Design and fabrication of quick responsive and highly sensitive LPG sensor using ZnO/SnO₂ heterostructured film

Zinc acetate dihydrate and tin chloride dihydrate were used as the starting materials. 0.2M aqueous solution of Zn(CH₃COO)₂.2H₂O and 0.1M aqueous solution of SnCl₂.2H₂O were mixed and stirred at room temperature for 30 min 0.05M NaOH solution was added dropwise to the mixture with continuous vigorous stirring till the pH of the solution became neutral. The solution was then transferred to a 100 ml Teflon-lined stainless-steel autoclave. Heat and pressure treatments were then given to the mixture by maintaining the temperature at 180 °C for 20 h. After cooling to room temperature, the obtained solution was centrifuged and washed with distilled water and ethanol, then dried at 80 °C for 5h in a hot air oven and calcined at 500 °C for 2h @ 10 °C min⁻¹ in a programmable furnace. The final product obtained was ZnO/SnO₂ heterostructure. The synthesis process of ZnO/SnO₂ is depicted with the help of a flowchart shown in figure 1.

Zoom In Zoom Out Reset image size

The glass substrate of dimension 1 × 1 cm² was taken and cleaned using distilled water and ethanol followed by acetone in an ultrasonic cleaner. Further, the substrates were dried at 80 °C on a hot plate for 15 min to remove VOCs. Then, the substrates were ready for film fabrication. A dilute solution of ZnO/SnO₂ was made in isopropyl alcohol (IPA) while sonicating by ultrasonic waves for 4 h. The dilute solution was obtained which was then spun on the rinsed substrates using spin coater at a speed of 3500 rpm for 30 s and dried at 50 °C for 10 min on a hot plate. The film fabrication process of coating and drying were repeated three times for achieving the required thickness.

XRD pattern of the prepared powder of ZnO/SnO₂ heterostructure was recorded by glancing angle x-ray Diffractometer (Bruker D8 advanced Ecosystem), equipped with monochromatic Cu-K_α as radiation source (40 kV and 20 mA). Surface morphological study of the material was done by the Field emission scanning electron microscope (JEOL, JSM 7610F) equipped with W (Tungsten) hairpin filament & LaB₆ gun operated at 20 kV. The particle size of the nanocomposite was analyzed by a Particle size analyzer (Nanozetasizer-ZS90). Optical characterization of the sample was performed by using UV–visible spectrometer (Evolution 201) to record the UV-Visible absorption spectra of ZnO/SnO₂ heterostructures. Also, Fourier transform infrared spectrum was obtained using FTIR spectrophotometer (Thermoscietiic Nicole 6700) in the spatial frequency regime of 4000–400 cm⁻¹.

For LPG sensing, a gas chamber made of borosil glass along with the inlet and outlet knobs was designed. The inlet of the gas chamber was connected to the concentration measuring unit for the exact vol% of the gas inserted inside the chamber and an outlet knob for removal of LPG. A detailed description of the gas chamber was reported in our previous publication [14]. Silver contacts were made on the fabricated film which was then used as a LPG sensing element. For electrical measurements, Keithley electrometer (Keithley-6517B) was used.

The x-ray diffraction pattern of ZnO/SnO₂ heterostructure is depicted in figure 2(a). The diffraction peaks well match with the JCPDS card no. 79–0207 revealing ZnO wurtzite hexagonal structure (P63mnc space group) and with JCPDS card no. 21–1250 revealing SnO₂ tetragonal structure (P42/mnm space group. The crystallite size (D_hkl) from every peak was calculated by using Scherrer's formula [26] given in equation (1).

$$\begin{eqnarray}&&{{\rm{D}}}_{{\rm{hkl}}}=\,\displaystyle \frac{0.9\lambda }{{\beta }_{{\rm{hkl}}}\,\mathrm{Cos}\,\left({\theta }_{{\rm{hkl}}}\right)}\end{eqnarray} \tag{ 1 }$$

Here λ is the x-ray wavelength, θ_hkl is the Bragg diffraction angle and β_hkl is the full width at half maximum (FWHM) of the main peaks in XRD pattern in radian. The average crystallite size of all peaks was obtained as 17.55 nm.

Zoom In Zoom Out Reset image size

The Williamson and Hall (W-H) plot for relating the size and microstrain broadening was analyzed to estimate the average size and the strain in grains [27, 28]. The W-H plot shown in figure 2(b) has a slope value of 4.81 × 10^–3, which reveals the compressive microstrain in ZnO/SnO₂ heterostructure. The crystallite size estimated from the intercept value of W-H plot is 17.33 nm which is well matching with that calculated from Scherrer's formula (17.55 nm).

