作者: Xavier Cama-Moncunill , Maria Markiewicz-Keszycka , Raquel Cama-Moncunill , Yash Dixit , Maria P. Casado-Gavalda
DOI: 10.1016/J.IDAIRYJ.2018.04.014
关键词: Coefficient of determination 、 Laser-induced breakdown spectroscopy 、 Atomic absorption spectroscopy 、 Analytical chemistry 、 Spectroscopy 、 Calibration 、 Emission spectrum 、 Laser 、 Partial least squares regression 、 Materials science
摘要: Laser-induced breakdown spectroscopy (LIBS) was employed to predict the sodium content of infant formula (IF) over range 0.5–4 mg Na g−1. Calibration models were built using partial least squares regression (PLS), correlating LIBS spectral data with reference quantified by atomic absorption (AAS). The main aim this study demonstrate ability as a rapid tool for quantifying in IF, but also explore strategies concerning acquisition and pre-processing spectra. A different techniques, measuring depths (repetition laser shots) accumulations conducted evaluated terms PLS performance. best calibration model developed third-layer spectra normalised H I 656.29 nm emission line, yielding coefficient determination (R2) 0.93, root-mean-square errors (RMSE) 0.37 0.13 g−1 cross-validation validation, respectively.