MnO₂-SnO₂ Based Liquefied Petroleum Gas Sensing Device for Lowest Explosion Limit Gas Concentration

Tin chloride dihydrate (SnCl₂·2H₂O; Mw = 225.65 g l⁻¹) and Manganese chloride dihydrate (MnCl₂·2H₂O; Mw = 161.87 g l⁻¹) are used as metal precursors, distilled water, and ethanol as a solvent. For the precipitation of the solution sodium hydroxide (NaOH; Mw = 39.997 g l⁻¹) was used. All the required chemicals are purchased from fisher scientific Pvt. Ltd. India and used without additional purification.

The MnO₂-SnO₂ nanocomposite was synthesized by a simple sol-gel process. Initially, 0.5 M of SnCl₂·2H₂O and 0.5 M of MnCl₂·2H₂O were dissolved in 30 ml distilled water or 30 ml ethanol solution in separate beakers and magnetic starring for 30 min at room temperature. Then the solution of MnCl₂·2H₂O was added slowly to SnCl₂·2H₂O solution with continuous vigorous stirring. After that, the aqueous solution of NaOH (0.5 M) was added dropwise to maintain pH 7. Then precipitate was found and kept at ambient temperature for 24 h for aging. The precipitated solution was centrifuged and washed with distilled water followed by ethanol alternatively three times. After that, the precipitate was dried at 70 °C in the oven for 5 h and obtained fine powder. Further, this powder was annealed at 500 °C for 2 h in the furnace.

The sensing film of MnO₂-SnO₂ nanocomposite is fabricated by spin coating techniques on the glass substrate (1 × 1 cm² dimension). Initially, the glass substrate was cleaned with distilled water, ethanol, and acetone by ultrasonication. After that, the glass substrate was dried at 60 °C for 10 min and a homogenous solution of MnO₂-SnO₂ nanocomposite was dropped on the substrate and spin coating at 1500 rpm for the 60 s. Then it was dried at 60 °C for 10 min and this process was followed three times for acquiring desired thickness and annealed at 350 °C for 1 h. The copper electrode is deposited on the sensing film by silver paste. The gas sensing setup has already been discussed in previously published work. ^13,31 The gas sensing setup consists of an inlet and outlet knob and the sensing device is connected through the 6517B electrometer. A hygrometer is kept inside the chamber to measure relative humidity. The LPG cylinder is connected to a mass flow controller to measure the concentration of LPG. The end of the electrometer is connected to the computer system and measures the electrical signal (current and resistance).

The sensor response of a gas sensor is an important parameter that can be defined by the ratio of current in presence of gas toair, whereas the sensitivity can be estimated by the slope of sensor response curve as given by Eqs1 and 2. ³²

$$\begin{eqnarray}&&SR=\displaystyle \frac{{I}_{g}}{{I}_{a}}\end{eqnarray} \tag{ 1 }$$

$$\begin{eqnarray}&&S=\displaystyle \frac{Sensor\,response}{Conecnetration}\end{eqnarray} \tag{ 2 }$$

The response time of a gas sensor is defined as the sensor reaching maximum change in current and becoming constant after the exposure of LPG, the 90% time taken in this process is called response time. Whereas, 90% time required by the sensor to come back to its primary stage is called recovery time.