Also, the texture coefficient T_C(hkl) for the domination of crystallite orientation in the grown sample has been calculated by using equation (2) from the XRD data. The T_C (hkl) value greater than one reveals the preferred orientation, as well as the random orientation, are confirmed for the 1 value of T_C. Among all texture coefficients, the value 1.59 associated with the plane (100) shows the most prominent growth of ZnO along (100) plane and in the case of SnO₂ shows the value 1.26 associated with the (101) plane.

$$\begin{eqnarray}&&{{\rm{T}}}_{{\rm{C}}}\left({\rm{hkl}}\right)=\displaystyle \frac{\displaystyle \frac{I\left(hkl\right)}{{I}_{o}\left(hkl\right)}}{\displaystyle \frac{1}{N}{\sum }_{n}\displaystyle \frac{I\left(hkl\right)}{{I}_{o}\left(hkl\right)}}\end{eqnarray} \tag{ 2 }$$

Here, I(hkl) is the measured relative intensity of the plane (hkl), I_o(hkl) is the standard intensity of the plane (hkl), N is the reflection number and n is the number of diffraction peaks.

The quantitative phase analysis of ZnO (Hexagonal) and SnO₂ (Tetragonal) structures was estimated approximately from the ratio of intensities of individual phase peaks to the intensities of all peaks in heterostructures using the relations given in equations (3) and (4):

$$\begin{eqnarray}&&{\rm{H}}\left( \% \right)=\displaystyle \frac{{\rm{Area}}\,{\rm{of}}\,{\rm{Hexagonal}}\left({\rm{ZnO}}\right){\rm{peaks}}}{{\rm{Area}}\,{\rm{of}}\,{\rm{all}}\,\mathrm{peaks}\,}\times 100\end{eqnarray} \tag{ 3 }$$

$$\begin{eqnarray}&&{\rm{T}}\left( \% \right)=\displaystyle \frac{{\rm{Area}}\,{\rm{of}}\,\mathrm{Tetragonal}\,\left({{\rm{SnO}}}_{2}\right){\rm{peaks}}}{{\rm{Area}}\,{\rm{of}}\,{\rm{all}}\,\mathrm{peaks}\,}\times 100\end{eqnarray} \tag{ 4 }$$

Here, H(%) denotes the percentage of hexagonal ZnO presence in ZnO/SnO₂ and T(%) denotes the percentage of tetragonal SnO₂ structure respectively. From the XRD data of ZnO/SnO₂ heterostructure, it was found that tetragonal SnO₂ has 53.18% whereas hexagonal ZnO has 46.91%.

The particle size of the ZnO/SnO₂ heterostructure was analyzed by using Nanozetasizer in which the average particle size was measured by dynamic light scattering (DLS) method provided the nanomaterials are dispersed in solution. This method assumes the spherical particle with an aspect ratio of 1[29].

The graph obtained from Nanozetasizer for particle size analysis is shown in figure 3(a). From figure 3(a), it can be observed that the hydrodynamic diameter lies in the range 50–80 nm and the average particle size was found 64.23 nm.

Zoom In Zoom Out Reset image size

Also, Zeta potential was measured using Nanozetasizer which defines the voltage at the edge of the shipping (shear) plane with respect to the bulk dispersion medium. Zeta potential predicts the long-term stability of nanoparticles. The measurement was done at 25 °C with iso-propyl alcohol solvent containing 1.0 mmol l⁻¹ solvent. From figure 3(b), the value of zeta potential and conductivity was found −10.3 mV and 0.0116 (mS cm⁻¹) respectively for ZnO/SnO₂ heterostructure in isopropyl alcohol at 25 °C.

The UV-Vis (250–900 nm) absorbance spectrum for ZnO/SnO₂ heterostructure is shown in figure 4(a). The absorption was observed in the UV region (∼352 nm). The slow increase in absorbance and broadening of absorbance peak may be attributed to defects and particle size distribution. The absorption coefficient (α) can be calculated by Beer–Lambert's relation shown in equations (5) and (6):

$$\begin{eqnarray}&&I=\,{I}_{0}{e}^{-\alpha x}\end{eqnarray} \tag{ 5 }$$

$$\begin{eqnarray}&&\alpha \left(\lambda \right)=\displaystyle \frac{\mathrm{ln}\left(\displaystyle \frac{{I}_{0}}{I}\right)}{x}=\left(\displaystyle \frac{lnT}{x}\right)=\,\displaystyle \frac{2.303\,A}{x}\end{eqnarray} \tag{ 6 }$$

Where I₀ is the initial photon intensity, I is instantaneous photon intensity and x is the thickness of the cuvette. The variation of α(λ) of ZnO/SnO₂ heterostructure with wavelength is depicted in figure 4(b) and it was seen that as wavelength was increased, the absorption coefficient was decreased. The reason for this decrease may be due to internal electric field or distortion of lattice due to strain of charge carrier's inelastic scattering by phonons. The higher value of α in UV-region (330–400 nm) for the heterostructure is due to transition among extended states in valence and conduction bands.

