Creating locally enriched single-atom catalysts via bio-inspired internalization process for interference-resistant NO2 detection in complex environments.

Intelligent gas-sensing technology that accurately and stably identifies gas categories in complex atmospheres is critical for protecting public safety and environment. However, competing interactions among different gases on sensing surfaces can trigger interference even poison of sensors. Here, we present a bio-inspired atomic internalization process that produces locally enriched single Pt species, enabling highly selective and interference-resistant gas detection. By engineering Sn/C precursors as homologous receptors, Pt3Sn alloys are two-step redistributed into high-density single Pt species within regionalized SnO2 surface. This structure constitutionally alters the distribution patterns of NO2 molecules and thus delivers accurate NO2 monitoring in multicomponent atmospheres and stable detection for over 550 days, surpassing state-of-the-art commercial devices. Further integrating characteristic-specific sensors into arrays, the resulting device further achieves 100% classification accuracy for single and mixed gases at ultralow cost. Moreover, we demonstrate an autonomous "cruise-monitoring" system by equipping the sensor on a robot to detect and identify NO2 in real time. Our findings thus guide the accurate analysis and interference-resistant detection in complex gas mixtures.