Non-destructive determination of wood-constituents by Fourier transform Raman spectroscopy

The feasibility of using FT-Raman spectroscopy for rapid determination of various wood constituents non-destructively was examined using five Eucalyptus species, including samples of various ages and colors of samples, which are of importance as a plantation source. Wood constituents which relate to pulp properties (holocellulose, alpha-cellulose, hemicellulose, lignin, extractives, alkali-extractives, total-extractives, and extractives-free (EF) wood constituents for holocellulose, alpha-cellulose, hemicellulose and lignin) were measured. The application of 2nd derivatives transformation of Raman spectroscopic data revealed highly significant correlations between wet chemical and Raman predicted values for all traits except EF-hemicellulose, with standard error of prediction (SEP) 0.8 points in the calibration (for known samples) and SEP 3.4 points in the prediction (for unknown samples), respectively. Consequently, this non-destructive method has proved its validity for analyzing various Eucalyptus native wood meal samples, regardless of their age and color, to determine wood constituents and EF-wood constituents except hemicellulose. Use of this method will reduce the costs of tree improvement programs and the minimal sample size needed will allow trees to be non-destructively sampled