Exploring Surface-Enhanced Raman Spectroscopy (SERS) Methods for Rapid Determination of Ofloxacin Residues in Egg White

In this investigation, surface-enhanced Raman spectroscopy (SERS) combined with multivariate analysis was investigated for detection and characterization of ofloxacin residues in egg white. Gold nanoparticles (AuNPs) were meticulously synthesized to serve as the substrate for enhancing Raman signals. The SERS characteristic peaks of ofloxacin were conducted using the density functional theory (DFT) calculations. In an effort to optimize the quantitative detection conditions for ofloxacin residues in egg white, the optimal detection parameters were determined via a single-factor experiment. The optimal combination involves the homogeneous mixing of 500 μL of AuNPs, 75 μL of magnesium sulfate solution, and 20 μL of the test solution, followed by a 1-minute adsorption period prior to detection. Following a meticulous comparison of seven distinct spectral preprocessing methodologies, the utilization of adaptive iterative re-weighted penalized least squares (air-PLS) combined with savitzky-golay (SG) preprocessing for the spectral datawithin 400 to 1800 cm-1 were finally selected. The principal component analysis (PCA) helped establish an optimal support vector regression (SVR) model, which was subsequently employed to quantitatively detect the concentration of ofloxacin residues in egg white. It is noteworthy that the SVR model demonstrated outstanding performance metrics, with an R-squared prediction (R2 p) value of 0.9896, root mean square error of prediction (RMSEP) of 1.6246, and residual predictive deviation (RPD) of 8.1720. The findings indicate that the SERS-based multivariate analysis technique can swiftly and precisely identify the ofloxacin residues in egg white.