Zinc oxide–black phosphorus composites for ultrasensitive nitrogen dioxide sensing

Abstract

Multi-layer black phosphorus (m-BP) has been successfully incorporated in zinc oxide (ZnO) hollow spheres, thus forming hetero-structured ZnO–BP composites; a study performed on their properties reveals that the BP environmental stability can be significantly enhanced by the introduction of ZnO, and the sensors based on ZnO–BP composites exhibit high response, fast response behavior, outstanding selectivity, and ultralow detection limit of 1 ppb towards NO₂ gas molecules. The ultrasensitive sensing performance is suggested to be due to the large surface area, excellent carrier mobility, and enhanced charge transfer of ZnO–BP in the presence of BP. Moreover, the mechanism of NO₂ sensing with ZnO–BP is confirmed by the calculations obtained from the first-principle studies.

---

Conceptual insights

Phospohorene, due to its ultrahigh surface–volume ratio, high chemical activity and excellent carrier mobility, has been shown to be promising for gas sensing. However, the fabricated BP sensor suffers from some disadvantages, such as poor environmental stability, low sensitivity and long response–recovery process, due to which it is difficult to use in practical environments or safety monitoring. Multi-layer black phosphorus (m-BP) is successfully incorporated in zinc oxide (ZnO) hollow spheres, thus forming hetero-structured ZnO–BP composites. Interestingly, the introduction of ZnO significantly enhances BP environmental stability, making it a suitable candidate to use in practice. Remarkably, ZnO–BP sensor exhibits higher response, outstanding selectivity, faster response and recovery dynamic process, and an ultralow detection limit of 1 ppb to NO₂ in contrast to graphene-, BP- and ZnO-based sensors, which were previously used. This study has provided a route in the search of BP-based materials with excellent properties for use in practice, in particular towards gas sensing.

Phosphorene, one of the novel elemental 2D materials, exhibits a single layer of phosphorus sheet exfoliated from the bulk black phosphorus (BP); it has recently received extensive attention due to its significant advantages over semi-metallic graphene. These advantages include thickness-dependent direct band gap variation from monolayer to bulk BP in the range of 1.5–0.3 eV and free-carrier mobility (∼1000 cm² V s⁻¹), which is better than those of other typical 2D semiconductors such as molybdenum disulfide, tungsten diselenide, and boron nitride (BN).¹ However, it has been reported that phosphorene can be irreversibly degraded to oxidized phosphorus compounds under ambient conditions.^2,3 The proper protection of its surfaces by encapsulating with graphene, BN, oxides or multi-layer black phosphorus (m-BP) can generally improve the structural and environmental stabilities.^4–7

Phosphorene, due to its ultrahigh surface–volume ratio, high chemical activity, and excellent carrier mobility, has been demonstrated to be a promising material for gas sensing applications.^8–10 A good sensing functionality of phosphorene and doped phosphorene with even better performances than those of graphene and MoS₂ was theoretically demonstrated.¹¹ However, to the best of our knowledge and based on few decades of experience in this field, we have found that reports on experimental studies about phosphorene-based sensors are scarce. Recently, Abbas et al. first experimentally explored the chemical sensing of nitrogen dioxide (NO₂) gas molecules using field-effect transistors (FET) that were based on m-BP.⁹ However, the fabricated BP sensor suffered from a few disadvantages: it exhibited low sensitivity and long response–recovery process, and it was not specifically selective to NO₂, thus making it difficult to apply in practical items and in safety monitoring systems.

