Related Experiment Video
Updated: Sep 9, 2026
Real-time Breath Analysis by Using Secondary Nanoelectrospray Ionization Coupled to High Resolution Mass Spectrometry
Published on: March 9, 2018
SnSe/SnO2 Heterojunction-Based Single-Sensor Virtual Electronic Nose with Low Reaction Barrier for Trace
Jiongyue Hao1, Hangyu Liu1, Chun Huang1
1Key Laboratory of Optoelectronic Technology and Systems of the Education Ministry of China, College of Optoelectronic Engineering, Chongqing University, Chongqing400044, China.
Abstract:
Illicit methamphetamine (MA, ice) trafficking poses a severe global threat to public security, while non-contact on-site detection of MA remains a grand challenge due to its ultra-low saturated vapor pressure at room temperature (25 °C). Volatile sulfur compounds (VSCs), including hydrogen sulfide (H2S), methanethiol (CH3SH), and dimethyl sulfide (C2H6S, DMS), are stable characteristic markers released throughout the entire MA production, purification, storage, and transportation chain. Herein, we develop a single-sensor virtual electronic nose based on SnSe/SnO2 p-n heterojunction for precise VSC identification in illicit MA scenarios. The SnSe/SnO2 heterojunction, synthesized via a facile liquid-phase ultrasonic method, exhibits significantly enhanced room-temperature H2S sensing adsorption kinetics, delivering a high response of 2.57 (response = Ra/Rg, where Ra is the resistance in air and Rg is the resistance in the atmosphere, including H2S and air) to 2.5 ppm H2S in air, a low detection limit of 75 ppb, excellent anti-sulfur poisoning ability, and long-term stability. Combined with density functional theory (DFT) calculations, we reveal that interfacial charge redistribution in SnSe/SnO2 optimizes reactant adsorption energy, lowers the energy barrier of electronic modulation during H2S interactions to only 0.157 eV, and amplifies the adsorption-desorption kinetic differences between distinct VSCs. By extracting 12-dimensional static features from single-device response curves without any external modulation, we construct a PCA-KNN machine learning framework that achieves an overall identification accuracy of 79.65% for six VSC systems, with an ultra-high accuracy of 96.7% for target H2S even in complex mixtures. By circumventing the inter-sensor variance and complexities typical of multi-sensor arrays, this single-sensor virtual electronic nose provides a highly integrated, physically compact, and high-precision solution for on-site non-contact traceability of illicit MA trafficking.
