Enhanced ammonia sensing using Pt-decorated single-crystalline SnO₂ nanowires prepared by sputtering

Abstract

Ammonia is a well-known toxic and colorless gas that is widely produced in various chemical and biological industries; thus, developing high-performance gas sensors for the detection of NH₃ is essential. Herein, we prepared a high-performance NH₃ gas sensor based on SnO₂ nanowires coated with Pt by the sputtering method and studied the effect of Pt coating on the gas-sensing performance. Single-crystal SnO₂ nanowires were grown by the CVD method, and sputtering was used to control the thickness of the Pt coating layer. We demonstrated that the NH₃ gas-sensing performance of Pt–SnO₂ nanowires was significantly enhanced compared to that of pristine SnO₂ nanowires. The 5 nm Pt–SnO₂ nanowire sensor exhibits a high response (R_a/R_g) of 19.98 toward 500 ppm NH₃ gas, which is a 19-fold enhancement compared with that of the bare SnO₂ nanowires. The sensor exhibited a fast response and recovery with excellent selectivity over the interfering species C₂H₄, H₂, and acetone. We also proposed a mechanism for the enhanced sensitivity of Pt–SnO₂ nanowires toward NH₃.

---

1. Introduction

Gas sensors, which can detect or measure trace amounts of toxic and/or explosive gases, have attracted considerable attention from researchers worldwide because of their broad potential applications in air pollution monitoring, healthcare, military, and automotive systems.^1–3 Especially, the need to monitor hazardous gases, including CO, CO₂, NO₂, SO₂, H₂S, and NH₃, has been increasing with rising concerns about the environment and human health. Among these hazardous gases, ammonia (NH₃) is highly toxic even at low concentrations (e.g., ppm levels).^4,5 Exposure to 400 ppm NH₃ for one hour may cause throat irritation, whereas concentrations as low as 25 ppm can be readily detected by odor. Accordingly, OSHA has set the permissible limit for NH₃ exposure at 50 ppm for an 8 hour workday. Therefore, real-time monitoring of NH₃ at low concentrations, at ppm levels, is vital for applications in air pollution control and food quality monitoring.⁶

Nowadays, gas detection tools include infrared absorbers, ion mobility spectrometers, and gas chromatographs. However, these methods have the disadvantages of being time-consuming with complicated operations and high costs, which make it difficult to meet the growing demands.^7,8 Therefore, the development of next-generation toxic and harmful gas detection techniques is imperative to protect humans from potential health risks.² Resistive gas sensors based on metal-oxide semiconductors were reported to exhibit significant advantages, including a simple device structure, compatibility with silicon technology and suitability for miniaturized platforms, making them ideally suitable for real-time monitoring and early warning of NH₃ leakage.⁹ For these reasons, the exploration and development of high-performance gas sensors have attracted widespread attention from scientists in recent years.⁴

In recent studies, different materials have been synthesized for gas sensor applications.^10–12 Metal-oxide semiconductor-based gas sensors (MOS) have attracted significant attention for NH₃ detection, especially SnO₂, ZnO, WO₃, and TiO₂.^13,14 MOS can be classified into n-type or p-type semiconductors depending on the deficiency or excess of oxygen in their composition, which consequently determines their gas-sensing properties. An increased density of active sites on the surface of the sensing material is expected to enhance sensor sensitivity.^14,15 The intrinsic properties of metal oxides, namely, low cost, thermal stability, nontoxicity, and high chemical sensitivity, are ascribed effectively for a chemiresistive gas sensor. Furthermore, the mechanism of gas sensors mostly depends on the size of metal-oxide particles and their crystallinity.¹⁵ Ultraviolet or visible light irradiation was also used to enhance the gas-sensing performance of pure metal oxides and CdS/WSe₂/CdS heterojunction.^16,17 Tin oxide (SnO₂), an n-type semiconductor with an energy gap of 3.6 eV, is one of the most widely used oxides for developing toxic gas detectors.¹⁸ The applications of SnO₂ for gas sensing have also attracted considerable attention due to its excellent gas response. The introduction of noble metal nanostructures (Pd, Pt, Ag, Cu, and Ni) is considered an effective way to enhance the gas response as well as lower the operating temperature of the sensor.¹⁸ The enhancement of the sensitivity of the metal-oxide gas sensor by surface decoration with noble metal nanoparticles was ascribed to the formation of a heterojunction at the metal-oxide interface, and the catalytic activity of noble metal nanoparticles.^18,19 For metal oxide-based gas sensors, room-temperature and low-temperature sensors have attracted great interest due to their low power consumption.^19,20 Noble functionalization on metal oxides can also reduce the operating temperature of gas sensors.^21,22 However, effective control of the decoration of noble metal nanoparticles on the surface of nanomaterials for enhanced gas-sensing performance is still a challenge.

