Ensuring the authenticity and quality of honey, particularly for monofloral type, is essential due to high market value and vulnerability to adulteration [1]. This study investigates the potential of high-capacity sorptive extraction (HiSorb), a novel and environmentally friendly tool [2-3], for profiling volatile organic compounds (VOC) in honey from diverse botanical sources and geographical regions.
Headspace VOCs were extracted using HiSorb probes on an automated Centri platform. Sampling conditions were optimised for three different sorbent phases by means of design of experiment (DoE) evaluating extraction time, temperature and sample weight. The combination of this systematic sampling strategy with gas chromatography separation and mass spectrometry detection (GC-MS) enabled the acquisition of highly informative chromatographic data.
Over 80 samples of Robinia pseudoacacia (L., acacia) and Coriandrum sativum (L., coriander) honey from different Italian regions were analysed to investigate differences between their VOC profiles and to find potential markers for botanical and geographical origin. A total of 113 compounds were putatively identified using linear retention index and mass spectra similarity to commercial libraries. These compounds were compiled into a custom-made library for data mining purposes. The combination of univariate and multivariate statistical analysis including analysis of variance (ANOVA), principal component analysis (PCA) and data reduction with random forest machine learning algorithms facilitated the separation of 30 key differentiating compounds based on botanical source and 21 potential markers for geographical origin. This study proves the effectiveness of HiSorb-GC-MS in discriminating honey aroma and identifying authenticity markers. Furthermore, it lays the foundation for constructing a comprehensive database of compounds to enhance the separation of high-quality honey from adulterated products. The integration of a novel green sampling technique, along with advanced statistical tools and correlation analyses, underscores the potential for streamlined workflows in honey authentication, thereby improving the robustness and reliability of the results.
N. Żak, A. Wilczyńska, Foods, 2023, 12, 3210.
L. Hearn, R. Cole, N.D. Spadafora, R. Szafnauer, Adv Sample Prep., 2022, 3, 100032.
D. Eggermont, N.D. Spadafora, J. Aspromonte, R. Pellegrino, G. Purcaro. Anal. Bioanal. Chem., 2023, 415, 4501.