Basically, NO₂ is a typical polluting gas with pungent odor; it is generated from the combustion of chemical plants and automotive emissions, and it plays a major role in the formation of acid rain, photochemical smog and ground-level ozone, which are potentially harmful to human health and the ecosystem. The Japanese standards strictly regulated the threshold limit value (TLV) for NO₂ to less than 40–60 ppb.^12,13 Therefore, the detection of trace levels of NO₂ becomes an exceedingly significant task for monitoring environmental pollution. Gas sensors based on metal oxide semiconductors (MOSs) such as In₂O₃, ZnO, SnO₂, Fe₂O₃, and WO₃ are predominant gas detecting devices for domestic, laboratorial and commercial applications, owing to their miniaturized dimensions, availability of easy fabrication methods, low-costs, and considerable chemical, structural and environmental stabilities.^14–19 However, in the last decade, the concentration detection limit of MOS sensors was still high, resulting in their threshold and resolution deficiencies to monitor accurately the variations of low concentrations of NO₂ in atmosphere.

Herein, we illustrate that the incorporation of m-BP nanosheets into ZnO hollow spheres can significantly improve NO₂ sensing properties (Fig. 1). The hetero-structured ZnO–BP composites exhibit high response, ultralow detection limit, quick response and recovery characteristics as well as excellent selectivity towards NO₂. After incorporation of BP, the detection limit of ZnO–BP sensor decreases to 1 ppb, and the response increases to 20 under 100 ppb NO₂, which is more than triple the responses of ZnO and BP sensors. First-principles density functional theory calculations suggest that there is a charge transfer from ZnO to BP via Zn–P bonds in the ZnO–BP heterostructure, which can increase the local chemical affinity of BP and enhance the sensing properties of ZnO–BP towards NO₂ molecules. Moreover, the existence of BP increases gas adsorption and diffusion, the surface area-to-volume ratio and carrier mobility. Our results also indicate that the introduction of ZnO can significantly enhance BP environmental stability, making it suitable to use in practice. We believe that this study will benefit further investigations of the fundamental science and applications of phosphorene in the context of chemical sensors.

Fig. 1 Sensor device configuration and the molecular structures of the materials.

The morphologies and structures of ZnO and ZnO–BP were characterized with scanning electron microscopy (SEM), transmission electron microscopy (TEM), high angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), high resolution TEM (HRTEM), corresponding selected area electron diffraction (SAED) and atomic force microscopy (AFM). The panoramic SEM image in Fig. 2a shows that the obtained ZnO was composed of numerous monodispersed spheres. The products were hollow in shape with an average diameter of ∼4 μm and thickness of spherical shell of about 400 nm. The external and internal surfaces of the spherical shells composed of abundant interlinked nanoparticles contributed to their rough and porous structure, which increased the specific surface area and the surface defects. The HRTEM image and SAED pattern (Fig. 2b) obtained from a random area on the ZnO hollow spheres indicated that the particles forming the structure were polycrystalline in nature. The lattice inter-planar spacing of 0.28 nm corresponding to the (100) plane of wurtzite ZnO was evidenced. We further investigated the elemental distribution of ZnO hollow spheres using HAADF-STEM (Fig. 2c), where clear peaks for O (red) and Zn (orange) on the whole sphere shells were detected. This profile coupled with composition line-scan (green line) across ZnO verified the hollow shape of the ZnO spheres. Fig. 2d shows the SEM image of the ZnO–BP composites. The morphology of ZnO was maintained even after ultrasonic treatment except that few nanosheets of BP were located on the surface of the hollow spheres. The TEM image shown in Fig. 2e confirmed the presence of 2D BP nanosheets. The 0.218 nm lattice spacing of the fringes was assigned to the (020) plane of orthorhombic BP. The SAED pattern result suggested that these nanosheets are single crystalline in nature.²⁰ AFM of ZnO–BP composites on Si/SiO₂ substrate demonstrated the presence of a range of shapes and sizes of phosphorene (Fig. 2f). Height-profiling of nanosheets revealed well-defined edges of 2.95 nm thickness. According to the AFM measurements of BP in a previous study, we can infer that these nanosheets predominantly consist of more than five-layered phosphorene.²¹

Fig. 2 (a–c) SEM, TEM with corresponding SAED and HRTEM, and HAADF-STEM, images of ZnO. (d–f) SEM, HRTEM with SAED, and AFM height profile of m-BP nanosheets images of ZnO–BP.