In the XRD pattern of MnO₂, the diffraction peaks at the 2θ angles of 18.530°, 36.215°, 43.665°, 48.059°, 58.366°, 63.748°, 67.281°, 75.427°, 76.196°, 80.810°, and 83.665° correspond to the Braggs planes of (111), (311), (400), (331), (511), (440), (531), (533), (622), (444), and (711) as shown in Fig. 1a. From Fig. 1b, in the XRD pattern of SnO₂, many peaks are found at the 2θ angles of 26.535°, 33.663°, 37.764°, 51.620°, 54.775°, 57.883°, 61.903°, 64.739°, 65.876°, 71.322°, 78.594°, and 83.704°, which corresponds to the Braggs plane of (110), (101), (200), (211), (200), (002), (310), (112), (301), (302), (321), and (222). The MnO₂-SnO₂ nanocomposite was analyzed by X-ray diffraction for confirming the crystal nature and the XRD pattern as shown in Fig. 1c. From the XRD analysis, it can be seen that the material is highly crystalline. In the XRD pattern of MnO₂-SnO₂, a combination of MnO₂ and SnO₂ peaks are found and well matched with standard JCDPDS#44–0141 (Cubic MnO₂) and JCPDS#21–1250 (Tetragonal SnO₂). ^33,34 The XRD peaks at the 2θ angles of 18.630°, 26.526°, 33.835°, 37.861°, 51.726°, 54.678°, 57.890°, 61.857°, 64.786°, 65.919°, 71.343°, 78.751°, and 83.789° represent the Bragg planes of (111), (110), (101), (001), (211), (411), (220), (310), (112), (203), (013), (321) and (222) respectively. The crystallite size (D) and the induced strain of all samples were calculated by the Williamson-Hall plot. ³⁵

$$\begin{eqnarray}&&\beta \,C\mathrm{os}\,\theta =\frac{k\lambda }{D}+4\varepsilon \,S\mathrm{in}\,\theta \end{eqnarray} \tag{ 3 }$$

Zoom In Zoom Out Reset image size

Where k is the Scherrer constant, λ is the wavelength of the X-ray source (Cu Kα; α = 0.15406 nm), β is the full width and half maxima (FWHM), ε is the induced strain, and θ is the Braggs angle. From the Williamson-Hall plot, the crystallite size and induced strain were calculated. The crystallite size was calculated by the intercept of the linear fitting of the graph plotted between 4×Sinθ and β×Cosθ along the x and y-axis, respectively whereas the induced strain was calculated by the slope of the graph. The observed crystallite size for the samples was found to be 13.306 nm, 13.344 nm, and 16.786 nm for the samples SnO₂, MnO_2, and MnO₂–SnO₂ as shown in Figs. 1d–1f respectively. The induced strain of the samples SnO₂, MnO_2, and MnO₂–SnO₂ was found to be 5.499 × 10⁻⁴, 1.270 × 10⁻³, and 2.627 × 10⁻⁴. The lattice constant and volume can be calculated using the following relations: $\tfrac{1}{{d}^{2}}=\tfrac{{h}^{2}+\,{k}^{2}+\,{l}^{2}}{{a}^{2}}$ andV = ${a}^{3}$ for cubic structure MnO₂ whereas $\frac{1}{{d}^{2}}\,=\,\frac{{h}^{2}\,+\,{k}^{2}}{{a}^{2}}+\frac{{l}^{2}}{{c}^{2}}$ and V = ${a}^{2}c,$ for tetragonal structure (SnO₂). The interplanar distance of pure MnO₂ and SnO₂ were found to be 4.784 Å and 3.356 Å, respectively corresponding to (111) and (110) diffraction planes. ³⁶ In the MnO₂-SnO₂ composite the interplanar distance of MnO₂ and SnO₂ were found to be 4.771 Å and 3.358 Å structures. The dislocation density calculates using the formula: $\delta =\,1/{D}^{2}\,$ (nm⁻²), the calculated structural parameters are depicted in Table I.

Table I. Structural parameters of MnO₂, SnO_2, and MnO₂-SnO₂ nanocomposite.