Herein, Pt–SnO₂ nanowire NH₃ sensors were fabricated by chemical vapor deposition combined with sputtering modification of Pt nanoparticles. We precisely controlled Pt deposition to optimize gas-sensing performance. Experiments showed that the Pt-decorated-SnO₂ (PS) sensors have high sensing response and short response/recovery as compared with pristine SnO₂ nanowires. Furthermore, the sensing mechanism of PS was studied and investigated. Our results indicate that the PS-based sensor exhibits a 19-fold higher performance than pure SnO₂ under the same experimental conditions. Additionally, the sensors exhibited excellent selectivity for ammonia over contaminants, such as ethylene and acetone.

2. Experimental

2.1 Materials

In the experimental process, the analytical substance consisting of pure tin powder (99.99% Sn) was used for the growth of SnO₂ nanowires, whereas a two-inch Pt (99.99%) target was used for the deposition of a Pt catalyst layer. The experimental process includes two main stages: (i) chemical vapor deposition (CVD) growth of SnO₂ nanowires and (ii) sputtering deposition of Pt nanoparticles, as shown in Fig. 1.

Fig. 1 Schematic of the (A) synthesis of SnO₂ nanowires and (B) decoration of Pt nanoparticles.

2.2 Preparation of SnO₂ nanowires

The preparation of SnO₂ nanowire gas sensors has been described in our previous articles as published elsewhere.²³ As reported, the SnO₂ nanowires were synthesized by a chemical vapor deposition (CVD) method using a lab-made system, where we studied the effect of growth temperature and growth time on the density and morphology of the nanowires. In this work, to ensure the growth of SnO₂ nanowires on a glass substrate without melting it, we used a low temperature of 730 °C with a growth time of 8 min. In a typical procedure, SnO₂ nanowires were grown at 730 °C with pure Sn powder as an evaporation source. The evaporation source and Pt interdigital electrodes were loaded at the center zone of the CVD system, and the temperature was increased to 730 °C within 30 min. The SnO₂ nanowire growth process was carried out for 8 min with a flow rate of 0.5 sccm oxygen, introduced evenly into the reaction chamber, and a pressure of about 10⁻¹ Torr. After the completion of the growth process of SnO₂ nanowires, the furnace was cooled down naturally to room temperature (RT).