Zoom In Zoom Out Reset image size

The optical energy band gap value for ZnO/SnO₂ heterostructure was investigated using 'Tauc's plot' (inset in figure 4(a)) [9]. The Tauc's relation given in equation (7) was used for obtaining the optical band-gap (E_g) [30].

$$\begin{eqnarray}&&\alpha =\displaystyle \frac{{\alpha }_{o}{\left(hv-{E}_{g}\right)}^{n}}{h\nu }\end{eqnarray} \tag{ 7 }$$

Here hν is photon energy, α is absorbance coefficient, α_o is a characteristic parameter, h is Planck's constant, and n is power factor. The value of n may vary depending upon the type of transition taking place. Here, the transition is a direct allowed type, therefore 'n' is taken as ½. The estimated value of E_g for ZnO/SnO₂ heterostructure is ∼3.85 eV.

FTIR spectroscopy operated at 4000–400 cm⁻¹ was used for the identification of the functional group in ZnO/SnO₂ heterostructure. The corresponding spectrum is shown in figure 5. The data reveals that the absorption peak at 485.26 cm⁻¹ is assigned to the Zn-O bonding, 534.18 cm⁻¹ for Zn-O-Sn bonding and 601.68 cm⁻¹ corresponds to O-Sn-O antisymmetric stretching. The broad absorption peaks observed at 3423.02 cm⁻¹ reveals the O–H stretching due to the absorption of water molecules. Some additional peaks at 1635.34 cm⁻¹, 1392.35 cm^−1, and 1116.58 cm⁻¹ are assigned to N–O stretching (nitro compound), O–H bending (due to alcohol), and C–O stretching (due to secondary alcohol). These peaks are occurring because in the synthesis process ammonia solution was used for precipitation and alcohol was used for washing.

Zoom In Zoom Out Reset image size

Raman spectroscopy is one of the effective tools to analyze the vibrational modes, sensitive defects, size, and structural phase purity of the nanomaterials. Figure 6 shows the Raman spectrum of ZnO/SnO₂ heterostructure at room temperature. Zinc oxide has a hexagonal crystal structure with a point group ${C}_{6v}^{4}$ and space group (P6₃mc) and tin oxide has a tetragonal crystal structure with point group ${D}_{4h}^{14}$ and space group P4₂/mnm. From figure 6 the Raman peaks observed at 334, 439, and 582 cm⁻¹ [31] which attributed to the E_2H-E_2L, E_2H, and E_1L vibrational modes of hexagonal ZnO respectively, whereas the peaks at 381, 480 and 763 cm⁻¹ can be described to the Eu, Eg and B_2g vibrational mode of tetragonal SnO₂, respectively [32]. The vibrational modes of ZnO as E_2H-E_2L, E_2H, and E_1L are associated with the vibrational of heavy Zn sublattice and an oxygen atom. Another mode such as Eu, Eg and B_2g represent IR active mode, doubly degenerates mode, and B_2g non-degenerate mode respectively. In non-degenerate Raman active modes (B_2g) vibrations in the plane perpendicular to the c-axis while the doubly degenerates Raman E_g mode is along to the direction of the c-axis. The Sn-atom is in rest position, whereas the O-atom vibrates. The same Raman active modes were observed in earlier literature which is in good agreement with standard hexagonal ZnO and tetragonal SnO₂ structure [33, 34].

Zoom In Zoom Out Reset image size

The morphological analysis of ZnO/SnO₂ heterostructure film was done at 1 μm and 100 nm scales using a FESEM and shown in figure 7(a) and (b). It can be observed from figures that the particles have formed spherical clusters leaving the voids on the film. These clusters together with voids serve as active centres for the adsorption/desorption of gas molecules. Also, the heterostructured material due to the composite formation provides enhanced sensing activity by changing the material physics at the electronic level [15]. Also, the composition of the sensing film was analyzed by Energy-dispersive x-ray spectroscopy. The obtained spectrum is shown in figure 8 which clearly showed the existence of Zn, Sn, and O elements. The % composition of elements is shown in the inset of figure 8.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The prepared film of ZnO/SnO₂ heterostructure was exposed to LPG and the variation in electrical resistance with time for different concentrations of LPG was recorded using an electrometer at room temperature. The sensing behaviour was studied by focusing on its sensing parameters with the exposure of LPG. For this, sensor response, response time, recovery time, sensitivity, and selectivity were studied.

