(146a) Characterization of a Volatile Gas Plume in Bubbling Fluidized Bed Biomass Gasifier with Single Point Asymmetric Fuel Injection

In general, the fluidized bed reactor presents near ideal conditions for mixing and heat transfer. Devolatilization (rapid release of volatile fuel components as a result of heating of the fuel particle) of fuel particles in a high temperature fluidized bed can occur rapidly (on the order of 1-2 sec.). Kunii and Levenspiel (1991) describe ?rising plumes? and ?plugs? of volatile gases rising from individual coal particles at the fuel injection point. While devolatilization is complete within a few seconds after the particle is introduced into the high temperature bed, the particle will continue to mix and react within the bed for up to about 100 seconds. However, if the volatile gases, which contain a sizable portion of the total gases produced during gasification (including most of the tar content), exit the bed immediately after being devolatilized, then there is a reduction in in-bed reaction time. In the case of long chain hydrocarbons, this could result in reduced physical (thermal cracking) and chemical breakdown (reactive degradation) resulting in high tar content syngas. In addition, the rapid release of volatile gases could alter fluidized bed hydrodynamic conditions resulting in decreased mixing performance in the bed (Gomez-Barea et al. 2009).

Experiments were performed at the University of Utah Institute for Clean and Secure Energy to investigate the presence of a volatile gas plume in a bubbling fluidized bed with a single fuel injection point. Initial characterization of the volatile gas plume was performed using an acrylic cold flow fluidized bed (scaled from pilot scale fluidized bed). A novel gas plume measurement method was devised which maps the plume using point scalar dispersion measurement (temperature and carbon dioxide concentration) throughout the bed under different hydrodynamic conditions. Results obtained from the cold flow experiments were then used to design, and were compared to, experiments in a pilot scale bubbling fluidized bed. The presence of a gas plume in the bed, emanating from the fuel injection point, was verified and characterized for various bed hydrodynamic conditions. Although no direct correlation can be made between plume presence and syngas tar content, the data presented provides a basis for future research into the relationship between the volatile gas plume and devolatilized gas (tar) dispersion in the bed.