The several peaks of ZnO–BP in X-ray diffraction (XRD) patterns observed at 31.64°, 34.30° and 36.12° (Fig. 3a) are indexed to the diffraction planes (100), (002), and (101) of wurtzite ZnO (JCPDS No. 36-1451). The other diffraction peaks at 16.81°, 26.61° and 52.40° are characteristic peaks for the orthorhombic phase of BP,²² suggesting that the structure of BP remains the same after liquid ultrasonication. Fig. 3b shows the Raman spectra of BP and ZnO–BP composites. Three strong peaks correspond to the out-of-plane vibration mode A¹_g, in-plane vibration mode B_2g and A²_g of BP. Noteworthily, ZnO–BP composites exhibit the A²_g mode at 462.57 cm⁻¹, which is 1.75 cm⁻¹ lower than that of BP; this is probably due to the stacking-induced anisotropic structure change from bulk to multilayer phosphorene after ultrasonic treatment.^3,21 Furthermore, the intensities of Raman peaks related to bulk phosphorus are smaller than those of ZnO–BP, and this is consistent with the results of MoS₂–graphene composites, which is attributed to the effects of optical interference.^20,23 These results demonstrate the successful synthesis of multilayered BP with anhydrous organic solvents in a sealed-tip ultrasonication system.

Fig. 3 (a) XRD patterns of BP, ZnO and ZnO–BP. (b) Raman spectra of BP and ZnO–BP.

To further understand the surface compositions and chemical states of the elements present in ZnO–BP composites, we studied X-ray photoelectron spectra (XPS) of BP, ZnO and ZnO–BP. As shown in Fig. 4a and b, the complete spectra of BP and ZnO–BP confirmed involvement of P, O, C and Zn, P, O, and C elements, respectively. The peak at 284.2 eV corresponded to the adventitious carbon-based contaminants adsorbed on the surfaces. P 2p electron core-level XPS spectra of BP and ZnO–BP composites are shown in Fig. 4c. The two peaks at 130.1 and 131.0 eV were assigned to 2p_3/2 and 2p_1/2 of P³⁺, respectively, and they were characteristics of crystalline BP.^24,25 Please note that compared to BP, ZnO–BP exhibited a higher XPS signal at 131.0 eV and a distinct lower signal at 134.8 eV, indicating more P³⁺ and less P⁵⁺ in ZnO–BP. It has been reported that the P⁵⁺ sub-bands at the higher binding energy are observed due to PO_x for oxidized phosphorus species, because of the degradation upon exposure to ambient conditions.^25–27 The indiscernible signal of P⁵⁺ ions in the XPS spectrum of ZnO–BP suggested that the introduction of ZnO hollow spheres can enhance BP environmental stability.²⁶ We also analyzed the time evolution of the P 2p signal, as shown in Fig. S1 (ESI†). The signals of P⁵⁺ ions in the XPS spectra of ZnO–BP were still indiscernible after 24, 48, and 96 hours, suggesting good environmental stability of ZnO–BP. In addition, the Zn 2p electron core level XPS spectra of pristine ZnO and ZnO–BP composites are given in Fig. 4d; the results indicated two symmetric peaks at 1022.77 and 1045.88 eV, which were attributed to 2p_3/2 and 2p_1/2, respectively, of Zn²⁺ in the pure ZnO hollow spheres. For the ZnO–BP composites, the Zn 2p peaks shifted to high binding energy, evidently after introduction of BP. The fact that the charge density of ZnO decreased, whereas that of BP increased, implied the existence of a charge transfer process from ZnO to BP in the ZnO–BP composites.

Fig. 4 (a and b) Survey XPS spectra of BP and ZnO–BP, respectively. (c) P 2p spectra of BP and ZnO–BP. (d) Zn 2p spectra of ZnO and ZnO–BP.

