Sciencemadness Discussion Board

DIY light phase continuous extraction apparatus configuration

SuperOxide - 10-6-2022 at 11:10

I was looking for glassware that could allow me to do continuous extraction of a chemical extracting it from the heavy phase to the light phase, even if the desired compound is more soluble in the heavy phase.

I found this diagram online, which makes it look like its basically a Dean-Stark apparatus, but with a funnel to act as a delivery tube to allow the fresh returned solvent to be delivered to the bottom of the extraction flask so it can bubble up through the denser phase, which would contain the desired compound.


Turns out, getting a funnel that small that would fit in a 24/40 or even 14/20 dean stark is pretty difficult. I did try to make one out of a test tube, but I suck at glass blowing.

But, after some thinking, I was able to come up with something I think would work just as well and with commonly used glassware that I already own.
Note: The Graham condenser obviously wouldn't be what's used, it was just what I had in front of me. A graham condenser would be the ideal choice)

(More pics here)

I haven't actually used it yet, but I see no reason this wouldn't work. Regardless, I thought I'd share it so if anyone has this same need then they would have this setup available. Or if someone has a better way to accomplish this, then I would love to hear.

Thanks.

P.S. One thing I thought about changing is adding a very short dispersion tube to the end of the solvent delivery tube, but since dispersion tubes are usually for gas, im not sure how well that would work out. I suppose it would have to be a course frit to not slow down the return so much.

[Edited on 10-6-2022 by SuperOxide]

Jenks - 10-6-2022 at 12:36

How about stirring the mixture in the extraction flask slowly enough that the lower phase doesn't make it to the top? That should be more efficient than passively bubbling the light phase through the heavy phase, especially given that the dense phase is shallower in the flask than it would have been in a Dean-Stark? You might then be able to simplify by eliminating the adapters numbered 1 and 2 and letting the light phase overflow in the distillation head (piece you are holding).

Is anything going to leach out of the plastic tubing on long exposure to hot solvent? The simplification above should also eliminate the need for the tubing.

SuperOxide - 10-6-2022 at 13:09

Quote: Originally posted by Jenks  
How about stirring the mixture in the extraction flask slowly enough that the lower phase doesn't make it to the top? That should be more efficient than passively bubbling the light phase through the heavy phase, especially given that the dense phase is shallower in the flask than it would have been in a Dean-Stark?

Well if that was the case then a simple liquid/liquid extraction with a sep funnel would suffice. But this is for when something is actually more soluble in the denser phase than in the light phase.

Quote: Originally posted by Jenks  
You might then be able to simplify by eliminating the adapters numbered 1 and 2 and letting the light phase overflow in the distillation head (piece you are holding).
That wouldn't actually work, because it needs to bubble up through the heavier phase to make as much contact as possible. Hence the vacuum adaptor extension
Quote: Originally posted by Jenks  
Is anything going to leach out of the plastic tubing on long exposure to hot solvent? The simplification above should also eliminate the need for the tubing.

Guess that depends on the solvent. Using different tubing would work fine if you chose it correctly.

Again, the above was just somewhat of a concept. I haven't used it yet, but I do plan to.

Organikum - 26-7-2022 at 11:14

Quote: Originally posted by SuperOxide  
...

(More pics here)

...
[Edited on 10-6-2022 by SuperOxide]

As I have some extensive experience with continous extractions I have to point out some small flaws in the setup as in the picture:

The overflow tube should have a blocking function for gas and vapors or most of the vapor will try to get through this hole leading to some smallish drame. By incorporating a siphon one prvents this from happening.

Generally I prefer to have a fat vapor exit and a small but separate delivery (overflow) tube. Additionally I want a bottom valve to withdraw solvent/product concentrate from time to time. All this under the assumption that the product might be a bit sensitive to everything and we try not to stress it in any way.

Thats why I prefer Di-Ethylether as extraction solvent. Least stress on the product and environmentally ok, also a small leak won´t cost you your last braincells but searching and repairing might come out to be a enjoyable evening.

The whole solvent transport up, and condensing is best done with metal parts, SS is great here and the condensing power of a tiny metal condenser is so much superior to a glass one of the same size, it is a whole new world. Polypropylene works too, but is better used for the reaction vessel. Tubing, the small parts needed call for virgin PTFE.
For other inspiration go and have a look for "corrugated or flexible SS tubing solar" and be surprised what is available. But for something like this it would be just one flexible tube insulated and maybe heated (resistance wire) in the first part and on top just bent downwards and fitted with a plastic cooling jacket - voila! Condenser!

Last not least for bigger setups the "atomisation" of the solvent when it enters the reaction vessel at the bottom is paramount for best performance. Good ways are to release the solvent below a horizontally fast spinning disk, this works very well or to use a pump to move the reaction broth around and to atomize the extraction solvent. Sadly the standard fountain pump impellers did not agree with Ether. At least not with pure or almost pure or high.concentration ether. The problem was solved by using a somewhat bigger fountain pump and to accelerate the reaction broth alone and to pump it through an spiral nozzle aspirator like device where then the ether gets sucked in at the side and mixed and atomized. Worked great.

/ORG

SuperOxide - 5-8-2022 at 18:13

Quote: Originally posted by Organikum  

As I have some extensive experience with continous extractions I have to point out some small flaws in the setup as in the picture:

The overflow tube should have a blocking function for gas and vapors or most of the vapor will try to get through this hole leading to some smallish drame. By incorporating a siphon one prvents this from happening.

