NO 2 Gas Sensing Properties of Ag-Functionalized Porous ZnO Sheets

Abstract

Herein, we report a simple and scalable synthesis route to prepare Ag-functionalized porous ZnO sheets and their enhanced NO₂ gas sensing properties. Porous ZnO sheets functionalized with well-dispersed submicron Ag particles were prepared by using a hydrothermal method-based one-pot synthesis route from Zn and Ag precursors. NO₂ gas sensing performance (response, selectivity, response time, and recovery time) was optimized at 200°C in the gas sensor fabricated with 3 at% Ag-functionalized porous ZnO sheets. We demonstrated a response ( $R_g$ / $R_a$ ) of 17.18 to 10 ppm NO₂ gas and also obtained a high response of 14.05 even at 60% relative humidity due to the synergetic effect of improved NO₂ gas adsorption in the presence of Ag particles and increased resistance by the formation of Schottky barrier at Ag-ZnO heterojunctions.

1. Introduction

Nitrogen dioxide (NO₂) is a red-brown highly oxidizing gas [1] and one of the major air pollutants [2], causes nose and respiratory problems, coughs, fatigue, eye irritation, and nausea, and affects patients with debilitating pulmonary system diseases [3–5]. It also makes severe environmental issues such as smog and acid rain [6, 7]; thus, the development of NO₂ gas sensors with high reliability and performance is urgently required.

ZnO is widely used gas sensing materials due to its semiconducting nature ( $E_g=3.37\text{eV}$ ) with high electron mobility (200 cm² V^-1 s^-1), nontoxicity, and high thermal stability [8, 9]. After the first report on the gas sensing properties of ZnO [10], many researches for gas sensing based on ZnO have been carried out [11, 12]. One of the main approaches to enhance the gas sensing properties of ZnO is morphological or dimensional control since gas adsorption is directly related to the surface area of gas sensing materials. Various forms of ZnO such as nanowires [13], nanorods [14], nanotubes [15], nanofibers [16], nanopetals [17], hierarchical architectures [18], nanosheets [19], and nanowalls [20] have been studied to improve gas sensing properties. However, a major obstacle in the application of ZnO-based gas sensor is poor selectivity of ZnO owing to the relatively good response to different gases [21]. To address this, decoration or functionalization of noble metal particles on ZnO surface, which improves the adsorption of gas molecules and forms a Schottky potential barrier [22, 23], is commonly used [24]. Among noble metal elements, Ag can be a good choice due to its low price, high electrical conductivity, and good catalytic performance [25, 26]. Yin et al. [27] and Wei et al. [22] reported the important roles of Ag particles on the sensing properties of one-dimensional ZnO structures to NO₂ and ethanol gases. In particular, there have been many studies on NO₂ gas sensing [28–32], but there are few reports on ZnO, which can simultaneously achieve a two-dimensional nanosheet cross-sectional area and a nanocomposite with Ag in a single process.

In the present study, we aimed to prepare hybrid materials of Ag particles and ZnO powders in an effort to improve the NO₂ gas sensing properties. Notably, by using hydrothermal method-based one-pot synthesis technique, well-dispersed submicron Ag particles (1, 3, and 5 at%) functionalized porous ZnO sheets were obtained from Zn and Ag precursors. NO₂ gas sensing properties including response, selectivity, response time, and recovery time were optimized in a sensor fabricated with 3 at% Ag-functionalized porous ZnO sheets at 200°C. Possible mechanisms for enhanced sensing properties in this hybrid materials were also discussed.

2. Experimental

2.1. Synthesis of Ag-Functionalized Porous ZnO Sheets

Initially, 0.02 mol Zn(NO)₃·6H₂O and 0.06 mol urea were completely dissolved in a 180 mL mixture of ethanol–water (1 : 2, v–v) via mixing for 30 min. Then, 0.06, 0.12, and 0.18 g AgNO₃ (1, 3, and 5 at% Ag) were added to the solution, and a transparent mixed solution (zinc nitrate hexahydrate + urea + silver nitrate in ethanol-water) was obtained. This solution was poured into a 250 mL Teflon-lined autoclave and maintained at 180°C for 24 h. After this, the white precipitate was collected and washed several times with deionized water and absolute ethanol by centrifugation. Next, the precipitate was dried under vacuum at 80°C for 6 h. Finally, the Ag-functionalized porous ZnO sheets were obtained after annealing under Ar flow at 550°C for 4 h. The corresponding schematic is shown in Figure 1.

Figure 1

Schematic of the synthesis method.

