The Role of Zn Ions in the Structural, Surface, and Gas-Sensing Properties of SnO2:Zn Nanocrystals Synthesized via a Microwave-Assisted Route

Although semiconducting metal oxide (SMOx) nanoparticles (NPs) have attracted attention as sensing materials, the methodologies available to synthesize them with desirable properties are quite limited and/or often require relatively high energy consumption. Thus, we report herein the processing of Zn-doped SnO2 NPs via a microwave-assisted nonaqueous route at a relatively low temperature (160 °C) and with a short treatment time (20 min). In addition, the effects of adding Zn in the structural, electronic, and gas-sensing properties of SnO2 NPs were investigated. X-ray diffraction and high-resolution transmission electron microscopy analyses revealed the single-phase of rutile SnO2, with an average crystal size of 7 nm. X-ray absorption near edge spectroscopy measurements revealed the homogenous incorporation of Zn ions into the SnO2 network. Gas sensing tests showed that Zn-doped SnO2 NPs were highly sensitive to sub-ppm levels of NO2 gas at 150 °C, with good recovery and stability even under ambient moisture. We observed an increase in the response of the Zn-doped sample of up to 100 times compared to the pristine one. This enhancement in the gas-sensing performance was linked to the Zn ions that provided more surface oxygen defects acting as active sites for the NO2 adsorption on the sensing material.


Introduction
Climate change has been the main topic covered in scientific forums to find groundbreaking strategies to minimize the influence of human activities on global environmental problems.Industrial activities and the burning of fossil fuels have been primarily responsible for the release of various gases into the atmosphere [1][2][3][4].Some of these gaseous species, in addition to contributing to the greenhouse effect, are also harmful to human health, such as NO 2 , O 3 , and CO [2,5,6].Unfortunately, people who live and/or work near these pollutants are more likely to contract certain diseases, such as respiratory and cardiovascular ones, and even cancer.Recently, the World Health Organization (WHO) reported that atmospheric pollution has been responsible for the death of approximately 7 million people worldwide per year [2,4,7].
Nitrogen dioxide (NO 2 ) is a toxic gas, and exposure to it can cause inflammation of the airways, asthma, and other respiratory sicknesses (e.g., coughing and difficulty in breathing) [6,8,9].The largest sources of NO 2 gas emissions are mainly linked to the Regarding the synthesis method, nonaqueous routes have attracted attention in terms of obtaining materials with diverse functional properties in a controlled and reproductive way [44,45].The microwave-assisted nonaqueous synthesis of organic solvents under the exclusion of water is able to overcome some of the major limitations of aqueous systems.The main advantages of this route are the simplicity, short synthesis times (few minutes), and relatively low temperatures (<200 • C) [45,46], mainly when compared to the data in Table 1.In the gas-sensing area, we have previously reported the potential of the microwave-assisted route to obtain ZnO/SnO 2 nanoheterostructures applied as ozone gas sensors.The experiments demonstrated that heterostructures were promising for detecting ozone gas under UV-light stimulation [47].
Motivated by these considerations, we conducted a detailed investigation of the effects of Zn ions on the structural, surface, and NO 2 gas-sensing properties of SnO 2 NPs synthesized via the microwave-assisted nonaqueous method.The properties of Zn-doped SnO 2 NPs were studied by X-ray diffraction (XRD), X-ray absorption near-edge structure spectroscopy (XANES), high-resolution transmission electron microscopy (HRTEM), and X-ray photoelectron spectroscopy (XPS).The gas-sensing experiments demonstrated good sensing activity at a mild temperature (150 • C), which could further be improved by increasing the Zn content.In addition, the Zn-doped nanocrystals revealed high stability of their NO 2 sensing performance under different humidity levels.

Synthesis Procedure
Pristine SnO 2 and Zn-doped SnO 2 NPs (Zn/(Zn + Sn) = 5 and 30 mol %) were synthesized using a microwave-assisted nonaqueous method in a CEM Discover reactor operating at a frequency of 2.45 GHz.The synthesis of the Zn-doped SnO 2 samples was performed by adding 12.3 mmol of tin tetrachloride and zinc acetylacetonate to 10 mL of toluene and then 30 mL of benzyl alcohol was added to the solution.Concerning the undoped SnO 2 sample, the procedure was similar except for the addition of a zinc precursor.Thereafter, the reaction mixture was transferred into a 35 mL glass tube and sealed with a Teflon cap.The microwave-assisted treatment was performed for 20 min at 160 • C, maintaining the reaction mixture in continuous stirring.The precipitate was separated from the liquid phase by centrifugation, washed three times with ethanol, and then dried overnight at 60 • C. The Zn-doped samples were labeled as SnZn1 (5 mol % Zn) and SnZn2 (30 mol % Zn).

