Back to search

Agricultural University of Athens

Προσδιορισμός χημικών δεικτών και ανάπτυξη χημειομετρικών μοντέλων ως εργαλεία ταυτοποίησης της αυθεντικότητας μελιών

Abstract

dc:description

The European Union's focus on research projects to promote and protect the iconic products of its member states has been strong in recent years. Greece is one of the regions of the Balkan Peninsula with intense vegetation and biodiversity, which contributes to the production of high-quality honey of great commercial and research interest. The present thesis aimed at the isolation, detection and identification of volatile and non-volatile fractions of endogenous Greek honeys through instrumental chemical analysis techniques and the use of the results for the development of chemometric authenticity models. The study focused on the optimization of volatile fraction extraction, the study of phenolic compounds from selected samples, the recording of spectroscopic data and the creation of a database that will address honey authenticity issues with the help of chemometrics. The study started with an extensive literature review, highlighting the limitations of traditional physicochemical methods for the detection of the botanical origin of honey and the potential of instrumental chemical analysis techniques. The review highlighted the need for accurate and reliable methods that standardize monofloral Greek honeys and do not compromise consumer confidence and the reputation of the honey industry. The sampling of 655 honey samples was carried out in the framework of two research projects in conjunction with professional beekeepers during the 2018 and 2021 beekeeping season. The samples were analyzed according to European and Greek legislation and stored according to good practice standards. In the development and optimization of the surface response methodology for volatile components, varieties with strong economic, export and research interest were selected. Of the at least 31 volatile compounds isolated from thyme honey using the SPME technique and GC-MS chromatography analysis, 10 were considered to be the most important. Similarly, of the at least 49 volatile compounds from the Erica honey, 13 were considered to be the most important contributors. In both cases, temperature, extraction time and water: honey ratio (v/m) were the most important factors with a positive effect on recovery for most volatile compounds. Equilibration time ensured qualitative and quantitative reproducibility, while sample volume was related to the isolation of minor compounds. Also, the effect of magnetic stirrer speed was found to be quite significant in the context of predictive models. In each case, a set of external validation tests was used to document the method. In the next part of the study, the use of GC-MS instrumental analysis in combination with the Stepwise-LDA chemometric technique helped to construct a tool for authentication of the four most commercial honeys in Greece. Fir, thyme, pine and citrus honeys were correctly classified at 85.5% with cross-validation of 82.3% and external validation of 88.5%. Also, the same chemical analysis was used to construct an gas-chromatographic database of 596 samples. In this way, the main volatile compounds that could give the "type" of honey were identified and compounds with a significant presence in more than one type of honey were detected, such as benzaldehyde, furfural, octane, nonanol, nonanal, 2-phenyl acetaldehyde, and some methyl esters. The analysis of phenolic constituents using LC-Qtof-MS on 82 pure varieties including honey from thyme, pine, fir, heather, cotton, chestnut, sunflower and orange showed the detection of at least 28 compounds. Of these compounds, 23 were found at high levels, while some could potentially be chemical indicators of botanical origin for some of them. FTIR spectroscopy recorded spectroscopic data and in combination with the Stepwise-LDA chemometric technique accurately (correct classification 93.5%, cross-validation 82.5%, external validation 84.6%) classified the most commercial honey samples based on their botanical origin. Raman spectroscopy, compared to traditional physicochemical analyses of honey, appeared clearly more powerful as a tool for botanical discrimination of fir, pine and thyme honeys. According to the results, the percentage of correct classification of honey using Raman was 95.3%, 90.6% and 84.3% in terms of correct classification, internal and external validation, respectively. Differentiation based on registration with right angle FS registration proved to be another reliable tool. A total of 87 honey samples from four botanical cultivars (21 thyme, 18 pine, 21 fir and 16 citrus) were classified with high classification rates based on two fluorometric ranges (λex = 330 and 360). In the first case (λex = 330), the percentage of correct classification, internal and external validation was 94.9%, 94.9% and 82.1%, respectively. While in the second case (λex = 360) it was 81.4%, 79.7% and 82.1%, respectively. In both cases, the results confirmed that the fluorescent phenolic compounds absorbing at these wavelengths are related to the botanical origin of the pure melliferous olives. In conclusion, this PhD study successfully developed chemometric authenticity models for Greek honey using both chromatographic and spectroscopic data. The developed chemometric models offer a practical tool for honey producers, regulatory authorities and quality control laboratories to verify the authenticity and identity of Greek honey. By applying these models, producers and traders can protect consumers and maintain the integrity of the honey industry. The study also highlights the importance of continued research and development in the field of honey authenticity in conjunction with instrumental chemical analysis.

Degree

thesis:*
Grantor dc:publisher
Agricultural University of Athens
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Xagoraris, Marinos
  • Ξαγοράρης, Μαρίνος

Subjects

dc:subject × 41

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/54764

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Xagoraris, Marinos; Ξαγοράρης, Μαρίνος. Προσδιορισμός χημικών δεικτών και ανάπτυξη χημειομετρικών μοντέλων ως εργαλεία ταυτοποίησης της αυθεντικότητας μελιών. Agricultural University of Athens, 2023. http://hdl.handle.net/10442/hedi/54764