SO₂-sensing properties of NiO nanowalls synthesized by the reaction of Ni foil in NH₄OH solution

Nickel oxide (NiO) is a $p-$type semiconductor with a wide bandgap energy of 3.6–4.0 eV. NiO is a stable $p-$type metal oxide with high carrier transport performance [1, 2] and thus has wide-ranging applications, such as in lithium-ion batteries [3], supercapacitors [4], biosensors [5], and catalysts for water splitting [6]. Moreover, many groups claim that NiO nanomaterials exhibit excellent sensing properties toward formaldehyde, ${{{\rm H}}_{2}}{\rm S}$, ethanol, ${\rm N}{{{\rm H}}_{3}},{{{\rm H}}_{2}}$, and ${\rm N}{{{\rm O}}_{2}}$ [1, 2, 7–9]. However, studies on the ${\rm S}{{{\rm O}}_{2}}{{\rm \text -}}$sensing behavior of NiO nanostructures are few.

NiO nanomaterials with wall-plate-like morphologies are predominantly synthesized [10–13] among others (e.g. nanorods, nanowires, nanoflowers, and nanotubes) [14–17]. Several groups introduce different approaches for the synthesis of NiO nanowalls on Ni metal surfaces. Zhan et al synthesized 3D NiO nanowalls on a Ni foam through a hydrothermal method [10]. Tang et al [13] used electrochemical corrosion approach to prepare NiO nanowalls grown on a Ni foam. Ni et al introduced an easy electrochemical corrosion method to fabricate NiO nanowalls [18]. NiO nanowalls can also be prepared by plasma-assisted oxidation [19]. Recently, our group introduced an easy, low-cost method of synthesizing ${\rm Ni(OH}{{)}_{2}}{\rm /NiO}$ nanowalls through the surface reaction of Ni foil in aqueous ${\rm N}{{{\rm H}}_{4}}{\rm OH}$ [20]. Nonetheless, the gas-sensing properties of the NiO nanowalls fabricated by these methods remain lacking. The advantage of NiO nanowalls over other morphologies like nanowire, nanoparticles or nanosheets is to provide better film porosity which can enhance the adsorption and desorption processes of the target gases over the film surface.

In this study, ${\rm Ni(OH}{{{\rm)}}_{2}}$ nanowalls were fabricated through the chemical reaction of Ni foil with aqueous ${\rm N}{{{\rm H}}_{4}}{\rm OH}$. The NiO nanowalls were obtained by dehydrating the ${\rm Ni(OH}{{{\rm)}}_{2}}$ nanowalls at $500\, {}^\circ{\rm C}$ using a hotplate. The ${\rm S}{{{\rm O}}_{2}},\,{\rm N}{{{\rm O}}_{2}},\,{\rm N}{{{\rm H}}_{3}}$ and ${{{\rm H}}_{2}}{\rm S}$ sensing properties of the as-prepared NiO nanowalls were investigated and compared at a working temperature of $50\, {}^\circ{\rm C}-300\, {}^\circ{\rm C}$.

We used a facile method in our previous study to synthesize the NiO nanowalls using a Ni foil [20]. Figure 1 illustrates our synthesis process. Commercial pure Ni foil (thickness $=~0.1\,{\rm mm}$; purity  =  99.99%; Aldrich) was cut into a $2\,{\rm cm}\times 2\,{\rm cm}$ plate (figure 1(a)). The plate was immersed in acetic acid for 10 min and then ultrasonically cleaned in a bath sonicator with acetone, ethanol, and distilled water for 5 min. After drying under a flow of ${{{\rm N}}_{2}}$, the plate was folded at the corners to form table-like plate and was added to 25 ml of ${\rm N}{{{\rm H}}_{4}}{\rm OH}$ solution in a 100 ml Duran laboratory bottle (figure 1(b)). Meanwhile, a cleaned glass ($2\,{\rm cm}\times 2\,{\rm cm}$) was positioned under the plate. The Duran bottle was kept at $70\,~ {}^\circ{\rm C}$ for 48 h in an oven. After treatment, the glass substrate was covered by a green film (figure 1(c)), dried in an oven at $70\,~ {}^\circ{\rm C}$ for 12 h, and annealed at $500\, {}^\circ{\rm C}$ for 1 h using a hotplate. As a result, the green film became a dark brown film (figure 1(d)). A simple parallel Au electrode was patterned on the film by thermal evaporation (figure 1(e)).

