Classification and authentication of quinoa, amaranth, and wheat flour mixtures by GC-MS combined with intelligent sensors

Quinoa and amaranth are gaining recognition as nutritious alternatives to traditional cereals, highlighting the need for reliable methods to authenticate their products. This study examined the volatile profiles of quinoa, amaranth, and wheat flours using GC-MS and an E-nose device. Twenty-four volatile compounds were identified, including limonene, 1R-α-pinene, and L-β-pinene, which were exclusive to pseudocereal flours, and hexanal, abundant in wheat flour as an oxidation indicator. The E-nose achieved 89.7 % accuracy in discriminating between quinoa, amaranth, and wheat flours and effectively separated double and triple mixtures. A PLS model revealed a strong correlation between E-nose data and concentrations of limonene, α-pinene, and β-pinene (R 2 CV = 0.94–0.95). The integration of GC-MS and E-nose proved highly efficient for flour authentication, with canonical discriminant analysis successfully identifying pseudocereal flours in mixtures with wheat flour, ensuring product quality and authenticity. • E-nose discriminated quinoa, amaranth, and wheat flours by aromatic signatures. • GC-MS differentiated flour profiles based on volatile organic compound composition. • Limonene, 1R-α-pinene, and L-β-pinene were key markers in pseudocereal flours. • PLS and PCA improved pattern recognition and VOC qualitative analysis. • E-nose validation showed it as a rapid, reliable tool for detecting adulteration.