Characterization Techniques
X-ray diffraction (XRD) patterns were measured using a Shimadzu XRD 6100 (Shimadzu Corporation, Kyoto, Japan) diffractometer with a monochromatic Cu Kα source, collected at room temperature, in a continuous scan mode with a speed of 2 • min −1 and a step of 0.02 • .The morphological properties were analyzed using a field emission scanning electron microscope (FE-SEM, Supra35, ZEISS Company, Jena, Germany) and a high-resolution transmission electron microscope (HRTEM, TECNAI G2 F30, FEI Company, Hillsboro, OR, USA) operating at 200 kV.The mean crystal size was estimated from the analysis of the TEM images through the measurement of approximately 100 NPs.X-ray absorption near-edge spectroscopy (XANES) experiments were carried out at the Brazilian Synchrotron Light Laboratory.The experiments were carried out at the Zn Kand Sn-L3 edges at room temperature using a flat Si(111) double crystal monochromator.The processing and analysis of XANES spectra were performed using the MAX software package [48].XPS analysis was performed on a Thermo Scientific K-Alpha spectrometer using a monochromatic Al Kα source.The as-obtained data were analyzed using Casa XPS software 2.3.25 (Casa Software Ltd., Teignmouth, Devon, UK), and the spectra were calibrated using the C 1s line (284.8eV) of the adsorbed carbon on the sample surface.

Gas-Sensing Experiments
The detailed procedure of the preparation of sensing devices is described in the Supporting Information.The gas-sensing experiments were carried out in a dynamic chamber that allowed us to control the operating temperature and the gas concentrations by using a set of mass flow controllers.Our workbench allowed exposure of the samples to distinct NO 2 levels, ranging from 0.1 to 2.0 ppm.The total gas flow was kept equal to 500 SCCM.The applied DC voltage was kept constant (1 V), while the electrical resistance was monitored using a sourcemeter (model 2540, Keithley Instruments, Cleveland, OH, USA).Further information can be found in our previous papers [17,49,50].The sensor response (S) was estimated following the procedure reported elsewhere [51].
The sensing performance of the Zn-doped nanocrystals in ambient moisture and with a reducing gas (here, CO gas) was investigated in another chamber (piezo-driven probe model, NEXTRON company; Busan, Republic of Korea).This chamber was connected to a system that allowed the control of temperature, relative humidity (% RH; from 0 to 40%), and target gas concentration.For the sake of comparison, the performance of the sample with NO 2 and CO gases was evaluated under the same conditions (e.g., sensing chamber, %RH, operating temperature, and gas flow).