Parameters	MnO2	SnO2	MnO2-SnO2 composites
			MnO2	SnO2
$a$ (Å)	8.286	4.746	8.284	4.748
b (Å)	8.286	4.746	8.284	4.748
c (Å)	8.286	3.212	8.284	3.188
Volume (Å)3	568.898	72.348	568.486	71.880
d (Å)	4.784	3.356	4.771	3.358
Dislocation density (δ) × 10−3 (nm−2)	5.616	5.648	3.54
Lattice strain (ε) × 10−3	1.27	0.549	0.262

The surface morphological investigations of all samples were carried out by using scanning electron microscopy. From Fig. 2a, SEM images of MnO₂ exhibited bunch-like coagulated nanostructures. SEM image of the SnO₂ nanomaterial shows the bunch of the small spheres collected such as shown in Fig. 2c. Fig. 2d the SEM image of MnO₂-SnO₂ depicts the bunches of the nanoparticles at the scale of 5 μm with the resolution of ×3000, whereas Fig. 2e is presented at the scale of 2 μm. Fig. 2f shows the comprehensive morphologies of the SnO₂ nano spheres and nanorods with brush-like aggregates with hemispherical ends projecting out. From the SEM images, it can also be observed that the particles were agglomerated. The agglomeration effect depends on the size of the particles such as particle's size moving towards the nanoscale, the agglomeration gets increased. The EDS spectra of the MnO₂ and SnO₂ are presented in Figs. 2g and 2h respectively, show the presence of all elements according to their content. The presence of the Pt element is due to the coating in process of imaging. In Fig. 2i, the energy dispersive X-ray (EDS) spectrum of MnO₂-SnO₂ shows that the sample contains the Mn, Sn, and O elements, and no other peaks are observed, which suggests the clear formation of the pure MnO₂-SnO₂ material. EDS spectrum has shown the presence of atomic weight percent of Sn with 7.55%, Mn with 6.37%, and O with 86.08%.

Zoom In Zoom Out Reset image size

The UV-visible absorption technique is very useful to analyze the optical properties of any material. For analysis of the optical properties of the material UV-visible absorbance spectrum was recorded from 250–900 nm as shown in Figs. 3a–3c. The optical bandgap of the as-prepared material was calculated by the Tauc's plot such as given in the relation Eq. 4. ³⁷

$$\begin{eqnarray}&&\left(\alpha h\upsilon \right)=K{\left(h\upsilon -{E}_{g}\right)}^{n}\end{eqnarray} \tag{ 4 }$$

Where K is the constant, v is transition frequency, and n defines the nature of transition such as for n = 1/2 and 3/2 corresponds to the direct allowed and direct forbidden transition, whereas n = 2 and 3 corresponds to the indirect allowed and indirect forbidden transition respectively. Here, a transition is direct allowed. The observed direct optical band gaps of MnO₂ and SnO₂ were found to be 3.407 eV and 3.037 eV respectively, whereas for MnO₂–SnO₂ nanocomposite, the band gap was found to be 3.202 eV. The observed band bap of the nanocomposite lies in the mid of both of its constituents, which signifies the band bending near the band edges.

Zoom In Zoom Out Reset image size

For confirming the functional groups associated with the as-prepared material, FTIR spectroscopy was used in the range of 4000–400 cm⁻¹ as shown in Figs. 3d–3f. From Fig. 3d, it can be observed that the stretching vibration of the Mn–O mode occurred at the wavenumber of 528.11 cm⁻¹, whereas Sn–O stretching vibration (Fig. 3e) was found at 615.53 cm⁻¹. In the FTIR spectrum of MnO₂–SnO₂, the peak at the wavenumber of 3403.74 cm⁻¹ represents the -OH stretching vibrations which is due to atmospheric humidity adsorbed on the surface of the nanocomposite. ²⁶ Another peak at the wavenumber of 1627.62 cm⁻¹ attributed to the C=O due to atmospheric CO₂ molecules. The peak at the wavenumber of 1155.15 cm⁻¹ represents the existence of some hydrates in the material. The peaks at the wavenumber of 615.18 and 528.39 cm⁻¹ are accredited to the vibration stretching of the Sn–O, and Mn–O vibrations respectively.

The LPG sensor works on the surface adsorption/desorption phenomenon in which the LPG interacts with the sensing surface. For the measurement of LPG sensing the device is fabricated by the drop cast method. The change in current/resistance was measured by Keithley electrometer 6517B. The sensing setup used was reported in previous work. ¹³ Generally, LPG consists of propane and butane and show a reducing nature which donates electron after the interaction. Before the measurement of LPG, the sensing device keeps in the ambient atmosphere for 24 for proper saturation.

