy on the MoS 2 surface, which may weaken the MoS 2 defects [40]. Equally important is that the MoS 2 surface might be spontaneously functionalized with the excessive OA molecules in the reaction process, and then the strong hydrophobic interaction [41,42] of the OA ligands on both the QDs and MoS 2 surfaces leading to the noncovalent binding of QDs to MoS 2 . However, the detailed mechanisms regarding how the OA ligands or molecules take part in the synthesis of MoS 2 nanosheets sensitized with QDs need further investigation. The efficient attachment and coverage of the QDs onto the MoS 2 nanosheets are further indicated by the high-resolution TEM image in Fig. 2d. Well-crystallized QDs with diameters of approximately 3.26 nm were uniformly separated on the surface of the MoS 2 nanosheets.

3
The lattice spacings of these spherically shaped QDs were 0.21 and 0.34 nm, corresponding to the (220) and (111) lattice planes of PbS, respectively. The edges of the MoS 2 nanosheets were not continuous, probably because some defects were generated in the synthesis processes. The typical elemental mapping data were characterized by EDS, as shown in Fig The surface elements and chemical states of the sensitized MoS 2 -based film were characterized by X-ray photoelectron spectroscopy (XPS) in the supporting information. As expected, Pb, Mo, and S were detected on the film, which was consistent with the EDS results. Figure S4a-c shows the high-resolution XPS spectra of Pb 4f, S 2p, and Mo 3d, respectively. Two peaks located at 142.7 and 137.8 eV correspond to the 4f 5/2 and 4f 7/2 of the Pb 2+ state exhibited in Fig.  S4a. Most of the Mo signal is from its Mo 4+ state at the peak positions around 228.5 and 229.2 eV, mainly corresponding to Mo 4+ 3d 5/2 (Fig. S4c). Two dominant S 2p peaks were observed around 161.5 and 162.2 eV (Fig. S4b), accompanied by a slightly flat peak at 163.8 eV, which were assigned to the divalent sulfide ions (S 2− ) of the MoS 2 and PbS.

NO 2 Gas-Sensing Properties
The NO 2 -sensing performance was measured using a homemade computer-connected source meter system under room temperature. We performed repeatability test for the both devices at the same time and measured the relative response to six and four successive cycles toward 10 ppm NO 2 for pristine MoS 2 nanosheets and the sensitized MoS 2 gas sensors, respectively (Fig. S5a). The pristine MoS 2 sensor showed the complete recovery at room temperature without any extra stimulus such as optical or thermal source; however, the completed response/recovery cycle required a slightly time. After sensitization by the PbS QDs, the sensitized MoS 2 sensor exhibited an obviously enhanced response to the same concentration of NO 2 gas, also with a fast response/recovery time and excellent reversibility. Transient resistance characteristic of MoS 2 nanosheets and the sensitized MoS 2 gas sensors to 10 ppm NO 2 is shown in Fig.  S5b, exhibiting p-type gas-sensing behavior for both sensors. The improved performance can be attributed to the excellent access of gas molecules adsorption by the PbS QDs as NO 2 receptors, as well as the favorable 0D-2D interface for charge transfer, which will be discussed in detail later. Three kinds of theoretical Mo to Pb molar ratio (2%, 5%, and 8%) were used in the precursor solutions during the synthesis, and we found that sensor response was much higher by a medium molar ratio of 5% (Fig. S6). Thus, we used this optimal molar ratio to sensor fabrication in this work. The representative time-resolved response and recovery curves of the pristine MoS 2 and the sensitized MoS 2 gas sensor were illustrated in more detail in Fig. 4a, b. In general, many defects may occur in the surface of MoS 2 , which can lead to a strong chemisorption between MoS 2 and gas molecules, so that NO 2 or other gases such as O 2 are difficult to desorb from the MoS 2 [43], resulting in a weakened recovery kinetics, as shown in Fig. 4a. The sensitized MoS 2 sensor exhibited a superior performance not only with an excellent response of 6.15 to 10 ppm NO 2 , which was almost five times greater than the pristine MoS 2 device, but also with an outstanding response/recovery ability, with the time improving from 50/233 to 15/62 s, respectively.