Characterization of Zn-Doped SnO 2 Nanocrystals
XRD patterns of the doped and undoped SnO 2 samples are presented in Figure 1.All reflections were indexed to a tetragonal SnO 2 phase, in accordance with JCDPS file #41-1445.It should be mentioned that no peaks related to Zn-rich phases were identified, indicating that Zn ions were homogenously incorporated into the SnO 2 lattice.Furthermore, the slight broadening of the XRD peak with increasing Zn content can be linked to the reduction in crystallographic domains.The average crystallite size was estimated from the FWHM of the (110) XRD peak using Scherrer's equation, and the obtained values were 3.2 nm (SnO 2 ), 3.0 nm (SnZn1), and 2.6 nm (SnZn2).Wang and co-workers observed a similar tendency for Zn-doped SnO 2 nanostructures obtained via a hydrothermal treatment, attributing this peak enlargement to the reduction in crystallinity and crystal size [37].Figure 2 presents the TEM analysis of pristine SnO2 and SnZn2, where it can be observed that the spherical morphology of both samples was quite similar, even for the Zndoped sample, with an average crystal size of 7 nm.HRTEM images, the inset of the respective TEM images, show that the distance between the planes was about 0.34 nm, corresponding to the (110) crystallographic plane of the tetragonal SnO2, in accordance with the XRD data.The HRTEM images present agglomerated NPs, indicating a coalescence Figure 2 presents the TEM analysis of pristine SnO 2 and SnZn2, where it can be observed that the spherical morphology of both samples was quite similar, even for the Zn-doped sample, with an average crystal size of 7 nm.HRTEM images, the inset of the respective TEM images, show that the distance between the planes was about 0.34 nm, corresponding to the (110) crystallographic plane of the tetragonal SnO 2 , in accordance with the XRD data.The HRTEM images present agglomerated NPs, indicating a coalescence between particles (illustrated by a red arrow) which can be related to the crystal growth mechanism [52][53][54].This is expected as previous studies have reported that the oriented attachment mechanism plays an important role in SnO 2 synthesis in hydrothermal conditions [52,55]. Figure 2 presents the TEM analysis of pristine SnO2 and SnZn2, where it can be observed that the spherical morphology of both samples was quite similar, even for the Zndoped sample, with an average crystal size of 7 nm.HRTEM images, the inset of the respective TEM images, show that the distance between the planes was about 0.34 nm, corresponding to the (110) crystallographic plane of the tetragonal SnO2, in accordance with the XRD data.The HRTEM images present agglomerated NPs, indicating a coalescence between particles (illustrated by a red arrow) which can be related to the crystal growth mechanism [52][53][54].This is expected as previous studies have reported that the oriented attachment mechanism plays an important role in SnO2 synthesis in hydrothermal conditions [52,55].To investigate the electronic structure of the short-range order of Zn-doped SnO2 NPs, XANES spectra were collected at the Sn-L3 and Zn-K edges, as displayed in Figure 3.For the sake of comparison, commercial micrometric powders (ZnO and SnO2) were used as references.XANES spectra at the Sn-L3 edge present three electronic transitions (labeled as P1, P2, and P3), as shown in Figure 3a.The physical origin of these peaks To investigate the electronic structure of the short-range order of Zn-doped SnO 2 NPs, XANES spectra were collected at the Sn-L3 and Zn-K edges, as displayed in Figure 3.For the sake of comparison, commercial micrometric powders (ZnO and SnO 2 ) were used as references.XANES spectra at the Sn-L3 edge present three electronic transitions (labeled as P1, P2, and P3), as shown in Figure 3a.The physical origin of these peaks corresponds to electronic transitions from 2p 3/2 to 5s 1/2 and nd 5/2 levels [28,56].It can be seen that the pristine SnO 2 , SnZn1, and SnZn2 spectra are quite similar to each other, presenting less significant oscillations compared to the reference sample.This feature reveals that these samples have a low short-range order around the Sn atoms that can be attributed to the rapid crystallization provided by the microwave-assisted route.Regarding the Zn addition, no significant influence on the short-range structure of the SnO 2 samples was observed within the zinc content evaluated here.
Sensors 2023, 23, x FOR PEER REVIEW 6 of 14 corresponds to electronic transitions from 2p3/2 to 5s1/2 and nd5/2 levels [28,56].It can be seen that the pristine SnO2, SnZn1, and SnZn2 spectra are quite similar to each other, presenting less significant oscillations compared to the reference sample.This feature reveals that these samples have a low short-range order around the Sn atoms that can be attributed to the rapid crystallization provided by the microwave-assisted route.Regarding the Zn addition, no significant influence on the short-range structure of the SnO2 samples was observed within the zinc content evaluated here.Figure 3b illustrates the Zn K-edge spectra of SnZn2 and the reference ZnO.Three electronic transitions are visible in the spectra, labeled as A, B, and C. Regarding the electronic transitions A and B, they are attributed to the 1s to 4p transition, and the C transition is due to multiple scattering contributions [17,57].The analysis of both spectra showed that they differ from each other, especially in the post-edge region between 9670 and 9780 eV.This means that the chemical environment of Zn atoms in SnO2 samples is not the same as in ZnO, suggesting their homogeneous incorporation in the SnO2 network.Thus, XANES and XRD analyses allow us to affirm that the addition of Zn ions did not favor the clustering of ZnO (at both long-and short-range) in the Zn-doped SnO2 nanocrystals.XPS analysis was carried out to further characterize the pristine SnO2 and Zn-doped SnO2 samples and illustrate their surface compositions and electronic states.In the XPS survey spectra of these samples, Figure S2a, the Sn and O were found to be the predominant elements, whilst the Zn was present only in the doped samples, as expected.The Figure 3b illustrates the Zn K-edge spectra of SnZn2 and the reference ZnO.Three electronic transitions are visible in the spectra, labeled as A, B, and C. Regarding the electronic transitions A and B, they are attributed to the 1s to 4p transition, and the C transition is due to multiple scattering contributions [17,57].The analysis of both spectra showed that they differ from each other, especially in the post-edge region between 9670 and 9780 eV.This means that the chemical environment of Zn atoms in SnO 2 samples is not the same as in ZnO, suggesting their homogeneous incorporation in the SnO 2 network.Thus, XANES and XRD analyses allow us to affirm that the addition of Zn ions did not favor the clustering of ZnO (at both long-and short-range) in the Zn-doped SnO 2 nanocrystals.
XPS analysis was carried out to further characterize the pristine SnO 2 and Zn-doped SnO 2 samples and illustrate their surface compositions and electronic states.In the XPS survey spectra of these samples, Figure S2a, the Sn and O were found to be the predominant elements, whilst the Zn was present only in the doped samples, as expected.The quantification of peaks from survey spectra revealed that the Zn atom percentages (at%) were 0.6% and 2.0% for the SnZn1 and SnZn2 samples.
Figure S2b shows the Sn 3d high-resolution XPS spectra, in which the Sn 3d 5/2 and Sn 3d 3/2 peaks were located at 486.6 and 495.0 eV, confirming the presence of Sn 4+ in all samples.No significant shift in the Sn 3d 5/2 and Sn 3d 3/2 peaks positions with Zn addition was observed, indicating that the zinc addition did not affect the local electronic structure of Sn.From the high-resolution Zn 2p XPS spectra of the Zn-doped SnO 2 samples, Figure S2c, two symmetrical peaks were identified at binding energies of 1022.7 and 1045.8 eV, with a spin-orbit splitting of 23.1 eV, corresponding to Zn 2p 3/2 and Zn2p 1/2 [58].
Regarding the high-resolution O 1s spectra, Figure 4a, a very asymmetric peak was found that indicates the presence of different oxygen species.These spectra were deconvoluted into two Gaussian-Lorentzian components, labeled as O I and O II .The main component at 530.2 eV (O I ) was typical of lattice oxygen anions bound to the metal cations, whilst the second component (O II ) was linked to hydroxyls and adsorbates [23,37,59,60].From the O 1s high-resolution spectra, the relative percentage area of the O II component was approximately 38.7% (SnO 2 ), 39.5% (SnZn1), and 49.3% (SnZn2), indicating that Zn addition increased slightly the concentration of hydroxyls on the surface of the samples.In addition, the metal-to-oxygen ratios (metal/oxygen) of the pristine and Zn-doped samples were also estimated from survey spectra only considering the metal-oxygen bond in O 1s spectra (530.2 eV), and the obtained data are displayed in Figure 4b.The analysis of this figure reveals a tendency to increase the metal/oxygen ratio with zinc content.Note that a stoichiometric sample (SnO 2 ) must present a metal/O ratio of 0.5, respectively.Thus, the behavior observed in Figure 4b reveals that samples prepared via the microwave-assisted nonaqueous route are oxygen deficient, and the zinc addition into the SnO 2 network favors the enhancement of surface oxygen defects, here induced by the partial substitution of Zn 2+ by Sn 4+ ions.Wang and co-workers using experimental and theoretical approaches also demonstrated that the increase in Zn content favored the formation of oxygen vacancies in SnO 2 nanostructures [37].

