Jet-bed interaction (JBI), in this thesis, refers to the interaction between an atomized liquid jet and a gas-solid fluidized bed. Industrial processes, such as granulation, fluid catalytic cracking (FCC) and fluid coking, use JBI to disperse liquid feeds over the surface of solid particles. In fluid coking, this thesis shows that feed dispersion devices should be designed to limit heat and mass transfer limitations caused primarily by the agglomeration of feed liquid and fluidized particles. This means limiting the formation of large granules with high levels of saturation. The objectives of this thesis are to develop an experimental technique to assess JBI for the fluid coking process and to improve upon conventional nozzle designs and nozzle operating conditions.

First, an experimental technique was developed to assess JBI. The technique involved an injection of a sugar solution into a bed of fluidized coke particles. Granules formed during liquid injection were recovered and analyzed for their size and moisture content. The moisture content was determined by measuring the amount of glucose contained in the granules.

Various novel nozzle designs were tested with the aforementioned experimental technique. A number of these designs produced small and drier granules than conventional nozzle technology.

To interpret the experimental results in terms of their benefits for fluid coking, a two-level theoretical model was developed to predict liquid yields in a commercial fluid coker, given the granule size and moisture content distributions. The model captured the cracking and devolatilization of liquid on the level of the agglomerate and the mixing and vapor cracking behavior on the level of a fluid coking reactor.

A study was conducted in which the air to liquid ratio, nozzle size and level of pulsation was varied in a conventional gas-liquid nozzle. Experimental results were combined with the theoretical model to estimate liquid yields in a fluid coking reactor as a measure of nozzle performance. While the results pertaining to pulsation were inconclusive, increasing air to liquid ratio and decreasing nozzle size were found to be beneficial. Good correlation was found between the pressure at the tip of the nozzle and liquid yields with a potential benefit of up to 1.3 wt.% feed.