The responses of ZnO–BP composites and different reference materials towards 100 ppb NO₂ were measured at various operating temperatures. As shown in Fig. 5a, the maximum response of the sensor based on ZnO–BP was 20 at 160 °C, and the response of ZnO sensor reached 5.2 at 140 °C. Clearly, the ZnO–BP sensor exhibited more than triple the response of that of the ZnO sensor toward NO₂ gas. Furthermore, no noticeable responses were generated from the sensors based on BP, GO, and ZnO–GO materials when they were exposed to 100 ppb NO₂. From the view of the practical application, a sensor should present a rather high selectivity. Therefore, the cross-sensitivities to 1 ppm oxidizing gases (SO₂) and reducing gases (H₂, CO, NH₃, CH₃COCH₃, CH₃CH₂OH, and CH₃OH) were also measured for the BP, ZnO and ZnO–BP sensors at 160 °C (Fig. 5b). It was seen that the ZnO–BP sensor exhibited the largest response to 100 ppb NO₂ compared with those to 1 ppm of seven other gases, although the concentrations of others were much higher. Meanwhile, relatively lower responses to the interferential gases were observed for the composites than those for pure ZnO; this indicated that the response of ZnO to interferential gases was suppressed after its combination with BP. Due to an enhanced response to NO₂ and the decreased responses to the other interferential gases, the as-prepared ZnO–BP composites exhibited an excellent selectivity to NO₂. It is well-known that the response and recovery times are also important parameters for evaluating the properties of a gas sensor. The transient characteristics of the sensors based on ZnO–BP and ZnO toward 100 ppb NO₂ were investigated at 160 °C (Fig. 5c). The ZnO–BP sensor demonstrated a rapid response, where the response and recovery times were about 9 and 15 s (Fig. S2, ESI†), respectively. Comparatively, the response and recovery times of ZnO were approximately 14 and 109 s, respectively. The ZnO–BP sensor even exhibited higher response–recovery values to NO₂ than the sensors based on as-prepared ZnO and previously reported ZnO.^28–31 Additionally, after BP decoration, the electrical resistance of ZnO hollow spheres increased, which was probably due to the formation of a heterojunction between p-type BP and n-type ZnO and the generation of an electrical barrier between crystal grains. Fig. 5d shows the transient responses of the ZnO–BP sensor sequentially exposed to 1, 5, 10, 50, 100, and 500 ppb NO₂ at 160 °C (Fig. S3, 1–10 000 ppb, ESI†). The response of the sensor increased with the increasing NO₂ concentration. It is worth noting that the present sensor exhibited a clear response of 1.307 to NO₂ concentration as low as 1 ppb. At lower NO₂ concentrations (1–500 ppb), the response increased rapidly with the increasing NO₂ concentration as most of the active sites on the surface of the sensing materials were available for chemical or physical interactions, whereas at higher gas concentrations, the sensor sensitivity gradually saturated because of lack of adequate active sites for further NO₂ adsorption. The Japanese standards strictly regulated the threshold limit value (TLV) for NO₂ to less than 40–60 ppb. Therefore, the ZnO–BP sensor can be suitable for the detection of low concentrations of NO₂ for monitoring environmental pollution (Fig. S4, ESI†). In addition, for 100 ppb NO₂, seven reversible cycles of the response curve maintained their initial value without a clear decrease, proving that the sensor exhibited good repeatability (Fig. S5, ESI†).

Fig. 5 (a) Responses to 100 ppb NO₂ of different sensors at different operating temperatures. (b) Cross-sensitive responses of ZnO, BP and ZnO–BP sensors to various gases at 160 °C. (c) The dynamic response and recovery characteristic curves of ZnO and ZnO–BP sensors to 100 ppb NO₂ at 160 °C. (d) Transient response of ZnO–BP sensor to different NO₂ concentrations at 160 °C, the inset shows the response at low NO₂ concentrations (1–10 ppb).