Gas-Sensing Measurements
The gas-sensing performances of the pristine SnO 2 , SnZn1, and SnZn2 samples were evaluated for NO 2 gas.To find the best working temperature of the samples, they were exposed to 1 ppm of NO 2 with a heating temperature under the sensors varying from 100 to 300 • C, Figure 5. From the analysis of the curves, the highest responses were achieved at around 150 • C for the Zn-doped samples.At this temperature, the response of the SnZn2 sample was around 25 times higher than that of SnZn1 and 100 times higher than the pristine sample.The observed order of sensing performance was SnZn2 > SnZn1 > SnO 2 , demonstrating that the Zn addition improved the sensitivity to NO 2 gas.This enhancement can be linked to the addition of Zn ions into the SnO 2 nanocrystals which favored the formation of surface oxygen defects, as above-mentioned.Many studies have highlighted the importance of oxygen defects for improving the sensitivity of SMOx [25,[61][62][63].Thus, Sensors 2024, 24, 140 7 of 13 both Zn ions and surface oxygen defects exert a positive influence on the sensing activity of SnO 2 NPs, acting as active sites for NO 2 sensing reactions.Some researchers have reported that nanosized features and the presence of unsaturated cations and oxygen defects may facilitate the interaction of the analyte at the MOX surface, thus increasing its sensitivity [16,64,65].
In addition, the metal-to-oxygen ratios (metal/oxygen) of the pristine and Zn-doped samples were also estimated from survey spectra only considering the metal-oxygen bond in O 1s spectra (530.2 eV), and the obtained data are displayed in Figure 4b.The analysis of this figure reveals a tendency to increase the metal/oxygen ratio with zinc content.Note that a stoichiometric sample (SnO2) must present a metal/O ratio of 0.5, respectively.Thus, the behavior observed in Figure 4b reveals that samples prepared via the microwave-assisted nonaqueous route are oxygen deficient, and the zinc addition into the SnO2 network favors the enhancement of surface oxygen defects, here induced by the partial substitution of Zn 2+ by Sn 4+ ions.Wang and co-workers using experimental and theoretical approaches also demonstrated that the increase in Zn content favored the formation of oxygen vacancies in SnO2 nanostructures [37].

