Some recent studies have shown that under some conditions there can be severe gas mal-distribution of gas in deep fluidized beds of Geldart group A materials. The conventional knowledge is that bubbling and near-turbulent beds of these materials are well mixed. Knowlton (2001) presented videos of fluidization tests conducted with 4 and 11% fines (<44 micron) FCC catalyst particles in a transparent 0.3-m-diameter unit. For the 4% fines catalyst particles, a significant gas bypassing was observed in the bed in a form of a high velocity air jet that preferentially flowed up near the column wall while the rest of the bed stayed defluidized or poorly fluidized. The air jet moved around the wall but it occasionally anchored or stayed longer at the cyclone dipleg return location. When the fines content in the bed was increased to 11% < 44 micron the gas bypassing disappeared and the fluidization became quite uniform.
The study of Wells (2001) using FCC catalyst particles in a semi-circular Plexiglas unit for bed heights of 2.4 to 4.9 m found that a “snake” of streaming gas formed a short distance above the grid and passed through the bed, bypassing the mostly stagnant bed. Issangya et al (2007) using FCC catalyst in a 0.3-m-diameter suggested that the standard deviation of differential pressure fluctuations measured across short intervals around the column can be used as a diagnostic tool for gas bypassing in deep fluidized beds.
Gas bypassing can lead to serious problems if it occurs in industrial beds. For example, in FCC regenerators and strippers or in CFB combustors it can result in afterburning, poor gas/solids contacting and poorly fluidized entrances to standpipes, poorly fluidized discharge regions of cyclone diplegs and poor solids flows around loop seals. The paper discusses results of tests conducted for FCC catalyst particles in a 0.9-m-diameter fluid bed unit in which horizontal baffles were employed to eliminate gas bypassing. The unit had a transparent Plexiglas section for visual observation of the fluidization behavior as well as instrumentation to measure differential pressure fluctuations and bubble void fraction profiles. Videos will be presented.