Gas Sensor Based on ZnO Nanostructured Film for the Detection of Ethanol Vapor

: In this paper, the ZnO<La> target was synthesized by the solid-state reaction method and a nanostructured thin ﬁlm was deposited by the RF (radio frequency) magnetron sputtering method on a Multi-Sensor-Platform. The obtained ZnO<La> nanostructured ﬁlm was investigated as the sensing material. Energy-Dispersive X-ray (EDX) analysis indicated the existence of La in the synthesized ZnO<La> material. Scanning Electron Microscope (SEM) images of the ﬁlm showed the grain sizes in the range of 20–40 nm. Sensor performance characteristics such as a dynamic response, response and recovery times, and ethanol detection range were investigated at 50–300 ◦ C. A sensitivity was observed at extremely low concentrations of ethanol (0.7 ppm). The minimum response and recovery times of the sensor corresponding to 675 ppm ethanol vapor concentration at 250 ◦ C were found to be 14 s and 61 s, respectively. The sensor showed a high response, good selectivity, fast response/recovery behavior, excellent repeatability toward ethanol vapor, and low sensitivity toward humidity. These characteristics enable the use of a ZnO<La> based sensor for ethanol detecting applications.


Introduction
Today, the detection of ethanol vapors (C 2 H 5 OH) and the accurate measurement of their low concentrations are the focus of researchers, as ethanol is considered the most widely used organic compound in areas such as fuel processing, agronomic, biomedical, food, pharmaceutical, hygienic, cosmetic, and chemical industries. In particular, it is known that the composition of chemical compounds emitted from food can be used to analyze its quality and freshness. Among the chemicals emitted from the food, ethanol is considered the most important substance and the detection of its concentration is widely used in modern electronic noses to continuously monitor the quality and freshness of stored groceries and meat. Besides, ethanol is known as a rather toxic and hazardous gas, which can greatly affect human health. It stands out some harmful effects such as vomiting, nausea, skin irritations, low blood pressure, low blood sugar level, etc. Excessive use of alcoholic beverages also leads to drunk driving accidents. Therefore, the rapid detection of ethanol is inevitably required and the ethanol gas sensing materials have been developed [1][2][3][4][5][6].
Choosing the right material with advanced physical and chemical properties for the design and production of nanoscale devices is one of the most important issues. As one of the most widely used and studied materials, it is essential to develop the ZnO-based nanostructured material as the building blocks for future nanoscale devices. ZnO crystals stand out for their excellent thermal, piezoelectric, optical, electrical as well as semiconductor properties such as a high isoelectric point, tunable resistance, controllable charge-transfer

Synthesis of ZnO<La> Material
The sensor manufacturing process started with purchasing materials. To obtain the ZnO<La> sputtering target, ZnO (99.9%) and La2O3 (99.9%) nanopowders were collected from Alfa Aesar (Figure 1a). A ZnO:La2O3 ceramic target was prepared by the solid-state reaction method as a magnetron sputtering target. Firstly, the required quantities of the corresponding metal oxide powders (ZnO + 1 at.% La2O3) were weighed and well mixed for 10 h. Then, the mixture was heat-treated at 900 °C for 8 h, whereas the initial heat treatment of mixed powders removed the wetness of the metal oxide raw materials, which led to better mixing and grinding of powders. After mixing for 25 h, when the mixture was completely homogeneous, it was pressed (with 2000 N/cm −2 pressure) into a cylindrical tablet with a diameter of 60 mm. The pressed ZnO:La2O3 based tablet was kept in the programmable furnace Nabertherm, HT O4/16 (with the controller of C 42) in the temperature range of 900-1450 °C for 25 h. In the final stage, after mechanical treatment for removing surface defects, a 2 mm thick cylindrical target with a diameter of 50 mm was obtained for magnetron sputtering (Figure 1b).