2.2. Characterization

Crystallinity and phases were examined by X-ray diffraction (XRD, SmartLab, Rigaku) with CuKα radiation ( $\lambda =1.541\text{Å}$ ). Morphology was analyzed using a transmission electron microscope (TEM, Talos F200X, FEI). An energy dispersive X-ray spectrometer (EDS, Talos F200X, FEI) incorporated into the TEM was used to verify chemical composition. X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific Co.) was employed to investigate the chemical states using the C 1 s peak (284.8 eV) as a reference. Photoluminescence (PL, Ram Boss, Dongwoo Optron) was evaluated using a He–Cd laser with 325 nm excitation wavelength.

2.3. Gas Sensing Test

Au interdigitated electrodes were sputter-deposited on the surface of an alumina substrate. Sputtering was performed at 80 mA for 12 min, and the final thickness of the electrodes was 300 nm. The fabricated gas sensor was placed on a ceramic heater, and its gas sensing properties were measured using a lab-made gas sensing system (Supplementary Materials (SM), Figure S1). Desired amounts of target gases were mixed with dry synthetic air as a background using mass flow controllers (MFCs). The resistance of the gas sensor in air ( $R_a$ ) and in the existence of the target gas ( $R_g$ ) was checked using a multimeter. The response was calculated as $R=R_g/R_a$ for oxidizing gases and $R=R_a/R_g$ for reducing gases. The response time was calculated as the time needed for the sensor to reach 90% of the final resistance in the presence of NO₂ gas. Furthermore, the recovery time was evaluated as the time required for the sensor to reach 90% of the final resistance after stoppage of the NO₂ gas flow.

3. Results and Discussions

3.1. Morphological, Chemical, and Structural Studies

Figures 2(a)–2(f) indicate the TEM images of 1, 3, and 5 at% Ag-functionalized porous ZnO sheets, respectively. Interestingly, sheet-like ZnO with highly dense pore, which is desirable for sensing applications due to the large surface area, was formed. Dark particles represent the Ag particles functionalized on the surface of ZnO. Figures 3(a)–3(o)show the TEM elemental mapping data of 1, 3, and 5 at% Ag-functionalized porous ZnO sheets, respectively. Zn and O are evenly distributed over the sheet surface, whereas Ag is localized in certain places. Specifically in Figures 3(e), 3(j), and 3(o), it can be clearly observed that the dark points are pores, and the yellow points are Ag particles, which ensures the formation of Ag-functionalized porous ZnO sheets. The TEM–EDS point analysis results of 1, 3, and 5 at% Ag-functionalized porous ZnO sheets are given in SM Figure S2(a–d), (e–h), and (i-l), respectively. The white area in SM, Figure S2(a and b) is mostly rich in Ag, whereas the gray area in SM, Figure S2 (c and d) is rich in ZnO. Similarly, in SM, Figure S2 (e and f) and SM, Figure S2(i and j), a white area with 99.8 wt% Ag and 98.9 wt% Ag is recorded, whereas in SM, Figure S2(g and h) and SM, Figure S2(k and l), a gray area with 88.6 wt% Zn and 90.0 wt% Zn is observed. Overall, the TEM–EDS point analysis results confirmed the formation of Ag-functionalized porous ZnO sheets. SM, Figure S3(a) shows the XRD patterns of 3 at% Ag-functionalized porous ZnO sheets. Standard JCPDS files of Ag and ZnO are also shown in SM, Figure S3(b and c) for comparison. In both cases, the peaks related to both Ag and ZnO according to the JCPDS file nos. 04-0783 and 36-1451 were observed [27] without impurities. The crystallinity in XRD shows good agreement with HRTEM (Figure 4). Figure 5(a) indicates the XPS survey of 3 at% Ag-functionalized porous ZnO sheets, revealing the presence of all expected elements, namely, Zn, Ag, and O (peak related to C is due to the environment). The high-resolution Zn 2p core-level region is shown in Figure 5(b), where the two peaks at the binding energies of 1022.38 and 1045.38 eV are attributed to Zn 2p _1/2 and Zn 2p _3/2, respectively, with a spin-orbit splitting of 23.03 eV, which is in accordance with the previously reported literature [33]. This indicates that Zn is in the Zn²⁺ oxidation state [34, 35]. Figure 5(c) shows the high-resolution Ag 3d core-level region, where the Ag 3d _3/2 and Ag 3d _5/2 peaks can be observed at the binding energies of 374.28 and 368.28 eV, in accordance with the literature [36].

Figure 2

TEM images of the (a, b) 1, (c, d) 3, and (e, f) 5 at% Ag-functionalized porous ZnO sheets.