To further investigate the NO 2 -sensing properties of the sensors, the dynamic response curves were recorded with the NO 2 concentration of 1, 2, 5, 10, and 20 ppm, respectively, shown in Fig. 4c. Both devices showed recoverable response under room temperature, and the response values gradually increased with the increasing NO 2 gas concentration. Obviously, the device based on MoS 2 nanosheets sensitized with QDs was more sensitive than the pristine MoS 2 for NO 2 gas detection and indicated potential for a lower limit of detection (LOD). The pristine MoS 2 had less of a response when exposed to 1 ppm NO 2 , while the sensitized MoS 2 sensor still performed 2.30 toward the same concentration with a rapid response/recovery rate, which even better than the measurement to 20 ppm of pristine MoS 2 device (details are shown in Fig. S7). Owing to this improvement, the theoretical LOD for NO 2 was calculated to be 174 and 94 ppb in the case of pristine MoS 2 and QD-sensitized MoS 2 , respectively (calculation details in Fig. S8). However, the measurement error of the LOD for both sensors is mainly from accuracy 1 3 of gas concentration and errors in test results. The dependence of the sensor response on gas concentrations range from 1 to 20 ppm is also analyzed in Fig. 4d. The fitting equation between the response value (S) and NO 2 concentration (C) can be illustrated as a power law relationship, and the exponent was estimated to be 0.1089 together with a coefficient of determination (R 2 ) value of 0.988 for the MoS 2 sensor, while the values were 0.4523 and 0.991 for the sensitized MoS 2 sensor. Importantly, the theoretical analysis of the relationship between the response values and gas concentrations was significant for the gas sensor, which will facilitate the determination of gas concentrations in practical applications. Selectivity is considered as an important parameter for gas sensors, and we compared the response of the sensitized MoS 2 gas sensors toward several gases in our lab. As shown in Fig. 5, the sensors exhibited high response to 10 ppm NO 2 gas and negligible response to 10 ppm H 2 , SO 2 , NH 3 and 200 ppm C 2 H 5 OH vapor, respectively, at room temperature. The inset showed the dynamic response curves upon gas exposure and release of the intervening gases, respectively. We also investigated the NO 2 -sensing performance of the sensitized MoS 2 sensors in the range of RH of 19%, 29%, 48%, 65%, and 85%. The sensor response toward 10 ppm NO 2 had a tendency to grow over the RH gradually increased (shown in Fig. S9a). While the functional relationship between relative humidity and response could be further an accuracy of 2%. The average sensitivity to 10 ppm NO 2 under RH ~ 65% was 6.19, which was close to the average sensitivity of 6.14 under RH ~ 62% (Fig. 4d). Therefore, the RH ranged from 62 to 65% was within the error range. Under high RH environments, we suspected that water molecules preadsorbed on the surface of the sensitized MoS 2 , dissociating into OH − and H + to form hydroxyl groups. Hydroxyl groups as an electron donor lead to increase in resistance of the materials [44]. NO 2 has strong adsorption properties compared with the physical adsorption of water molecules. When NO 2 injected, they could kick out the physical adsorption of water molecules and cause a further decrease in resistance, thus achieving a higher response. Actually, it is reported that the hydroxyl groups could improve the NO 2 -sensing performance in recent study [35,[45][46][47]. Compared to other MoS 2 -based NO 2 sensors (Table 1), our MoS 2 nanosheets-based sensor only maintained a general level at room-temperature (RT) operation; however, under the same conditions, the sensitized MoS 2 sensor had a superior performance with no thermal treatment or UV illumination [21,48]. Compared to the most MoS 2 -based gas sensors in the current published papers, the sensitized MoS 2 gas sensor exhibited an excellent response from 6.15 to 10 ppm NO 2 at room temperature, accompanied by a rapid response/recovery time of 15/62 s, indicating high sensitivity  and outstanding recovery ability. In addition, as shown in Fig. S9c, long-term stability test of the sensitized MoS 2 sensor upon 10 ppm NO 2 was consistent with the effect of relative humidity on the sensing performance. Furthermore, QD-sensitized MoS 2 nanosheets with excellent solution processability are particularly attractive for next-generation gas sensors compatible with silicon-based or flexible substrates.