2.3 Decoration of Pt nanoparticles

The Pt decorated SnO₂ nanowires were prepared using the sputtering method. Briefly, a 50 W DC power source was used to bombard the Pt target in a plasma-containing environment, ejecting atoms from the target. Then, these ejected atoms were deposited on the surface of SnO₂ nanowires. In this process, the density of Pt nanoparticles decorated on the surface of SnO₂ nanowires was controlled by varying the sputtering time. The samples were denoted as PS-30, PS-50, and PS-75 corresponding to the sputtering times of 30 s, 50 s, and 75 s, respectively (Table 1). The Pt thickness was measured on a thin film sample deposited on a Si wafer using a stylus profilometer (Dektak Pro, Bruker, Germany). The morphology, composition, structure, and gas-sensing properties of the Pt-decorated SnO₂ materials were investigated by scanning electron microscopy (SEM; JEOL 7600F), energy-dispersive X-ray spectroscopy (EDS), high-resolution transmission electron microscopy (HRTEM, JEOL, JEM-2100), selective area electron diffraction (SAED), and a gas-sensing system based on the change in resistance.²³ The electrical properties and gas-sensing characteristics of the fabricated sensors were studied by using a homemade gas-sensing measurement system, as reported in ref. 24. Briefly, the sensor was placed on a hot plate with controlled temperature ranging from 250 to 450 °C, while its electrical properties were measured. The electrical properties were studied by current–voltage (I–V) measurements under a bias voltage of ±5 V. The sensor resistance was monitored while exposing the devices to NH₃ with different concentrations at various temperatures. The sensor response was defined as the resistance ratio of the sensor in air (R_a) to that in the target gas (R_g). The response time and recovery time are defined as the time required for the sensor signal to reach 90% of the total change after exposure to the target gas (response) and after removal of the target gas (recovery), respectively.²⁵

Table 1 Samples at different sputtering times

Sample	Sputtering time (seconds)	Pt thickness (nm)
SnO2 NWs	None	None
PS-30	30 s	3 nm
PS-50	50 s	5 nm
PS-75	75 s	8 nm

---

3. Results and discussion

3.1 Characterization of materials

To investigate the influence of the synthesis process on the morphologies, crystal structures and gas-sensing properties of SnO₂ nanowires and Pt/SnO₂ nanowires, we compared their SEM, EDS, HRTEM, SAED, I–V curves, and NH₃ gas response, as shown below.

The SEM images of SnO₂ and Pt-decorated SnO₂ samples are shown in Fig. 2, confirming the successful growth of nanowires on the entire Pt electrodes. As shown in Fig. 2(A1 and C1), the SnO₂ nanowires have an average length of about 8 micrometers, whereas the average diameter is about 80 nm. The nanowires grew on the surface of the Pt electrodes, but none of the nanowires grew on the bare SiO₂ substrate due to the catalytic activity of the Pt electrode. The size distribution histograms of SnO₂ nanowires and Pt–SnO₂ nanowires are shown in the insets in Fig. 2(C1 and C2), respectively. It is clear that the nanowires have diameters ranging from 50 to 200 nm, suggesting the growth process was not entirely uniform. The number of nanowires with small diameters of about 50–100 nm is higher than others. Pt was decorated on the surface of SnO₂ nanowires by the sputtering method, as shown in Fig. 2(A2 and C2). There was a slight difference in the morphology of the sample after the deposition of Pt nanoparticles. However, we could not observe the individual Pt nanoparticle in the SEM images because of the limited resolution.

Fig. 2 SEM images of sensors with different magnifications: (A1) and (C1) SnO₂ nanowires and (A2) and (C2) Pt-decorated SnO₂ nanowires. Insets in (C1) and (C2) show the corresponding nanowire-diameter distributions.

As the SEM images did not provide clear evidence of Pt nanoparticles on the surface of the SnO₂ nanowires, we used the EDS analysis to determine the composition of the PS-50 sensor, as shown in Fig. 3. The EDS spectrum acquired at 15 keV displays the presence of Sn, O and Pt elements with weight percentages of 28.29 wt%, 70.26 wt% and 1.45 wt%, respectively. Therefore, the EDS results further verify the successful fabrication of the PS-50 sample with Pt decoration on the surface of SnO₂ nanowires.

Fig. 3 EDS spectrum of the PS-50 sensor.

Additionally, elemental mapping was performed to evaluate the elemental distribution in the PS-50 sample, as shown in Fig. 4A–D. The EDS mapping data confirm the uniform distribution of O, Sn, and Pt elements over the entire surface of the sample, indicating the effective decoration of Pt on the surface of SnO₂ nanowires. Furthermore, the Pt content is significantly lower than that of Sn and O due to the short sputtering time (50 s).

Fig. 4 (A) SEM image of PS-50 and the elemental mappings for (B) O, (C) Sn, and (D) Pt elements in the sensor.