(b) (c) (d) (e) (f)

Figure 3

TEM elemental mapping of the (a–e) 1, (f–j) 3, and (k-o) 5 at% Ag-functionalized porous ZnO sheets.

(b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) (n) (o)

Figure 4

HRTEM image of the 3 at% Ag-functionalized porous ZnO sheets.

Figure 5

(a) XPS survey of the 3 at% Ag-functionalized porous ZnO sheets and high-resolution XPS core-levels of (b) Zn 2p and (c) Ag 3d.

(b) (c)

SM, Figure S4 shows the PL spectrum of 3 at% Ag-functionalized porous ZnO sheets. Based on literature, the PL spectrum of ZnO indicates a band emission in the ultraviolet (UV) region and a broad defect emission in the visible range [37]. The UV emission is owing to the near-band-edge emission of ZnO, which results from the recombination of free excitons through an exciton–exciton collisions [38]. Emissions in the visible region are because of the recombination of different defects [37].

3.2. Gas Sensing Studies

We measured the resistance of the gas sensors at 100, 200, and 300°C and found the optimum operating temperature of 200°C. Figures 6(a)–6(c) show the dynamic resistance curves of (Figure 6(a)) 1, (Figure 6(b)) 3, and (Figure 6(c)) 5 at% Ag-functionalized porous ZnO sheet gas sensors at various concentrations of NO₂ gas at 200°C, respectively. The sensors exhibit $n$ -type semiconducting behavior due to the $n$ -type nature of ZnO. Different sensor parameters including response, response time, and recovery time were calculated and shown in Figure 7 and Table 1. The response of the 3 at% Ag-functionalized porous ZnO sheet sensor (8.04, 17.18, and 59.26 for 2, 10, and 20 ppm NO₂ gases, respectively) was higher than those of the 1 at% (7.11, 16.29, and 28.16 for 2, 10, and 20 ppm NO₂ gases, respectively) and 5 at% (6.62, 10.86, and 11.40 for 2, 10, and 20 ppm NO₂ gases, respectively) Ag-functionalized porous ZnO sheets sensors at all NO₂ gas concentrations (Figure 7(a)). Interestingly, the response time of the 3 at% Ag-functionalized porous ZnO sheet sensor (203, 199, and 102 s for 2, 10, and 20 ppm NO₂ gases, respectively) was shorter than that of the 1 at% Ag-functionalized porous ZnO sheet sensor (210, 294, and 375 s for 2, 10, and 20 ppm NO₂ gases, respectively) but longer than that of the 5 at% Ag-functionalized porous ZnO sheets sensor (189, 23, and 14 s for 2, 10, and 20 ppm NO₂ gases, respectively), at all NO₂ gas concentrations (Figure 7(b)). A comparison of the recovery times of three gas sensors is shown in Figure 7(c). As the 3 at% Ag-functionalized porous ZnO sheet sensor demonstrated a higher response to NO₂ gas, it was selected for further studies. In fact, ppb-based low concentration of NO₂ gas can be detected but their responses were too low as seen in Figure 8. Figures 9(a) and 9(b) show the dynamic response graphs of ZnO-Ag 3at% sensor and presents the relevant selectivity to 10 ppm of various gases at 200°C. The response of the gas sensor to 10 ppm of NO₂, H₂S, C₆H₆, SO₂, and C₇H₈ gases was 17.18, 3.75, 2.74, 2.32, and 2.22, respectively. This demonstrates the good selectivity of optimal gas sensor. In addition, Figures 9(c) and 9(d) show the resistance of the patterned ZnO-Ag 3at % sensor to 10 ppm of NO₂, 0, 30, and 60% relative humidity at 200°C. The basic condition is that all gases are dry. However, in an actual environment, the target gas is inevitably a gas containing moisture that is always present in the atmosphere. Therefore, dry air was allowed to pass through the bubble system to control the humidity of the dry gas. That is, by controlling the humidity in the RH probe hygrometer, wet air can meet dry target gas, so the change in response of gas sensing can be observed at all gas humidity. However, it should be noted that the initial resistance may change, as seen in Figure 9(c), depending on the humidity. In other words, even if the gas sensing process temperature (200°C) is the same, since the relative humidity of the initial gases entering the chamber is different, it is possible to have different initial resistances. Due to the following reactions, electron and oxygen vacancy ( $V_o$ ) increase in $n$ -type ZnO, so the initial base line further decreases compared to the dry gas state (Figure 9(c)) [39, 40]. $\begin{array}{}\text{(1)}& {\text{H}}_2\text{O}+\text{Zn}+\text{O}⟶\left(\text{HO}-\text{Zn}\right)+\text{OH}+{\text{e}}^{-},\text{(2)}& \text{H}+{\text{O}}^{2-}⟶{\left(\text{OH}\right)}^{-}+{\text{e}}^{-},\text{(3)}& {\text{H}}_2\text{O}+2\text{Zn}+\text{O}⟶2\left(\text{HO}-\text{Zn}\right)+{\text{V}}_{\text{o}}.\end{array}$

Figure 6

Dynamic resistance curves of the (a) 1, (b) 3, and (c) 5 at% Ag-functionalized porous ZnO sheet gas sensor at 200°C to 2, 10, and 20 ppm NO₂ gas.