In the case of metal oxide, the gas sensors work on the principle of adsorption and desorption of air and target gas. Generally, sensitivity and % sensor response of the gas sensor for n-type material are defined by the relations given in equations (8) and (9) respectively [24]. Also, sensitivity is the change in the resistance of the film corresponding to the change in time and % sensor response is the percentage ratio of variation in resistance of the film after interaction with analyte gas to the resistance of the film in the air.

$$\begin{eqnarray}&&S=\displaystyle \frac{{R}_{g}}{{R}_{a}}\,\end{eqnarray} \tag{ 8 }$$

$$\begin{eqnarray} &&\% SR=\displaystyle \frac{{R}_{g}-{R}_{a}}{{R}_{a}}\times 100\end{eqnarray} \tag{ 9 }$$

Here, R_a is stabilized resistance in air and R_g is sensor resistance after LPG injection in the gas chamber. Also, when the sensor reaches 90% of the maximum resistance of the film during adsorption of LPG, the time taken is called response time. Similarly, recovery time is the time required to come back 90% of its initial value during the desorption [14].

To study of the gas sensing properties of ZnO/SnO₂, the LPG concentration (in vol%) was varied, and correspondingly different sensing curves were plotted and shown in figure 10. The curves represent the sensing characteristics for 0.5, 1.0, 1.5, and 2.0 vol% of LPG. Figure 9(a) shows the change in resistance of the film with the exposure time for different concentrations of LPG. From figure 9(b), it can be observed that % sensor response increases almost linearly with the increasing concentration of LPG which is good for an efficient gas sensor, and the maximum value was obtained as 276.51 for 2.0 vol% LPG. Figure 9(c) exhibits the sensitivity for different concentrations of LPG and the maximum sensitivity was found as 3.78 correspondings to 2.0 vol% LPG. The minimum response time was found for 0.5 vol% of LPG and the values of which are 10 s and 15 s respectively. The sensing characteristics of the film were repeated after two months of fabrication, a minute change (± 4%) was observed indicating the stability and reliability of the sensor.

Zoom In Zoom Out Reset image size

The repeatability curve for ZnO/SnO₂ film was observed by alternatively exposing 0.5 vol% LPG and air for three consecutive cycles at room temperature is shown in figure 9(e). The sensing response curve shows an almost similar response (R_a to R_g and again R_g to R_a) repeated to three consecutive cycles which display good repeatability. The repeatability of the reported LPG sensor after 3 weeks showed 97.59% repeatably. Selectivity is one of the most imperative and challenging parameters for gas sensors and sensor response towards a specific gas needs to be noticeably higher than those of other gases for selective gas detection. To study the selective behaviour of the ZnO/SnO₂ heterostructure operating at room temperature, the gas sensor response towards LPG, ethanol and acetone with concentration of 0.5 vol% each were measured. The corresponding results are shown in figure 9(f). The ZnO/SnO₂ heterostructure exhibits a higher % sensor response to LPG (∼138.51), whereas it shows a considerably lower sensor response to CO₂ (∼79.17), acetone (∼56.18), and ethanol (∼28.56). In order to quantify the selectivity to LPG, the selectivity coefficient (K) of LPG to another gas is defined by equation (10).

$$\begin{eqnarray}&&{\rm{K}}=\displaystyle \frac{{S}_{LPG}}{{S}_{gas}}\end{eqnarray} \tag{ 10 }$$

Here S_LPG and S_gas are the sensor response of LPG and another gas respectively. The selective coefficient for the ZnO/SnO₂ heterostructure was 3.54 towards CO₂, 8.56 towards acetone and 16.18 towards ethanol. Higher K values imply more selective detection of LPG in presence of other gases, e.g. K value of 16.18 for ethanol indicates that the sensor responses 16.18 times higher for LPG than that for ethanol. Thus the experimental results indicate that the ZnO/SnO₂ heterostructure-based sensor has good selectivity towards LPG. The required sensor parameters for different concentration of LPG has been tabulated and shown in table 2.

