Cryogenic liquification of air (Linde process)

[rppearso]

Does anyone know what pressure air has to be compressed too so that the JT effect causes the air to liquify? I am running a simulation and I have the pressure up to 9000 psig and 0F and the drop in temp is only to -125F. The requisite temp to liquify air is ~ -320F from my simulation results. I can't imagine a 9000 psig compressor is very affordable but I am not sure if my simulator is telling me the truth or does the Linde process really have exotic compression in order to JT to a liquid?

 

[SteamKing]

Does anyone know what pressure air has to be compressed too so that the JT effect causes the air to liquify? I am running a simulation and I have the pressure up to 9000 psig and 0F and the drop in temp is only to -125F. The requisite temp to liquify air is ~ -320F from my simulation results. I can't imagine a 9000 psig compressor is very affordable but I am not sure if my simulator is telling me the truth or does the Linde process really have exotic compression in order to JT to a liquid?

I'm not sure the details of your simulation, but this article describes how the JT effect can be used to liquefy nitrogen (which is 79% of the atmosphere):

http://en.wikipedia.org/wiki/Joule–Thomson_effect

The problem with your simulation is that you might be assuming that the liquefaction of the air due to the JT effect occurs in a single expansion. The JT effect can be applied more than one time to the same gas in order to produce the phase change. See the article for more details.

 

[rppearso]

That is very true, I am now looking into interstage cooling by using an unrelated gas on the cold side of a heat exchanger. What is the best gas for the Joule-Thompson throttling process. I know hydrogen and helium will heat up so they are aweful, perhaps some of these commercial refrigerants? The good thing is I live in an area where I can get a lot of free cooling for a good portion of the year.

The reason I am doing this is the extreme cost of Xenon gas for other research purposes, its so outragiously expensive that I think it would be worth pouring a small concrete pad and setting up some low volume equipment and design it so that I can just let it run and come pick up a bottle of Xe after a few weeks.

 

[SteamKing]

That is very true, I am now looking into interstage cooling by using an unrelated gas on the cold side of a heat exchanger. What is the best gas for the Joule-Thompson throttling process. I know hydrogen and helium will heat up so they are aweful, perhaps some of these commercial refrigerants? The good thing is I live in an area where I can get a lot of free cooling for a good portion of the year.

The reason I am doing this is the extreme cost of Xenon gas for other research purposes, its so outragiously expensive that I think it would be worth pouring a small concrete pad and setting up some low volume equipment and design it so that I can just let it run and come pick up a bottle of Xe after a few weeks.

It's not clear how much xenon you are hoping to produce. The reason xenon is expensive is that it occurs in such a small fraction of the atmosphere (approx. 87 ppb).

 

[rppearso]

So is Neon and Kypton and they are a fraction of the cost of Xenon so I don't think it has anything to do with the process but rather the supply and demand. All of the nobel gases come from a fractionation process (or in some cases Helium is produced with natural gas). But that is beside the point, I need Xenon for research purposes and don't want spend huge amounts of money for every single bottle. Becasue Xenon has a high dew point it should condense out before all the other gases so I may not need a full on fractionation but I will also need krypton so I might want to be able to separate Xenon and Krypton but I don't really care about the nitrogen and oxygen and if the full inlet stream was liquified the nitrogen would overwhelm the column making a much larger (and more expensvie) column necessary. Krypton is cheap today but since it is a trace gas it should not take too much to design in a way to separate it out as well, possibly in a second stage of cooling/condensation. Just in case the price sky rockets and I still need it.

So I was thinking of designing a process that would chill inlet air to a temperatuer between the dew point of Xenon/Krypton and Nitrogen/Oxygen/Argon. Even if I did need Argon, it is an industrial gas and I can purchase locally for not too horrible cost. I might want Helium and Neon as well but I don't think they are terribly expensive either, I mean they fill kids balloons with Helium.

Just doing compression and let down is not sufficient so I would need an interstage cooler that had a heat transfer medium capable of really low JT temperatures.

Do you know if there is a specific refrigerant or other gas that is used for cryogentic cooling? If all I am worried about is condensing out nobel gases and just venting nitrogen and oxygen as waste gas the process can be drasticly miniturized. I realize there is a market for liquid nitrogen but I don't think I could absorb the upfront equipment costs to handle all that extra volume.

The only big piece of equipment I might need would be the chiller/condenser to make sure there was enough residence time for the nobel gasses to drop out before the waste gas is vented.