Generally I prefer to have a fat vapor exit and a small but separate delivery (overflow) tube. Additionally I want a bottom valve to withdraw solvent/product concentrate from time to time. All this under the assumption that the product might be a bit sensitive to everything and we try not to stress it in any way.

Yes, I thought of this as well but I can't think of a way around it. But how would a siphon work for that? A siphon would pull all the solvent through (like on a soxhlet), and this setup is just to slowly return the lighter phase as more solvent is distilled over.
I know it's less than optimal, but I think if I have the solvent flask hot enough that the vapour also makes its way all the way up to the condenser, then that would be sufficient.
If you have any alternative suggestion that would work, please let me know.
Quote: Originally posted by Organikum  
Thats why I prefer Di-Ethylether as extraction solvent. Least stress on the product and environmentally ok, also a small leak won´t cost you your last braincells but searching and repairing might come out to be a enjoyable evening.

Yes. One good use for this setup would be for purifying hypophosphorous acid using diethyl ether. I haven't actually done it yet, but that is one thing I have planned for it.
Quote: Originally posted by Organikum  
The whole solvent transport up, and condensing is best done with metal parts, SS is great here and the condensing power of a tiny metal condenser is so much superior to a glass one of the same size, it is a whole new world. Polypropylene works too, but is better used for the reaction vessel. Tubing, the small parts needed call for virgin PTFE.

The tubing in the picture is just what I had on hand. I actually do have some teflon tubing that would work fine for this. I would only use it for the overflow connection (which is convenient because the two ports facing each other would make it easy to hold the teflon tubing in place). The other vertical tubing in the left flask won't be there, I will probably purchase something like this and then shorten the glass tube just enough to clear the bottom of the flask.
Quote: Originally posted by Organikum  
Last not least for bigger setups the "atomisation" of the solvent when it enters the reaction vessel at the bottom is paramount for best performance. Good ways are to release the solvent below a horizontally fast spinning disk, this works very well or to use a pump to move the reaction broth around and to atomize the extraction solvent. Sadly the standard fountain pump impellers did not agree with Ether. At least not with pure or almost pure or high.concentration ether. The problem was solved by using a somewhat bigger fountain pump and to accelerate the reaction broth alone and to pump it through an spiral nozzle aspirator like device where then the ether gets sucked in at the side and mixed and atomized. Worked great.

Well, you're thinking way past the scale im working on, buddy. I will probably use this once, then never again. For getting any apparatus with horizontally fast spinning disks is a little outside my price range, lol.

Mateo_swe - 11-8-2022 at 00:32

You can make a very small funnel by enlarging the end of a glasstube.
Just heat the end of the glass tube with a torch and enlarge with a tool in the shape of a pencil, use the sharp end.

SuperOxide - 13-8-2022 at 17:12

Quote: Originally posted by Mateo_swe  
You can make a very small funnel by enlarging the end of a glasstube.
Just heat the end of the glass tube with a torch and enlarge with a tool in the shape of a pencil, use the sharp end.


I think this is what I will have to do. Won't the wood on the pencil catch on fire?.. lol. Or will it just char and form a protective layer of carbon?

Mateo_swe - 15-8-2022 at 02:12

Using a pen maybe isnt the best, just use a tool in that shape made of metal.
Experiment a little, glass tubes are dirt cheap.

Organikum - 16-8-2022 at 12:07

Quote: Originally posted by Mateo_swe  
Using a pen maybe isnt the best, just use a tool in that shape made of metal.
Experiment a little, glass tubes are dirt cheap.


Graphite toold for forming hot glass are better, graphite is not very expensive and easy to work on. A graphite rod in a drill which is fixed to a bench and a knife is all whats needed to make a conical graphite piece ideal for widening glasstubes. Just be aware that graphite dust is obnoxious, I strictly recommend to do this and other graphite shaping with high speed tools wet.

SuperOxide - 20-1-2023 at 20:43

I finally found the answer I was looking for.... "thistle tube". Had I known what a thistle tube was, I would have known exactly what to setup.

For those of you who don't know, thistle tubes are funnels with narrow openings at the top, and very long stems. They're used to deliver a liquid to an exact spot.... Which is exactly what I was trying to accomplish, lol.

Pictures: https://imgur.com/a/IQkSGxr

The funnel is about 24mm in diameter and 300mm long. And using a 24/40 => 45/50 and then 45/50 => 24/40, I can fit the funnel part right below the drip tip of a condenser, while the stem goes down to the bottom of a dean stark apparatus.


I feel like half the battle of coming up with a solution is just knowing the right words to Google. If I had known to Google "Thistle tube", that would have been the end of it. lol. None of the light phase continuous extraction diagrams/videos said that the funnel they were pointing at was actually called a Thistle tube :mad:

SuperOxide - 27-1-2023 at 15:00

Quick update, I was able to find (what I think) is the perfect piece of glassware to go with the thistle tube:



What I may do is put in a bunch of 4mm glass beads in the bottom of the collection part, so when the liquid floats up out the bottom of the thistle tube, it will hit a bunch of beads and get broken into smaller globules.
Or, alternatively I could burn some small holes into the sides of the bottom of the stem... But I like the bead idea more.

You can see how the thistle tube is just long/wide enough to be able to catch the condensation, yet not so wide as to block any of the vapours from reaching the condenser.

Amazing! Can't wait to put this thing to work :-)

SuperOxide - 28-1-2023 at 01:12

Other than purifying hypophosphorous acid using diethyl ether, does anyone have an example setup I can use as s test?

I was thinking DCM and water, since they're insoluble in one another and DCM is denser so should stay on the bottom while the water is passed through the DCM then overflowed back into the boiling flask.

Any other mixtures I could try?