ABSTRACT

Furthermore, if it can be determined whether reaction kinetics or heat transfer limit the progress of the torrefaction within the particle, further simplification of the analysis is possible. Otherwise, both heat transfer (dependent on particle size) and reaction rates (dependent on the temperature within the particle) need to be taken into account simultaneously. Mass transfer during torrefaction relates solely to the transport of product vapors and gases within the solid particle and eventually into the bulk, gas phase surrounding the biomass particles. The dominant mechanism of mass transport is convection, driven by the pressure built up inside the particle. Darcy’s law, for the movement of fluids in porous media, is usually seen as being sufficient to describe convection if information regarding the pressure profile within the particle is available. Increasing the partial pressure of volatiles within the particle would also tend to suppress desorption of material into the vapor phase. 13.2.2 Reaction Models and KineticsSince torrefaction may be seen as a mild form of slow pyrolysis, many of the applied reaction models have been taken from slow pyrolysis literature. As indicated in Fig. 13.1, numerous individual reactions take place during torrefaction. As no exact model is possible, simplifications are necessary and this results in numerous, plausible approximations. Reactions may compete with one another, include intermediate products, or may take place independently. The initial reactant may be considered as one homogeneous material (raw biomass) or the model may be refined further to take into account the varying reactivities of the underlying biomass constituents (cellulose, hemicellulose, and lignin). The thermal decomposition rates of biomass or the individual biomass constituents are frequently determined through thermogravimetric analysis (TGA), where the overall mass loss during the course of a predetermined temperature program indicates the extent to which these reactions take place. This approach has its limitations in that some overlapping reactions may not be easily distinguished. Furthermore, the presence of any heat transfer limitations during analysis may retard the observed mass loss thereby resulting in unreliable data. This necessitates the use

of small biomass samples for TGAs. Care must also be taken to avoid systematic errors in temperature readings. A thermogravimetric curve for the thermal decomposition of biomass constituents is given in Fig. 13.3. One can clearly see that the different individual biomass constituents start reacting at different temperatures and also produce different fractions of solid residues after the completion of the heating program. In addition, it is generally found that the behavior of the individual biomass constituents does not exactly add up to the behavior of the overall biomass; some interactions do occur.