The schematic of LPG sensing using ZnO/SnO₂ heterostructure film is shown in figure 10. LPG is a reducing gas that donates an electron to n-type metal oxide semiconductor. Here ZnO is n-type semiconductor and also SnO₂ n-type semiconductor with a wide optical bandgap. ZnO is coupled with SnO₂ grains to form a n-n heterojunction. In this junction, electron transfer occurs from a semiconductor with low work-function (SnO₂) to the other with high work-function (ZnO) until Fermi level equalizes. This creates an electron-depleted layer at the interface of ZnO and SnO₂ which bends the energy band. The enhanced sensing performance of ZnO/SnO₂ heterostructure is attributed to the combined effect of the formation of a depleted layer at the surface of individual ZnO and SnO₂ as well as at the formation of hetero-junction between ZnO and SnO₂. The electron density in the conduction channel of SnO₂ is more in comparison to that of ZnO. After contact with each other, the electrons from the conduction channel of SnO₂ move towards the conduction channel of ZnO forming the depletion region. This depleted region works as a barrier and is called the potential barrier. The energy band diagram is shown in figure 10. The LPG sensing using ZnO/SnO₂ heterostructure thin film takes place in two stages i.e. oxidation and reduction [24, 25]. In the first stage, air oxygen oxidizes the ZnO/SnO₂ heterostructure sensor surface by capturing conduction electrons. This gives rise to depletion region formation beneath the sensor surface and corresponding conduction channel narrows down. This depletion width depends on the number of air oxygen molecule adsorption and the number of conduction electrons available in the ZnO/SnO₂ heterostructure at that temperature. This depleted ZnO/SnO₂ heterostructure film offers additional resistance besides grain boundary resistance due to the increase in electric Schottky barrier potential (qV_b) in the presence of air. This adsorption of oxygen continues until the equilibrium is reached between the interfaces due to the interactions of oxygen molecules with the chemisorption sites at that particular temperature. This progress of adsorbed oxygen species can be described by equations (11)–(13) as follows:

$$\begin{eqnarray}&&{{\rm{O}}}_{2}\left({\rm{gas}}\right)\to {{\rm{O}}}_{2}\left({\rm{ads}}\right)\end{eqnarray} \tag{ 11 }$$

$$\begin{eqnarray}&&{{\rm{O}}}_{2}\left({\rm{ads}}\right)+{{\rm{e}}}^{-}\to {{{\rm{O}}}_{2}}^{-}\left({\rm{ads}}\right)\end{eqnarray} \tag{ 12 }$$

$$\begin{eqnarray}&&{{{\rm{O}}}_{2}}^{-}\left({\rm{ads}}\right)+{{\rm{e}}}^{-}\to 2{{\rm{O}}}^{-}\left({\rm{ads}}\right)\end{eqnarray} \tag{ 13 }$$

Zoom In Zoom Out Reset image size

The negatively charged species present on the surface of the material play a crucial role in detecting the LPG. However, the reaction mechanism of LPG is complex and still not fully explained. The main component in the LPG is propane (C₃H₈) and butane (C₄H₁₀). There is an exchange of electrons between the LPG molecule and oxygen species adsorbed on chemisorption centres as shown by the reaction given in equations (14)–(15).

$$\begin{eqnarray}&&{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2{\rm{n}}+2}+2{{\rm{O}}}^{-}\to {{\rm{H}}}_{2}{\rm{O}}+{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2{\rm{n}}}:{\rm{O}}+{\rm{e}}-\end{eqnarray} \tag{ 14 }$$

$$\begin{eqnarray}&&{{\rm{C}}}_{{\rm{n}}}{{\rm{H}}}_{2{\rm{n}}}:O+{{\rm{O}}}^{-}\to {{\rm{H}}}_{2}{\rm{O}}+{{\rm{CO}}}_{2}+{\rm{e}}-\end{eqnarray} \tag{ 15 }$$

Here, C_nH_2n+2 represents the various hydrocarbon [14]. In the second stage, LPG is exposed to ZnO/SnO₂ heterostructure. The LPG donates the electron after interaction with oxygen species and these ejected electrons recombine to form the electron-hole pair. In the case of n-type semiconductor, the resistance drastically increases in the beginning due to the rapid adsorption of hydrocarbon, afterwards, it increases slowly and finally gets saturated. When the flow of the LPG is stopped for the recovery characteristics, the oxygen molecules in the air will be adsorbed on the surface of the film. The depletion width of ZnO/SnO₂ heterostructure sensor decreases by gaining electrons from oxygen ion species. Hence the corresponding channel width increases. Consequently, the resistance of the sensor decreases.