Wednesday, February 1, 2012

Process Wednesday: Vortexes aren't good?

Undesirable? (Credit: eHow)
More on the engineering of reactors, this time from Mr. Ozzie Frietas, the technical sales manager for De Dietrich Process Systems:
The patented design of PharmaCore’s new, glasslined OptiMix reactors incorporates three vertical baffles on the walls of the reactor shell. Side baffles are common in metal tanks. They prevent the formation of vortexes that spin the entire fluid mass but do very little actual mixing or shearing. In baffled tanks, on the other hand, the baffles create turbulence that enhances mixing. They also optimize heat transfer, suspension and distribution of solids and gases, and mass transfer.  
Typical glass-lined reactors do not use straight flat baffles due to the challenge of attaching them to the reactor wall and encapsulating them with glass. Instead, they drop one or more baffles from the top head into the reactor. Under some conditions, these top-mounted baffles may not provide enough turbulence to suppress vortex formation. Vortices not only do a poor job of mixing, but they also splash liquid against reactor walls. In heated reactors, splashes soon turn into dried products that are difficult to remove from the reactor wall during
cleanup. 
De Dietrich developed the process to build wall-baffled, glass-lined reactors a few years ago. The company welds the baffles to the reactor wall prior to coating all steel surfaces with glass. The three baffles suppress vortexing and splashing. The symmetry of the three baffle/three mixing blade combination also minimizes bending loads that can deflect the agitator shaft and reduce seal life.
I'm terribly amused to learn that vortices "do a poor job of mixing." In the lab, the presence of a healthy-looking vortex is a sign to a chemist that things are mixing well, not badly. 

But then again, our friend Neal Anderson tells us this, too:
Without baffles, low-viscosity fluids tend to create vortices and swirl with poor mixing; baffles disrupt vortices and promote flow patters that lead to good mixing.
While I'm firmly convinced of the undesirability of removing dried product off the walls of reactors (boy, am I ever!) and the prevention of such, I'd really like to learn more about mixing and what's the best kind of mixing.  Vortices are poor at mixing? Could have fooled me.

Fascinating video from commenter marvinthefish -- I'm convinced!:

13 comments:

  1. Probably why James Bond always has his drinks "shaken, not stirred".

    ReplyDelete
  2. Try vortexing at a party and see how you get on! Better mixing (WOTS).

    ReplyDelete
  3. I am intrigued, too. I guess in my 100 mL rbf heat transfer and distribution probably isn't so much of a problem...

    ReplyDelete
  4. Paul Mavros has got quite some publications related to mixing on it's name. I suggest you check his website out . https://docs.google.com/Doc?id=dc2ptgt2_112cn4dxwdn

    ReplyDelete
  5. Assuming things are isothermal and laminar, the stuff that's stuck in a vortex can only get out by diffusion.

    I've seen this sort of stuff in the gas phase where recirculation can result in very long residence times compared to what you'd expect based on flows and volumes.

    ReplyDelete
  6. this "vortex does a lousy job at mixing" downed on me after running anionic polymerizations where one of the commonly-used bases (organopotassium that makes deeply-colored dark solutions) is added via canula to a stirred solution of the initiator until a hint of the organopotassium color persists - just like a titration. So if you stir vigorously with a healthy-looking vortex in the middle as you canulate the organopotassium in and if you aim onto the center of vortex around the stirbar (instead of the periphery of your vortex) you wil have a swirlingh colorless reaction mix with a deep-colored column of unmixed organopotassium right around the vortex center, looking like a miniature tornado. Very pretty/

    ReplyDelete
  7. This video gave me a good idea of the benefits of a baffle to mixing!

    http://www.youtube.com/watch?v=6E2-y96uYR4&noredirect=1

    ReplyDelete
    Replies
    1. Well, I guess the question is now: How do you set up a simple baffle for small (<250mL) reactions?

      Delete
    2. I don't know if the placement is correct, but like Anon9:52a says below, thermocouples or a thermal well (closed glass tube, mounted in adapter) should help at least a little.

      Delete
    3. Our local glassblower suggested crimping the walls of the RBF slightly, each baffle 120 degrees from the next. Apparently she does this all the time for larger (~1L) erlenmeyer flasks.

      I've asked her to crimp two 100mL RBFs, to see how well they work.

      Delete
  8. It all depends... on the goals. If the process goal is to efficiently mix liquid(s) then vortex is bad and baffles are good. If the process goal is to mix liquids and gases (hydrogenation) then the vortex is very effective and baffles are bad.

    We used to use baffles in RBFs. Back in the day they were called thermometers. Now we have thermocouples - 1/8", 1/16", 1/32"...

    ReplyDelete
  9. I must know: isn't it "vortices?"

    I mean, there aren't any matrixes (matrices, yes?). However, there are duplexes, sixes, and perplexes is still a verb. So, I'm stuck.

    ReplyDelete
    Replies
    1. The Oxford Dictionary says both plural forms are acceptable.

      Delete