Matrix-Guided Size Selection of Plasmonic Nanoprobes for Improving Quantitative Robustness of SERS Lateral Flow Immunoassays

Surface-enhanced Raman scattering-based lateral flow immunoassays (SERS-LFIAs) have emerged as a promising platform for rapid food safety analysis. However, their quantitative reliability is often compromised by matrix-dependent variations during lateral flow, and the influence of probe size on analytical performance remains insufficiently understood. In this work, Au@DTNB immunoprobes with diameters of 15, 30, and 50 nm were prepared to systematically investigate the effect of nanoparticle size on the quantitative performance of SERS-LFIA for aflatoxin B1 (AFB1) detection. All three probe sizes exhibited excellent concentration-dependent responses and generated four-parameter logistic calibration curves with correlation coefficients (R2) of 0.997, 0.994, and 0.992, respectively. Although the 50 nm probes produced the strongest Raman signals, recovery experiments in rice and soybean samples revealed that the 15 nm probes provided the highest quantitative accuracy, with recoveries of 90.1–109.2% and 100.4–108.9%, respectively, and relative standard deviations below 7.2%. In contrast, larger probes exhibited progressively greater deviations from the spiked concentrations despite their stronger SERS responses. These results demonstrate that maximizing Raman signal intensity alone does not necessarily improve quantitative performance in complex food matrices. Instead, probe size should be optimized according to matrix characteristics to achieve reliable quantitative analysis. This work establishes a matrix-oriented strategy for probe size selection and provides practical guidance for the rational design of quantitative SERS-LFIA platforms for food safety analysis.