The microstructure and crystallinity of Pt–SnO₂ nanowires investigated by HRTEM images are shown in Fig. 5. The diameter of the large nanowire is about 80 nm, whereas the small nanowire is about 50 nm (Fig. 5A). We could observe many tiny Pt nanoparticles of about 5–10 nm in size decorated on the surface of the nanowires. By counting the number of Pt nanoparticles and their average size, we estimated that the density of Pt nanoparticles is about 500 particles per µm² surface of SnO₂ nanowires. The SEAD pattern (inset of Fig. 5A) exhibits sharp diffraction spots, confirming the single crystallinity of SnO₂ nanowires. The HRTEM image (Fig. 5B) shows lattice fringes with interplanar spacing matching well with Pt crystalline planes, particularly the Pt (111) plane (0.28 nm). The result is in good agreement with a recent report.²⁶ The HRTEM observation and SEAD analysis confirm the successful decoration of Pt nanoparticles on the surface of single-crystal SnO₂ nanowires.

Fig. 5 (A) TEM and (B) HRTEM images of the Pt–SnO₂ nanowires.

3.1.1 Electrical and gas-sensing properties. The electrical properties of the fabricated sensors were investigated by measuring their I–V characteristics at 250–450 °C using a source meter (Keithley 2602). At each temperature, the I–V characteristics were measured in air for 100 data points within a bias range of ±5 V with a 0.3 s interval between observations.²³

The I–V plots of the SnO₂ nanowires and PS-30, PS-50, and PS-75 sensors are shown in Fig. 6A–D respectively, indicating the linear dependence of current (I) on applied voltage (V), which is regarded as Ohmic contact between the Pt electrode and the sensing materials. Herein, the Ohmic behavior at the Pt/SnO₂ interface can be attributed to the direct growth of SnO₂ nanowires on Pt, which ensures clean contact with low interfacial resistance, as well as to the high carrier concentration induced by oxygen vacancies, which enables electron tunneling across the thin depletion region. An Ohmic contact in resistive gas sensors ensures that the device response is dominated by the surface sensing mechanism instead of interfacial contact effects.²⁷

Fig. 6 The current–voltage characteristics of (a) SnO₂ nanowires, (b) PS-30, (c) PS-50 and (d) PS-75 sensors measured in the range of 250–450 °C in air.

In this study, the as-prepared sensors with different sputtering times of Pt (30 s, 50 s, and 75 s) were tested toward NH₃ gas under dynamic conditions. NH₃ concentrations ranging from 25 to 500 ppm were tested at various working temperatures. Prior to the measurements, the sensors were stabilized on a hot plate with an air flow rate of 400 sccm for 1 h until a stable resistance was achieved. Once the resistance reached a steady state, NH₃ was introduced into the measurement chamber by switching the gas valve for 200 seconds per pulse. The sensor was then allowed to recover to its baseline resistance under an air flow of 400 sccm.²⁴ Moreover, it is crucial to figure out the optimal sensor and investigate their gas-sensing characteristics. Three different types of sensors corresponding to different sputtering times of Pt nanoparticles, namely, PS-30, PS-50, and PS-75, were tested, and we found that the PS-50 sensor demonstrated a superior response towards NH₃ as compared to the others (data not shown).

As depicted in Fig. 7A, PS-50 was exposed to increasing concentrations of NH₃ gas at different operating temperatures. Notably, PS-50 demonstrated a fast response and recovery towards 500 ppm NH₃ with response/recovery times of 6/9 seconds at 450 °C. However, the response and recovery times of the sensor show an upward trend as the operating temperature decreases. The phenomenon occurred due to the activity of oxygen ions adsorbed on the surface of the material and the gas adsorption and desorption mechanism. The increase in response and recovery times at lower operating temperatures can be attributed to slower surface reaction kinetics and reduced desorption rates of reaction products. Fig. 7B displays the sensor response as a function of NH₃ concentration at 250–450 °C. The response increases with increasing NH₃ concentration, and a tendency toward saturation is observed as the NH₃ concentration approaches 500 ppm. In addition, the sensor exhibited the highest response at 300 °C (response value of 19.98 for 500 ppm NH₃), which was identified as the optimal working temperature. The optimal sensing temperature of 300 °C is attributed to the dominant presence of highly reactive chemisorbed O⁻ species, which promote efficient surface redox reactions with target gas molecules. At lower temperatures, oxygen activation and reaction kinetics are limited, while at higher temperatures, rapid gas desorption reduces the probability of surface reactions and decreases the sensor response.

