[5-1015-A] Food Safety (2)

[5-1015-A-02] Application of Fluorescence Spectroscopy for the Classification of honey based on Geographical Origin

*Abdullah Iqbal^1,2, Mizuki Tsuta¹ (1. Food Research Institute, National Agriculture and Food Research Organization 2-1-12 Kan-nondai, Tsukuba, Ibaraki 305-8642 Japan (Japan), 2. Dept. of Food Technology & Rural Industries, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh(Bangladesh))

Keywords:Honey, chemometrics, classification, geographic origin

The Front-face fluorescence spectroscopy was applied in this study for the classification of honey based on geographical origin. Honey samples (Robinia pseudoacacia and Blended floral source) of different origin (i.e., China, Hungary and Japan etc) used in this study were collected from their production sites. Before the fluorescence measurement, the samples were put in shaking water bath at 60℃ for 30 min with 100 rpm shaking speed. Then after stirring to obtain the homogeneity, the honey samples were diluted to 100 times with the addition of 20% (v/v) ethanol solution. The front-face fluorescence excitation-emission matrices were then recorded from 200nm to 800nm (at an interval of 1 nm) whereas excitation spectra were recorded between 200nm to 500nm (with an interval of 5nm). With the application of necessary pre-processing (i.e., normalization, mean centering, autoscaling and/or combination thereof) and digital smoothing polynomial filters (i.e., Savitzky-Golay smoothing filters) for smoothing out the noisy signals, the rayleigh scattering rays were removed from the spectra. The chemometric analysis were then applied to the spectral data using principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) for classification of the honey samples. A reasonable sensitivity (ranging from 0.90 to 1.000) and specificity (ranging from 0.795 to 1.000) for class predictions was obtained from the PLS-DA model. The results showed that front-face fluorescence spectroscopy has potential for the discrimination of Robinia pseudoacacia honey based on geographical origin. But it is not possible to discriminate the blended samples based on geographical origin.