Gas-Sensing Measurements
The gas-sensing performances of the pristine SnO2, SnZn1, and SnZn2 samples were evaluated for NO2 gas.To find the best working temperature of the samples, they were exposed to 1 ppm of NO2 with a heating temperature under the sensors varying from 100 to 300 °C, Figure 5. From the analysis of the curves, the highest responses were achieved at around 150 °C for the Zn-doped samples.At this temperature, the response of the SnZn2 sample was around 25 times higher than that of SnZn1 and 100 times higher than the pristine sample.The observed order of sensing performance was SnZn2 > SnZn1 > SnO2, demonstrating that the Zn addition improved the sensitivity to NO2 gas.This enhancement can be linked to the addition of Zn ions into the SnO2 nanocrystals which favored the formation of surface oxygen defects, as above-mentioned.Many studies have highlighted the importance of oxygen defects for improving the sensitivity of SMOx [25,[61][62][63].Thus, both Zn ions and surface oxygen defects exert a positive influence on the sensing activity of SnO2 NPs, acting as active sites for NO2 sensing reactions.Some researchers have reported that nanosized features and the presence of unsaturated cations and oxygen defects may facilitate the interaction of the analyte at the MOX surface, thus increasing its sensitivity [16,64,65].Considering the superior performance of SnZn2 to detect NO2 gas, it was fur vestigated at different NO2 levels ranging from 0.1 to 2 ppm at a working temper 150 °C. Figure 6a reveals that this sample was sensitive to all of these levels, with dence of saturation.It should be mentioned that these results demonstrate the p feasibility of Zn-doped SnO2 NPs as a sensing material since NO2 gas is harmful to health in concentrations higher than 0.1 ppm [8,66].This figure also depicts the g peatability of the response and recovery of the SnZn2 sample which was able t low concentrations even after consecutive exposure cycles.
The long-term stability of SnZn2 was also evaluated, exposing it repeatedl Considering the superior performance of SnZn2 to detect NO 2 gas, it was further investigated at different NO 2 levels ranging from 0.1 to 2 ppm at a working temperature of 150 • C. Figure 6a reveals that this sample was sensitive to all of these levels, with no evidence of saturation.It should be mentioned that these results demonstrate the practical feasibility of Zn-doped SnO 2 NPs as a sensing material since NO 2 gas is harmful to human health in concentrations higher than 0.1 ppm [8,66].This figure also depicts the good repeatability of the response and recovery of the SnZn2 sample which was able to detect low concentrations even after consecutive exposure cycles.
health in concentrations higher than 0.1 ppm [8,66].This figure also depicts the good repeatability of the response and recovery of the SnZn2 sample which was able to detect low concentrations even after consecutive exposure cycles.
The long-term stability of SnZn2 was also evaluated, exposing it repeatedly to 0.1 ppm of NO2 gases over 14 days at an operating temperature of 150 °C, as seen in Figure 6b.It can be noted that the sample was able to detect NO2 gas over the whole period, revealing that its surface remains active after several exposure cycles.Long-term stability of this sample upon exposure to 0.1 ppm of NO2 gas for a period of approximately 2 weeks.All measurements were performed under dry air (0% RH).
The influence of relative humidity on the NO2 sensing performance was also investigated.To this end, the SnZn2 NPs were kept at 150 °C and then exposed to 0.2, 0.25, 0.5, 0.7, and 1 ppm of NO2 gas under different relative humidity values (0, 20, and 40% RH).It can be seen that the NO2 sensor response was enhanced with increasing relative humidity, as shown in Figure 7a.Note that the sample exhibited high repeatability of its The long-term stability of SnZn2 was also evaluated, exposing it repeatedly to 0.1 ppm of NO 2 gases over 14 days at an operating temperature of 150 • C, as seen in Figure 6b.It can be noted that the sample was able to detect NO 2 gas over the whole period, revealing that its surface remains active after several exposure cycles.
The influence of relative humidity on the NO 2 sensing performance was also investigated.To this end, the SnZn2 NPs were kept at 150 • C and then exposed to 0.2, 0.25, 0.5, 0.7, and 1 ppm of NO 2 gas under different relative humidity values (0, 20, and 40% RH).It can be seen that the NO 2 sensor response was enhanced with increasing relative humidity, as shown in Figure 7a.Note that the sample exhibited high repeatability of its response, confirming its reliability even under ambient moisture, as demonstrated by the nine exposure cycles displayed in Figure 7b.response, confirming its reliability even under ambient moisture, as demonstrated by the nine exposure cycles displayed in Figure 7b.
According to the findings, the sensor response of SnZn2 to 1 ppm NO2 increased from S = 66.4 ± 0.4 (RH = 0%) to S = 285.9± 0.4 (RH = 40%).A low interference of humidity on the sensing performance is a desirable characteristic in the development of sensing materials [16,[66][67][68][69][70].Several investigations demonstrate that the presence of moisture in the ambient may impair the sensing activity [51,68,71,72].In 2021, we reported that the sensing response of α-Fe2O3 to BTEX gases (Benzene, Toluene, Ethylbenzene, and Xylenes) was reduced as a function of humidity level [51].This negative effect was linked to the competition of water and BTEX molecules for the same surface-active sites [51].In contrast, Yan and co-workers found that humidity can improve the NO2 sensing performance of WS2/graphene composites working at room temperature [66].Shooshtari and co-workers observed an increase in the ethanol sensing response of TiO2 nanowires for RH of up to 50% [73].They explained that this behavior was a result of the presence of hydroxyl groups and oxygen adsorbates.For higher humidity levels (>50% RH), the hydroxyl groups cover almost all of the TiO2 surface, thus limiting the oxygen adsorption and consequently reducing the sensitivity toward ethanol [73].The SnZn2 sample was also evaluated to detect a reducing analyte, specifically CO gas, using the optimal temperature of 150 °C. Figure S4 shows that the same sample exhibited a very low sensitivity to CO gas.Despite detecting the CO gas, the sensing response of SnZn2 NPs toward NO2 was superior compared to CO gas, as seen in Figure 8.According to the findings, the sensor response of SnZn2 to 1 ppm NO 2 increased from S = 66.4 ± 0.4 (RH = 0%) to S = 285.9± 0.4 (RH = 40%).A low interference of humidity on the sensing performance is a desirable characteristic in the development of sensing materials [16,[66][67][68][69][70].Several investigations demonstrate that the presence of moisture in the ambient may impair the sensing activity [51,68,71,72].In 2021, we reported that the sensing response of α-Fe 2 O 3 to BTEX gases (Benzene, Toluene, Ethylbenzene, and Xylenes) was reduced as a function of humidity level [51].This negative effect was linked to the competition of water and BTEX molecules for the same surface-active sites [51].In contrast, Yan and co-workers found that humidity can improve the NO 2 sensing performance of Sensors 2024, 24, 140 9 of 13 WS 2 /graphene composites working at room temperature [66].Shooshtari and co-workers observed an increase in the ethanol sensing response of TiO 2 nanowires for RH of up to 50% [73].They explained that this behavior was a result of the presence of hydroxyl groups and oxygen adsorbates.For higher humidity levels (>50% RH), the hydroxyl groups cover almost all of the TiO 2 surface, thus limiting the oxygen adsorption and consequently reducing the sensitivity toward ethanol [73].
The SnZn2 sample was also evaluated to detect a reducing analyte, specifically CO gas, using the optimal temperature of 150 C. Figure 8 shows that the same sample exhibited a very low sensitivity to CO gas.Despite detecting the CO gas, the sensing response of SnZn2 NPs toward NO 2 was superior compared to CO gas, as seen in Figure 8.It is clear that the results here obtained revealed that the addition of Zn ions into the SnO2 nanocrystals significantly enhanced the sensing performance toward sub-ppm NO2 levels.Nevertheless, it would be interesting in future works to find rational strategies to reduce the influence of moisture on gas sensitivity and to obtain a fair sensing performance (e.g., complete recovery, stability, and reproducibility) at working temperatures closer to room temperature.