Gas-Sensing Mechanisms
As previously noted, the gas sensor based on MoS 2 nanosheets sensitized with QD had a good NO 2 -sensing performance at room temperature, which was quite possible for the combinational effects between the PbS QDs and MoS 2 nanosheets. Therefore, we proposed three basic factors of receptor function, transducer function and utility [49], as well as an interface energy band diagram to investigate the sensing mechanism of QD-sensitized MoS 2 nanosheets. As illustrated in Fig. 6a, PbS QDs always exhibited p-type conduction behavior in air atmosphere because of physisorbed O 2 molecules, which consumed electrons and introduced lots of holes as well. When exposed to NO 2 gas, according to our previous research [32,34,35], due to the strong binding energy compared to O 2 , NO 2 kicks out the originally physisorbed O 2 molecules and binds to Pb 2+ through O, introducing more charge-transfer-driven p-type doping and developing a hole concentration in the p-type PbS QDs. For pristine p-type MoS 2 nanosheets, the defects mainly on the edge sites of the MoS 2 acted as active sites for NO 2 molecules, and these defects dominated process contributing to the poor response, slow rates of response, or even incomplete recovery due to high energy binding sites [50], especially operation at room temperature without any illumination. Thus, the inevitable receptor-transducer function [51] conflict cannot be well addressed in the pristine MoS 2 -based gas sensor. After sensitization with QDs (illustrated in Fig. 6b), most of the high energy binding sites on the surface of MoS 2 were occupied by the highly active QD receptors which had larger surface-to-volume ratio as well as abundant surface defects (mainly from dangling bonds, surface Pb sites, sulfur vacancies, etc.) capable of active interaction with NO 2 gas molecules adsorption, contributing to a marked enhancement in the response. Furthermore, the adsorption energies of NO 2 on the MoS 2 and PbS were calculated based on the density functional theory (DFT) in the previous literature, indicating that the adsorption energy of NO 2 on the PbS is significantly larger than that on MoS 2 [52]. Therefore, PbS QDs may serve as receptors of NO 2 molecules and enhance the receptor function of the MoS 2 sensors. Combining with Fig. 6b, we also used an interface energy band diagram to further study the sensing mechanism. To simulate the actual environment, Kelvin probe measurement was carried out in ambient air. The work function (W F ) of the PbS QD and MoS 2 nanosheet was approximately 4.61 and 4.96 eV, respectively. Next, we used ultraviolet photoelectron spectroscopy (UPS) to confirm the valence-band edge (E v ) [53] and the scan of the spectra for both as shown in Fig. S10. The E v value was calculated to be 5.20 and 5.46 eV for the PbS QD and MoS 2 nanosheet, respectively. We also introduced a UV-Vis-NIR absorption spectrum mainly to evaluate the energy bandgaps (E g ) of the MoS 2 nanosheet and PbS QD. As shown in the MoS 2 spectrum in Fig. S11a, the characteristic absorption peaks that appeared in the visible regions were consistent with the general features of TMDs with trigonal prismatic coordination, which confirmed the 2H polytype of the MoS 2 nanosheet [54]. The intercept was interpolated inside giving the value to E g of 1.50 eV for MoS 2 through the Kubelka-Munk transformed reflectance spectra, indicating that the prepared MoS 2 with few-layer nanosheets possesses a bandgap larger than the bulk materials. Figure S11b shows that an exciton absorption peak appeared in 992 nm, from which we could obtain the calculated E g of 1.25 eV of PbS QD. It exhibited a significantly broadened bandgap compared to the bulk PbS (0.41 eV), confirming a conservation of strong quantum confinement effect [55]. Taking together the above experimental parameters, the initial condition (before mutual contact) of the energy band structure for PbS QD and MoS 2 nanosheet could be illustrated in Fig. S12. Because of the difference in work functions (4.61 vs. 4.96 eV), when the PbS and the MoS 2 were brought into contact, the electrons pass from the PbS to MoS 2 , creating a positive charge region closed to the PbS surface and opposite one near the MoS 2 surface. Finally, interface band structure was developed for both sides as band bending occurred and a potential barrier of 0.35 eV F = W F(PbS) − W F(MoS 2 ) formed in the contact position, which was accompanied by the balanced E F . As exhibited in the diagram in Fig. 6c, a majority of the NO 2 molecules adsorbed on the surface of the QD receptors may form donor-like surface states in general, and a direct electron extraction from the conduction band of QD into the NO 2 molecules, which also meant hole injection from the NO 2 into the valence band of QD. Anyway, a mass of holes will accumulate at the interface closed to the side of the PbS QDs during its receptor function process. Equally important was that the MoS 2 nanosheets served as the conductive path in the system, leading the NO 2 -induced holes flow to the electrode for collection, easily overcoming the relatively low potential barrier generated at the interface of the valenceband edge. DFT calculation results recently demonstrated that the diffusion barrier is only dozens of meV for NO 2 on MoS 2 , which also proved that NO 2 gas molecules may easily diffuse rapidly on MoS 2 surface [40]. Thus, MoS 2 nanosheets can serve as the charge transport highway for the effective transducer function of the sensitized surface adsorption of NO 2 gas molecules into an electrical resistance change of the sensor.
