In the previous post, I showed that there were very little differences in the performance of models with a CSTR or plug flow emulsion. For both of those models, I assumed that there is a bubble phase above some known air rate. This is the Davidson-Harrison  model assumption. If instead we eliminate the bubble phase and allow all air, regardless of rate, to pass through the emulsion bed, then we see a difference in parametric trends.
With the plug flow model as the basis, the effect of the bubble phase assumption (i.e. with or without) is shown in the following figures.
The above figure shows that the bubble phase assumption leads to different optimum air rates using gasoline yield as the objective. With a bubble phase, the optimal air rate is at the onset of the bubble phase, the minimum rate on the graph. Without a bubble phase, increasing air rate increases the gasoline yield.
The above figure shows that coke level decreases as air is increased for the case without a bubble phase. The reduced coke level lowers the catalyst decay in the riser which leads to the increase conversion and gasoline yield discussed above. The constant amount of air to the emulsion phase in the Davidson-Harrison model results in fairly constant coke levels, as shown in the graph on the left.
Finally, the two models show different responses to changes in catalyst rate.
These results demonstrate the importance of the bubble phase assumption. They also provide an example of the use of parametric trends to distinguish between models. Models can be validated by comparison with similar plant data trends.
 Davidson, J.F., and D. Harrison, "Fluidized Solids", Cambridge U. Press, London, 1968.