Reliable quantification of structurally similar flavonoids in complex matrices remains challenging due to spectral overlap and matrix interference. Here, we report a semiconductor surface-enhanced Raman scattering (SERS) sensing platform based on an interface-engineered N-doped anatase/rutile TiO2 phase junction that enables multiplex quantification through a charge-transfer enhancement mechanism. The engineered interface synergistically modulates the electronic structure, facilitating efficient charge-transfer resonance under 785 nm excitation and thereby improving Raman sensitivity. The optimized platform enables quantitative detection of representative flavonoids with limits down to 5 × 10–6 M, which are much below the typical concentration levels of flavonoids in nutraceuticals, and exhibits good signal reproducibility. To resolve overlapping spectral features in multicomponent systems, partial least-squares regression is integrated with SERS, enabling accurate multiplex quantification of structurally similar flavonoids in complex mixtures. The sensing strategy is further validated using crude extracts of Scutellaria baicalensis, where the quantified compositions show strong agreement with liquid chromatography–mass spectrometry results, demonstrating robustness in real-world samples with minimal pretreatment. This work establishes an interface-engineered semiconductor SERS platform for multiplex chemical sensing and highlights its potential for rapid analysis in complex natural systems.
Song et al. (Thu,) studied this question.