Concluded from the above discussion, the sensitized MoS 2 sensor had a good response and recovery kinetics even at room temperature because of the favorable 0D QD-2D MoS 2 interface, combining the improvement of both receptor function and transducer function [49,51,56]. Beyond that, the utility factor is one of the important factors which concerns the gas-sensing performance and goes up with the smaller pore size as well as thinner gassensitive film [49]. We took characterization about SEM cross-section morphology of the sensitized MoS 2 based on alumina ceramic substrate. However, it was difficult to observe the thickness of such nanothin film clearly on the rough ceramic substrate because it was hard for cutting. Hence, we employed the comparative smooth silicon substrate for material deposition. Figure S13a displays the cross section of the three-layer QD-sensitized MoS 2 thin film on silicon substrate, revealing a conformal film deposition, and the film thickness was estimated to be 135 nm. Thus, the utility factor could be benefited greatly from the relatively porous thin-film features, which enhanced the accessibility of inner sulfide grains to the NO 2 molecules, leading to enhanced gas diffusion and reaction, thereby achieving higher response along with shorter response/ recovery time. We further provided more details in Fig.  S13b about NO 2 -sensing performance of different deposited layers and finally found that the three-layer thin-filmbased sensors had a stable response together with a fast recovery time. In brief, our sensitized MoS 2 gas sensors exhibited a better NO 2 gas-sensing performance at room temperature than that of the pristine MoS 2 sensors. The sensitized MoS 2 architecture overcome the receptor-transducer function conflict limitation, as well as enhanced the utility factor by sensitizing MoS 2 nanosheets with QDs. More importantly, a deeper understanding of the 0D-QDs with tunable bandgaps will further promote progress in the engineering of energy band alignment at the 0D-2D heterojunction interface, paving a promising way to develop gas-sensing performance of 2D layered materials.

Conclusions
In summary, we proposed a facile synthesis strategy for sensitizing MoS 2 nanosheets with PbS quantum dots as NO 2 gas molecules. The sensitized MoS 2 gas sensor exhibited sensitive and recoverable response at room temperature, with the response/recovery time shortened from 50/233 to 15/62 s upon 10 ppm of NO 2 exposure/ release cycle, respectively, compared to the pristine MoS 2 nanosheets. The gas-sensing mechanism was attributed to the fundamental factors of receptor function, transducer function and utility, as well as the favorable 0D-2D interface between QDs and MoS 2 nanosheets. Through the surface sensitization of MoS 2 nanosheets with PbS QDs as sensitive and selective NO 2 receptors, combined with the favorable charge transfer at interfaces and excellent charge transport, the receptor and transducer function as well as the utility factor were desirable enhanced, thereby achieving the enhanced performance for NO 2 gas sensing. This work demonstrated a novel sensitized MoS 2 gas sensor with superb sensitivity and extremely low power consumption. The solution-processable and room-temperature operable gas sensors could be integrated with silicon-based or even flexible substrates to achieve smart on-chip electronic nose.

Supplementary Material
Homemade sensor setup; SEM image of the flowerlike MoS 2 nanosheets; EDS elemental mapping of QD-sensitized MoS 2 nanosheets; XPS characterization of QD-sensitized MoS 2 nanosheets; repeatability curves and transient resistance characteristic of the MoS 2 nanosheets and QD-sensitized MoS 2 nanosheets sensors; sensor response of QD-sensitized MoS 2 with different Pb:Mo; transient relative response of MoS 2 sensors toward different NO 2 concentrations; LOD calculation of MoS 2 nanosheets sensor and QD-sensitized MoS 2 sensor; sensor response at different relative humidity and long-term stability of the QD-sensitized MoS 2 gas sensors; UPS characterization of MoS 2 nanosheets and PbS QDs; UV-Vis-NIR spectra of MoS 2 nanosheets and PbS QDs; the initial energy band structure of PbS QD and MoS 2 nanosheet; SEM cross-section morphology of QD-sensitized MoS 2 thin film; and NO 2 -sensing properties of QD-sensitized MoS 2 with different deposition layers.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.