I was involved in the research, design, and successful commercial operation of a large bubble tower in the late 1970s, so I was struck by the Business Concentrate about Evonik doing research on bubble columns (C&EN, June 30, page 13). Our system involved liquid, gas, and wax particles in which foaming was a potential problem.
Propane dewaxing, used in the petroleum industry to provide wax-free lubricants that will flow at low temperature, normally uses cooling from room temperature to about –40 °C by autorefrigeration of liquid propane in a pair of very large batch chillers. With careful control of the chilling by reducing the pressure, wax crystals of a particle size that can be filtered are precipitated.
We wanted to increase the plant capacity by precooling in a large autorefrigerated bubble tower to reduce the work of the batch chillers. We started by looking up the available literature, and I was surprised by the dearth of available studies on such devices. What was available then was on nitrogen bubbling through water in small columns of a few inches diameter.
The glass column we had available in the engineering laboratory was about 2 feet in diameter and several feet high. We filled it with hexane and, as I recall, bubbled nitrogen gas up through it to get away from the surface tension of a nitrogen-water system.
We altered both liquid and gas rates and observed the flow regimes, which varied considerably. The key factor for us was the pressure gradient from top to bottom of the tower, which was affected by the relative volumes of liquid and vapor, both moving upward. As the hydraulic pressure was reduced in the plant column, more and more propane was vaporized and the temperature dropped. So the commercial tower, about 50 feet high, was built in three sections with the largest-diameter one at the top.
Amazingly, given all the complications, the scale-up worked, and the large tower did its job without foaming over or back-mixing. The Evonik tower shown in the C&EN photo appears to be made of steel, in which case it would be difficult to visualize the various flow regimes.
James D. Bushnell
Palm Harbor, Fla.