Fig. 7 (A) Resistance of the PS-50 sensor at various working temperatures and (B) sensor response as a function of the NH₃ concentration.

To demonstrate the enhancement in NH₃ sensing performance after decorating with Pt nanoparticles, the pure-SnO₂ nanowires and Pt-decorated SnO₂ nanowires-based sensors were tested toward 500 ppm NH₃ at 250–450 °C. As illustrated in Fig. 8A, the Pt–SnO₂ nanowire sensor demonstrates significantly higher response values than the pure SnO₂ nanowire sensor across the entire temperature range. In addition, Fig. 8A illustrates an upward trend in the sensor's response with a decrease in temperature from 450 °C to 300 °C. The Pt–SnO₂ nanowire sensor shows the highest response at 300 °C. In contrast, when the temperature decreases to 250 °C, the sensor response declines because, at lower temperatures, the number of active chemisorbed oxygen species (O⁻) is reduced and the surface reaction kinetics is slower. Furthermore, the PS-50 sensor showed a 19-fold higher response than the SnO₂ nanowire-based sensor under identical conditions, confirming the significant catalytic role of Pt nanoparticles.

Fig. 8 (A) Comparison of the responses of sensor based on pure-SnO₂ nanowires and PS-50 sensor toward 500 ppm NH₃ at different temperatures and the (B) selectivity of the PS-50 sensor in the presence of contaminant gases.

We investigated the selectivity of the PS-50 sensor against the contamination of 500 ppm C₂H₄, H₂ and acetone at the optimal working temperature of 300 °C. The response of the PS-50 sensor to 500 ppm of different gases is plotted in Fig. 8B. The response values were 2.67, 3.34, 6.86 and 19.98 for acetone, H₂, C₂H₄, and NH₃, respectively. Herein, the sensor exhibits the highest response toward NH₃, which may be attributed to the fact that NH₃ is a polar molecule with a lone pair of electrons on the nitrogen atom. This enables strong interaction with the Lewis acid sites (Sn⁴⁺) on the SnO₂ surface. In contrast, H₂ is nonpolar, while C₂H₄ and acetone require more complex oxidation mechanisms. Thus, the results demonstrated that the PS-50 sensor exhibits an excellent selectivity toward NH₃ detection.

The long-term and short-term stability of the PS-50 sensor were studied at 300 °C and the data are shown in Fig. 9. The short-term stability was assessed by repeatedly switching the atmosphere between air and 25 ppm NH₃ for nice pulses. As illustrated in Fig. 9A, the sensor response demonstrated high repeatability over most cycles, and the resistance fully recovered to its baseline value upon exposure to fresh air.

Fig. 9 (A) Short-term stability of PS-50 towards 25 ppm NH₃ and (B) their long-term stability over 15 days.

The long-term stability of the sensor was tested for 15 days, and the results reveal the good stability of the sensor with negligible sensor response degradation (Fig. 9B). Nevertheless, the PS-50 sensor exhibited a stable response over the entire experimental period, suggesting its potential for practical applications. The good stability of the sensor was attributed to the high crystallinity of the SnO₂ nanowires, and the effective decoration of Pt nanoparticles without aggregation or oxidation of Pt.¹⁸

We compared our results with other reports on NH₃ sensors. As summarized in Table 2, our sensor showed higher response than SnO₂/Pt thin film and SnO₂/PPy, but lower than others, regardless of the operating temperature. However, we demonstrated that the PS-50 sensor significantly enhanced the response by 19-fold compared to the pristine SnO₂ sensor.

