Room-Temperature Nitric Oxide Gas Sensors Based on NiO/SnO₂ Heterostructures

Abstract

Nitric oxide (NO) is a very well-known indoor pollutant, and high concentrations of it in the atmosphere lead to acid rain. Thus, there is great demand for NO sensors that have the ability to work at room temperature. In this work, NiO/SnO₂ heterostructures have been prepared via the polyol process and were tested against different concentrations of NO gas at room temperature. The structural and morphological characteristics of the heterostructures were examined using X-ray diffraction and scanning electron microscopy, respectively, while the ratio of NiO to SnO₂ was determined through the use of energy-dispersive spectrometry. The effects of both pH and thermal annealing on the morphological, structural and gas-sensing properties of the heterostructure were investigated. It was found that the morphology of the heterostructures consisted of rod-like particles with different sizes, depending on the temperature of thermal annealing. Moreover, NiO/SnO₂ heterostructures synthesized with pH = 8 and annealed at 900 °C showed a response of 1.8% towards 2.5 ppm NO at room temperature. The effects of humidity as well as of stability on the gas sensing performance were also investigated.

1. Introduction

Nitric oxide (NO), a well-known dangerous environmental pollutant, is a product of fossil fuel combustion and is found in emissions. Thus, its detection is crucial in terms of monitoring air quality. It is responsible for a number of human diseases, such as irritation to the skin or eyes and respiratory malfunctions, such as asthma, etc. [1,2]. Specifically, NO acts as a biomarker for asthma detection because its production by airway cells is strongly related to diseased cells. It has been found that the NO concentration exhaled by breath increased from 33 ppb for a healthy individual to about 98 ppb for an asthma patient [3,4,5]. In addition, the threshold limit value (TLV) for NO is 25 ppm for an 8 h exposure, according to the Occupational Safety and Health Organization (OSHA) [6]. As a result, there is a high demand for NO gas sensors that can be prepared using cost-effective methods and work at room temperature (RT).

In order to meet the above-mentioned requirements, a great number of materials have been tested as sensing elements against NO gas and using various sensing techniques, such as looking for a change in electrical resistance (chemoresistive sensor) [7], optical sensors [8] or surface acoustic wave devices [9,10]. Among them, chemoresistive sensors are by far the most investigated due to the simplicity of the measurement as well as the great variety of materials that can be used in this technique; however, most of them require a high operating temperature [1,2,11,12,13,14] or UV irradiation [15,16,17] in order to detect NO gas concentrations below the TLV, leading to extra energy consumption. Thus, only a few works have been reported on NO detection at room temperature. Specifically, Chang et al. [18] used N-doped reduced GO to detect 1 ppm NO with a 1.7 response, while Gaussian et al. [19] succeeded in detecting 100 ppm NO with a response of 80.7% and a response time of 300 s. Finally, Kuchi et al. [20] grew a TiO₂-rGO heterostructure in order to detect 2.75 ppm NO for a 7.1 response and a 440 s response time.

In this work, NiO/SnO₂ heterostructures synthesized using a polyol process were tested as room temperature NO gas sensors at concentrations far below the TLV. SnO₂ is a well-known n-type semiconductor that has been extensively studied as a gas sensor [21,22,23,24] in relation to numerous gases, such as NO₂ [25], SO₂ [26], H₂ [27], CH₄ [28], etc. Moreover, SnO₂ was the first material that was used in commercial chemoresistive gas sensors due to its stability and reliability [24]. Recently, it has been used in combination with a p-type metal oxide semiconductor, forming a p-n heterostructure, in order to enhance gas sensing performance. More specifically, p-n heterostructures of CuO/SnO₂ [29], Co₃O₄/SnO₂ [30] and NiO/SnO₂ [31,32] have successfully detected H₂S, NH₃, SO₂ and formaldehyde, respectively. Apart from these, NiO/SnO₂ heterostructures have also been tested against ethanol [33], triethylamine [34] and n-butanol [35]. Another issue that the gas sensors have to deal with is the lack of selectivity due to the interference of various gases [36]. Herein, the heterostructures were also tested against other gases, showing no response, indicating a selectivity toward NO.

Furthermore, NiO/SnO₂ p-n heterostructures have been applied to a great number of applications, such as anode materials for lithium-ion batteries [37], energy storage devices [38], photocatalysis [39], UV photodetectors [40], etc.; moreover, they can also be synthesized using a variety of methods, including sputtering [31], hydrothermal reactions [32], and electrospinning [35], which leads to different morphologies. Herein, NiO/SnO₂ heterostructures grown using a simple and cost-effective chemical method were deposited via the spin coating method on interdigitated electrodes (IDEs)/glass substrates. The NiO/SnO₂ sensor successfully detected 1 ppm NO at room temperature, making it a possible NO gas sensing material.