Gas Sensor Fabrication
The gas sensing layer was deposited on the Multi-Sensor-Platform by the RF magnetron sputtering method using synthesized ZnO<La> target by the VTC-600-2HD DC/RF Dual-Head High Vacuum Magnetron Plasma System ( Figure 2). The Multi-Sensor-Platforms were acquired from Tesla Blatna (Blatná, Czech Republic). The sensor substrate was designed for heating, temperature measurement, and resistance recording. Using heat resistance, the chip can be kept at a constant temperature, or a temperature cycle can be operated. The sensor platform consists of a temperature sensor (Pt 1000), a heater, and interdigitated electrode structures (IDES) in a platinum thin film on a ceramic substrate. The platinum heater and temperature sensor were covered with a passive glass layer (Figure 3). The gas sensitive ZnO<La> layer was deposited onto the non-passivated electrode structure, so the Multi-Sensor-Platform was converted into a gas sensor. The fabrication steps of the sensor are presented in Figure 4.

Gas Sensor Fabrication
The gas sensing layer was deposited on the Multi-Sensor-Platform by the RF magnetron sputtering method using synthesized ZnO<La> target by the VTC-600-2HD DC/RF Dual-Head High Vacuum Magnetron Plasma System ( Figure 2). The Multi-Sensor-Platforms were acquired from Tesla Blatna (Blatná, Czech Republic). The sensor substrate was designed for heating, temperature measurement, and resistance recording. Using heat resistance, the chip can be kept at a constant temperature, or a temperature cycle can be operated. The sensor platform consists of a temperature sensor (Pt 1000), a heater, and interdigitated electrode structures (IDES) in a platinum thin film on a ceramic substrate. The platinum heater and temperature sensor were covered with a passive glass layer ( Figure 3). The gas sensitive ZnO<La> layer was deposited onto the non-passivated electrode structure, so the Multi-Sensor-Platform was converted into a gas sensor. The fabrication steps of the sensor are presented in Figure 4.
The magnetron sputtering process was carried out under the following technological regimes: 60 W input power, 200 • C deposition temperatures, 70 mm distance between the target and substrate, 2 × 10 −3 Pa base pressure, 6 × 10 −1 Pa deposition pressure, and 20 min duration of sputtering. Then, the ZnO<La> based sensor was sensitized by the deposition of palladium (99.9%) catalytic particles on the surface of the sensing layer using the ion-beam sputtering method (the cathode current and the anode voltage were equal to 65 A and 25 V, respectively). In the final stage of the sensor production, it was annealed for 4 h at 300 • C for the stabilization of gas sensing parameters and the improvement of the long-term stability.

Gas Sensing Test
Using the laboratory-designed (computer-controlled) gas testing system [22], the gas sensing behavior of the ZnO<La> sensor were studied in the air and in the presence of ethanol vapors. To get a required concentration of C 2 H 5 OH in the 2 L chamber, the aqueous solution of ethanol was injected into the measuring chamber on a specially designed heated container for the rapid conversion of the ethanol in the liquid phase to the gas phase. The real concentrations in ppm unit of ethanol, acetone, water, toluene, dimethylformamide (DMF), and ammonia vapors were determined by the method described in [23]. During the measurement of the resistance change of the sensing layer 5 V DC voltage bias was applied across the sample. For the heating of the active surface of the sensor in the range of 25-300 • C, voltage biases in the range of 0-5 V were applied to the platinum heater. As a quantitative characteristic of the sensitivity, the response of the fabricated sensor is defined, as the resistance ratio of the sensitive film (R = R air /R gas ), measured in air (R air ) and in the presence of ethanol gas (R gas ).

Characterization
The thickness of the interdigitated electrode of the Multi-Sensor-Platform was measured by the Alpha-Step D-300 (KLA Tencor, Milpitas, CA, USA) profiler. The measured result of the electrode-substrate transition profiles is shown in Figure 5a. The thickness of the electrodes varies in the range of 800-900 nm, and the width of the gaps between the electrodes is about 15 microns.