Zoom In Zoom Out Reset image size

The morphologies of samples were characterized by field-emission scanning electron microscopy (FESEM; JEOL JSM-7610F). The structures of the samples were identified using x-ray diffraction (XRD; XPERT-PRO) with Cu Kα radiation of $\lambda =1.5418\,{{\rm }\mathring{\rm A} }$. The NiO nanomaterial was extracted from the annealed sample and was dispersed on a molybdenum grid for transmission electron microscopy (TEM; Philips Telnai G2F20 S-TWIN) observations. The gas-sensing properties of the as-synthesized materials were measured using a dynamic gas-testing system. During measurement, the chamber was pumped and was maintained at a vacuum pressure of 10 Torr. The carrier gas was dry air and the total flux rate was 200 sccm. The operating temperature was from $50\,{}^\circ {\rm C}$ to $300\,{}^\circ {\rm C}$ ($300\,{}^\circ {\rm C}$ is the maximum value of our gas-testing system). Gas response was defined as $S=\frac{({{R}_{g}}-{{R}_{a}})}{{{R}_{a}}}\%$, where ${{R}_{g}}$ and ${{R}_{a}}$ represent sensor resistance in the target gas and air, respectively. Sensor response $({{\tau }_{{\rm Response}}})$ and recovery time $({{\tau }_{{\rm Recovery}}})$ are determined by the time for sensor resistance to reach 90% of its steady-state value from ${{R}_{a}}$ to ${{R}_{g}}$ and ${{R}_{g}}$ to ${{R}_{a}}$, respectively.

Figure 2 shows the XRD patterns of the films before and after annealing at $500\,{}^\circ {\rm C}$. For the pristine film, major diffraction peaks at $11.6{}^\circ ,\,33.6{}^\circ ,\,35.9{}^\circ$ and $60.1{}^\circ$ can be assigned to the (001), (110), (111), and (301) planes of $3{\rm Ni(OH}{{{\rm)}}_{2}}\cdot 2{{{\rm H}}_{2}}{\rm O}$ [JCPDS file no. 00-022-0444, hexagonal structure, space group P–31 m (162)] (figure 2(a)). Minor peaks at $2\theta$ values of $19.4{}^\circ ,\,32.5{}^\circ$ and $70.9{}^\circ$ correspond to (001), (100), and (103) planes of ${\rm Ni(OH}{{{\rm)}}_{2}}$ [JCPDS file no. 00-014-0117, hexagonal structure, space group P–31 m (164)]. No other impurities form in the film. Figure 2(b) shows the XRD pattern of the annealed film, indicating that ${\rm Ni(OH}{{{\rm)}}_{2}}$ was transformed to NiO. The peaks located at $36.8{}^\circ ,\,42.8{}^\circ$ and $62.1{}^\circ$ well agree with (101), (012), and (110) planes of NiO [JCPDS file no. 98-016-6115, hexagonal structure, space group R–3 m (166)]. In this pattern, the second phase of ${\rm N}{{{\rm i}}_{{\rm 3}}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ [JCPDS file no. 00-006-0144, hexagonal structure] is also detected by the existence of (100), (006), and (105) planes at the diffraction peaks of $34{}^\circ ,\,37.1{}^\circ$ and $46.2{}^\circ$, respectively. The board peaks observed in both patterns suggest that the film is composed of small crystallites. The crystallite size of NiO calculated through the Scherrer equation [21] with the $2\theta$ of $62.1{}^\circ$ is 2.2 nm.

Zoom In Zoom Out Reset image size

The surface morphologies of the film before and after annealing at $500\,{}^\circ {\rm C}$ for 1 h were similar (figures 3(a) and (c)). The films comprise curvy nanowalls with an average thickness of 15 nm. Cross-sectional FESEM images indicated that the thickness of both films is ~1450 nm (figures 3(b) and (d)). The sample before annealing (figure 3(b)) comprises a film layer (~314 nm) and a nanowall layer (~1122 nm). The sample after annealing is dense with a nanowall thickness of approximately 561 nm (figure 3(d)). The synthesis of NiO nanowalls, in which Ni can react with ${\rm N}{{{\rm H}}_{4}}{\rm OH}$ to form ${\rm Ni(OH}{{{\rm)}}_{2}}$, which was dispersed in the solution, was proposed in our previous study [20]. The nanowalls are probably formed by an oriented assembly of ${\rm Ni(OH)}_{4}^{2-}$ units onto the ${\rm Ni(OH}{{{\rm)}}_{2}}$ nanoseeds. The transformation from ${\rm Ni(OH}{{{\rm)}}_{2}}$ to NiO without morphology change at $500\,{}^\circ {\rm C}$ is well documented [20, 22, 23].

Zoom In Zoom Out Reset image size

Figures 4(a) and (b) show the TEM and HR-TEM images of the extracted NiO nanowalls, respectively. The nanowalls were composed of many small crystallites (figure 4(b)). The dimensions of all crystallites are analyzed in a histogram graph (figure 4(c)). The average diameter of the crystallites is 2.2 nm, which is similar to the calculated value using the XRD pattern equation above.