2. Materials and Methods

2.1. Synthesis of NiO/SnO₂ Heterostructure

NiCl₂·6H₂O (>98%, Sigma Aldrich, St. Louis, MI, USA) and SnCl₂·2H₂O (98%, Alfa Aesar, Haverhill, MA, USA) were used as Ni and Sn sources, respectively; ethylene glycol, C₂H₆O₂ (≥99.5%, Merck, Rahway, NJ, USA) was used as a solvent, while ammonia NH₃, (25% in water, Sigma Aldrich, St. Louis, MI, USA) was used in order to regulate the pH. All of the purchased reagents were of analytical grade and used without further purification. In a typical procedure, 0.475 g NiCl₂·6H₂O and 0.521 g SnCl₂·2H₂O were solved in 30 mL of ethylene glycol. The solution was stirred for 1 h, after which the pH was 0.3, and the solution a transparent green color. By adding ammonia, solutions with 2 different pH values, namely 6 and 8, were prepared, having the colors green and light blue, respectively, and they were semitransparent. After 1 h of stirring, the solutions were centrifuged for 15 min at 4000 rpm, and the precipitate was washed two times with ethanol and dried for 1 day at 70 °C. The obtained powder was thermally annealed at 300 °C, 600 °C and 900 °C for 2 h with a step of 2 °C/min.

2.2. Fabrication of NiO/SnO₂ Sensor

The gas sensor was prepared by milling 0.04 of NiO/SnO₂ powder with a binder consisting of an organic solution of terpineol (Sigma Aldrich, St. Louis, MI, USA) and ethyl cellulose (5% in toluene and ethanol solution, TCI, Tokyo, Japan). The milling duration was 30 min, after which the solution was deposited through the use of the spin coating technique (700 rpm for 10 s and 3000 rpm for 30 s) on commercial interdigitated electrodes (IDEs) (Ω Metrohm DropSens, Oviedo, Spain) on a glass substrate. The electrodes were platinum (Pt), and their characteristic distance was 5 μm. After deposition, the sensors were dried at 80 °C for 10 min. Finally, the sensing elements were thermally annealed at 400 °C for 2 h with a 2 °C/min step.

2.3. Materials Characterization

The structure of NiO/SnO₂ was examined via the X-ray Diffraction (XRD) technique using a Bruker AXS D8 Advance copper anode diffractometer (Cu-Kα radiation) equipped with a nickel foil monochromator. It operates at 40 kV and 40 mA over the 2θ/θ collection range of 10°–80° with a scan rate of 0.05°/s. From the XRD pattern, the crystallite size (d) is calculated using Scherer’s Equation (1), as below:

$$\mathrm{d}=\frac{0.9·\mathsf{\lambda }}{\left(\mathrm{FWHM}\right)·\cos \mathsf{\theta }}$$

(1)

where λ = 0.154 nm is the wavelength of X-ray radiation, FWHM = Full Width at Half Maximum of the peak corresponding to 2θ angle, and θ is the Bragg angle. The surface morphology was investigated via scanning electron microscopy (SEM) employing a JEOL 7000 microscope (JEOL Ltd., Akishima, Tokyo, Japan) operating at 15 keV, equipped with an energy-dispersive X-ray spectrometer (EDS), which was used for the stoichiometric analysis.

The gas sensing performance of the NiO/SnO₂ heterostructure was studied in a homemade stainless-steel chamber with a volume of 0.7 L. A mechanical pump was used to initially evacuate the chamber as well as to regulate the total pressure in the presence of gases. Two mass flow controllers were used to insert the gases in the chamber with a constant flow of 500 sccm (standard cubic centimeters per minute). Each measurement cycle consisted of a 5 min exposure of the sensor to NO gas, in which the electrical resistance decreased, followed by a 5 min exposure to nitrogen (absence of NO) for the recovery. The NO gas concentrations were 2.5, 5, 7.5 and 10 ppm in nitrogen, which is far below the TLV. The total pressure during the measurements was constant and equal to 800 mbar. In order to monitor the electrical current changes of the sensor upon its interaction with NO gas, a voltage of 1 Volt was applied using a Keithley 6517A electrometer. Τhe procedure was controlled and monitored with a computer using the LabVIEW program. The response R (%) of the sensor is defined by Equation (2)

$$\mathrm{R}\left(\%\right)=\frac{{\mathrm{R}}_{{\mathrm{N}}_2}-{\mathrm{R}}_{\mathrm{NO}}}{{\mathrm{R}}_{{\mathrm{N}}_2}}·100\%$$