Initially, the comparative LPG sensing of MnO₂, SnO_2, and MnO₂–SnO₂ nanocomposite film was performed at room temperature (32 °C and 60%RH) for 0.5 and 2.0 vol%, respectively. The LPG characteristics curve of MnO₂, SnO_2, and MnO₂–SnO₂ nanocomposite for 0.5 and 2.0 vol% LPG are shown in Figs. 4a–4c. The sensor response of MnO₂, SnO_2, and MnO₂–SnO₂ nanocomposite was calculated by using Eq. 1 for 0.5 and 2.0 vol% respectively, and the histogram is shown in Fig. 4d. For lower concentrations 0.5 vol%, the sensor responses of MnO₂, SnO_2, and MnO₂–SnO₂ were found to 1.058, 1.095, and 1.44 respectively whereas for higher concentrations these were found to be 1.321, 1.380, and 2.420 respectively. The LPG sensing behaviour of MnO₂–SnO₂ nanocomposite was found better in comparison to pure SnO₂ and MnO₂.

Zoom In Zoom Out Reset image size

Further, the LPG sensing characteristics of MnO₂–SnO₂ nanocomposites performed at RT for below LEL 0.5, 1.0, 1.5, and 2.0 vol% with time as shown in Fig. 5a. Initially, the current of sensing film in the air was saturated at 3.069 μm whereas in 0.5 vol% LPG the current suddenly increased around 4.45 μm. Because of this, LPG is a reducing gas and after the interaction current increases. The sensor response for each cycle is calculated by using Eq. 1 and plots the graph between sensor response and LPG concentration. For the lower concentration of 0.5 vol% of LPG, the sensing curve depict in Fig. 5b, in which the sensor response was found to be 1.44. Similar to other concentrations the sensor response was calculated. The maximum sensor response of MnO₂–SnO₂ nanocomposite film was found to be 2.42 for 2.0 vol%. The sensor response curve with LPG concentration is represented in Fig. 5c. The slope of linear fit data gave the values of sensitivity which is 0.646 sensor response/vol% of the sensing device. Additionally, the R² factor was found to be 0.99% which suggests that the device is properly working below LEL.

Zoom In Zoom Out Reset image size

The response and recovery time were calculated and mentioned in Table II. For the lowest concentration 0.5 vol%, the response and recovery time calculated as shown in Fig. 5b are found to be 17.02 and 23.85 s respectively. The comparative work reported on nanostructured-based LPG sensors are presented in Table III. The repeatability of MnO₂–SnO₂ sensing film was performed after 4 weeks for 0.5 vol% of LPG and found to 98.36% repeatable with the same result. The reproducibility curve of LPG sensing is represented by five consecutive cycles of LPG sensing as shown in Fig. 5e. The sensing film was found to have a 1.64% error because the atmospheric humidity will affect the sensing film. Another, the selectivity of a sensor is also a vital parameter that represents the ability of the sensing device for a particular gas.

In this work, selectivity of the MnO₂–SnO₂ nanocomposite sensing film was performed using various gases such as ethanol, acetone, ammonia, and CO₂ for the same concentration (0.5 vol%) and operating temperature. The selectivity histogram is shown in Fig. 5f. The sensor response of sensing film in presence of ethanol, ammonia, acetone, CO_2, and LPG was found as 1.08, 1.10 1.12, 1.13, and 1.44 respectively. The MnO₂–SnO₂ nanocomposite sensing film is highly sensitive to LPG in comparison to other gases.

