Dynamic working temperature modulation has received considerable attention as a way to enable metal oxide-based gas sensors to recognize the detected gases. Previous reports mainly focused on heating-voltage waveform manipulation and algorithm improvement, not involving the insight of the temperature modulation sensing behavior correlated with fine nanostructures of metal oxide. Here, three types of SnO2 nanoparticles with different percentages of high-energy {221} and low-energy {110} crystal facets were synthesized and assembled as gas sensors. Under a working temperature modulation by applying a square wave pulse heating voltage, their dynamic response behaviors toward volatile organic compounds (VOCs) including alcohols, aldehydes, ketones, amines, and aromatic compounds were systematically explored. The results indicated that they exhibited different characteristic response curves toward different VOCs, presenting the crystal facet-dependent temperature modulation sensing behaviors. This effect was analyzed further and explored in conjunction with the characteristic response curves. Based on their different characteristic response curves that occurred on the three types of SnO2 nanoparticles, the detected gases, and even some congeners, were well discriminated along with a simple PCA recognition algorithm. Additionally, their quantitative analysis was also achieved.
With the advancement of smart detection technologies, gas sensors have gained increasingly widespread applications in environmental monitoring, industrial safety, medical diagnostics, and intelligent agriculture. The response curve of a single gas sensor is conventionally used to calculate sensor performance evaluation parameters of response/recovery time and response value. The selectivity of a single sensor is currently evaluated by a selectivity coefficient based on the response value, which faces difficulty in distinguishing gases with different concentrations. In this study, a new metric of fill factor (FF) derived from response curve morphology features was introduced to offer a novel solution for single-sensor selectivity analysis. Three variants of the novel metrics, FFres (response phase fill factor), FFrec (recovery phase fill factor), and FF (total fill factor), can be calculated from the response curve. As a proof-of-concept, the FFs of WS2, WS2@PANI, and Au/PANI/WS2 sensors at different concentrations of ammonia, acetone, methanol, ethanol, and water vapor were calculated and used for gas selectivity analysis. Results demonstrated distinct FF ranges for ammonia and other gases, validating FF as a potential selectivity indicator. This approach provides a multidimensional analysis framework for enhanced single-sensor discrimination by constructing a parameter space incorporating curve morphology features.
危险品化学品经营许可证（带存储）