(2)

where ${\mathrm{R}}_{{\mathrm{N}}_2}$ is the sensor’s resistance value after a 5 min exposure to nitrogen (N₂) and ${\mathrm{R}}_{\mathrm{NO}}$ is the sensor’s resistance value after a 5 min exposure to NO of different concentrations. Moreover, the response (tresp) and recovery (trec) times are defined as the time that is required for the sensor’s resistance to equal 10% of the ${\mathrm{R}}_{{\mathrm{N}}_2}$ in the presence of NO and equal to 90% of the ${\mathrm{R}}_{{\mathrm{N}}_2}$ in the presence of N₂ (absence of NO). In Figure 1, a typical measurement of gas sensing response under a 5 min exposure to NO followed by a 5 min exposure to N₂ at room temperature for the NiO/SnO₂ heterostructure, is presented.

It should be noted at this point that the recovery of the sensor has taken place using N₂ instead of air in order to avoid the formation of NO₂ as a product of the reaction between NO and O₂. In addition, it has been reported [41] that there is no difference in the sensors’ behavior if the recovery takes place using N₂ or air.

Finally, in the case of the synthetic procedure, we synthesized the materials under the same conditions more than 10 times, and we used powders with the same structural morphological and sensing properties. The synthetic procedure is reproducible. Moreover, in all cases, the materials were measured 3 times and showed exactly the same results.

3. Results

In Figure 2, the XRD patterns of the NiO/SnO₂ heterostructures prepared from solutions with pH equal to 6 (a) and 8 (b) before and after their thermal annealing are presented. It can be seen that the as-prepared material was of low crystallinity, while after the thermal annealing, a mix of NiO and SnO₂ phases with strong peaks was observed, confirming the formation of the heterostructure, which is also in accordance with the literature [32]. The crystalline planes were identified using JCPDS card No. 41-1445 and 78-0643 for SnO₂ and NiO, respectively. In particular, for pH = 6, at 300 °C, the peaks that are observed at 2θ = 26.62°, 34.02° and 51.86° correspond to the (110), (101) and (211) planes of SnO₂, respectively, while the one at 2θ = 43.11° corresponds to the (202) plane of NiO. For the heterostructures that were annealed at 600 °C, the peaks at 2θ = 26.63°, 33.98°, 37.98°, 51.86°, and 54.62° correspond to the (110), (101), (111), (211), and (220) planes of SnO₂, respectively, while those at 2θ = 37.29°, 43.32°, and 62.96° correspond to (202) (021) and (024) planes of NiO, respectively. Finally, for heterostructures that were annealed at 900 °C, the peaks at 26.66°, 33.92°, 38.00°, 51.82°, 54.82°, 57.91°, 61.05°, 64.78°, 66.04° and 69.25° correspond to (110), (101), (111), (211), (220), (002), (310), (112), (301) and (311) planes of SnO₂, respectively, while those at 2θ = 37.25°, 43.28°, 62.92°, 71.36° and 75.36° correspond to (021), (202), (024), (312) and (223) planes of NiO, respectively. In the same way, for the heterostructures prepared with pH = 8, the observed peaks and the corresponding planes were 2θ = 26.74° (110), 34.02° (101), 52.01° (211) of SnO₂ and 2θ = 43.09° (202) at the annealing temperature of 300 °C, while for the annealing temperature of 600 °C, the observed peaks and planes were 2θ = 26.60° (110), 33.97° (101), 54.61° (220) και and 62.62° (021) for SnO₂ and 2θ = 37.34° (021), 43.26° (202) and 62.95° (024) for NiO. Finally, for the annealing temperature of 900 °C, the peaks and the corresponding planes are observed at 2θ = 26.66° (110), 33.95° (101), 38.05° (111), 51.87° (210), 54.81° (220), 57.93° (002), 62.03° (310), 64.86° (112) and 66.08° (301) for SnO₂ and at 2θ = 43.31° (202), 37.30° (021), 62.87° (024), 71.33° (312) and 75.32° (223), for NiO. A summary of the planes of each heterostructure material is presented in

Table 1. Crystallographic planes of the NiO/SnO₂ heterostructure.

pH	Annealing Temperature (°C)	Material	Planes
6	300	NiO	(200)
		SnO2	(110), (101)
	600	NiO	(202), (021), (024)
		SnO2	(110), (101), (111), (211), (220)
	900	NiO	(202), (021), (024), (312), (223)
		SnO2	(110), (101), (111), (210), (220), (002), (310), (112), (301)
8	300	NiO	(202)
		SnO2	(110), (101), (211)
	600	NiO	(202), (021), (024)
		SnO2	(110), (101), (220), (021)
	900	NiO	(202), (021), (024), (312), (223)
		SnO2	(110), (101), (111), (210), (220), (002), (310), (112), (301)

.