Zoom In Zoom Out Reset image size

Figure 5 shows the gas responses of the as-prepared NiO nanowalls toward ${{{\rm H}}_{2}}{\rm S},\,{\rm N}{{{\rm O}}_{2}},\,{\rm S}{{{\rm O}}_{2}}$ and ${\rm N}{{{\rm H}}_{3}}$. The device is highly sensitive to ${\rm S}{{{\rm O}}_{2}}$. The ${\rm S}{{{\rm O}}_{2}}$ response of the device gradually increases from 6.3% at $100\,{}^\circ {\rm C}$ to 8.7% at $250\,{}^\circ {\rm C}$ and saturates to 8.8% at $300\,{}^\circ {\rm C}$. The optimal operating temperature of the NiO nanowalls was $200\,{}^\circ {\rm C}$ to ${{{\rm H}}_{2}}{\rm S}$ and ${\rm N}{{{\rm O}}_{2}}$ and $100\,{}^\circ {\rm C}$ to ${\rm N}{{{\rm H}}_{3}}$. Figure A1 (appendix) shows the transient curves of the device at different operating temperatures. The positive response of the device to reduced gases (${{{\rm H}}_{2}}{\rm S},\,{\rm N}{{{\rm O}}_{2}}$ and ${\rm S}{{{\rm O}}_{2}}$) and to oxidizing gas $({\rm N}{{{\rm O}}_{2}})$ suggests that the NiO nanowalls exhibit a $p-$type semiconductor sensing characteristic. The gas-sensing mechanisms of NiO to both types of gases are well documented [7, 24]. Table 1 shows a comparison of NiO nanowall-based ${\rm S}{{{\rm O}}_{2}}$ sensor with other materials. The sensing performance, 'gas response' and 'response/recovery time' of the synthesized NiO nanowalls are not better than those of reported nanomaterials. However, the sensing performance of a gas sensor strongly depends on many factors, such as electrodes (material, gap, size, shape, and position), testing system set-up (gas inlet position and flux rate), and test chamber (chamber volume, medium with/without oxygen, and humidity). Thus, the limit of detection (LOD) to ${\rm S}{{{\rm O}}_{2}}$ should be considered for comparison. For this specification, the NiO nanowalls can detect down to 1 ppm ${\rm S}{{{\rm O}}_{2}}$ gas, which is better than many of the reported ${\rm S}{{{\rm O}}_{2}}$ sensors.

Table 1. Comparison of NiO nanowall-based ${\rm S}{{{\rm O}}_{2}}$ sensor and other reported ${\rm S}{{{\rm O}}_{2}}$ sensors.

Material	Operating temperature (°C)	${\rm S}{{{\rm O}}_{2}}$ concentration (ppm)	Gas response	Response time (s)	Recovery time (s)	LOD	Ref.
BiFeO3 nanoparticles	300	5	2.03	20	50	5 ppm	[25]
Graphene oxide  +  ${\rm Sn}{{{\rm O}}_{2}}$ nanocomposites	60	10	1.25	~150	~150	10 ppm	[26]
${\rm Ru/A}{{{\rm l}}_{2}}{{{\rm O}}_{3}}{\rm /ZnO}$ nanocomposites	350	5	~10 %	60	360	5 ppm	[27]
${\rm Ti}{{{\rm O}}_{2}}$ nanotubes	200	10	~15 %	~100	~100	10 ppm	[28]
${\rm Sn}{{{\rm O}}_{2}}$ thin film	180	500	55	80	70	50 ppm	[29]
${\rm W}{{{\rm O}}_{3}}$ nanoparticles	260	5	~1.18	~180	~600	5 ppm	[30]
${\rm Sn}{{{\rm O}}_{2}}$ nano-dodecahedrons	183	5	1.5	~10	~15	200 ppb	[31]
NiO nanowalls	300	5	4 %	40	57	1 ppm	This work

Zoom In Zoom Out Reset image size

The ${\rm S}{{{\rm O}}_{2}}-$sensing behavior of the NiO nanowalls is related to the reaction of the adsorbed oxygen adatoms ${{{\rm O}}^{-}}$ on the NiO surface with the ${\rm S}{{{\rm O}}_{2}}$ molecules, which can be described as equation (1):

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {\rm S}{{{\rm O}}_{2}}({\rm g})+{{{\rm O}}^{-}}(ads)\to {\rm S}{{{\rm O}}_{3}}-{{h}^{+}}.\nonumber \end{align} \tag{ 1 }$$