Table 2 Comparison of NH₃ gas-sensing performance between the current work and previous reports

Material	Method	Concentration	Response (Ra/Rg)	Temperature	References
a Response defined as Ra/Rg.b Response defined as (Ra − Rg)/Ra × 100(%).
PANI/Fe doped CeO2 nanocomposite	In situ polymerisation	25 ppm	500^b	RT	28
CuO/WO3	Thermal evaporation and sputtering	100 ppm	673^b	300	29
SnO2/Pt/WO3	Sputtering	5 ppm	2.36^a	250	13
ZnO	CVD	1000 ppm	22.5^b	300	30
SnO2/PPy	Hydrothermal	200 ppm	200^b	RT	31
Pt/SnO2	Sputtering	450 ppm	25.7^a	230	32
Pt/SnO2	CVD and sputtering	500 ppm	20^a	300	This work

---

3.1.2 Gas-sensing mechanisms. The gas-sensing mechanism of SnO₂ nanowires and Pt–SnO₂ nanowire sensors can be explained by the surface depletion model. When measured in air as a reference, oxygen molecules are adsorbed on the metal-oxide surface to form O₂⁻, O⁻, and O²⁻ species, depending on the operating temperatures. The reaction process can be described by the following equations:^33–35

O₂ (gas) → O₂ (ads) (1)

O₂ (ads) + e− → O₂⁻ (ads) (2)

O₂⁻ (ads) + e−→ 2O⁻ (ads) (3)

O− + e⁻→ O²⁻ (ads) (4)

Adsorbed oxygen species extract electrons from the SnO₂ semiconductor thereby generating an electron depletion layer, Fig. 10A.

Fig. 10 Schematic of the NH₃ gas-sensing mechanism of (A) pristine SnO₂ and (B) Pt/SnO₂ in the presence of air and NH₃. (C) and (D) Energy band diagram of the PS sensor in the presence of air and NH₃.

The absorbed oxygen species capture electrons from the SnO₂ semiconductor and form an electron depletion layer (Fig. 10A). In the moderate temperature range of 200–300 °C, the predominant oxygen species is O⁻. Therefore, upon exposure to NH₃, the reaction between the adsorbed oxygen species and NH₃ molecules occurs according to the following equation:

2NH₃ + 3O⁻ (ads) → N₂ + 3H₂O + 3e⁻ (5)

According to the obtained results, the PS-50 sensor exhibits an improved gas-sensing response compared to the pure SnO₂ sensor. The enhancement of gas-sensing properties of the Pt/SnO₂ nanowire sensor is explained by (i) the spillover effect and (ii) the formation of a Schottky barrier at the interface between the noble Pt metal and the SnO₂ metal-oxide semiconductor.¹³ The catalytic activity of Pt nanoparticles sensitizes the dissociation of oxygen molecules into active oxygen and enhances the pre-adsorbed oxygen species on the surface of SnO₂, the so-called spillover effect. In addition, when Pt is decorated on the SnO₂ surface, electrons are transferred from SnO₂ to Pt because its work function (4.6 eV) is lower than that of Pt (work function 5.6 eV). This phenomenon leads to the formation of the Schottky barrier or an electron depletion layer at the Pt/SnO₂ interface.¹³ As a result, the Pt decoration facilitates the adsorption of oxygen on the surface layer of SnO₂ nanowires and increases the base resistance of the sensor. NH₃ gas introduced into the chamber can react with the pre-adsorbed oxygen species, as described in eqn (5). Therefore, the captured electrons are released back into the conduction band of SnO₂. As a result, the width of the depletion layer decreases, and the resistance of the sensor decreases.³² Here, we defined the sensor response as the ratio of the sensor resistance in air to that in NH₃ gas. Therefore, the high initial resistance in air, induced by the catalytic Pt nanoparticles, significantly enhances the sensor response. Thus, our study documented that by effectively controlling the decoration of Pt nanoparticles on the surface of SnO₂ nanowires, where the correlation between Pt catalytic activity and the sensing performance was optimized, we could develop an excellent NH₃ gas sensor.