(b) (c)

Figure 7

(a) Response, (b) response time, and (c) recovery time of the 1, 3, and 5 at% Ag-functionalized porous ZnO sheet gas sensors to NO₂ gas at 200°C.

(b) (c)

Table 1

Response, response time, and recovery time of 1, 3, and 5 at% Ag-functionalized porous ZnO sheets gas sensors to different concentrations of NO₂ gas at 200°C.

	ZnO-Ag 1 at%			ZnO-Ag 3 at%			ZnO-Ag 5 at%
	2 ppm	10 ppm	20 ppm	2 ppm	10 ppm	20 ppm	2 ppm	10 ppm	20 ppm
Response ( $R_g$ / $R_a$ )	7.11	16.29	28.16	8.04	17.18	59.26	6.62	10.86	11.40
Response time (s)	210	294	375	203	199	102	189	23	14
Recovery time (s)	160	116	100	323	75	49	35	43	638

Figure 8

Response curves of the 3 at% Ag-functionalized porous ZnO sheets with 200, 400, and 600 ppb NO₂ gas concentration at 200°C.

Figure 9

(a) Dynamic response curves of the 3 at% Ag-functionalized porous ZnO sheet gas sensor to 10 ppm of various gases at 200°C. (b) Selectivity pattern of (a, c). Dynamic resistance curve of the 3 at% Ag-functionalized porous ZnO sheet gas sensor in the presence of 30% RH and 60% RH. (d) Response comparison of 10 ppm NO₂ gas sensor in dry and humid atmospheres. (e) Dynamic response curve of the 3 at% Ag-functionalized porous ZnO sheet gas sensor after three months. (f) Comparison between the response of the gas sensor to 10 ppm NO₂ in a fresh state and after three months.

(b) (c) (d) (e) (f)

Therefore, in order to investigate the effect of humidity, it is reasonable to compare the difference in response as shown in Figure 9(d). The responses to 10 ppm NO₂ gas at 0, 30, and 60% relative humidity at 200°C were 17.18, 15.21, and 14.05, respectively. Even in the humid environment, the sensor can reliably detect low concentrations of NO₂ gas. Finally, regarding the performance of gas sensing, even if measured after 3 months, the response value decreases, but there seems to be no problem with the long-term use of gas sensing (Figures 9(e) and 9(f)). Then, the responses to 10 ppm NO₂ gas at 200°C in a fresh state and after three months were 17.18 and 4.37, respectively.

3.3. Sensing Mechanisms

The basic sensing mechanism conductometric gas sensors are the modulation of the resistance of the sensing layer in the existence of the target gas [41, 42]. Initially, in air, when gas sensors are exposed to an oxygen environment, oxygen adsorbs electrons from the surface of ZnO because of its high electron affinity and is converted to ionic species as follows [43, 44]: $\begin{array}{}\text{(4)}& O_2\left(g\right)⟶O_2\left(\text{ads}\right),\text{(5)}& {\text{O}}_2\left(\text{ads}\right)+{\text{e}}^{-}⟶{\text{O}}_2^{-}\left(\text{ads}\right),\text{(6)}& {\text{O}}_2^{-}\left(\text{ads}\right)+{\text{e}}^{-}⟶{\text{O}}^{-},\text{(7)}& {\text{O}}^{-}+{\text{e}}^{-}⟶{\text{O}}^{2-}.\end{array}$

Due to the abstraction of electrons by the adsorbed oxygen species, an electron depletion layer (EDL) is developed on the ZnO surface, where a higher resistance relative to that of the core parts is expected owing to the lower concentrations of electrons. When NO₂ gas is injected into the gas chamber, the surface EDL expands because NO₂ has a higher electron affinity (2.28 eV) than oxygen (0.43 eV) and leads to an increase in the resistance of the gas sensor [45]. The relevant reactions are as follows [46]: $\begin{array}{}\text{(8)}& \text{N}{\text{O}}_2+{\text{e}}^{-}⟶{\text{NO}}_2^{-},\text{(9)}& \text{N}{\text{O}}_2+{\text{e}}^{-}⟶\text{NO}+{\text{O}}^{-},\text{(10)}& {\text{NO}}_2^{-}+{\text{O}}^{-}+2{\text{e}}^{-}⟶\text{NO}+2{\text{O}}^{2-}.\end{array}$

Furthermore, in the homojunctions between ZnO sheets, potential barriers are initially formed in air, and upon exposure to NO₂ gas, the height of these potential barriers increases due to the further abstraction of electrons, contributing to the sensing signal.

