Heterointerfacial Phonon Engineering for Source-LevelNoise Reduction in Gas Sensing

Abstract Reliable gas sensing in dynamic environments is essential for safety, environmental surveillance, and health monitoring. Motion-induced noise easily masks weak signals and compromises the low limit of detection (LOD). Existing circuit- and algorithm-based methods mainly reduce noise during propagation or postprocessing. However, their effectiveness becomes limited when weak signals overlap with nonstationary noise. Herein, we propose a source-level noise reduction strategy based on continuous nanoscale heterointerfacial phonon engineering. The continuous heterointerface promotes interfacial scattering and relaxation of nonequilibrium phonons, thereby reducing their perturbation into charge transport. The MoBTx/ZnO sensor exhibits an ultralow noise of 0.0005%, achieving one order lower than the current methods. It detects 5 parts-per-billion (ppb) NO2 with a high signal-to-noise ratio of 7200. The theoretical LOD is 0.0088 ppb. This strategy embeds intrinsic noise resilience into the sensing mechanism and provides a generalizable nanoscale route toward robust weak-signal detection in resistance-based sensing under dynamic operation.