Conclusions
We report here a versatile approach for preparing Zn-doped SnO2 NPs via a microwave-assisted nonaqueous route for use as promising chemoresistive NO2 sensors.XRD, XANES, and HRTEM analyses confirmed the presence of nanocrystalline SnO2 and the absence of spurious phases.Sn-L edge XANES spectra revealed a local disorder around Sn atoms caused by the rapid crystallization provided by the microwave-assisted route.Zn-K edge XANES indicated that the Zn ions were incorporated into the SnO2 network.XPS analyses revealed an increase in the oxygen defects at the SnO2 surface, probably due to the partial replacement of Sn 4+ by Zn 2+ ions.The presence of these defects was linked to the high sensor response of SnZn2 (2 at% of Zn) toward NO2 gas which was able to detect sub-ppm levels, i.e., from 0.1 ppm.Furthermore, the addition of Zn ions increased the moisture tolerance of SnO2 NPs, presenting a higher sensing response in a more humid environment (40% RH) compared to a dry one.The enhanced sensor properties were linked to the presence of Zn ions that favored the formation of surface oxygen vacancies, thus contributing to the adsorption of NO2 molecules.These findings highlight the promising properties of the microwave-assisted nonaqueous route for obtaining Zn-doped SnO2 NPs for chemoresistive NO2 sensors in practical applications.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1  It is clear that the results here obtained revealed that the addition of Zn ions into the SnO 2 nanocrystals significantly enhanced the sensing performance toward sub-ppm NO 2 levels.Nevertheless, it would be interesting in future works to find rational strategies to reduce the influence of moisture on gas sensitivity and to obtain a fair sensing performance (e.g., complete recovery, stability, and reproducibility) at working temperatures closer to room temperature.

