(481f) Chemistry Modeling of Biomass Pyrolysis: Application to the CFD Simulation of a Laminar Entrained Flow Reactor

A current limitation to the efficiency of biomass gasification is the high level of tars usually found in the product gas, causing fouling and hazardous waste. Costly cleaning treatments to control the tar content of the exiting gas are usually applied downstream. Another appealing approach would be to directly optimize the gasifier design and operating conditions to the biomass properties for minimal tar production. However, such an approach requires a better understanding of the detailed chemical and physical processes occurring simultaneously in the reactor, including mixing, heat and mass transfer, and chemical reactions. Computational Fluid Dynamics (CFD) offers a unique opportunity to incorporate fundamental results, for example from detailed chemical modeling, into a larger scale framework, allowing to gain some valuable insight on tar formation pathways.

In this work, an extensive experimental investigation of biomass pyrolysis in a laminar entrained flow reactor is used as a basis for the development and validation of a reactive CFD model. The three-dimensional, cylindrical configuration remains simple enough to allow for a relatively detailed description of the chemical processes occurring in the reactor. Particles considered in the experiments are finely milled to a maximum size of 180 μm, which is below the size of a typical cell structure. Therefore, intra-particle transport processes are neglected in the CFD model. Particles are assumed to remain spherical in shape as they shrink during conversion, allowing the use of well-characterized correlations for the drag and heat transfer models. Biomass density is kept constant, and a volumetric approach is adopted to evaluate the chemical source terms for the solid phase.

A preliminary chemical mechanism describing biomass volatilization and further conversion of the products into permanent gases and polycyclic aromatic hydrocarbons has been assembled from our previous work on soot formation and data available in the literature. To reduce the computational cost associated with solving the chemistry in CFD, automatic reduction techniques have been applied to this detailed mechanism containing several hundreds of species to obtain a low-order scheme able to correctly reproduce the dynamics of the full chemical system. Since tar formation is the focus of this study, emphasis in the reduced model has been placed on the accurate description of aromatic rings formation from small hydrocarbon species and their subsequent molecular growth. The obtained chemical scheme has been integrated into CFD, and simulation results have been compared to the experimental data collected in the LEFR for a wide range of reactor temperatures and residence times. An improved description of the early stages of biomass volatilization based on carefully chosen model compounds has been derived from the analysis of the computations and the principal component results obtained in the experimental work.