It is known that the performance of resistive sensors is greatly influenced by the thickness value of the sensing film. The optimal thickness of the sensitive film can be decisive in the fabrication of a gas sensor with enhanced parameters. Sensors based on the thick film (>300 nm) sometimes show insufficient speed and sensitivity due to their high ratio of the film thickness and the electron depletion layer (they have an unaffected bulk area), while sensors with extremely thin films (<40 nm) demonstrate the instability of parameters over time. This is mainly due to the creation of crystalline cracks in a thin layer over time and the formation of agglomerations of nanoparticles. The choice of the optimal value of the thickness is decisive for improving the performance parameters of the sensor [24][25][26][27][28]. The thickness of the obtained ZnO<La> thin film was also measured by the Alpha-Step D-300 (KLA Tencor) profiler and it was equal to 72 nm corresponding to 60 W input power and 20 min duration of sputtering process (Figure 5b). SEM images were obtained using secondary and back-scattered electron signals (SE and BSE, respectively) through a VEGA TS 5130MM (with microanalysis system of INCA Energy 300) instrument (Figure 6a,b). The weight and atomic percent of different elements in the given ZnO<La> material were determined by EDX analysis and shown in Figure 6c. It is confirmed that La is almost homogeneously dispersed in the synthesized ZnO material expressed in the form of white dots (Figure 6b,d). The presence of Zn, La, and O peaks are clearly visible in the quantitative analysis of the EDX spectrum, which indicates the existence of La in the synthesized ZnO<La> material. The actual concentration of La (0.84 at.%) in the ZnO<La> target material is slightly different from the initial calculated concentration (1 at.%), which is assumed due to the certain losses of the material during the mixing and milling of the powders and the error of the measuring instruments.
The morphological structure of the deposited ZnO<La> film was analyzed by scanning electron microscopy using the VEGA TS 5130MM instrument (Figure 7a,b). It is obvious that the film has a granular structure with high homogeneity. Gas sensing parameters are largely determined by the sizes of the grains in the sensitive film, while their dimensions can be actually controlled by changing the power of the magnetron generator. Typically, at a higher power of the generator, argon ions strike the target with higher kinetic energy, which results in a larger size of grains released from the target. At a lower power of the generator, the kinetic energy of the bombarding ions decreases, and consequently, the size of the particles released from the target decreases, while excessive reduction of the power can stop the deposition process. Therefore, it is obvious from the SEM images that the average grain sizes in the film are in the range of 20-40 nm corresponding to the power of 60 W. SEM images were obtained using secondary and back-scattered electron signals (SE and BSE, respectively) through a VEGA TS 5130MM (with microanalysis system of INCA Energy 300) instrument (Figure 6a,b). The weight and atomic percent of different elements in the given ZnO<La> material were determined by EDX analysis and shown in Figure 6c. It is confirmed that La is almost homogeneously dispersed in the synthesized ZnO material expressed in the form of white dots (Figure 6b,d). The presence of Zn, La, and O peaks are clearly visible in the quantitative analysis of the EDX spectrum, which indicates the existence of La in the synthesized ZnO<La> material. The actual concentration of La (0.84 at.%) in the ZnO<La> target material is slightly different from the initial calculated concentration (1 at.%), which is assumed due to the certain losses of the material during the mixing and milling of the powders and the error of the measuring instruments. The morphological structure of the deposited ZnO<La> film was analyzed by scanning electron microscopy using the VEGA TS 5130MM instrument (Figure 7a,b). It is obvious that the film has a granular structure with high homogeneity. Gas sensing parameters are largely determined by the sizes of the grains in the sensitive film, while their dimensions can be actually controlled by changing the power of the magnetron generator. Typically, at a higher power of the generator, argon ions strike the target with higher kinetic energy, which results in a larger size of grains released from the target. At a lower power of the generator, the kinetic energy of the bombarding ions decreases, and consequently, the size of the particles released from the target decreases, while excessive reduction of the power can stop the deposition process. Therefore, it is obvious from the SEM images that the average grain sizes in the film are in the range of 20-40 nm corresponding to the power of 60 W.
The EDX elementary analysis of the sputtered ZnO<La> based thin film is also well examined by the VEGA TS 5130MM instrument (Figure 7c,d). Figure 7d shows that there are clearly depicted peaks of Zn, Al, La, and O, where the presence of Al is conditioned by the Al2O3-based substrate. In order to separate the values of the concentrations of the elements contained in the film from the elements of the substrate, the obtained results were processed by the StrataJem program. The actual concentration of La in the ZnO<La> based film is 0.81 at.%, which is not significantly different from its concentration in the ZnO<La> based target material (0.84 at.%). Elemental mapping analysis of the ZnO<La> film was carried out by EDX spectroscopy. The SEM image, the corresponding results of elemental mapping analysis of the ZnO<La> film, and the X-ray intensity of characteristic lines of different elements are presented in Figure 8. It is obvious that the distribution of elements Zn, Al, O, and La is even, differed from each other by colors in the mapping.