In addition, the effects of Ag on the gas sensing characteristics of ZnO sheets were investigated. Owing to the differences between the work functions of Ag (4.72 eV) and ZnO (4.45 eV) [47], the electrons transfer from ZnO to Ag, resulting in intimate contacts between Ag and ZnO to equate the Fermi levels. Accordingly, a Schottky barrier is formed near the interfaces of ZnO and Ag in air. This increases the resistance of the gas sensor as compared to that of pristine ZnO sheets. Upon exposure to NO₂ gas and further extraction of electrons from the surface of ZnO, the width and height of the Schottky barriers increase, leading to higher resistance of the gas sensor in the presence of Ag particles. This effect is an electronic effect or electronic sensitization of Ag. In other words, as the content of Ag increases, the initial resistance of the ZnO sheets may increase because more Schottky barriers are formed as seen in Figure 6. In another effect, similar to Au [48], Ag catalytically decomposes the oxygen molecules into atomic species when they come in contact with the surface of Ag particles. Then, owing to the so-called spill-over effect, the oxygen atoms move to the surface of neighboring ZnO and easily adsorb on this surface [49]. Therefore, more oxygen species adsorb on the surface of ZnO, and their more reactions with NO₂ lead to a higher response of the sensor to NO₂ gas. In addition, because of the process temperature of 200°C, oxygen species can react with Ag to form Ag₂O, but even at this time, electrons are transferred from ZnO to Ag₂O to play a role in increasing the resistance of ZnO [50]. Figure 10 schematically shows the sensing mechanism of the 3 at% Ag-functionalized sensor. Table 2 presents a comparison between the NO₂ sensing capacities of ZnO-based gas sensors reported in the literature and the present optimized sensor [24, 46, 51–57]. It can be observed that the present sensor has good performance when compared with the performances of other relevant gas sensors mainly because of the catalytic effect of Ag, formation of Ag/ZnO Schottky contacts, and high surface area due to the porous nature of the synthesized sheet, where the gas molecules can diffuse into the deep parts of the sheets.

Figure 10

Schematic of the NO₂ sensing mechanism of Ag-functionalized porous ZnO sheets.

Table 2

Comparison between the NO₂ gas sensing properties of ZnO-based gas sensors with the present optimized gas sensor.

Sensor	NO2 conc. (ppm)	Response ( $R_g$ / $R_a$ )	Temp. (°C)	Ref.
Au-decorated ZnO-branched SnO2 NWs	10	13.07	300	[24]
Co-decorated ZnO branched SnO2 NWs	10	7.48	300	[51]
ZnO@Au miscrospheres	1	17	250	[46]
1 mol% Pd-ZnO NWs	1	13.5	100	[52]
Ni-doped ZnO thin film	5	2.08	200	[53]
ZnO nanosheet arrays	100	19	180	[54]
ZnO nanoparticles	100	6.86	150	[55]
ZnO nanorods	0.001	1.3	150	[56]
Al-doped ZnO nanorods	5	0.86	175	[57]
3 at% Ag-functionalized porous ZnO sheets	2	8.04	200	Present work
	10	17.18
	20	59.26

4. Conclusions

We proposed the hydrothermal route to prepare hybrid materials of Ag and ZnO. Submicron Ag particles functionalized porous ZnO sheets were formed during the chemical reaction between Zn and Ag precursors in a mixed solution of ethanol and water. From the designed measurements of NO₂ gas sensing properties of the sensors fabricated with Ag-functionalized porous ZnO sheets, we found that response and recovery time can be simultaneously improved by the hybridization of Ag particles due to the chemical and electrical sensitization effects even under high-humidity atmosphere. The 3 at% Ag-functionalized porous ZnO sheet sensor represents response of 17.18 and response time of 199 s under 10 ppm NO₂ gas at 200°C and shows good selectivity. We believe that researches based on our synthetic approach will provide enhanced sensing properties in any oxide-based gas sensing materials and expand the applications of semiconductor metal oxide gas sensors.