Development of an Integrated HS-SPME/GC–MS and a Chemometric Method for the Classification of Specialty Arabica Coffee Beans

Abstract The study and characterization of specialty coffees are essential to ensure the superior quality and traceability of these premium products, given growing market demand, the lack of standardized quality, and the risk of fraud. This study investigates the volatile profiles of specialty coffee beans from different producing countries, obtained with various post-harvest processing methods and different roasting levels. To analyze the samples, an integrated approach using headspace solid-phase microextraction combined with gas chromatography-mass spectrometry (HS-SPME/GC–MS) and chemometric techniques, including principal component analysis (PCA) and linear discriminant analysis (LDA), was developed and optimized. Considering the limited number of samples for each class, the LDA models achieved a 100% correct classification rate in training for all classes, except for Ethiopian green beans, which were classified correctly with a rate above 85%. The results of the validation process are also acceptable, with a minimum correct classification of 50%. Furthermore, some volatile compounds could be considered potential product or process markers, such as furan compounds, mainly associated with the degree of roasting. These compounds also contribute to the flavor of the final beverage. Preliminary results suggest that the optimized method could represent an important tool for authenticity assessment and environmental valorization of specialty coffees, with potential applications in quality control and product or process certification.