4. Conclusion

In this study, we successfully synthesized SnO₂ nanowires and Pt-decorated SnO₂ nanowire-based gas sensors with different sputtering times of Pt for NH₃ detection. The morphology, crystal structure, and electrical and gas sensing properties of the fabricated sensors were investigated. The Pt-decorated SnO₂ nanowire materials with a Pt sputtering time of 50 s (PS-50) exhibited the highest response as compared with others, where the highest response value of 19.98 for 500 ppm NH₃ at 300 °C was achieved. The PS-50 sensor also showed fast response/recovery times of 11/200 seconds and 6/9 seconds at 300 °C and 450 °C, respectively. The Pt–SnO₂ sensor could detect the lowest NH₃ concentration of 25 ppm. In addition, the Pt–SnO₂ sensor exhibited a 19-fold enhancement in response value as compared with that of the bare SnO₂ nanowires. The exceptional high response of the PS-50 sensor to NH₃ was attributed to the synergistic effect of the Schottky junction at the interface of Pt and SnO₂ and the catalytic spillover effect of Pt nanoparticles. Our results showed the potential of the Pt–SnO₂ nanowire sensor for applications in toxic gas detection.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article.

Acknowledgements

This work was financially supported by the Hanoi University of Science and Technology under project no. T2024-TĐ-008.

