Fast spark discharge-laser-induced breakdown spectroscopy method for rice botanic origin determination

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Accurate botanical origin determination is crucial for rice quality control and authentication.
• Traditional methods for rice variety identification can be time-consuming and labor-intensive.
• Developing rapid, efficient, and non-destructive analytical techniques is essential.

Purpose of the Study:

• To develop a simple, fast, and efficient spark discharge-laser-induced breakdown spectroscopy (SD-LIBS) method for determining rice botanical origin.
• To apply predictive modeling using support vector machine (SVM) for classifying rice varieties.
• To evaluate the robustness and eco-friendliness of the developed SD-LIBS method.

Main Methods:

• Analysis of 72 rice samples from four varieties (Guri, Irga 424, Puitá, Taim) using SD-LIBS.
• Selection of spectral lines (C, Ca, Fe, Mg, N, Na) as input variables for the SVM prediction model.
• Optimization of SVM algorithm parameters using a central composite design (CCD) for enhanced classification performance.

Main Results:

• The optimized SVM model achieved 96.4% correct predictions on test samples.
• Sensitivities and specificities per class ranged from 92% to 100%.
• The method demonstrated consistent and reproducible results, indicating robustness.

Conclusions:

• The developed SD-LIBS method is a robust, eco-friendly, and efficient tool for rice botanical identification.
• The technique offers high accuracy and reliability in discriminating between different rice varieties.
• This approach avoids chemical waste, aligning with green analytical chemistry principles.