Near infrared (NIR) spectroscopy coupled with partial least squares (PLS-1) regression was used to predict the lignin contents and monosaccharide compositions of milled wood of Pinus radiata. The effects of particle size and moisture content were investigated by collecting NIR spectra of four sample types: large (<0.422mm) and small (<0.178mm) particles, in both ambient and dry conditions. PLS-1 models were constructed using mixtures of compression wood (CW) and opposite wood (OW) that provided a linear range of cell-wall compositions. Our results show that lignin contents and monosaccharide compositions of pure CWs and OWs can be successfully predicted using NIR spectra of all four sample types. However, large particles in ambient conditions have the most efficient preparation and the standard error (SE) values for lignin (2.10%), arabinose (0.34%), xylose (1.33%), galactose (2.54%), glucose (6.98%), mannose (1.48%), galacturonic acid (0.22%), glucuronic acid (0.06%), and 4-O-methylglucuronic acid (0.25%) were achieved.
Keywords: Compression wood; Moisture content; Particle size; Polysaccharides.
Copyright © 2018 Elsevier B.V. All rights reserved.