The ZnO<La> film was analyzed by FTIR spectra by NEXUS FT-IR (Nicolet) on two different substrates with different optical transmission ranges (Substrate 1: 1100-230 cm −1 and Substrate 2: 4000-2300 cm −1 ). The strong intensity peak at around 400 cm −1 shows the presence of Zn-O [29] and the almost imperceptible peak in the wavenumber range of 500-600 cm −1 is supposed to be attributed to lanthanum [30], which is due to its very low concentration in ZnO material (Figure 9a). FTIR study was also carried out in the range of higher wavenumbers (4000-2300 cm −1 ) to detect hydroxide groups on the film surface. The peaks corresponding to the pronounced hydroxide states (O-H) are mainly in the range of 3200-3600 cm −1 [31]. As shown in Figure 9b, there are no strong intensity peaks in the indicated range, which confirms the absence or insignificant amount of hydroxide groups in the film. Elemental mapping analysis of the ZnO<La> film was carried out by EDX spectroscopy. The SEM image, the corresponding results of elemental mapping analysis of the ZnO<La> film, and the X-ray intensity of characteristic lines of different elements are presented in Figure 8. It is obvious that the distribution of elements Zn, Al, O, and La is even, differed from each other by colors in the mapping.   Elemental mapping analysis of the ZnO<La> film was carried out by EDX spectroscopy. The SEM image, the corresponding results of elemental mapping analysis of the ZnO<La> film, and the X-ray intensity of characteristic lines of different elements are presented in Figure 8. It is obvious that the distribution of elements Zn, Al, O, and La is even, differed from each other by colors in the mapping.  500-600 cm −1 is supposed to be attributed to lanthanum [30], which is due to its very low concentration in ZnO material (Figure 9a). FTIR study was also carried out in the range of higher wavenumbers (4000-2300 cm −1 ) to detect hydroxide groups on the film surface. The peaks corresponding to the pronounced hydroxide states (O-H) are mainly in the range of 3200-3600 cm −1 [31]. As shown in Figure 9b, there are no strong intensity peaks in the indicated range, which confirms the absence or insignificant amount of hydroxide groups in the film.

Gas Sensing Studies
To study the gas sensing properties and to select the optimal operating temperature of the ZnO<La> sensor, the dependence of the sensor resistance on the temperature in the air was removed (Figure 10a), and the response values of the sensor were measured in the

Gas Sensing Studies
To study the gas sensing properties and to select the optimal operating temperature of the ZnO<La> sensor, the dependence of the sensor resistance on the temperature in the air was removed (Figure 10a), and the response values of the sensor were measured in the temperature range of 50-300 • C toward 675 ppm of ethanol vapor (Figure 10b). As the temperature increased, the resistance of the sensor decreased due to the generation of free carriers in the semiconductor, and a further decrease in temperature was accompanied by the recovery of the resistance almost to its initial value. Despite the increase in temperature, the resistance of the sensor hardly changes in the temperature range of 160-270 • C. It is assumed that in this temperature range the rate of free charge carriers generation in the semiconductor due to temperature rise is almost equal to the rate of adsorption of oxygen species on the active surface. It is revealed from Figure 10b that the response first increased with the temperature up to 250 • C and reduced in higher temperatures. The sensor showed an imperceptible response toward ethanol vapor at 50 • C and farther approached more than 117 at 250 • C. At a higher temperature (300 • C), a sharp decrease in response to the value of 22 was recorded. Therefore, the temperature of 250 • C was selected as the optimum temperature to investigate the gas sensing characteristics of the ZnO<La> sensor. It is assumed that at temperatures smaller than 250 • C, the rate of ethanol adsorption and chemical reaction on the active surface is slowed down because at these temperatures the activation energies are not sufficient for kinetic processes to take place completely. At higher temperatures (>250 • C), the rate of desorption of gas molecules exceeds that of adsorption, which leads to a decrease in sensitivity. Gas sensing parameters of the ZnO<La> sensor at different operating temperatures in the presence of 675 ppm of ethanol vapor are summarized in Table 1.
