Volatiles as chemical markers suitable for identification of the geographical origin of green Coffea arabica L

Abstract

Traceability of geographical origin is a challenging matter for producers and consumers. This research focuses on identifying green Coffea arabica from Africa, South and Central America using volatiles and aqueous-soluble chemical compounds for the first time. Using Linear Discriminant Analysis, we created two models. The first one focused on aqueous soluble compounds and their properties (pH, total antioxidant capacity, total polyphenolic content, caffeine, and chlorogenic acids) showed 91.30% accuracy of identification and, during cross-validation, predicted 82.61% correct identification. The 83.36% of the variability was explained with caffeine and TAC. The second focused on volatiles correctly identifying 100% of testing samples and predicted 86.96% accuracy in cross-validation. 91.17% of the variability between African, South, and Central American coffees was explained based on ketones, aldehydes, organic acids and esters, nitriles, alcohols, and alkenes, while ketones appeared as the strongest parameter among volatiles.