On-Chip Terahertz Interferometric Sensing Based on Focused Surface Plasmonic Field and Spatial Spectral Fusion Methodology

Terahertz trace sensing is inherently constrained by weak light-matter interactions. To address this fundamental bottleneck, we propose a novel sensing paradigm based on surface plasmonic (SP) field interference, transitioning from conventional free-space spectroscopic sensing to on-chip surface optical field sensing. Leveraging the extraordinary confinement and focusing capabilities of SP fields, the localized light intensity is substantially enhanced, thereby enhancing the efficiency of light-matter interactions. More importantly, by employing orthogonal slit antenna pairs as coherent surface wavelet sources and a three-stage coherent superposition mechanism, two sets of focused surface waves construct an interferometric field with precisely tunable optical path differences. Ultrasensitive detection is realized through the analysis of focal interference spectra. Experimental validation based on the hydrolysis of acetylcholinesterase demonstrates that the sensing platform attains a detection limit (LOD) as low as 3.125 μg/mL. Furthermore, by fusing 2D surface spectral data with machine-learning algorithms, accurate prediction of enzyme concentrations in the range of 3.125–50 μg/mL is successfully achieved. This work establishes an on-chip precision interferometric sensing technique integrating photonic integration technology and data fusion, representing a breakthrough advancement in the on-chip integration level, detection accuracy, and operational robustness of terahertz sensing systems.