The effect of the biomass components lignin, cellulose and hemicellulose on TGA and fixed bed pyrolysis

Thermochemical conversion of biomass has been studied extensively over the last decades. For the design, optimization and modeling of thermochemical conversion processes, such as fixed bed pyrolysis, a sound understanding of pyrolysis is essential. However, the decomposition mechanism of most biomass types into gaseous, liquid, and solid fractions is still unknown because of the complexity of pyrolysis and differences in biomass composition. The aim of this study was to find characteristic differences in the pyrolysis behavior of three widely used biomass feedstocks to optimize the performance of industrial fixed bed pyrolysis. This aim was achieved in three steps. First, devolatilization kinetics during pyrolysis of three biomass types was investigated in a thermogravimetric analyzer (TGA). Then, a one-step multi-component pyrolysis model with three independent parallel reactions for hemicellulose, cellulose and lignin was derived to correlate the kinetics with single component decomposition and to identify their amount in the biomass sample. In a final step, the findings were tested in a fixed bed reactor at laboratory scale to prove applicability in industrial processes. Three types of biomass were chosen for this investigation: wheat straw, rape straw and spruce wood with bark. They represent biomass with a high cellulose, hemicellulose and lignin content, respectively. Since lignin is the most stable and complex of these three biomass components, its amount is assumed to be the main controlling factor in the thermochemical decomposition process. The thermogravimetric (TG) curve of spruce wood with bark was found to shift to about 20 K higher temperatures compared to the TG curves of straw and rape straw. This result indicates that a higher activation energy is needed to decompose woody biomass, which contains a higher amount and a different type of lignin than straw. Three wood decomposition phases were distinguished from the negative first derivatives curves (DTG): a shoulder during hemicellulose decomposition, a peak during cellulose decomposition and a smaller rise during lignin decomposition. By comparison both herbaceous biomass types decomposed in only two phases at lower temperatures. The decomposition of the herbaceous, and woody biomass samples was completed at about 830 K and 900 K, respectively, leaving only a solid residue of ash. The derived pyrolysis model estimated the composition and described the devolatilization curves of each biomass with sufficient accuracy for industrial processes, although the same activation energy set, taken from the literature, was used for each biomass. In the fixed bed pyrolysis experiments similar characteristics were found to those in the TGA experiments. Herbaceous biomass with a higher cellulose and hemicellulose content decomposed faster and produced a larger fraction of gaseous products than woody biomass with a higher lignin content. According to the assessment of the product distribution, performed after each experiment, woody biomass pyrolysis led to a larger fraction of solid products than herbaceous biomass pyrolysis. We conclude that industrial fixed bed pyrolysis can be optimized for different biomass feedstocks with a specific composition of cellulose, hemicellulose and lignin. (C) 2013 Elsevier B.V. All rights reserved.