Black phosphorene is a layered material similar to graphene. For graphene, good sensing properties have already been illustrated by both theoretical and experimental investigations.^32,33 Therefore, a comparison of NO₂ sensing performances using ZnO–BP composites as sensing materials and other reported graphene-based, BP-based and ZnO-based sensing materials is presented in Table 1. The ZnO–BP composite sensor displays much lower detection limit, faster response and recovery behavior and higher sensitivity compared to other reported materials.^9,34–43

Table 1 Gas responses to NO₂ in the present study and those reported in the literatures

Samples	Detection limit [ppb]	Response time [s]	Recovery time [s]	Response	Ref.
a Where ΔG = Ggas − G0; Ggas and G0 are the instantaneous conductances of the sensor detected under target gas exposure and before exposure to target gas, respectively.
BP FETs	5	∼280–350	130	2.9% (5 ppb, ΔG/G0)^a	9
Suspended BP	25 000	∼20	>20	45% (200 ppm, ΔR/R0)	34
PNSs	20	—	—	190% (20 ppb, ΔG/G0)	35
RGO	5000	3600	600	12% (5 ppm, ΔR/Ra)	36
RGO	1000	420 (5 ppm)	1680 (5 ppm)	11.5% (5 ppm, ΔR/Ra)	37
ZnO	10	270	240	37% (10 ppb, ΔR/Ra)	38
ZnO/RGO	5000	96 (50 ppm)	1552 (50 ppm)	<2% (5 ppm, ΔR/Ra)	39
ZnO/m-SWCNTs	2500	208	>208	∼52% (2.5 ppm, ΔR/Ra)	40
ZnO NR's	1000	48	180	41% (1 ppm, ΔR/Ra)	41
ZnO BNW's	1000	26	162	7.5% (1 ppm, ΔR/Ra)	41
RGO/SnO2	500	135 (5 ppm)	200 (5 ppm)	3.31 (5 ppm, Rg/Ra)	42
RGO/Cu2O	64	—	500 (2 ppm)	67.8% (2 ppm, ΔG/G0)	43
ZnO–BP	1	2	16	130.7% (1 ppb, ΔR/Ra)	This study

---

The possible reasons for the improved gas sensing properties of the ZnO–BP composite sensor are discussed. First, the introduction of BP could improve the electron transfer rate due its excellent carrier mobility, leading to fast response and recovery behavior to the target gas. Second, the large surface area derived from the existence of BP may promote gas adsorption and diffusion on the active surface. Third, the Schottky barrier around the interface of p–n junction can result in the specific migration of charge from ZnO to BP, as proved by XPS data. BP acting as a charge mediator may facilitate the charge transfer from ZnO to NO₂ molecules, similar to the observations for WO₃–GO composites reported elsewhere.^44,45 Therefore, the significant enhancement in charge transfer, excellent carrier mobility and the promoted gas adsorption and diffusion due to the presence of BP account for the superior sensing performance of ZnO–BP composites, which exhibit high response, ultralow detection limit, fast response and excellent selectivity.

To understand the mechanism of the sensing properties at the atomic level, we performed first-principle calculations based on the density functional theory (DFT). Theoretical calculations were performed using the Quantum Espresso package,⁴⁶ employing the GGA–PBE exchange correlation functional.⁴⁷ The non-local van der Waals dispersion (vdW) corrections according to the vdW-DF2 procedure were taken into account.^48,49 The geometry optimization was considered to be converged when the residual force on each atom was smaller than 0.01 eV Å⁻¹. The calculated structures for NO₂ adsorptions on ZnO–BP, ZnO and BP are shown in Fig. 6a, b and c, respectively, where NO₂ molecules physisorb on ZnO–BP and BP and chemisorb on ZnO. The adsorption energy of NO₂ molecules is calculated by the following formula:

E_ad = −(E_total − E_substrate − nE_molecule)/n (1)

