Discrimination between domestic and imported red chili powders using laser-induced breakdown spectroscopy: variable...

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Consumer trust in Korean red chili powder relies on authenticity.
• Distinguishing domestic from imported chili powder is a significant challenge.

Purpose of the Study:

• To develop a rapid, cost-effective, and reproducible method for authenticating domestic red chili powder.
• To classify commercial red chili powder samples using Laser-Induced Breakdown Spectroscopy (LIBS) and machine learning.

Main Methods:

• Collected LIBS spectra from 79 commercial red chili powder samples (43 domestic, 36 imported).
• Selected nine representative elemental emission lines (Mg, Li, Rb, Ca, K, Na, H, C, O) using interclass-distance analysis.
• Modeled data using LDA, k-NN, NB, SVM, and RF algorithms with systematic variable selection.
• Validated models using leave-one-out cross-validation (LOOCV) and bootstrapped random-split validation.

Main Results:

• Highest LOOCV accuracy of 97.5% achieved with only two or three variables (e.g., Mg + Li).
• Robust discrimination confirmed by consistent mean accuracies across 100 independent validation partitions.
• Mg, Li, and Rb (occasionally Ca) were consistently selected, indicating agro-environmental and geological differences.

Conclusions:

• LIBS combined with principled variable selection offers a reliable framework for differentiating domestic and imported red chili powders.
• A compact feature set is sufficient for accurate classification, enabling rapid and cost-effective authentication.
• The method supports consumer trust by ensuring the authenticity of Korean red chili powder.