Electronic Thesis and Dissertation Repository

Bubble Dynamics in a Sand Fluidized Bed in the Presence of Biomass Pellets

Nicolas Torres Brauer, The University of Western Ontario

Abstract

Given the worldwide concerns regarding fossil fuel availability, global warming, water, and air pollution and economy needs, there is an urgent requirement to develop and implement alternative energies. Biomass gasification can consume vegetal waste to produce syngas, which is a mix of carbon monoxide, hydrogen, and carbon dioxide. When biomass gasification is carried out in a fluidized bed, the process is highly efficient.

The experimental study of the present Ph.D. dissertation took place in a Plexiglas 0.44 m diameter unit, equipped with a CREC Optiprobe system and a video camera. The unit was filled with 200-900 µm sand particles and biomass pellets measuring 2.7 cm in length and 0.8 cm in diameter. This was used to take experimental measurements of bubbles size, velocity, and shape.

Single bubbles injected in a sand bed at minimum fluidization with and without biomass were first considered. Later, bubbles evolving in a sand bed, operating in the dense phase bubble regime were studied. On this basis, the capabilities of a developed geometrical model to predict the BAC (Bubble Axial Chord) and the BACs correlated with the BRVs (Bubble Rising Velocity) were examined. The bubble dynamics model considered includes an adjustable bubble wake parameter, with predictions providing the bubble chord, the bubble frontal-radius, and the bubble rising velocity. Following this step, the effects of biomass pellet addition on the BAC, the BRV, and the bubble assigned shape were evaluated.

Computational Fluid Dynamics Multiphase Particle-in-Cell (CFD-MPPIC) simulation using a CPFD Barracuda VR® v17.4.1 software was considered, and this for the various cases studied. The proposed CFD-MPPIC model involved the following: a) an equivalent 1 cm3 biomass pellets, b) an adjusted particle drag model, c) a bubble-wall interaction, d) a selection of computational cells, e) a Multiphase Particle-in-Cell population sizing, and f) a cell distribution implemented throughout the column.

Experimental results in the Plexiglas 0.44 m diameter unit, validated the proposed CFD-MPPIC model with good predictions of both BACs and BRVs. This was the case for all conditions studied, with the only exception of bubbling beds with added biomass, where calculated BRVs were found for some operating conditions as less trustable.