Conclusions
We report here a versatile approach for preparing Zn-doped SnO 2 NPs via a microwaveassisted nonaqueous route for use as promising chemoresistive NO 2 sensors.XRD, XANES, and HRTEM analyses confirmed the presence of nanocrystalline SnO 2 and the absence of spurious phases.Sn-L edge XANES spectra revealed a local disorder around Sn atoms caused by the rapid crystallization provided by the microwave-assisted route.Zn-K edge XANES indicated that the Zn ions were incorporated into the SnO 2 network.XPS analyses revealed an increase in the oxygen defects at the SnO 2 surface, probably due to the partial replacement of Sn 4+ by Zn 2+ ions.The presence of these defects was linked to the high sensor response of SnZn2 (2 at% of Zn) toward NO 2 gas which was able to detect sub-ppm levels, i.e., from 0.1 ppm.Furthermore, the addition of Zn ions increased the moisture tolerance of SnO 2 NPs, presenting a higher sensing response in a more humid environment (40% RH) compared to a dry one.The enhanced sensor properties were linked to the presence of Zn ions that favored the formation of surface oxygen vacancies, thus contributing to the adsorption of NO 2 molecules.These findings highlight the promising properties of the microwave-assisted nonaqueous route for obtaining Zn-doped SnO 2 NPs for chemoresistive NO 2 sensors in practical applications.

Sensors 2023 , 14 Figure 1 .
Figure 1.XRD patterns of the undoped and Zn-doped SnO2 NPs.At the bottom, the vertical bars correspond to the data obtained from the JCPDS file #41-1445.

Figure 1 .
Figure 1.XRD patterns of the undoped and Zn-doped SnO 2 NPs.At the bottom, the vertical bars correspond to the data obtained from the JCPDS file #41-1445.

