Characterization of Refuse Derived Fuel Using Thermogravimetric Analysis and Chemometric Techniques

Abstract

An approach for characterization of refuse derived fuel (RDF) using thermogravimetry and chemometric techniques was developed. For this purpose, a series of samples coming from lignocellulosic products (wood, cardboard, paper, newspaper) and plastics (polyethyleneterephthalate, high density polyethylene, polypropylene, polypropylene, nylon and polyvinylchloride), as well as their mixtures, were investigated by means of thermogravimetry (TG) in a temperature range between 25 and 800°C. The datapoints of TG diagrams (weight loss) were then subjected to principal component analysis in order to unravel similarities/ dissimilarities of the investigated samples. A classification was obtained according to their woody/petroleum derived origination. This classification was more evident if partial least square discriminant analysis was employed. Finally, a partial least square analysis was carried out for the determination of lignocellulosic content in the sample. The model was validated by application to samples with known mass fraction of lignocellulosic products. Finally, the model was applied to two RDF samples using fractions of their particle sizes from 1 mm to less than 0.032 mm, and the results were compared with their ultimate and proximate analysis.