Classification of honeys from calabria using diffuse-light absorption spectroscopy

Optical spectroscopy has emerged as a rapid and non-destructive technique for distinguishing, classifying, and authenticating honeys according to their floral origin 1,2 Indeed, it is an ideal candidate for “green” analyses, since it does not use reagents or chemical treatments, thus avoiding the problem of waste discharge and the related environmental side effects. This work considers a collection of 13 honeys produced in Calabria, a region of Southern Italy. They belonged to three floral origins: 3 of acacia, 5 of citrus, and 5 of chestnut, respectively. They were measured using diffuse-light absorption spectroscopy in the visible spectroscopic range. This technique makes use of an integrating sphere that contains the honey sample in a vial. Almost all the light impinging on the sphere surface is diffusely reflected, and the detector can be placed anywhere in the sphere in order to gather the average flux. Because of the intrinsic optical absorption of honey, a reduction of the radiance in the sphere occurs. The reduction is related to the absorption of honey and to its volume, and is independent of non-absorbing objects within it, such as suspended scattering particles. In practice, diffuse-light absorption spectroscopy provides spectroscopic information which is intrinsically independent from crystallization phenomena, which are very frequent in honey. The Principal Component Analysis (PCA), which is one of the most popular techniques for explorative analysis and data dimensionality reduction, was used for processing the spectroscopic data. Figure 3 shows the score maps in the PC1-PC2 space which successfully grouped the honey samples in three distinct clusters, corresponding to the acacia, citrus, and chestnut floral variety, respectively.