Laser-induced breakdown spectroscopy for food safety analysis: sample presentation, data interpretation, and...

Food-safety decisions require rapid screening of adulteration, origin fraud, elemental contamination and freshness loss. Laser-induced breakdown spectroscopy (LIBS) provides direct multi-element fingerprints with short acquisition times and little reagent consumption. Its performance in foods, however, depends on a coupled analytical chain: sample presentation controls laser coupling and mass removal; plasma evolution and acquisition govern emission; calibration and modelling determine interpretation; and validation establishes reliability beyond the development dataset. This critical review uses that chain to assess food-related LIBS workflows. It examines powders, moist tissues, fatty materials, liquids and spatially heterogeneous samples, then considers plasma-aware acquisition, preprocessing, chemometrics, deep learning and multimodal fusion. Task-specific comparisons with ICP-based methods, near-infrared spectroscopy, Raman spectroscopy and hyperspectral imaging clarify where LIBS adds value. Pelletization provides a reproducible baseline for many powders, whereas moist, fatty and liquid matrices often require flattening, phase control, immobilization or enrichment. Readiness is strongest for selected mineral-related authenticity screens and elemental warning at concentrations clearly above a decision threshold, provided that independent validation and quality control are available. Closed-set provenance and portable sorting are generally less mature, while regulatory trace confirmation, cross-instrument transfer and realistic microbial analysis require stronger evidence. Progress depends on end-to-end timing, threshold-focused validation, interpretable models, drift monitoring and selective confirmation with complementary techniques.