Efficiency of Near-Infrared Spectroscopy in Quantifying Lignin in Black Liquor-Impregnated Reforestation Wood

Cellulose and lignin are biopolymers with significant potential for chemical synthesis and energy production; however, their heterogeneous composition presents challenges for their use as raw material sources. This study employed near-infrared (NIR) spectroscopy coupled with partial least-squares regression (PLSR) to predict cellulose and lignin content in sapwood and heartwood of Eucalyptus urophylla and sapwood of Pinus taeda, all impregnated with black liquor under high pressure. Samples were analyzed across three longitudinal sections (top, middle, base), with no significant compositional variation detected. Near-infrared spectral data (1100-2500 nm) and pre-processed using the standard normal variate (SNV) method, yielded high predictive accuracy: R² values of 0.98-0.99 for cellulose and 0.94-0.96 for lignin, with root mean square error (RMSE) values of 0.2-0.3 and 0.1, respectively. Principal component analysis (PCA) explained 98% of sample variance, revealing clear distinctions between E. urophylla sapwood and heartwood. These findings confirm the efficacy of NIR-PLSR as a nondestructive, reliable alternative to conventional chemical analyses, with implications for improved quality control and decision-making in the wood treatment industry.