Wednesday, January 25, 2012

Process Wednesday: retreat curve agitators

Retreat curve agitator (Credit: Southern Glass)
I don't know what you think of when you imagine the agitator in a large-scale reactor, but I think I always imagined something that looked like a boat propeller. So it's come as a surprise to me to become familiar with the graceful lines of the retreat curve impeller that comes with a glass-lined reactor.

For a little tutorial on the retreat curve impeller, I refer us to "Understand Flow Patterns in Glass-Lined Reactors", an article by Dickey et al. in the November 2004 issue of Chemical Engineering Professional:
Glass-lined reactors are essential process equipment in the pharmaceutical and speciality chemicals industries. A typical glass-lined reactor included a retreat curve impeller (RCI) near the bottom of the vessel and usually a single baffle mounted through a nozzle in the vessel head. The RCI with rounded blade corners may limit harmful turbulence effects while maintaining circulation throughout the vessel. Glass lining (the term "lining" is used to refer to the glass coating on the agitator and the inside of the tank) provides corrosion resistance, is easy to clean, and eliminates product contamination.  
The retreat curve of the RCI blades provids better radial flow than radial flow impellers with similar power characteristics. The impeller is placed near the bottom of the vessel to maximize the allowable range of liquid levels and to produce circulation from the bottom to the top of the vessel. The baffle (occasionally two baffles) is mounted from a nozzle in the top head because mounting to the side of a glass-lined vessel is difficult. The impeller and baffle always have a rounded cross-section without sharp corners because high stresses in the glass can cause the brittle coating to fail. 
There's a lot in those paragraphs that I don't really understand (radial flow? turbulence effects are harmful?), but it's interesting to me nonetheless.


  1. I usually think of turbulence as being helpful to mixing, but I'm guessing that laminar flow is desired for vertical set up a recirculation pattern that extends over the entire height of the vessel.

  2. Ditto on the turbulence, unless long-chain molecules (polymer, proteins and other macromolecules) are involved. In those cases, the divergent flow patterns can lead to degradation. One part of the molecule is being dragged to the left, while the other is being dragged to the right. Not a good outcome. In laminar flow, everything is being dragged to the left, but some parts are going there just a little faster than others, much like cars in the left-hand lane get to their destination faster than the cars in the right-hand lane (unless you're in Minnesota where people don't know how to drive and think that because they are doing the speed limit, they have a God-given right to the left-hand lane).

    Radial flow is flow directly from the central stirring rod out to the wall (axial would be parallel to the axis; circumferential would in a circular direction).

  3. Just don't crack the enamel, you won't be very popular at all!