(21f) Evaluation of a pelletized PSA bed while operating under bed fluidization and expansion conditions

An 80 inch long by 2.4 inch diameter single PSA bed (with a plastic translucent wall) was utilized to evaluate the dynamic behavior of the absorbent when operating under bed fluidization and expansion conditions. To mimic a bed retention system, an additional layer of 5 mm glass beads of variable depth was placed above the adsorbent layer at the top of the column. There was no spring or any other element holding the adsorbent bed down above the glass beads, which ensured capturing any bed movement or the occurrence of any other phenomenon related to operating at fluidization conditions (i.e., churning, coning, and dusting). Instead, there was only 5 inches of empty space above the glass beads with a camera and a light sensor positioned to detect movement of the bed. A four step PSA cycle was utilized in this study that included a feed step (upward flow), cocurrent depressurization step, countercurrent depressurization step and a countercurrent pressurization step. This PSA cycle operated between 15 and 70 psia at room temperature, with variable feed flow rates and step times. Based on pressure drop characterization of the bed and static force balances, modeling showed that the bed was capable of operating way beyond fluidization conditions with no movement of the particles. However, conditions were found that caused particle movement, lifting and bed expansion during the feed and cocurrent depressurization steps. The lack of movement when operating beyond fluidization conditions was most likely due to the small diameter of the bed coupled with wall friction. The weight of the glass beads had some but no significant impact on preventing the bed from moving. The results revealed the discrepancies between the observed and predicted flow rates at different pressures that should have caused bed movement of some sort. Videos of the top of the bed also corroborated the conditions that caused bed movement were well above predicted fluidization conditions.