References

1. N. Tamaekong, C. Liewhiran, A. Wisitsoraat and S. Phanichphant, Sensors, 2010, 10, 7863–7873.
2. S.-J. Kim, I.-S. Hwang, Y. C. Kang and J.-H. Lee, Sensors, 2011, 11, 10603–10614.
3. I.-C. Chen, S.-S. Lin, T.-J. Lin, C.-L. Hsu, T. J. Hsueh and T.-Y. Shieh, Sensors, 2010, 10, 3057–3072.
4. M. Van Damme, L. Clarisse, S. Whitburn, J. Hadji-Lazaro, D. Hurtmans, C. Clerbaux and P.-F. Coheur, Nature, 2018, 564, 99–103.
5. L. Liu, X. Zhang, W. Xu, X. Liu, Y. Li, J. Wei, M. Gao, J. Bi, X. Lu, Z. Wang and X. Wu, J. Agric. Food Chem., 2020, 68, 3354–3361.
6. R. Zhang, B. Wang, X. Wang, K. Zeng and C. Guo, Colloids Surf., A, 2024, 702, 135037.
7. S. Li, Y. Yuan, Z. Shang, X. Yin, A. Sampaolo, P. Patimisco, V. Spagnolo, L. Dong and H. Wu, Photoacoustics, 2023, 33, 100557.
8. D. Ripepi, R. Zaffaroni, M. Kolen, J. Middelkoop and F. M. Mulder, Sustain. Energy Fuels, 2022, 6, 1945–1949.
9. L. T. Zegebreal, N. A. Tegegne and F. G. Hone, Sens. Actuators, A, 2023, 359, 114472.
10. D. Zhang, J. Wu, P. Li and Y. Cao, J. Mater. Chem. A, 2017, 5, 20666–20677.
11. D. Wang, D. Zhang, Y. Yang, Q. Mi, J. Zhang and L. Yu, ACS Nano, 2021, 15, 2911–2919.
12. D. Zhou, Z. Kang, X. Liu, W. Yan, H. Cai, J. Xu and D. Zhang, Sens. Actuators, B, 2023, 392, 134072.
13. N. Van Toan, C. M. Hung, N. D. Hoa, N. Van Duy, D. Thi Thanh Le, N. Thi Thu Hoa, N. N. Viet, P. H. Phuoc and N. Van Hieu, J. Hazard. Mater., 2021, 412, 125181.
14. C. Liu, H. Tai, P. Zhang, Z. Ye, Y. Su and Y. Jiang, Sens. Actuators, B, 2017, 246, 85–95.
15. Q. Li, W. Zeng and Y. Li, Sens. Actuators, B, 2022, 359, 131579.
16. W. Zheng, G. Lu, X. Liu, S. Fan, Y. Hu, N. Pinna and J. Zhang, Matter, 2025, 8, 101914.
17. J. Li, N. Zhao, X. Liu, X. Chang, W. Zheng and J. Zhang, Nano Lett., 2026, 26, 668–675.
18. Z. Wang, L. Zhu, S. Sun, J. Wang and W. Yan, Chemosensors, 2021, 9, 1–32.
19. S. Sen, A. Nilabh and S. Kundu, Microchem. J., 2021, 165, 106111.
20. Y. Xu, J. Xie, Y. Zhang, F. Tian, C. Yang, W. Zheng, X. Liu, J. Zhang and N. Pinna, J. Hazard. Mater., 2021, 411, 125120.
21. Y. Xu, W. Zheng, X. Liu, L. Zhang, L. Zheng, C. Yang, N. Pinna and J. Zhang, Mater. Horiz., 2020, 7, 1519–1527.
22. J. Shen, F. Li, B. Yin, L. Sun, C. Chen, S. Wen, Y. Chen and S. Ruan, Sens. Actuators, B, 2017, 253, 461–469.
23. (a) T. M. Ngoc, N. Van Duy, C. M. Hung, N. D. Hoa, N. N. Trung, H. Nguyen and N. Van Hieu, RSC Adv., 2018, 8, 36323–36330; (b) T. T. N. Hoa, N. V. Duy, C. M. Hung, N. V. Hieu, H. H. Hau and N. D. Hoa, RSC Adv., 2020, 10, 17713–17723.
24. N. Van Hieu, L. T. B. Thuy and N. D. Chien, Sens. Actuators, B, 2008, 129, 888–895.
25. D. T. T. Le, S. Iannotta, N. V. Hieu, C. Corradi, T. Q. Huy, M. Pola and M. Tonezzer, J. Nanosci. Nanotechnol., 2014, 14, 5088–5094.
26. S. Wang, L. Sang, Z. Jiao, F. Zhang, K. Wang, B. Xu, P. Zhang, B. Liu, Y. Wang, Y. Li and R. Hu, Appl. Surf. Sci., 2025, 687, 162257.
27. S. H. S. Pai, A. Mondal, R. Barathy T, B. Ajitha, J. J. Samuel E and Y. A. K. Reddy, Int. J. Hydrogen Energy, 2024, 50, 928–941.
28. C. Esmaeili, S. Ashtiani, C. Regmi, A. Laposa, J. Voves, J. Kroutil, K. Friess, V. Povolny and S. Lotfian, Heliyon, 2024, 10, 0–11.
29. S. Park, S. Park, J. Jung, T. Hong, S. Lee, H. W. Kim and C. Lee, Ceram. Int., 2014, 40, 11051–11056.
30. S. J. Chang, W. Y. Weng, C. L. Hsu and T. J. Hsueh, Nano Commun. Netw., 2010, 1, 283–288.
31. Y. Li, H. Ban and M. Yang, Sens. Actuators, B, 2016, 224, 449–457.
32. M. Shahabuddin, A. Sharma, J. Kumar, M. Tomar, A. Umar and V. Gupta, Sens. Actuators, B, 2014, 194, 410–418.
33. Y. Masuda, Sens. Actuators, B, 2022, 364, 131876.
34. N. V. Toan, C. M. Hung, N. D. Hoa, N. V. Duy, D. T. T. Le, N. T. T. Hoa, N. N. Viet, P. H. Phuoc and N. V. Hieu, J. Hazard. Mater., 2021, 412, 125181.
35. N. P. Hung, N. V. Duy, C. T. Xuan, D. T. T. Le, C. M. Hung, H. Jin and N. D. Hoa, RSC Adv., 2024, 14, 12438–12448.

---