Modeling of dense gas-solid reactive mixtures applied to biomass pyrolysis in a fluidized bed

A model is presented for mathematically describing the thermofluid dynamics of dense, reactive, gas-solid mixtures. The model distinguishes among multiple particle classes, either on the basis of their physical properties (diameter, density) or through their thermochemistry (reactive versus inert particles). A multifluid approach is followed where macroscopic equations are derived from the kinetic theory of granular flows using inelastic rigid-sphere models, thereby accounting for collisional transfer in high-density regions. Separate transport equations are constructed for each of the particle classes, allowing for the description of the independent acceleration of the particles in each class and the interaction between size classes, as well as for the equilibration processes whereby momentum and energy are exchanged between the respective classes and the carrier gas. Aimed at high-density suspensions, such as fluidized beds, the relations obtained for the stress tensor are augmented by a model for frictional transfer, suitably extended to multiple-class systems. This model, previously derived, is here enlarged to include heat and mass transfer, as well as chemical reactions and is therefore applicable to general gas-solid combustion systems. The noteworthy novelties of the model with respect to other derivations in the literature include: (i) a systematic and consistent derivation of the solids transport equations and transport properties within the multifluid concept, allowing for non-equilibrium effects between the respective particle classes, (ii) the ability to explicitly account for the possibility of porous solid fuel particles, and (iii) the modeling of multiple chemical reactions in both gas and solid phases and the associated effects of heat and mass transfer. The model, which includes a separately validated chemistry model, is applied to high-temperature biomass particle pyrolysis in a lab-scale fluidized bed reactor and is used to obtain yield of reaction products. The results indicate that, at fixed initial particle size, the fluidizing gas temperature is the foremost parameter influencing tar yield. The biomass feed temperature, the nature of the feedstock, and the fluidization velocity all have minor impact on the yield. It is also shown that the fluidizing gas temperature can be optimized for maximizing the tar yield. (C) 2001 Elsevier Science Ltd. All rights reserved.