ture, the resistance of the sensor hardly changes in the temperature range of 160-270 °C. It is assumed that in this temperature range the rate of free charge carriers generation in the semiconductor due to temperature rise is almost equal to the rate of adsorption of oxygen species on the active surface. It is revealed from Figure 10b that the response first increased with the temperature up to 250 °C and reduced in higher temperatures. The sensor showed an imperceptible response toward ethanol vapor at 50 °C and farther approached more than 117 at 250 °C. At a higher temperature (300 °C), a sharp decrease in response to the value of 22 was recorded. Therefore, the temperature of 250 °C was selected as the optimum temperature to investigate the gas sensing characteristics of the ZnO<La> sensor. It is assumed that at temperatures smaller than 250 °C, the rate of ethanol adsorption and chemical reaction on the active surface is slowed down because at these temperatures the activation energies are not sufficient for kinetic processes to take place completely. At higher temperatures (>250 °C), the rate of desorption of gas molecules exceeds that of adsorption, which leads to a decrease in sensitivity. Gas sensing parameters of the ZnO<La> sensor at different operating temperatures in the presence of 675 ppm of ethanol vapor are summarized in Table 1.   Metal oxide based sensors mainly operate in three temperature ranges. At higher temperatures (400-700 • C), kinetic processes on the semiconductor surface take place rapidly enough that equilibrium is immediately established between the bulk oxide and the environment. Due to the reduced diffusion coefficient of oxygen species at medium temperatures (200-400 • C), the chemical composition of metal oxide does not reach equilibrium and the kinetic processes take place mainly on the semiconductor surface. For sensors operating in this temperature range, the use of thin films is preferable, as the main processes are performed mostly on the active surface. The ZnO<La> sensor with a thickness of 72 nm confirms this claim, as it works more efficiently at 250 • C. At lower operating temperatures  • C), the rate of ethanol adsorption and chemical reactions onto the sensor surface decrease, thus resulting to a reduction in sensor response.
The real-time responses of the ZnO<La> based sensor were measured at 250 • C operating temperatures toward different concentrations of ethanol vapor. Figure 10c shows the actual response-recovery curves of the sensor for ethanol vapor with concentrations varied from 0.7 to 34 ppm (at six different concentrations) at 250 • C. The typical response/recovery times were 14 and 60 s, respectively, confirming the fairly high speed of the sensor. A sensitivity was observed at extremely low concentrations of ethanol (0.7 ppm) and toward the higher concentration (34 ppm) the sensor response reached more than 19. The sensor response to the ethanol vapor increased linearly toward ethanol concentration varied from 0.7 to 34 ppm at 250 • C (Figure 10d). This will allow us to detect the ethanol vapor traces in real environments and accurately measure the low concentrations of this gas.
The repeatability of the sensor parameters is crucial for the long-term operation of the sensor, which is best characterized by a repeatability test of the response in real-time. Multiple measurements of the sensor response were performed toward 6.7 ppm ethanol vapor at 250 • C operating temperature to test the sensor drift over time. Figure 10e shows the repeatability test of the sensor performed for seven cycles of response and recovery curves, which represents almost the same response/recovery behavior and nearly matching response values confirming the fairly high repeatability.
The response and recovery times were estimated at different temperatures to compile the information on the sensor speed. The response and recovery times of the ZnO<La> sensor were calculated as the time is taken for rising and falling to 90% of the maximum response value after exposure to ethanol vapor and air, respectively. The response and recovery times decreased with the increase of the operating temperature approaching tens of seconds at higher temperatures (>200 • C) ( Table 1). In the temperature range of 100 to 250 • C, the recovery time was definitely longer than the response time, while their values were quite close at higher temperatures (300 • C). It is assumed that the rate of gas desorption at the semiconductor surface exceeds the rate of adsorption at higher temperatures, which leads to a decrease both in the recovery time and in the sensor response. The dynamic resistance curve of the sensor showed that the response and recovery times of the sensor were found to be 14 and 61 s, respectively, for 675 ppm of ethanol vapors at an operating temperature of 250 • C (Figure 11a). The high speed of the sensor confirms its suitability for real-time detection of ethanol vapors.
