Characterization of rice cultivars using Raman spectroscopy and multivariate analysis

Abstract

Rice is a widely consumed cereal known for its high carbohydrate content and the presence of essential nutritional elements such as proteins and lipids. The previous Raman spectroscopy (RS) studies on whole brown rice kernel have only been able to partially reveal nutritional content. This study employs RS to accurately assess the nutritional profiles of forty rice cultivars in powdered form, facilitating their characterization. The combination of RS and advanced chemometric techniques enables the discrimination of rice cultivars based on key biochemical attributes, including amylose and starch content, aromatic properties, protein content, and total antioxidant capacity, resulting in an impressive accuracy rate of 73%. Significant observations in the fingerprint region included specific Raman spectral features associated with glucose ring stretching at ∼478 cm⁻¹ and indicative stretching vibrations for glycosidic bonds and amylose structures at ∼409 cm⁻¹, ∼439 cm⁻¹, ∼521 cm⁻¹, and ∼868 cm⁻¹. Furthermore, prominent spectral bands associated with amino acids such as phenylalanine (∼1004 cm⁻¹, ∼1580 cm⁻¹, ∼1606 cm⁻¹), tryptophan (∼872 cm⁻¹, ∼1360 cm⁻¹, ∼1543 cm⁻¹, ∼1620 cm⁻¹), and tyrosine (∼853 cm⁻¹, ∼1176 cm⁻¹, ∼1556 cm⁻¹) were observed. Additionally, characteristic peaks related to protein content, carotenoids, and amino acids were identified at ∼1084 cm⁻¹, ∼1111 cm⁻¹, ∼1162 cm⁻¹, ∼1263 cm⁻¹, ∼1310 cm⁻¹, and ∼1404 cm⁻¹. These findings underscore the potential of RS as a fully automated, rapid, and accurate tool for evaluating the nutritional content of rice cultivars.