Browsing by Author "Salin, Jarno"

Honey is viscose nutrient solution which is produced by honey bees from the nectar of the plants or excretions of plant sucking insects. For food use honey is collected from beehives in honeycombs which are centrifuged and the separated solution is canned. Honey consists mainly of carbohydrates and water but there are also in small amounts numerous other compounds such as proteins, vitamins, organic acids and phenolic compounds. Most of the carbohydrates in honey are fructose and glucose but also a great number of di-, tri-, and oligosaccharides can be found in smaller amounts. Composition of honey and therefore the carbohydrate profile can vary greatly depending on the honeys botanical origin, soil, climate and other factors affecting the growing environment. The goal of this study was to develop a chromatographic method using high performance anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD) to determine carbohydrate profiles for domestic unifloral and mixed honeys. Principal component analysis (PCA) was used to investigate the differences in determined carbohydrate profiles. Equipment used in this study consisted of chromatographic separations module, electrochemical detector and Dionex CarboPac PA1 column. Milli-Q water, 100 mM sodium hydroxide solution and 100 mM sodium hydroxide solution with 0,5 M sodium acetate buffer were used as eluents. 33 domestic unifloral and mixed honeys of 12 different botanical origins were used as samples. Approximately 0,5 g of samples were weighed in three replicates to 50 ml volumetric flasks which were filled with Milli-Q water. From these solutions 1:5 and 1:200 dilutions were made and they were filtered to HPLC vials through 0,45 µm PVDF-filter before analysis. Carbohydrates in samples were identified using a standard mixture with 16 carbohydrate standards. Carbohydrates were quantified from chromatograms using the height of detected peaks. Ten different carbohydrates could be separated from the honey samples for determination of carbohydrate profiles. Most of the samples contained fucitol, fucose, glucose, fructose, turanose, nigerose and maltose. Some samples contained also trehalose and rhamnose. Raffinose was quantitated from one honeydew sample. In addition melezitose/isomaltose, kojibiose/1-kestose and β-gentiobiose were detected in samples but quantitation of these carbohydrates could not be done because of poor resolution. Concentrations of the quantitated carbohydrates varied greatly even between honey samples of same botanical origin. PCA revealed that honeydew honeys differed clearly from other unifloral honeys according to their carbohydrate profiles. The PCA model explained circa 64 % of variance in the whole dataset. Differed honeydew samples were removed from second PCA model. In this model Himalayan Balsam, cloudberry and willowherb honeys could be separated as individual groups from the other unifloral honeys. This model explained circa 40 % of variance in the whole dataset. However in PCA most of the unifloral honeys formed a group where no clear differences could be assessed between samples of different botanical origin. Moreover some samples aroused a suspicion over their actual botanical origin. The developed method could be used to separate honeydew honey samples from blossom honey samples. The differences among other unifloral honey samples were not clear although some of the unifloral honeys showed some signs of unique carbohydrate profile. On the other hand the number of samples used in this study was quite small. Therefore strong conclusions about the differences among carbohydrate profiles of different unifloral honeys could not be made. For future research greater number of samples is required in order to verify the potential differences found in this study.