Moreover, in

Table 2. Stoichiometric analysis and crystallite size (calculated at (110) plane of SnO₂) of the NiO/SnO₂ heterostructure.

pH	Annealing Temperature (°C)	NiO/SnO2 (%)/(%)	Crystallite Size (nm)
6	300	35.66/64.34	14.0
	600	40.85/59.15	9.4
	900	35.71/64.29	30.1
8	300	35.31/64.69	14.1
	600	36.10/63.90	9.9
	900	43.64/56.36	33.3

, the crystallite size of the NiO/SnO₂ heterostructure, which was calculated for the (110) plane of SnO₂ using Equation (1), is presented. It can be seen that the crystallite size remains unaffected by the pH of the solution, while it is strongly dependent on the annealing temperature. Specifically, there is a decrease in crystallite size between 300 °C and 600 °C, which can be attributed to the formation of more planes from both materials (Table 1). In contrast, there is a notable increase in the crystallite size at the annealing temperature of 900 °C, which, in combination with the increase in the number of different crystallographic planes formed at this temperature (Table 1), confirms the high crystallinity of the heterostructure.

The surface morphology of the heterostructure was investigated via scanning electron microscopy, the images of which are presented in Figure 3a–f. From these, the two different materials, NiO and SnO₂, can be distinguished, indicating the formation of the heterostructure. The morphology of the heterostructure prepared with pH = 6 and annealed at 300 °C as well as of those prepared with pH = 8 and annealed at 300 °C and 600 °C consist mainly of rods (SnO₂) with different sizes depending on both the pH and annealing temperature, accompanied with nanosized cubes of NiO. This is in agreement with the stoichiometric analysis presented in Table 2, in which the ratio NiO/SnO₂ for the above-mentioned heterostructures is almost equal to 35%/65%, thus containing the least amount of NiO. In the other heterostructures, a sponge-like morphology can be observed; however, the two materials can be distinguished, even in the heterostructure in which the ratio NiO/SnO₂ is almost 44%/56%, which is the highest NiO content.

In order to characterize the electric behavior of the heterostructure, a voltage between −5 V and 5 V was applied, and the electrical current was monitored. In Figure 4, the I–V characteristic curve of the heterostructure under different vacuum atmospheres and nitrogen concentrations at 10 ppm NO is presented. It can be seen that there is a linear relationship between the electrical current and the applied voltage, indicating Ohmic contact between the NiO/SnO₂ and the Pt metal of the IDEs, independent of the environmental conditions. Moreover, in the inset of Figure 4, the part of the I–V curve, when varying the applied voltage from 0 V to 5 V, is presented in order to differentiate the electrical current values under 10 ppm NO.

All NiO/SnO₂ heterostructures were tested against NO gas at room temperature. In Figure 5a,b, the resistance variation with time of the heterostructures that were prepared at pH = 6 and 8 and annealed at 900 °C is presented. The cycle duration consisted of 5 min exposure to different concentrations of NO, followed by a 5 min exposure to N₂ for recovery at room temperature. It can be seen that both materials had the ability to detect NO at 2.5 ppm at room temperature. In particular, the sensor prepared with pH = 6 showed a response of 2.2% and 0.5%, while the one prepared with pH = 8 showed a response of 6.9% and 1.8% at 10 ppm and 2.5 ppm, respectively. In addition, in Figure 6a, the response of each heterostructure, as a function of NO concentration, is presented. It is obvious that the heterostructure that was prepared at pH = 8 shows a greater response than that with pH = 6. Moreover, the response of the former increased at a higher rate than the latter, indicating that the pH affects the response to NO gas, probably due to higher porosity (Figure 3f) that appeared to have the heterostructure that was prepared at pH = 8. The structural parameters, such as crystallite size, play a major role in gas sensing performance. However, in our case, crystallite sizes for the heterostructures prepared with different pH are similar and, as a result, the enhanced response in terms of NO can be attributed to the higher NiO content in the heterostructure or to the different morphology.

