Sciencemadness Discussion Board

Building an O2 concentrator & N2 & argon extractor

RogueRose - 13-5-2019 at 00:02

Well I've been working on this since I got interested in glass blowing and the prices for on-site O2 production was ridiculous - like $2500 - $20,000 ($20K for a heavy duty one person shop). I can make it for a mere fraction and I've designed 4 different version, currently on MKIV (only minimal production of previous versions).

What I need to do that I'm trying to figure out is the following:
How to dry the air (@ 5-10CFM input) to at least 90-95%
Remove any hydrocarbon (from oil in compressor)
Remove particulate (1um might be over-kill, 5um might be fine).

I'm considering using 3 different methods to dry the air:
Anhydrous CaCl2
Anhydrous MgCl2
NaOH
I've found that MgCl2 draws water from the air MUCH faster and better than CaCl2. The CaCl2 can only be used up to the tetahydrate b/c over 86F it looses water. The MgCl2 can make use of the hexahydrate which decomposes at 243F and after it reaches hexahydrate it turns to a saturated solution and drips off.

NaOH I am more unclear of. I think it will absorb water up to 50/50 NaOH/H2O, so if I start with solid NaOH, once it becomes a liquid, the air can be bubbled through and the water removed.

The MgCl2 is about $14/50lbs so a go0d price and should last a long time.

I was thinking of using a 5 gallon bucket as the air filter filled with the salt/hydroxide, and pass air up through it. At the bottom there would be a space with a mesh to allow for the air to pass evenly through the entire surface area of the container.

I could do the same with activated carbon but I also have 2 of these:
http://www.sciencemadness.org/talk/post.php?action=newthread...

water filters.jpg - 34kB

With these I could add 5um 4.5" diameter x 20" long activated carbon filters and then a 1um or 5um particulate filter in the other canister. I could use 1-2 of these and use one for MgCl2 and another for NaOH to remove water.


Doing the above would allow for the use of any air compressor and not just oil-less or higher priced units (ones I don't already have). I have all of the above already, so I'd like to try it as a proof of concept.


One of the more difficult thing I've had finding are cylinders to house the N2 absorber (zeolite). I'm thinking of using old fire extinguishers - I need 2 matching units that can handle 30-35psi so even old ones that don't pass inspection should be fine. Other options are old gas tanks like CO2, Argon, etc but they are usually heavy steel and finally there are old scuba tanks, often steel as well, but some are Al, carbon fiber, fiberglass, etc. Maybe some firefighter SCBA tanks would work - they are usually fiberglass as well and maybe I can find decommissioned ones. Any suggestions would be helpful. I'm looking for 3L-12L tanks

i'm looking at this for use with propane and other fuels for melting very high temp materials and as a source for my oxyacetylene torch instead of buying O2 as this will be much less expensive.

Next will be to either get H2 tanks or build a H2/O2 generator for a torch and possibly reactions that need H2.

Oh, and if I decide I want to capture N2 or argon, that is an additional addon that can be done, but ATM, I don't need to plan to save those.




Chemetix - 13-5-2019 at 01:15

You need a specialised group, like the Concentrated Lampworkers facebook group. It depends on where you are in the world but there are all sorts of resources to make an O2 concentrator. 35psi is easily done with aluminium tubing and a welder, a machine shop to put a thread in an end piece if you have to. The N2 isn't pure though, the beds have to bleed off some oxy along with some N2 before the O2 is concentrated enough for collection. I've thought about making one and I go through a lot of oxy (i'm a scientific glassblower) but the time to build and perfect one is a lot of down time...the economics are subjective to the individual I guess.

RogueRose - 13-5-2019 at 18:22

Quote: Originally posted by Chemetix  
You need a specialised group, like the Concentrated Lampworkers facebook group. It depends on where you are in the world but there are all sorts of resources to make an O2 concentrator. 35psi is easily done with aluminium tubing and a welder, a machine shop to put a thread in an end piece if you have to. The N2 isn't pure though, the beds have to bleed off some oxy along with some N2 before the O2 is concentrated enough for collection. I've thought about making one and I go through a lot of oxy (i'm a scientific glassblower) but the time to build and perfect one is a lot of down time...the economics are subjective to the individual I guess.


Yeah, I know the N2 isn't pure enough at the exit stage, but it is in higher concentration than normal, so it's a good starting point if I wanted to add on a N2 purification & capture section - as well as argon.

One of the biggest problems I'm having is deciding which zeolite to use as there are basically 3 kinds to choose from, and then there is size/shape (cylinder, sphere & powder). I found a place that sells surplus stuff that has a trailer load of (I think X13??) @ $.50/lb which is awesome or $500/ton which is $.25/lb. The problem is I have no need for 2000lbs and I'd like to try this stuff before I buy since the expiration date on the pallets says 04/12/2015. Now IDK how this stuff can go bad, especially if not being used, but it is almost 5 years past expiration date so I'm a little worried.




Twospoons - 13-5-2019 at 19:23

Quote: Originally posted by RogueRose  

What I need to do that I'm trying to figure out is the following:
How to dry the air (@ 5-10CFM input) to at least 90-95%
Remove any hydrocarbon (from oil in compressor)
Remove particulate (1um might be over-kill, 5um might be fine).


- use an oil-free compressor
- use a chiller on the low pressure side and chill the air to 1C to condense out any water

Now you no longer need a dessicant supply or an oil mist filter.

RogueRose - 14-5-2019 at 07:29

Quote: Originally posted by Twospoons  
Quote: Originally posted by RogueRose  

What I need to do that I'm trying to figure out is the following:
How to dry the air (@ 5-10CFM input) to at least 90-95%
Remove any hydrocarbon (from oil in compressor)
Remove particulate (1um might be over-kill, 5um might be fine).


- use an oil-free compressor
- use a chiller on the low pressure side and chill the air to 1C to condense out any water

Now you no longer need a dessicant supply or an oil mist filter.


That would be what I would do if I had to buy new equipment but I'm using what I have now. It'd be expensive to replace a setup that does 20+ CFM @120PSI with an oiless compressor even though I can't use that much output in O2.

As for the chiller I'm wondering whether an enclosed dehumidifier would work or maybe a mini fridge (one with a real compressor) cranked to the coldest - I could duct the air in at the top and have it exit at the bottom, maybe some coiled copper tubing (1/2 - 3/4") to keep the flow somewhat normal until it has cooled down. Another option is a small freezer. The chiller units I've seen are pretty expensive and IDK if they can handle the volume of air that will come in. But if they are meant to keep the temp at -20 to -50, then that cold temp will help with the greater volume and maybe hit the ~32-33F mark.

I wonder which is more cost effective, cooling the air first or using a rechargeable desiccant (maybe sodium silicate cat litter??) that needs heated every XX cubic ft. I would think there might be some benefit to compressing the cold air as it will be more dense upon compression, so maybe you get a little more volume overall and it would also help keep the compressor from overheating if that is a problem.

Twospoons - 15-5-2019 at 20:41

I thought the chiller would be more convenient, wasn't considering cost effectiveness

Sulaiman - 15-5-2019 at 20:49

Quote: Originally posted by RogueRose  

I wonder which is more cost effective, cooling the air first or using a rechargeable desiccant (maybe sodium silicate cat litter??) that needs heated every XX cubic ft. I would think there might be some benefit to compressing the cold air as it will be more dense upon compression, so maybe you get a little more volume overall and it would also help keep the compressor from overheating if that is a problem.

I guess that if your overall consumption is low, and your local atmosphere is 'dry'
then dessicants would be the best choice,
but
for high volumes of humid air, maybe refrigeration is better ?