Separating air

[Nick1911]

An interesting thought occurred to me.

I know that in industry, to get oxygen, nitrogen and argon, industrial gas suppliers often liquefy air under pressure and cooling, then use fractional distillation to pull out the separate components.

This makes sense.

I also know that in nuclear science, gaseous centrifuges spin uranium hexafluoride and are able to thus separate different isotopes of uranium by the slight difference in molecular weight.  Centrifuging a gas apparently works under similar principle as liquids.

Question:  If air were to be fed into a centrifuge, would it separate into it's component parts according to density?  If so, why isn't this used?  If not, why not?

Thanks!

[Myself]

I thought that different gasses were separated in a cascade refrigeration system.  Able to deal with the volumes needed much more efficiently.  Just keep dropping the temperature and collect what freezes.

[AmbulanceDriver]

A centrifuge system for uranium enrichment yields an incredibly small "enrichment" per centrifuge unit.  So much so in fact that massive cascades of centrifuges are used to "purify" the desired isotopes.  It's a very inefficient system, but it's one of the few mechanisms for enriching uranium.  There are easier, more efficient means for separating other gases.

[280plus]

I'm going to imagine if you get it spinning fast enough it might. Best I can do on that one. 

[Nick1911]

I'm going to imagine if you get it spinning fast enough it might. Best I can do on that one. 

Hmm... maybe if I put it in different terms...

"A stupid newbie just used your almost-full 50lb recovery cylinder full of R12 for recovering R134a.  Can you use the masses of the different refrigerants to separate them so you can sell your R-12 instead of paying to have the mixed refrigerant destroyed?"

And no, you can't just "find a problem" with one of the valves resulting in all the refrigerant leaking out overnight.


 

[280plus]

How much 134A we talking?

[280plus]

I did some poking around. Unfortunately the boiling point of r-12 is lower than the boiling point of 134A so unless my brain is scrambled there is no way to fractionate by temp. The r-12 would boil off first. The querstion is do you know the ratio of 12/134 in the can? Was it only a few ounces to 30+ lbs? What you can do is set the can somewhere in a stable temp for 24 hours. With a calibrated thermometer measure the tank temp and with a good set of gauges measure the tank pressure. See how close it is to corresponding to the temp pressure charts. If it's close you may get away with it. If it's way off you might be screwed. This is how they check the contents when they accept cans for recycle at the supplier BTW. Actual readings against what the chart says for the supposed contents.

[230RN]

OP:

If air were to be fed into a centrifuge, would it separate into it's component parts according to density?

I'd have to dig around a little (which I can't do now) but I seem to recall that some gases have different magnetic properties from others.  If I recall correctly, the oxygen molecule (I'm not talking about the ion) is slightly magnetic.  I wonder if a separation method might be based on this "half fact," or maybe it can be used as an enhancement for other separation methods... like centrifugal or diffusion.

Just some trivium I picked up some time ago in a galaxy far, far away.

Terry, 230RN

[vaskidmark]

I'm surprised you folks have not commented on the difference between the phyisical sizes of the molecules of various gasses.  My oxygen concentrator works via that principle - zeolite of pore size x = O².

There are other zeolites that "filter" other gasses - literally sucking them out of the air, through the pores, and putting them in a bottle pretty much by themselves.

How fine a separation are you looking for?  My )² concentrator, a simple one-stage machine, gives me approximately 98.5% O².  I'm fairly sure if you ran the filtrate through another stage or so you could come out with a much better concentration.

stay safe.

[280plus]

He wants to separate r-12 from r-134A. I'm not sure there's a filter for that. I guess there could be.

[Nick1911]

I'm posing the question simply as one of theory.

Using pore sizes that differentiate the molecules is certainly clever. 

[MechAg94]

Bulk industrial gas production is done through distillation at cryogenic temperatures.  Nitrogen boils at a lower temperature (-320F) than Oxygen (-297F) so the if you have a good liquid/gas mix, the oxygen concentrates in the liquid.  It is my understanding that the cryo-distillation is still more efficient for bulk production and high purity.  Also, Argon has a boiling point close to oxygen and can be pulled out as well in small quantities.  

I believe the oxygen concentrators use membrane technology to separate air.  They basically use bundles of capillary tubes with molecule sized pores that selectively allow smaller N2 molecules to pass through.  It takes some differential pressure to force the separation and it is very difficult to get decent purity.  My company has a membrane division that sells units for specific applications.  

