USE OF VISIBLE AND NEAR-INFRARED SPECTROSCOPY FOR DISCRIMINATION OF EUCALYPT SPECIES BY EXAMINATION OF SOLID SAMPLES

摘要

Eucalypt species have high importance in the forestry sector in Brazil. The wood has wide range of uses, including pulp, charcoal, plywood panels and treated boards. However, eucalypt species are hard to identify in lumber, which can cause problems in industrial use. So, the aim of this study was to combine visible and near-infrared spectroscopy to discriminate wood samples of two Corymbia spp. and five Eucalyptus spp. Visible spectroscopy was performed according to the CIELab standard. The data were analysed using descriptive statistics and regression analysis. Near-infrared analyses were based on ASTM E1655-05. The data were analysed with the Unscrambler X chemometric program, in the raw form and after pre-processing with second derivative of Savitzky-Golay. Visible and near-infrared spectroscopy can be successfully applied to separate Eucalyptus and Corymbia species. In the visible spectroscopic analyses, it was possible to separate the wood samples into two colour groups, i.e. red/rose and grey/yellow/brown. In the near-infrared spectra it was possible to discriminate samples, and three groups associated with density of the species were formed. Both techniques were appropriate to distinguish species and their industrial use might solve practical problems.