This study is focused on the prediction of fungal weight loss (WL) and high heating value (HHV) from raw and torrefied waste lignocellulosic feedstocks, according to their exposure duration to wood-destroying fungi, using near infrared spectroscopy (NIR) and chemometrics models. Sugarcane bagasse, coffee husk, eucalyptus and pine shavings were torrefied at 290 °C in a screw reactor, during 5, 7.5, 10, 15 or 20 min. Raw and torrefied biomasses were submitted to a water leaching step and then exposed to white and brown-rot fungi, simulating various storage conditions. WL and HHV were determined and NIR spectra were obtained from each sample modality (including raw and torrefied), after 2, 4, 8 and 12 weeks of fungal degradation. The NIR spectra were used to (i) classify the four different residual biomasses in raw and torrefied forms according to fungal degradation by white-rot and brown-rot fungus; (ii) to predict the WL during fungal exposure and to (iii) predict the HHV of the samples according to their decay exposure types and duration. Partial least squares regression (PLSR) or discrimination analysis (PLS-DA) models were built to perform these predictions from NIR spectra. PLS-DA models classify successfully the four different residual biomasses in raw and torrefied forms according to fungal decomposition. PLSR models to predict the HHV during their decay deterioration showed potential utility in an industrial context as a standardized continuous method. Otherwise, PLSR models to estimate the WL due to fungal degradation did not present good accuracy but can be useful for screening in decision making. Further studies are required to improve and develop more efficient models to predict the fungal degradation level of stored raw and torrefied biomasses. These results highlight the potential of NIR spectroscopy as a simple, fast, and efficient tool to analyse the fungal decomposition process over the time.

Download document (649 kb)
free for the members of IRG. Available if purchased.

Purchase this document