Computational study on biomass fast pyrolysis: Hydrodynamic effects on the performance of a laboratory-scale fluidized bed reactor

Fluidized bed reactors are utilized in a wide range of chemical industries, including petroleum refining, pharmaceutical and commodity chemicals production, and biomass conversion to fuels and higher-value chemicals. Such reactors are useful where multiple fluids (gases or liquids) and particulate solids are brought together in intimate contact to promote heat and mass transfer and chemical reactions. Recently fluidized-bed research includes computational simulations that provide new insights into the dominant physics and chemistry processes that control reactor performance. Computational simulations were utilized to understand how bubbling-bed hydrodynamics and fast-pyrolysis chemistry interact to control biomass pyrolysis reactor performance. The scope of this work is limited to bubbling bed conditions, designed and operated for lab scale studies of biomass fast pyrolysis, in a bed of inert Geldart Group B sand. Biomass particulates are injected near the bottom of the reactor and rapidly heated to release volatile compounds. The devolatilized biomass particles (char) and released volatile gases transit through the bed (at time scales depending on the hydrodynamic mixing state) and elutriate from the top of the reactor. The bubbling-bed hydrodynamics were simulated with MFiX, an open- source software package based on the two-fluid (continuum) approach for representing the bubbling bed multiphase flows. Processes of interest included the transport of biomass char and released volatile gases, and how these change with fluidizing gas flow, low-intensity bubbling, to slugging, to high- intensity turbulence. A key observation is that fast-pyrolysis tar yield can be increased by reducing the residence time in the freeboard by shortening the freeboard length or by adding secondary air. Also, of interest was how these transport processes are expected to affect the selectivity of product species exiting the top of the reactor. One promising concept for monitoring hydrodynamics in bubbling bed reactors are high- speed pressure measurements to quantify key mixing and transport properties. Computational simulations were utilized to identify quantitative statistical features in high-speed pressure measurements in the upper section of the bed, below the static bed height, to use as process monitoring tools. Other promising directions were identified for future experimental and computational studies.