After the reaction, the reduced number of holes in the accumulation zone results in increased sensor resistance. The sample exhibits higher response toward ${\rm S}{{{\rm O}}_{2}}$ than toward ${{{\rm H}}_{2}}{\rm S}$ although both gases possessed sulfide in the molecules. This behavior can be attributed to the reaction of ${\rm S}{{{\rm O}}_{2}}$ with NiO as mentioned by Tyagi et al [31]. NiO can react with ${\rm S}{{{\rm O}}_{2}}$ gas to produce NiS:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {\rm S}{{{\rm O}}_{{\rm 3}}}{\rm (g)}+{{{\rm O}}^{-}}(ads)\to {\rm S}{{{\rm O}}_{2}}+{{{\rm O}}_{2}}({\rm g})-{{h}^{+}}.\nonumber \end{align} \tag{ 2 }$$

The produced ${\rm S}{{{\rm O}}_{3}}$ gas molecules react with the remaining oxygen adatoms on the NiO surface:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {\rm S}{{{\rm O}}_{{\rm 3}}}{\rm (g)}+{{{\rm O}}^{-}}(ads)\to {\rm S}{{{\rm O}}_{2}}+{{{\rm O}}_{2}}({\rm g})-{{h}^{+}}.\nonumber \end{align} \tag{ 3 }$$

Under the presence of oxygen, NiS can be transformed to NiO:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {\rm NiS}+\frac{3}{2}{{{\rm O}}_{2}}\to {\rm NiO}+{\rm S}{{{\rm O}}_{2}}.\nonumber \end{align} \tag{ 4 }$$

The released ${\rm S}{{{\rm O}}_{2}}$ in equations (3) and (4) further reacts with oxygen adatoms following equation (1), thereby decreasing the number of holes in the accumulation. Thus, the NiO nanowalls are more sensitive to ${\rm S}{{{\rm O}}_{2}}$ than to ${{{\rm H}}_{2}}{\rm S}$.

Studies on the gas-sensing behavior of ${\rm N}{{{\rm i}}_{3}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ are rare. However, ${\rm N}{{{\rm i}}_{3}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ remained in the sample after annealing. This phase is possibly remained at the bottom of the sample because of the thick/dense film. ${\rm N}{{{\rm i}}_{3}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ can be easily transformed to NiO by giving off water molecules above $200\,{}^\circ {\rm C}$ [32]. The decomposition of ${\rm N}{{{\rm i}}_{3}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ may be prohibited at the bottom of the film because the film was thick and dense. Thus, the ${\rm N}{{{\rm i}}_{3}}{{{\rm O}}_{2}}{{({\rm OH})}_{4}}$ layer may not influence the sensor performance during the test.

Device stability is confirmed by the transient curves after exposure to 5 pulses of 20 ppm ${\rm S}{{{\rm O}}_{2}}$ (figure 6(a)). Sensor resistances under 'gas on' and 'gas off' are retained. At $300\,{}^\circ {\rm C}$ operating temperature, the response and recovery times of the device to 20 ppm ${\rm S}{{{\rm O}}_{2}}$ are 40 and 57 s, respectively (figure 6(b)). Figure 7 reveals the influence of ${\rm S}{{{\rm O}}_{2}}$ concentration on sensor resistance and sensor response. The minimum ${\rm S}{{{\rm O}}_{2}}$ concentration that the device can detect is 1 ppm (figure 7(a)) with a response of just above 2.5%. The relationship of the increase in sensor response with ${\rm S}{{{\rm O}}_{2}}$ concentration is linear (figure 7(b)). The linear correlation coefficient ${{R}^{2}}$ is 99%, indicating the high potential of the as-synthesized nanomaterial as a ${\rm S}{{{\rm O}}_{2}}$ sensor.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Curvy NiO nanowalls were grown on a glass substrate through the reaction of Ni foil in aqueous ${\rm N}{{{\rm H}}_{4}}{\rm OH}$. The nanowalls are found to be uniform with an average thickness of 15 nm. The nanowalls are extremely sensitive to ${\rm S}{{{\rm O}}_{2}}$ compared with ${{{\rm H}}_{2}}{\rm S},\,{\rm N}{{{\rm O}}_{2}}$ and ${\rm N}{{{\rm H}}_{3}}$. The highest response of the device to ${\rm S}{{{\rm O}}_{2}}$ is 8.8% at $300\,{}^\circ {\rm C}$, and this value is approximately two, six, and two times higher than the responses of the device to ${{{\rm H}}_{2}}{\rm S},\,{\rm N}{{{\rm H}}_{3}}$ and ${\rm N}{{{\rm O}}_{2}}$, respectively. The as-prepared NiO nanowalls exhibit good stability (in terms of sensor resistance), reasonable response/recovery time (below 1 min), and linear variation of resistance with ${\rm S}{{{\rm O}}_{2}}$ concentration. The LOD of the NiO nanowalls to ${\rm S}{{{\rm O}}_{2}}$ is 1 ppm. Thus, the NiO nanowalls synthesized through our simple method are a high-potential ${\rm S}{{{\rm O}}_{2}}$ sensor.

This research is funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED), under grant number 103.02-2016.20.