Another way I have heard of to get enriched O2 is using vacuum adsorption.  You have beds full of molecular sieve material that tends to hold certain molecule types.  Changes in pressure or heat causes them to release the stuff.  So the beds have to be cycled to adsorb and purge.  Bulk air separation plants uses molecular sieve to remove CO2 and water from the air before dropping the temperature down to cryogenic temperatures.  CO2 and water would just freeze out in the heat exchangers.  

http://www.uigi.com/compair.html
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6TG3-4313YC6-3&_user=9503488&_coverDate=05%2F31%2F2001&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1552852964&_rerunOrigin=google&_acct=C000070294&_version=1&_urlVersion=0&_userid=9503488&md5=57250d1c74018d632cfdf82da315ec6d&searchtype=a
These links look like they have more detail.

[MechAg94]

I think the membrane tech was originally developed by DuPont, but it pretty widespread these days.  It is used in a lot of applications.  The plant I work at uses a membrane system to pull excess hydrogen out of a syngas stream.  The hydrogen molecules are much smaller than the CO and methane molecules. 

[MillCreek]

I'm posing the question simply as one of theory.

Using pore sizes that differentiate the molecules is certainly clever. 

This was the same technology used to produce enriched uranium during WW II for the Manhattan Project.  The Project was experimenting with calutrons, gaseous diffusion and centrifuge enrichment to see which method worked the best.

[280plus]

I thought you said some noob got 134A in their can of 12?

[Nick1911]

I thought you said some noob got 134A in their can of 12?

Sorry, I should have been more clear; it's a hypothetical situation.

Same concept of how we would go about separating that out apply.

[280plus]

Best way is to fractionate by temp but the r-12 would boil out first. I thought you accidentally put a couple pounds of 134 into your 12 jug and wanted to separate it out. I figured boiling off 40 lbs of 12 to separate out 2 lbs of 134 would be a PITA. It's have to be a pretty complicated machine to do it by centrifuge I would think.

[Nick1911]

Best way is to fractionate by temp but the r-12 would boil out first. I thought you accidentally put a couple pounds of 134 into your 12 jug and wanted to separate it out. I figured boiling off 40 lbs of 12 to separate out 2 lbs of 134 would be a PITA. It's have to be a pretty complicated machine to do it by centrifuge I would think.

A thought on fractional distillation....  in the case of R12 and R134a, I wonder if the two would produce an azeotrope?

That might explain why even industrial refrigerant reprocessing places will destroy mixed refrigerant.

[280plus]

I do not know. I was under the impression mixed refrigerant was distilled and recycled. Having never actually turned in anything mixed I don't know what the process is. I know lots of guys do without telling anyone cause they don't really give a hoot. I'm not sure at what point the supplier (who takes it) checks to see what's actually in there and whether it comes back to bite the person who brought it in mixed. Good questions I'll have to ask. And so I did.

When I purchase a 50 lb recovery tank I pay ~$35. That covers my returning it with unmixed refrigerant for reclaim. If I were to mix say 134 and 12 I would then pay a $3 per pound premium for disposal. Some guys bring it in mixed and don't say anything. The supplier does not care, they just ship it to the reclaimer. They will toss it in with everything else that says oh, R-22 on it and then it has to come out to something like 97.7% pure so what theyll do is take a mess of it, like 20,000 lbs he said, and add virgin refrigerant to it till it meets that standard. The simply dilute it out till it meets spec. So I asked what happens to it after that and he said it goes overseas. So then I said, "So all refrigerant sold in the US is virgin?" and he said no, there are companies here that do sell reclaimed refrigerant but his particular company, United Refrigeration, does not. 

[MechAg94]

If one boils at a higher temp, you might set up a cooling coil to let the gas come off and then re-condense it in another container.  

[Brad Johnson]

Technically you're seperating air just by being alive and breathing.

Brad

[geronimotwo]

they wouldn't be of differing masses where they would naturally separate?

you could use a dyson vacuum with "cyclone" technology!

[vaskidmark]

Technically you're seperating air just by being alive and breathing.

Brad

Thought that was membrane technology.

And somehow I can't remember it specifically separating out & capturing R-whatever.

stay safe.