The effect of humidity on the RT-operated gas sensor is an important issue because according to climate the humidity will be changed so the detection of LPG in presence of rhumidity was carried out. ^44,45 For this, relative humidity inside the chamber was increased from 10% to 90% using humidifier (K₂SO₄ solution). ⁴⁶ Fig. 5g shows the response curve of MnO₂–SnO₂ thin film for 0.5 vol% of LPG in presence of relative humidity (10%–90%). At the lower humidity range (10%–30%) the sensor response is high and in the mid-range (30%–60%) minute changes are observed, whereas in the higher humidity range (60%–90%) the sensor response is very low. Because at higher humidity the physisorption is saturated and found low sensor response. Hence, the MnO₂–SnO₂ nanocomposite sensing film is good for the detection of LPG in the mid-range.

Mostly, the LPG sensing mechanism of metal oxide semiconducting materials is described by the adsorption/desorption surface phenomena. ⁴⁷ Due to the adsorption/desorption of analytes, the resistance or current of sensing film may be increased or decreased depending upon the type (n or p) of materials and target gas (reducing or oxidizing). ^48,49 In the present study, MnO₂ and SnO₂ are n-type semiconductors and formed n–n heterojunction. From UV spectrum, we have calculated the bandgap of SnO₂ and MnO₂ as 3.03 and 3.40 eV respectively. The low bandgap materials (SnO₂) have more electron density in comparison to high bandgap materials (MnO₂). From the conduction of low bandgap materials, the electrons move towards the conduction band of high bandgap materials, whereas the holes are migrated to the opposite. Due to the movement of electrons and holes, the band is shifted to its own position and the Fermi level is equilibrated and creating an electronic depletion layer at the interface of both semiconductors. Fig. 6a shows the band bending diagram before and after equilibrium of Fermi level and formation of n–n heterojunction. This junction is responsible for the adsorption/desorption of oxygen air and LPG. The sensing mechanism of MnO₂–SnO₂ nanocomposite thin film is shown in Fig. 6b. Earlier to exposure of LPG, environmental oxygen become adsorbed on the surface of the sensing film, electrons are trapped from the conduction band and oxygen species (O₂ ⁻, O⁻, and O²⁻) are formed, which can be explained as the following reactions (Eqns. 5, 6 and 7): ^50,51

$$\begin{eqnarray}&&{O}_{2\left(gas\right)}\to {O}_{2\left(ads\right)}\end{eqnarray} \tag{ 5 }$$

$$\begin{eqnarray}&&{O}_{2\left(gas\right)}+{e}^{-}\to {O}_{2(ads)}^{-}\end{eqnarray} \tag{ 6 }$$

$$\begin{eqnarray}&&{O}_{2(ads)}^{-}+{e}^{-}\to 2{O}_{\left(ads\right)}^{-}\end{eqnarray} \tag{ 7 }$$

Zoom In Zoom Out Reset image size

Due to the formation of oxygen ions on the surface of sensing film, the electronic depletion layer is formed which creates a depletion region and acts as a potential barrier. So, the current of sensing film was decreased and saturation after the adsorption of oxygen molecules. After that, the LPG was exposed inside the chamber, it interacted with chemisorbed oxygen species and formed an intermediate complex (C_nH_2n:O) with releasing the electrons. Chemical reactions can be represented by Eqs. 8 and 9.

$$\begin{eqnarray}&&{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2n+2}+2{{\rm{O}}}^{-}\to {{\rm{H}}}_{2}{\rm{O}}+{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2{\rm{n}}}:{\rm{O}}+{{\rm{e}}}^{-}\end{eqnarray} \tag{ 8 }$$

$$\begin{eqnarray}&&{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2{\rm{n}}}:{\rm{O}}+{{\rm{O}}}^{-}\to {{\rm{H}}}_{2}{\rm{O}}+{{\rm{CO}}}_{2}+{{\rm{e}}}^{-}\end{eqnarray} \tag{ 9 }$$

where n = 3 and 4. This intermediate complex is highly unstable and interacts with oxygen species (O⁻) and releases CO₂ and H₂O contents in the end product through the dehydrogenation mechanism. The electron density in the conduction band is increased, so the current is increased in presence of LPG as shown in Fig. 6a. When the concentration of LPG increases then the reaction rate increases, further as a result, the current has increased and becomes constant.