Wednesday, March 9, 2011

Process Wednesday: Scale-up questions

The Kilomentor blog is a really great and under-referenced resource in the chemistry blogosphere. He has a great list of questions for anyone looking to scale up a reaction. He asks incisive questions about difficulties that might arise during crystallization, filtration, drying and today's topic, extraction. Does the process have an extraction step(s)? If so:
  1. Are the phases susceptible to emulsion formation (vigorous shaking)?
  2. What provisions for promoting separation are known?
  3. Are the phases very dark and hard to tell apart? What to do?
  4. What about interfacial material? What should be done with it?
  5. Have you reduced the size of the liquid phases as much as possible to increase throughput?
I have worked under conditions of question 3, where phases are extremely difficult to tell apart. While it's difficult, it's not too tough on 22-liter scale (especially when you can watch/hear the solution coming out of the separatory funnel.) I cannot imagine what one does when the scale is increased significantly (shudder.)

Interfacial material is another weird one. What to do with the stuff? It's both organic and aqueous. I've never found a good answer to this question. (I guess it depends on its purity; there's gotta be a lot of inorganic salts in there, but if you're short on product, you might be tempted to try to recover it.) Readers?

5 comments:

  1. There's a feeling around our company that you don't really understand all the ins-and-outs of a reaction until you try it on kilo scale. That's where you see strange temp fluctuations, massive color changes, trigger precipitates, etc.

    So, regarding #4, the interfacial material - for me, it's usually left-behinds from the previous reaction steps that I didn't see on pilot. Activated carbon, tosylates, calcium / cesium salts, and nasty iodide byproducts tend to dominate my "interfacial" inventory.

    ReplyDelete
  2. I think that thought is correct.

    ReplyDelete
  3. I have worked under conditions of question 3, where phases are extremely difficult to tell apart...well, I am not a process chemist, but I can relate to you and how I resolved the same. these observations were made in a separatory funnel. I switch off the light in the hood and make a lab semi-dark. Then I use the flash light and light up from the back side of the funnel. One can clearly see the two layers! Not the beat but worked for us.

    ReplyDelete
  4. Thanks CJ for the links to Kilomentor as have found full of excellent reflections and advice.

    IMO #1 emulsions has often been most difficult at scale so if there are any hints in the lab then you best work out avoidance and/or remedies.

    For the dark layers separations it's not that different in the plant than Anon 10:07/lab. A 6-12 inches (or bigger) clear glass sight tube is installed on the out flow and then use a flashlight/lamp to visualize. Typically known amounts of each solvent used so can rapidly take off lower portion until get close to zone where expect layers then slow flow down and watch closely. If planning back extractions can use similar watch and listen clues to take rough cuts. Another possible option is to resort to adding small amount of celite which floats on water but sinks in organics so easy to see barrier (and remove later by filtration).

    For interfacial materials this is commonly viewed as opportunity for process improvements. Must do an effort/benefit assessment and then work toward reduction/elimination procedures up front or introduce reprocess/recovery modes to maximize yields. Old adage is the larger one goes the more one desires to minimize waste as even small percent increases/loses translate to big dollars.

    CMCguy

    ReplyDelete
  5. You can use the difference in conductivity between the aqueous and organic phase. Automated plants have such probes routinely present. It does not work if you do not have large amounts of inorganic salts present.

    One of the routine things I do during development work is to note the volumes and weights of the two phases. Gives you way more confidence in the first pilot runs. The phase separations are also always the critical parts of the process - recovering your product phase from a waste tank is not fun.

    The interfacial material (I like the German expression "Mulm") can be beneficial in phase seps as it is usually a good indicator that the other phase is well on its way.

    Especially when having great emulsifiers such as enzymes present in your reaction mixture, the work-up can be really tedious and time consuming. Cycling the temperature up and down may be an option, cycling the two phases through a loop if you have one present at your reactor may be another one. If all things fail, a coalescer will help you break the emulsion.

    ReplyDelete

looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20