A neighboring acquaintance was a long-time smoker and contracted COPD (chronic obstructive pulmonary disease) as a result. After collapsing and spending a few weeks in a hospital, and giving up smoking, this person went home and now finds it necessary to breathe pure oxygen fairly often. This is being supplied in the form of smallish pressure tanks (about one meter long each). It can also be supplied in liquid form (rather more expensively).

There is an alternate device called an "oxygen concentrator" which this person tried, but declared wasn't good enough. It is not a very large appliance, maybe a tenth of a cubic meter in size, and uses perhaps 400 watts of power to do its job of increasing the percentage of oxygen in the air it processes. I don't know what the percentage of oxygen is, that it outputs (ordinary air is about 20% O2), but like I said, it doesn't seem to be high enough for it to be helpful to this particular patient.

Normally oxygen is produced in quantity by a process that first liquefies air and then allows it to partly evaporate (which separates O2 from N2 and other gases; they all have different boiling points). But so far as I know, this process does not scale well to small sizes for low output (suitable for just one person, and I don't know what process the oxygen concentrator uses).

So, time for a HalfBaked suggestion. At the moment I'm going to ask you to think about the following processes in terms of doing them at 100% efficiency. We'll get back to that later.

Start with an ordinary water-electrolysis cell. It consumes a fair amount of electric power, but it can produce pure oxygen from water. Also, of course, it produces hydrogen, which we don't want.

However, now consider a "fuel cell", which come in a variety of types, depending on the fuel, and one type of which is a device that combines hydrogen and oxygen to make electricity and water (the opposite of electrolysis).

Suppose we therefore take the hydrogen output from our electrolysis cell and feed it directly into a fuel cell. We specifically want to use the kind of fuel cell that is being developed for use in automobiles, because it gets its oxygen from ordinary air; it does not need a supply of pure oxygen.

We now use the electrical output of the fuel cell (AND its water output) to feed the electrolysis cell. The net effect is that this pair of devices, in a fairly compact space, extracts pure oxygen from the air, at a rate suitable for just one person who medically requires it. At 100% efficiency of fuel-cell operation and electrolysis-cell operation, we also would not need any other power to be supplied to this appliance (and we should never need to add any water, either).

Of course, we cannot actually get 100% efficiency in those devices, so we would need to supply SOME power, and perhaps a tiny amount of water, just enough to make up for the efficiency losses (after supplying an initial amount to generate an initial amount of hydrogen, of course). Note that because it is rather difficult to make something totally leakproof for hydrogen, a small amount could be expected to leak (not a safety problem; it will leak out of the house faster than it leaks out of a well-built unit), and need to be replaced by adding a little distilled water occasionally. Anyway, the fuel cell can indeed supply most of the power needed by the electrolysis cell; we merely need to make up the difference from the power line. I suspect this will be rather less power than the 400 watts used by that wimpy oxygen concentrator....
-- Vernon, Oct 27 2010

Some oxygen concentators http://www.airsep.c...nconcentrators.html
The ones with wheels are about half a meter tall. [Vernon, Oct 27 2010]

Efficient Electrolysis Efficient_20Electrolysis_3f
As mentioned in an annotation [Vernon, Oct 28 2010]

Gibbs free energy http://en.wikipedia...i/Gibbs_free_energy
The fly in the thermodynamic ointment [8th of 7, Oct 28 2010]

This sounds Bakeable, but might run into Product Liability and safety issues.

The idea of "re-burning" the hydrogen in a fuel cell is ingenious and probably practical, worthy of a bun in itself. Alternatively the hydrogen could be used for cooking, heating, or inflating party balloons.

It could also run a small internal-combustion generator.

Purity of the oxygen might be an issue; the electrolysis cell will require noble metal electrodes, and a suitable electrolyte. Potassium hydroxide would probably be suitable.

Note that pure oxygen is quite dangerous in itself, being capable of causing all sorts of runaway combustion effects, hence [+]
-- 8th of 7, Oct 27 2010


When I picture water electrolysis I imagine a timidly bubbling strip and gradual accumulation of gas. I suspect an electrolysis outfit big enough to produce volumes of O2 suitable for huffing would be big.

I read on wikipedia how oxygen concentrators work - by adsorbing nitrogen under pressure, with nitrogen then released with next cycle and zeolite restored for next. I wonder if this could be done iteratively, using a sequence of concentrators to produce higher and higher O2 concentrations? Of course 2 or more devices would also have a larger footprint. But anyone with access to these devices could do the experiment - if the outflow from one is fed to the next, is the resulting gas product higher in O2? If so, one could accomodate higher O2 needs without the need for air liquefaction apparatus etc.
-- bungston, Oct 27 2010


// outfit big enough to produce volumes of O2 ... would be big. //

No, it's all about the current you put in to it. Size isn't important (Although electrode surface area is).
-- 8th of 7, Oct 27 2010


[8th of 7], a combustion generator is MUCH less energy-efficient than a fuel-cell generator.
-- Vernon, Oct 28 2010


// people who need .. oxygen supply .. for health //

That would be all of your species, then ? We have observed you do not do well in low oxygen levels.
-- 8th of 7, Oct 28 2010


From memory, hydrogen fuel cells have a typical efficiency of about 50% and electrolysis has a typical efficiency of 30%, so this system would have an efficiency of 15%. So if your device runs at 100W, you get 85W heat and 15W creating oxygen. But if this is more efficient/smaller that a regular oxygen concentrator (or a series of oxygen concentrators one feeding into the next), then great.
-- xaviergisz, Oct 28 2010


[xaviergisz], it is my understanding that the efficiency of both these processes is at least 66% --and that was before modern efforts to enhance it. For example, see "efficient electrolysis" link, an old Idea I posted.
-- Vernon, Oct 28 2010


You could go the other route of using refrigeration to liquify and separate the oxygen, then obtain energy from its subsequent evaporation. I think gas compression cycles tend to be more efficient than electrolysis.
-- Aq_Bi, Oct 28 2010


You're probably right, if the system is well insulated - which isn't difficult.

But you need to think carefully about the // obtain energy // thing - entropy rules. You can use LOX to absorb ambient energy, and use the resulting pressure as an energy source, but go look up Gibbs Free Energy <link>
-- 8th of 7, Oct 28 2010


[21 Quest], the OBOGS link clearly states that zeolites are used to absorb nitrogen, much as [bungston] described in his anno. It doesn't say how much is used (could be a rather larger amount than in the average home oxygen concentrator), and it doesn't say how much power is used to run it. Also, note that a higher power usage (and an aircraft generally has lots of power available) could allow a relatively small unit to still accomplish a lot of oxygen enrichment.
-- Vernon, Oct 29 2010


posted in 2010. Did you ever try to make it?
-- pashute, Dec 20 2016



random, halfbakery