 

[SteamKing]

The properties of interest for neon, krypton, and xenon are here:

Gas      Conc.           BP      Rel. Abundance
Neon      18.2 ppm       27 K         209
Krypton    1   ppm      120 K          11.5
Xenon     87   ppb      165 K           1

Oxygen                   90 K
Nitrogen                 77 K

From Wiki, the production of xenon is described briefly:

"Xenon is obtained commercially as a byproduct of the separation of air into oxygen and nitrogen. After this separation, generally performed by fractional distillation in a double-column plant, the liquid oxygen produced will contain small quantities of krypton and xenon. By additional fractional distillation steps, the liquid oxygen may be enriched to contain 0.1–0.2% of a krypton/xenon mixture, which is extracted either via adsorption onto silica gel or by distillation. Finally, the krypton/xenon mixture may be separated into krypton and xenon via distillation.[50][51] Worldwide production of xenon in 1998 was estimated at 5,000–7,000 m3.[52] Because of its low abundance, xenon is much more expensive than the lighter noble gases—approximate prices for the purchase of small quantities in Europe in 1999 were 10 €/L for xenon, 1 €/L for krypton, and 0.20 €/L for neon;[52] the much more plentiful argon costs less than a cent per liter."

You can see that even the commercial production of xenon does not rely on liquefication of air alone to extract xenon. Neon is relatively cheap because it is more abundant than xenon and krypton and it boils at a much lower temperature than the other two gases. By the time you lower the temp. to the BP of neon, everything else in the air has already been removed, so no separation of different elements is required as in the case of krypton and xenon.

 

[rppearso]

If you notice, Xenon is just as rare as Krypton yet 10 times more expensive. The dew point of the following gases are

Xenon -162F
Krypton -251F
Argon -302F
Neon -410F
Helium -452F
Oxygen -297F
Nitrogen -320F

So Krypton and Xenon will condense at higher temperatures than the rest of the gases. So if all I wanted was Xe and Kr I would get the temp to around -280F then just do a simple 2 component distillation. Or just condense out the Xe at the inner stage around -200 and the Kr on the second stage that way all the rest of the gases just vent off.

Argon is tricky because it is right inbetween N2 and O2. Helium and Neon would be easy because they would boil off long before anything else. But Argon, Neon and Helium are relativly cheap so I am not really worried about processing those gasses.

My question is what gas/refrigerant is the best for JT effects to do interstage cooling?

 

[rppearso]

Also when I was getting quotes the price was much higher than 10 euro/pounds or what ever that symbol is per liter. A bottle of Xe was anywhere from 4,000 to 10,000 dollars, don't recall what the pressure was but regardless that WAY too much money. If it is only 10 euro/liter maybe I need to find a new distributor, its not like its a hazerdous substance.

 

[SteamKing]

IDK what table you are reading, but krypton is about 10 times more abundant than xenon in the atmosphere. That's what 'Relative Abundance' means. If xenon = 1 and krypton = 11.5, there's 11.5 times as much krypton as xenon.

You don't specify the size of the bottle of xenon for which you were quoted $4000-$10000. The price of xenon is going to depend on the purity you desire and the quantity you wish to purchase.

Alibaba has some price quotes for xenon gas, which appear to be more affordable:

http://www.alibaba.com/showroom/xenon-gas.html

Be careful, because minimum quantities for purchase are also specified.

There are different varieties of xenon which are useful in research. If you desire to purchase one of the radioactive xenon isotopes, I'm sure this material will command a rather high price.

 

[rppearso]

The stuff I was looking at was just regular research grade Xe. The bottle was a DOT class 3AA-2015 (class 3 bottle) so basically the smallest refillable bottle you can buy. The quotes I got were from Air Liquid specialty gas division.

I got your link to work on firefox, thank you so much. Looks like that is really really pure stuff, I don't think I need Xe that pure, so if I could get it down to like 10$/liter then I could just buy it.

 

[streety]

https://www.physicsforums.com/threads/liquid-nitrogen-generator-n2-gas-generator.756538/

Have you seen this?

 

[SupaVillain]

As far as the joule-thomson effect check out... i think it's diyliquidnitrogen.com or something very close. He uses a 4500 psi scuba compressor (so it's oilless) at like 3 or 4 scfm straight into a molecular sieve and a nitrogen separator. He has videos online and his website describes it all well. there's also benNbuilds on youtube with a video something along the lines of diy air liquefier. He uses a scuba compressor of about the same strength i believe. You don't need 9000 psig unless youre the prince of all saiyans

 

[SupaVillain]

Im also considering building a cryo setup to accumulate Xenon because i want to make xenon difluoride and use it for MEMS specifically thermopile bridges that can serve as a FLIR (forward long range infrared camera) but it is expensive and xenon is so rare. Check out cascade cpu cooling and the cascade refriegeration system schematics on google images it might be strong enough to do what you want

 

[rppearso]

Very cool I will check it out, I have decided to first try my hands at a CO2 laser so that the gas medium is not such a cost burden or massive technical challenge. But I will be revisiting Xcimer lasers. Thank you.