29-8 |
Use of high-performance liquid chromatography and chemometrics for analysis of honey samples |
L. M. REID1, N. C. Shilton1, and G. Downey2. (1) Dept. of Agricultural & Food Engineering, Univ. College Dublin, Earlsfort Terrace, Dublin, 2, Ireland, (2) National Food Centre, Dunsinea, Castleknock, Dublin, 15, Ireland The desire of consumers for high quality food has never been greater due to public health scares relating to the authenticity and safety of foods. This has resulted in the need for effective laboratory techniques capable of determining the authenticity and origin of foods. Honey is a widely produced food and several different varieties of it are produced from various botanical sources to meet the tastes of the consumer. High-performance liquid chromatography (HPLC) has proved useful for detecting specific marker compounds in honey samples. However the possibility of using chemometrics to analyse the chromatographic data and to differentiate between samples has received little attention in the past. The objective of this study was to investigate the possibility of using HPLC in conjunction with chemometric data analysis to analyse and categorise 82 honey samples from different botanical origins. HPLC was carried out on the 82 honey samples and chromatograms were collected using a UV detector (260, 290, 310 and 340 nm) and a refractive index detector. The chromatographic data was then subjected to Principal Components Analysis (PCA) for classification studies. The refractive index chromatograms allowed the sugar profile of the honeys to be seen, and all honeys contained mainly fructose and glucose. When the chromatographic data was subjected to chemometric analysis, the samples of heather honey were distinguished from other types of honey. This distinction between samples was most clear at 260 nm, an absorption wavelength for fructose. However only 90% of the variance between samples was accounted for at this wavelength, suggesting that fructose differences were not the only source of variance between the samples. The results suggest that the combination of HPLC and PCA analysis offers potential for the categorisation of honey samples from different botanical sources.
Session 29, Carbohydrate
|