Here, E_total is the total energy of the substrate with the NO₂ adsorbates, E_substrate is the energy of the substrate without molecular adsorbates, E_molecule is the energy of the isolated NO₂ molecule, and n is the number of NO₂ molecules adsorbed on the substrate. The adsorption energies of NO₂ on the ZnO–BP, ZnO and BP substrates were 0.66, 1.39 and 0.51 eV, respectively. The adsorption energies were much higher than the thermal fluctuation energy (0.026 eV) at room temperature. Thus, the adsorption processes of NO₂ on ZnO–BP composite and BP were found to be stable. The adsorption energy of NO₂ on ZnO–BP composite was 1.29 times that of NO₂ on BP. The adsorption energy is a parameter that describes the binding degree between molecules and semiconductors. The charge transfer can change the conductivity of semiconductors, and it is a more important parameter for sensing properties. The charge transfer between NO₂ molecules and substrates is analyzed by the differential charge density distribution equation,

Δρ = ρ_total − ρ_molecule − ρ_substrate (2)

where ρ_total is the charge density of substrate with molecule adsorbed, ρ_molecule is the charge density of molecule, and ρ_substrate is the charge density of the substrate. The differential charge density distributions are shown in Fig. 6d–f. There was charge transfer between NO₂ molecules and the substrate, and NO₂ molecules obtained electrons from the substrate. The values of charge transfer between NO₂ and ZnO–BP, ZnO and BP substrates were 0.65, 0.53 and 0.36 e, respectively, as indicated by the Bader analysis. The charge transfer of NO₂ on ZnO–BP composite was 1.81 times that of NO₂ on BP. The difference in charge transfer explains why the sensor properties of ZnO–BP are better than those of ZnO and BP. Besides, the presence of BP increased the surface area-to-volume ratio; also, BP has good carrier mobility. For the ZnO–BP heterostructure, there was charge transfer from ZnO to BP via Zn–P bonds, which increased the local chemical affinity of BP and enhanced the sensing properties of ZnO–BP towards NO₂ molecules.

Fig. 6 Calculated structures of NO₂ molecules on (a) ZnO–BP, (b) ZnO and (c) BP substrates. Isosurfaces of differential charge density distributions of NO₂ molecules on (d) ZnO–BP, (e) ZnO and (f) BP substrates. For the isosurfaces of differential charge density, orange and green means attaining and losing electrons, respectively.

Conclusions

In summary, ZnO–BP composites have been successfully prepared by the ultrasonic treatment of liquid dispersion of BP in the presence of ZnO hollow spheres synthesized by the microwave-assisted hydrothermal method. Interestingly, the introduction of ZnO significantly enhances BP environmental stability, making it a suitable candidate to use in practice. Remarkably, ZnO–BP sensor exhibits higher response, outstanding selectivity, faster response and recovery dynamic process, and an ultralow detection limit of 1 ppb to NO₂ in contrast to graphene-, BP- and ZnO-based sensors, which were previously used. These characteristics are suggested to arise from significantly enhanced charge transfer as well as the promoted gas adsorption and diffusion due to the presence of BP. This study has provided a route in the search of BP-based materials with excellent gas-sensing properties, in particular towards NO₂.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The research is supported by Natural Science Foundation of Guangdong (2016A030313059), Science and Technology plan project of Guangdong (2017A010101018), Science and Technology plan project of Shenzhen (JCYJ20160307111047701, JCYJ20160226192754225, JCYJ20160307145209361, JCYJ20160307142444674).