To check the selectivity behavior in sensing performance, the ZnO<La> based sensor was examined in the presence of other interfering environmental gases. The sensor responses to 675 ppm ethanol, 785 ppm acetone, 3600 ppm water vapor, 650 ppm toluene, 700 ppm DMF, and 750 ppm ammonia were investigated at the 250 • C (Figure 11b). As it is evident, the sensor showed the highest response value to ethanol gas than other control gases. The dynamic response curves of the sensor for the above-mentioned gases confirm the high selectivity of the sensor towards ethanol vapor and the faster response/recovery behavior compared to that of the other gases (Figure 11c). Moreover, as a quantitative representation of the selectivity, the response ratios of ethanol to other gases (S ethanol /S other gases ) were calculated, and the values were in the range of 7.7-49.7. It is noteworthy that the sensor showed an extremely low response toward humidity (S = 2.37 at 3600 ppm water vapor) allowing it to work in conditions of high relative environmental humidity.
gases. The dynamic response curves of the sensor for the above-mentioned gases confirm the high selectivity of the sensor towards ethanol vapor and the faster response/recovery behavior compared to that of the other gases (Figure 11c). Moreover, as a quantitative representation of the selectivity, the response ratios of ethanol to other gases (Sethanol/Sother gases) were calculated, and the values were in the range of 7.7-49.7. It is noteworthy that the sensor showed an extremely low response toward humidity (S = 2.37 at 3600 ppm water vapor) allowing it to work in conditions of high relative environmental humidity. The sensing performance of the ZnO<La> sensor has been compared with reports related to ZnO-based C2H5OH sensors published in recent years. The corresponding results are summarized in Table 2 including the most important characteristics such as sensor material type, operating temperature, ethanol concentration, and response value. It is obvious from the comparison that the performance of the proposed ZnO<La> sensor is The sensing performance of the ZnO<La> sensor has been compared with reports related to ZnO-based C 2 H 5 OH sensors published in recent years. The corresponding results are summarized in Table 2 including the most important characteristics such as sensor material type, operating temperature, ethanol concentration, and response value. It is obvious from the comparison that the performance of the proposed ZnO<La> sensor is clearly comparable to that of other published reports. The most noticeable in the comparison is the extremely low detection limits of our sensor (0.7 ppm) at a relatively low operating temperature attributed to the discernible response value of 1.5.  It is noteworthy that the sensor based on pure ZnO material showed a poor sensitivity to ethanol vapors justifying the introduction of La into the main ZnO material. Besides, it is assumed that the introduction of La into ZnO material imports additional energy states in between the valence band and conduction band of zinc oxide reducing its bandgap energy. This reduces the sensor resistance by tuning it to a measurable range, while at the same time positively contributing to the gas sensing processes [30].

Gas Sensing Mechanism
The gas sensing mechanism for conductometric sensors based on semiconducting oxides works due to the adsorption and chemical reactions of target gas molecules on the active surface, as a result of which the sensor resistance changes. The chemisorbed oxygen species play a major role in the gas sensing mechanism. Thus, when the surface of the ZnO<La> layer contacts with air, oxygen molecules with sufficient activation energy adsorb on the sensor surface by trapping electrons from the conduction band of ZnO and form chemisorbed oxygen species (O − (ads), O 2− (ads) and O 2 − (ads)), leading to the formation of the electron depletion layer. The supposed reactions can be described as follows [40][41][42]: The state of adsorbed oxygen species on the semiconductor surface is highly dependent on operating temperature. In a low temperature range (<200 • C), O 2− states predominate on the active surface, whereas in the range of 200 to 400 • C, the number of O − states prevails. At higher operating temperatures (>400 • C), O 2− species begin to play a major role in the kinetic processes. For the ZnO<La> sensor, 250 • C was selected as the operating temperature, therefore the Ospecies predominantly contributed to the sensing behavior. The oxygen chemisorption results in a modification of the space charge region toward depletion. The resistance corresponding to this state is considered the base resistance. For n-type semiconductors, the majority of charge carriers are electrons and upon interaction with a reducing gas, an increase in conductivity occurs. When the sensors are exposed to alcohol vapor, the ethanol molecules react with surface oxygen species and produce free electrons, resulting in an increase in electronic conductance. The supposed reactions can be described as follows [43][44][45]: 6O − (adsorbed) + C 2 H 5 OH (gas) → 3H 2 O (gas) + 2CO 2 (gas) + 6e − , 3O 2 − (adsorbed) + C 2 H 5 OH (gas) → 3H 2 O (gas) + 2CO 2 (gas) + 3e − , 6O 2− (adsorbed) + C 2 H 5 OH (gas) → 3H 2 O (gas) + 2CO 2 (gas) + 12e − .