Figure 1 .
Figure 1.XRD patterns of the undoped and Zn-doped SnO2 NPs.At the bottom, the vertical bars correspond to the data obtained from the JCPDS file #41-1445.

Figure 2 .
Figure 2. TEM images of selected samples: (a) pristine SnO2 and (b) SnZn2 NPs.Inset shows the HRTEM images of their respective samples.(Left) White arrows indicate the atomic distance, and red arrow the coalescence between particles.(Right) White arrows indicate the nanocrystal size.

Figure 2 .
Figure 2. TEM images of selected samples: (a) pristine SnO 2 and (b) SnZn2 NPs.Inset shows the HRTEM images of their respective samples.(Left) White arrows indicate the atomic distance, and red arrow the coalescence between particles.(Right) White arrows indicate the nanocrystal size.

Figure 3 .
Figure 3. XANES spectra of undoped and Zn-doped SnO2 NPs.(a) Sn L3-and (b) Zn-K edge.For the sake of comparison, the spectra of commercial samples are inserted in panels (a,b).

Figure 3 .
Figure 3. XANES spectra of undoped and Zn-doped SnO 2 NPs.(a) Sn L3-and (b) Zn-K edge.For the sake of comparison, the spectra of commercial samples are inserted in panels (a,b).

Figure 4 .
Figure 4. O1 s XPS spectra of the pristine SnO2, SnZn1, and SnZn2 NPs.(a) High-resolution scan and (b) variation in (metal/oxygen) ratio as a function of zinc content.

Figure 4 .
Figure 4. O1 s XPS spectra of the pristine SnO 2 , SnZn1, and SnZn2 NPs.(a) High-resolution scan and (b) variation in (metal/oxygen) ratio as a function of zinc content.Sensors 2023, 23, x FOR PEER REVIEW

Figure 5 .
Figure 5. Sensor response of pristine and Zn-doped SnO2 NPs exposed to 1 ppm NO2 at working temperatures.The inset shows in detail the response of SnO2 and SnZn1 samples.A urements were performed under dry air (0% RH).

Figure 5 .
Figure 5. Sensor response of pristine and Zn-doped SnO 2 NPs exposed to 1 ppm NO 2 at different working temperatures.The inset shows in detail the response of SnO 2 and SnZn1 samples.All measurements were performed under dry air (0% RH).

Figure 6 .
Figure 6.(a) Gas-sensing performance SnZn2 at 150 °C exposed to 0.1 to 2.0 ppm of NO2 gas.(b)Long-term stability of this sample upon exposure to 0.1 ppm of NO2 gas for a period of approximately 2 weeks.All measurements were performed under dry air (0% RH).

Figure 6 .
Figure 6.(a) Gas-sensing performance of SnZn2 at 150 • C exposed to 0.1 to 2.0 ppm of NO 2 gas.(b) Long-term stability of this sample upon exposure to 0.1 ppm of NO 2 gas for a period of approximately 2 weeks.All measurements were performed under dry air (0% RH).

Figure 7 .
Figure 7. Sensor response of SnZn2 at 150 °C (a) exposed to different NO2 levels under different relative humidity and (b) exposed to a sequence of cycles of 0.25 ppm of NO2 under 40% RH.

Figure 7 .
Figure 7. Sensor response of SnZn2 at 150 • C (a) exposed to different NO 2 levels under different relative humidity and (b) exposed to a sequence of cycles of 0.25 ppm of NO 2 under 40% RH.

Sensors 2023 , 14 Figure 8 .
Figure 8.Comparison of the sensor response of SnZn2 exposed to NO2 (0.2 to 1 ppm) and CO (2 to 10 ppm).These measurements were performed at 150 °C in a dry air atmosphere (0% RH).
: (a) Photograph of sensing platforms based on SnO2 and SnZn2 samples compared to a U.S one tenth-dollar coin.(b) FESEM image of the SnZn2 sample onto sensing platform.; Figure S2: XPS spectra of the Zn-doped SnO2 NPs synthesized via microwave-assisted

Figure 8 .
Figure 8.Comparison of the sensor response of SnZn2 exposed to NO 2 (0.2 to 1 ppm) and CO (2 to 10 ppm).These measurements were performed at 150 • C in a dry air atmosphere (0% RH).

Table 1 .
Gas-sensing performance of Zn-doped SnO 2 synthesized by different methodologies.