References

1. P. Lu, A. Carvalho, J. Wu, H. W. Liu, E. S. Tok, A. H. C. Neto, B. Özyilmaz and C. H. Sow, Adv. Mater., 2016, 28, 4090.
2. J. O. Island, G. A. Steele, H. S. J. V. D. Zant and A. Castellanos-Gomez, 2D Mater., 2015, 2, 011002.
3. A. Castellanos-Gomez, L. Vicarelli, E. Prada, J. O. Island, K. L. Narasimha-Acharya, S. I. Blanter, D. J. Groenendijk, M. Buscema, G. A. Steele, J. V. Alvarez, H. W. Zandbergen, J. J. Palacios and H. S. Zant, 2D Mater., 2014, 1, 025001.
4. N. Gillgren, D. Wickramaratne, Y. M. Shi, T. Espiritu, J. W. Yang, J. Hu, J. Wei, X. Liu, Z. Q. Mao, K. Watanabe, T. Taniguchi, M. Bockrath, Y. Barlas, R. K. Lake and C. N. Lau, 2D Mater., 2014, 2, 011001.
5. Y. Q. Cai, G. Zhang and Y.-W. Zhang, J. Phys. Chem. C, 2015, 119, 13929.
6. X. Luo, Y. Rahbarihagh, J. C. M. Hwang, H. Liu, Y. C. Du and P. D. Ye, IEEE Electron Device Lett., 2014, 35, 1314.
7. Y. Q. Cai, G. Zhang and Y.-W. Zhang, Sci. Rep., 2014, 4, 6677.
8. Y. Q. Cai, Q. Q. Ke, G. Zhang and Y.-W. Zhang, J. Phys. Chem. C, 2015, 119, 3102.
9. A. N. Abbas, B. L. Liu, L. Chen, Y. Q. Ma, S. Cong, N. Aroonyadet, M. Köpf, T. Nilges and C. W. Zhou, ACS Nano, 2015, 9, 5618.
10. J. Xu, L. Chen, Y.-W. Dai, Q. Cao, Q. Q. Sun, S. J. Ding, H. Zhu and D. W. Zhang, ACS Appl. Mater. Interfaces, 2017, 9, 1602246.
11. L. Z. Kou, T. Frauenheim and C. F. Chen, J. Phys. Chem. Lett., 2014, 5, 2675.
12. X. Wang, J. Su, H. Chen, G.-D. Li, Z. F. Shi, H. F. Zou and X. X. Zou, ACS Appl. Mater. Interfaces, 2017, 9, 16335.
13. A. Maity and S. B. Majumder, Sens. Actuators, B, 2015, 206, 423.
14. H. M. Fahad, H. Shiraki, M. Amani, C. C. Zhang, V. S. Hebbar, W. Gao, H. Ota, M. Hettick, D. Klriya, Y.-Z. Chen, Y.-L. Chueh and A. Javey, Sci. Adv., 2017, 3, 1602557.
15. X. H. Liu, T. T. Ma, N. Pinna and J. Zhang, Adv. Funct. Mater., 2017, 27, 1702168.
16. H.-R. Kim, A. Haensch, I.-D. Kim, N. Barsan, U. Weimar and J.-H. Lee, Adv. Funct. Mater., 2011, 21, 4456.
17. D. Buso, M. Post, C. Cantalini, P. Mulvaney and A. Martucci, Adv. Funct. Mater., 2010, 18, 3843.
18. Q. Li, Y. Du, X. J. Li, G. Y. Lu, G. W. Y. Wang, Y. F. Geng, Z. Q. Liang and X. Q. Tian, Sens. Actuators, B, 2016, 235, 39.
19. M. T. Kelly, J. K. M. Chun and A. B. Bocarsly, Nature, 1996, 382, 214.
20. C. G. Lee, H. G. Yan, L. E. Brus, T. F. Heinz, J. Hone and S. Ryu, ACS Nano, 2010, 4, 2695.
21. W. L. Lu, H. Y. Nan, J. H. Hong, Y. M. Chen, C. Zhu, Z. Liang, X. Y. Ma, Z. H. Ni, C. H. Jin and Z. Zhang, Nano Res., 2014, 7, 853.
22. L. Ge, X. Y. Jing, J. Wang, J. Wang, S. Jamil, Q. Liu, F. C. Liu and M. L. Zhang, J. Mater. Chem., 2011, 21, 10750.
23. Y. H. M. S. T. Maruyam, Solid State Commun., 1982, 44, 877.
24. A. Brown and S. Rundqvist, Acta Crystallogr., 2010, 19, 684.
25. J. D. Wood, S. A. Wells, D. Jariwala, K. S. Chen, E. Cho, V. K. Sangwan, X. Liu, L. J. Lauhon, T. J. Marks and M. C. Hersam, Nano Lett., 2014, 14, 6964.