Ethanol sensing mechanisms on the ZnO-based material were investigated by nearambient-pressure X-ray photoelectron spectroscopy (NAP-XPS) [46], which showed the dehydrogenation of ethanol molecules, leading to ethoxy and acetaldehyde, where the detached hydrogens interact with the chemisorbed oxygen. These experimental results can be attributed to the support of the above-mentioned reactions.
If the sensing layer is fabricated from the metal oxide in nanoparticles form, the system is described by structural and band models of conductive mechanism. Without ethanol vapor, a high potential barrier (Schottky barrier) is formed between the adjacent nanograins leading to an increase in the film resistance. In the presence of ethanol molecules, as a result of the Reactions (5)- (7), the electrons are returned to the semiconductor lattice, which reduces the depletion region width, so the Schottky barriers are also lowered. A variation of the sensor resistance or electrical conductivity takes place as a result of the barrier height modulation [35]. The proposed schematic description of the C 2 H 5 OH sensing mechanism of the ZnO<La> material is shown in Figure 12.
Ethanol sensing mechanisms on the ZnO-based material were investigated by nearambient-pressure X-ray photoelectron spectroscopy (NAP-XPS) [46], which showed the dehydrogenation of ethanol molecules, leading to ethoxy and acetaldehyde, where the detached hydrogens interact with the chemisorbed oxygen. These experimental results can be attributed to the support of the above-mentioned reactions.
If the sensing layer is fabricated from the metal oxide in nanoparticles form, the system is described by structural and band models of conductive mechanism. Without ethanol vapor, a high potential barrier (Schottky barrier) is formed between the adjacent nanograins leading to an increase in the film resistance. In the presence of ethanol molecules, as a result of the Reactions (5)- (7), the electrons are returned to the semiconductor lattice, which reduces the depletion region width, so the Schottky barriers are also lowered. A variation of the sensor resistance or electrical conductivity takes place as a result of the barrier height modulation [35]. The proposed schematic description of the C2H5OH sensing mechanism of the ZnO<La> material is shown in Figure 12.  Pd compounds and Pd-based nanoparticles are mainly used as good catalysts for various gases [47]. Mainly, noble Pd nanoparticles form spillover zones around themselves on the semiconductor surface and promote O 2 dissociation leading to an increase in the number of chemisorbed oxygen species. The increased quantity of chemisorbed oxygen species and their reaction with the target gas molecules improve the sensing behavior. The sensitizing effect of palladium nanoparticles with the interaction of ethanol gas has been observed [32].
The high sensitivity for a metal oxide-based gas sensor is attributed to the grain size of the sensing film, which is comparable to, or smaller than, the twice Debye length (d ≤ 2L D ), while the L D is in the range of 20-22 nm for ZnO at 250-300 • C [48]. The grain sizes in the ZnO<La> film are in the range of 20-40 nm, which definitely proves the high sensing behavior of our sensor.

Conclusions
In summary, the ZnO<La> sputtering target was prepared. A gas sensor based on the ZnO<La> thin film was fabricated by the RF magnetron sputtering method. The fabricated sensor device was investigated for ethanol sensing applications. The analysis of the EDX spectrum of the ZnO<La> material confirms the presence of Zn, La, and O peaks. The SEM images of the ZnO<La> based thin film (72 nm) prove its granular structure (nanoparticles with size: 20-40 nm) with a high homogeneity corresponding to 60 W of power generated and 20 s of deposition time. The elemental mapping analysis of the ZnO<La> film confirms that the elements Zn, Al, O, and La are uniformly distributed. Reasonable response values of 117 and 1.5 were measured at 250 • C toward 675 and 0.7 ppm ethanol concentration, respectively. The results indicated that the ZnO<La> based sensor has enhanced gas sensing properties, including a high response, superb selectivity, and a fast response/recovery time towards ethanol. The sensor also exhibits a dynamic transient response and repeatability for ethanol.

Data Availability Statement:
The data presented in this study are available on request to the corresponding author.