Though fluidized beds are used for a whole variety of physical and chemical processes, there is still room for improvement. Large bubbles may have a negative effect on the performance of fluidized beds due to mass-transfer limitations, inhomogeneity and back mixing. Results of video analysis and residence time distribution (RTD) experiments are presented for a two-dimensional fluidized bed with secondary gas injection, i.e., with gas injected at various locations throughout the bed and not only via the bottom distributor. Video analyses of secondary injection experiments using 550 mm glass beads indicate that secondary injection reduces bubble size and the total volume of gas in the form of bubbles, while also hindering bubble coalescence. This implies that the gas-solid contact is improved and that less back mixing occurs. The residence time experiments aimed to prove these points explicitly, and to determine what effect secondary injection has on the residence time of the gas. The latter is not obvious because the mixing and flow conditions inside the reactor are significantly changed by the secondary injection. Further experiments using single injection points were conducted in order to elucidate the mechanism by which the above effects are produced.

The RTD experiments were conducted by injecting helium tracer pulses into either the primary or secondary air streams of a two-dimensional air-fluidized bed with 210-280 mm sand particles. The helium exiting the bed was measured with a thermal conductivity detector. The data from helium injections in the two separate streams were then combined and a total residence time distribution was determined. These experiments were conducted for different total flow rates as well as for different secondary to primary flow ratios, but always maintaining sufficient primary flow to keep the entire bed fluidized.

Results of the RTD experiments indicate that the flow behavior of the gas is much closer to plug flow than for a conventional fluidized bed, while the mean residence times are only slightly reduced. The reduction in residence time is much less than would be expected given that a significant portion of the flow is introduced much higher in the bed. The bubbles in the bubble phase are much smaller and thus rise with a much lower velocity. Smaller bubbles cause less agitation and, thus, less back mixing. This, along with the lower total bubble volume, contributes to the greater plug flow-like behavior and better gas-solid contact in the fluidized bed with secondary injection.

Video analyses of single injection point experiments indicate that, in the presence of a bubble, the injected gas causes the bubble to break up into smaller units. In the absence of a surrounding bubble, secondary gas is injected directly into the emulsion phase, creating a super-saturated zone from which small bubbles nucleate. At any height, these bubbles are much more uniformly distributed (thanks to the distribution of injection points), which reduces the likelihood that they will encounter another bubble and coalesce into larger bubbles.