FTIR-based models for assessment of mass yield and biofuel properties of torrefied wood

• Published in: Wood science and technology 2018, Vol.52 (1), p.209-227
• Main Authors: So, Chi-Leung, Eberhardt, Thomas L.
• Format: Article
• Language: English
• Subjects: Anaerobic conditions; Assessments; Atmospheric models; Biodegradation; Biodiesel fuels; Biofuels; Biomass; Biomedical and Life Sciences; Calorific value; Carbon content; Carbonization; Ceramics; Composites; Crucible furnaces; Drying; Drying ovens; Fourier transforms; Fuels; Gasification; Glass; Grinding; High temperature; Infrared analysis; Infrared heating; Infrared spectroscopy; Life Sciences; Machines; Manufacturing; Natural Materials; Original; Pretreatment; Principal components analysis; Process parameters; Processes; Pulp; Pyrolysis; Regression analysis; Temperature; Temperature effects; Thermal degradation; Wood; Wood chips; Wood Science & Technology
• ISSN: 0043-7719; 1432-5225
• DOI: 10.1007/s00226-017-0970-1

Biofuel properties can be improved through torrefaction, whereby the biomass is treated with moderately elevated temperatures (200–300 °C) under conditions that are essentially anaerobic and at atmospheric pressure. Varying the torrefaction conditions of temperature and treatment duration, as well as any feedstock pretreatments (drying, grinding), can generate products having varying degrees of thermal degradation, thus impacting performance in subsequent biofuel applications (e.g., gasification, pyrolysis, combustion). Pulp-grade pine wood chips were processed through a laboratory-scale crucible furnace retort with the remaining solid carbonized products subjected to ultimate, proximate, and spectroscopic analyses. Compared to air-dry wood chips, lower mass yields (i.e., greater thermal degradation) resulted from oven drying or grinding pretreatments. Regarding the torrefaction conditions, the degree of thermal degradation varied to a greater extent as a function of applied temperature than treatment duration. Furthermore, mass yield explained 95% of the variation in higher heating value. Fourier transform infrared (FTIR) spectroscopy combined with principal component analysis gave plots in which sample points clustered by torrefaction temperature and time, as well as by sample type (i.e., with or without pretreatment). Partial least squares regression was able to accurately predict mass yield and to a lesser extent carbon content. Results suggest that FTIR monitoring of finely ground samples in a production environment could be used off-line as a rapid assessment technique for mass yield and therefore facilitate adjustments to process parameters of temperature and/or treatment duration.