The response time of the heterostructures as a function of NO concentration is presented in Figure 6a,b. The heterostructure prepared at pH = 6 showed an almost constant response time of about 4 min, indicating that the amount of NO that is adsorbed on the surface is the same, independent of NO concentration, which is probably attributed to low porosity. In contrast, the higher porosity of the other heterostructure (pH = 8) resulted in decreased response time with concentration due to the fact that a greater amount of NO can be adsorbed on the surface. Porosity plays an important role in the gas sensing performance since the higher it is, the greater the gas content that will interact with the sensing element. As a result, the response will be increased, while the response time will be decreased. As far as the recovery time is concerned, it was found to have a mean value of about 4 min for both heterostructures.

The sensing mechanism that governs the interaction between NO gas and the heterostructure can be summarized in Equations (3)–(7) [11,42]

$$N{O}_g+e^{-}\leftrightarrow N{O}_{ads}^{-}$$

(3)

$$2N{O}_{ads}^{-}\to N_2{O}_{ads}^{2-}$$

(4)

$$N_2{O}_2^{2-}{}_{ads}\to N_2{O}_{ads}+O_{ads}^{-}$$

(5)

$$N_2{O}_{ads}^{-}\to N_2{O}_g+e^{-}$$

(6)

$$O_{ads}^{-}+N{O}_g\to N_2{O}_g+e^{-}$$

(7)

which are based on adsorption and desorption of NO gas on the surface of the heterostructure.

In order to examine the repeatability as well as the stability of the sensor, it was tested against 10 ppm NO for four cycles of 5 min NO — 5 min N₂, both the as-prepared sensor and after it had been stored in an ambient environment for 6 months. In Figure 7a, it can be seen that the sensor showed excellent repeatability as well as stability. Moreover, the enhanced response of the sensor after 6 months of storage can be observed, probably due to the O₂ adsorption on the surface of the heterostructure.

Additionally, the sensor prepared at pH = 6 was evaluated under different values of relative humidity, RH, as presented in Figure 7b. The response decreased as the RH increased from 5% to 30% and 45%; however, it increased when the RH was equal to 65%. The decrease in response in relation to relative humidity can be related to the fact that water molecules decrease the active sites in which the NO are adsorbed [7].

In

Table 3. Summarize of NO gas sensing elements.

Material	Operation Temperature (°C)	Response	Concentration (ppm)	Response Time (s)	Ref.
ZnO:ZnGa2O4	400	28.6 *	100	8	[1]
CuO NPs	50	3 **	100	<2.5	[2]
Ag:ZnO	100	53.28% ***	0.0216	-	[6]
SnO2	160	33.3 *	0.5	214	[7]
N:rGO	RT	1.7 *	1	-	[18]
WO3	150	3.22 *	0.1	-	[11]
Pd loaded Co3O4	200	1.16 *	0.2	456	[12]
PCDTBT Conductive polymer	RT	80.6% ***	100	300	[19]
TiO2-rGO	RT	7.1 *	2.75	440	[20]
Tb2O3/ZnO	180	28.3 *	1	208	[13]
NiO/SnO2 (pH = 6)	RT	0.5 *	2.5	223	This work
NiO/SnO2 (pH = 8)	RT	1.8 *	2.5	257	This work

* R_g/R_a, ** R_a/R_g,*** [(R_g − R_a)/R_a] × 100%.

, the gas-sensing characteristics of materials that have been used as NO-sensing elements are summarized. It can be seen that only a few of them operated at room temperature. From this, it can be concluded that the NiO/SnO₂ heterostructure can be a possible candidate sensing element for NO detection at room temperature.

Finally, it should be noted that NiO/SnO₂ heterostructures were also tested against other gases, such as ozone, hydrogen and carbon dioxide, showing no response, indicating a selective response to NO gas. However, more experiments must be performed in the future.

4. Conclusions

In the present work, the NiO/SnO₂ heterostructure was examined for the first time as a NO gas sensor at room temperature, showing encouraging results. In particular, the heterostructure was grown using a polyol process, varying the pH of the solution, and the received powder was further annealed at 300 °C, 600 °C and 900 °C. The heterostructures were characterized using the X-ray diffraction technique, confirming the two different phases of SnO₂ and NiO, while their ratio was determined via energy-dispersive spectroscopy. In addition, the surface morphology was studied using scanning electron microscopy, revealing that the formed morphologies are dependent mainly on the annealing temperature and less on the pH. The heterostructures were tested toward NO gas at room temperature, showing a response of 1.8% at 2.5 ppm, far below the TLV, and a response time of 4.3 min. Compared to other NO gas sensors operating at room temperature, it can be concluded that NiO/SnO₂ is a candidate sensing material for NO detection in areas that are burdened by NO emissions caused by automobiles, factories, etc. Thus, by controlling the NO concentrations, diseases such as asthma can be avoided.