26. A. Ziletti, A. Carvalho, D. K. Campbell, D. F. Coker and A. Castro Neto, Phys. Rev. Lett., 2015, 114, 046801.
27. H. U. Lee, S. C. Lee, J. Won, B. C. Son, S. Choi, Y. Kim, S. Y. Park, H.-S. Kim, Y.-C. Lee and J. Lee, Sci. Rep., 2015, 5, 8691.
28. H. W. Kim, J. K. Yong, A. Mirzaei, S. Y. Kang, M. S. Choi, J. H. Bang and S. K. Sang, Sens. Actuators, B, 2017, 249, 590.
29. J. Liu, S. Li, B. Zhang, Y. Xiao, Y. Gao, Q. Y. Yang, Y. L. Wang and G. Y. Lu, Sens. Actuators, B, 2017, 249, 715.
30. J. K. Yong, A. Mirzaei, S. Y. Kang, M. S. Choi, J. H. Bang, S. K. Sang and H. W. Kim, Appl. Surf. Sci., 2017, 413, 242.
31. R. Zhang, W. Pang, Z. H. Feng, X. J. Chen, Y. Chen, Q. Zhang, H. Zhang, C. L. Sun, J. J. Yang and D. H. Zhang, Sens. Actuators, B, 2017, 238, 357.
32. O. Leenaerts, B. Partoens and F. M. Peeters, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 77, 125416.
33. A. Miyamoto, Y. Kuwaki, T. Sano, K. Hatakeyama, A. Quitain, M. Sasaki and T. Kida, ACS Omega, 2017, 2, 2994.
34. G. Lee, S. Kim, S. Jung, S. Jang and J. Kim, Sens. Actuators, B, 2017, 250, 569.
35. S. M. Cui, H. H. Pu, S. A. Wells, Z. H. Wen, S. Mao, J. B. Chang, M. C. Hersam and J. H. Chen, Nat. Commun., 2015, 6, 8632.
36. J. D. Fowler, M. J. Allen, V. C. Tung, Y. Yang, R. B. Kaner and B. H. Weiller, ACS Nano, 2009, 3, 301.
37. P.-G. Su and H.-C. Shieh, Sens. Actuators, B, 2014, 190, 865.
38. E. Oh, H. Y. Choi, S.-H. Jung, J. C. Kim, K.-H. Lee, S.-W. Kang, J. Kin and S.-H. Jeong, Sens. Actuators, B, 2009, 141, 239.
39. N. Kumar, A. K. Srivastava, H. S. Patel, B. K. Gupta and G. D. Varma, Eur. J. Inorg. Chem., 2015, 1912.
40. X. Li, J. Wang, D. Xie, J. L. Xu, Y. Xia and L. Xiang, Mater. Lett., 2017, 206, 18.
41. Y. H. Navale, S. T. Navale, N. S. Ramgir, F. J. Stadler, S. K. Gupta, D. K. Aswal and V. B. Patil, Sens. Actuators, B, 2017, 251, 551.
42. H. Zhang, J. C. Feng, T. Fei, S. Liu and T. Zhang, Sens. Actuators, B, 2014, 190, 472.
43. S. Z. 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.
44. X. Geng, J. J. You, J. Wang and C. Wang, Mater. Chem. Phys., 2017, 191, 114.
45. X. Q. An, J. C. Yu, Y. Wang, Y. M. Hu, X. L. Yu and G. J. Zhang, J. Mater. Chem., 2012, 22, 8525.
46. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. D. Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Samos-Martin, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. Seitsonen, A. Smogunov, P. Umari and R. M. Wentzcovitch, J. Phys.: Condens. Matter, 2009, 21, 395502.
47. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
48. M. Dion, H. Rydberg, E. Schroder, D. C. Langreth and B. I. Lundqvist, Phys. Rev. Lett., 2004, 92, 246401.
49. K. Lee, É. D. Murray, L. Kong, B. I. Lundqvist and D. C. Langreth, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 081101.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8nh00052b

---