See the link for a description of an oxygen generator that combines hydrogen with atmospheric oxygen, and then uses electrolysis to separate the pure oxygen.

While the other method will work, there is a significant question regarding its overall energy efficiency. So, the goal here is to consider some other chemical reactions. Just for fun, let's start with the element mercury.

It happens that you can heat mercury in air, and it will form an oxide. It also happens that if you heat the oxide, it will decompose back into mercury and oxygen. That sounds pretty simple, doesn't it? Just do the first thing in part of the overall device, move the resulting oxide to another part of the device, and then obtain pure oxygen just by heating (presumably electrically) --and return the mercury to the first part of the device.

Of course that particular system will have a problem in that mercury vaporizes fairly easily, which means the person breathing the oxygen is likely to get poisoned by the mercury. So, what else is there?

It happens that gold oxide also decomposes with heat, and gold is not going to yield dangerous vapors like mercury. But making gold oxide is not a simple task, alas. So, what else is there?

How about natural hemoglobin? This iron compound is such that in the lungs, oxygen latches onto it easily, and elsewhere in the body, oxygen can be removed from it easily (both events at body temperature). I'd stop right here if I knew the stuff was stable enough for what we might call "small scale industrial use", getting shoved around in an endless loop inside a home oxygen generator. But what if it isn't that stable?

So, what I'd like to suggest is some molecular engineering, starting with thinking about the hemoglobin molecule, which is a fairly complex combination of proteins that apparently surround an iron atom in a way that lets it attract oxygen without making a strong chemical bond with the oxygen. Our goal is to build a different container (perhaps something as simple as a buckyball) that can do the same sort of thing, without being vulnerable to such things as bacteria eating hemoglobin. Of course we want it to be nicely stable in terms of thermal and mechanical manipulations, as well.

We might even do the above with some atom other than iron. The mollusc family of organisms uses a copper-based compound ("cyanoglobin") for oxygen transport. And who knows, perhaps if a mercury atom was appropriately surrounded, it could still work, and the weight of the surrounding molecules would keep it from escaping the oxygen generator, and poisoning someone.
-- Vernon, Jan 22 2016

Old Idea Oxygen_20generator
As mentioned in the main text. [Vernon, Jan 22 2016]

Commercial Oxygen gnerators http://www.onsitega...gen-generators.html
This gets to 95% using a zeolite system. They have "portable" medical ones as well. It only takes 4 people to carry it. [scad mientist, Jan 22 2016]

Palladium oxide https://en.wikipedi...Palladium(II)_oxide
When hot, combines with oxygen; when hotter, decomposes. Palladium is not volatile like mercury. Kind of expensive, though. [Vernon, Jan 27 2016]

In my high school (or maybe Jr. high) chemistry class we did a conservation of mass experiment where we heated iron filings in open air to create iron oxide. We verified that the mass had increased. Then we re-heated the iron oxide to release the oxygen, but put a lid on it while it cooled so very little oxygen would recombine with the iron. After that the mass returned to the original value.

Of course it's possible I'm remembering that wrong or there were additional undesirable reactions occurring, but if not, it seems to me we could just use straight iron without having to worry about a fancy unstable hemoglobin molecule. Of course hemoglobin would allow this to work at lower temperatures. I think I remember that for this experiment we had to use something hotter than our alcohol burners.

I guess with all that heat it would probably not be considered energy efficient unless we had a very good heat exchange mechanism.
-- scad mientist, Jan 22 2016


I don't understand this whatsoeveratall.

You seem to be describing systems that can absorb oxygen and then release it on demand, but how do they generate oxygen?
-- MaxwellBuchanan, Jan 22 2016


[MaxwellBuchanan], there exist things called "oxygen concentrators", which were discussed in the linked/older Idea. In both Ideas the goal is pure oxygen, not partly- concentrated oxygen. The "releasing" method in both is equivalent to "generation". How is that not clear?
-- Vernon, Jan 23 2016


So this is an oxygen concentrator.
-- notexactly, Jan 23 2016


[Vernon], interesting thoughts.

I have no idea of the energy efficiency involved, but how about: electrolysis of water to produce H2 and pure O2.

Run the H2 through a fuel cell that combines it with diffuse atmospheric oxygen.

Use the power produced by the fuel cell to partially offset the power consumed by the electrolysis of water, and run the exhaust water back to the electrolysis stage.
-- TIB, Jan 23 2016


// How is that not clear?//

Ah, right, so by "generation" you mean "purification".

An oxygen generator would actually generate oxygen, for instance by electrolysis.
-- MaxwellBuchanan, Jan 23 2016


[TIB], see the linked "Old Idea".

[MaxwellBuchanan], if heating gold oxide cause it to decompose and release oxygen, as mentioned in the main text, how is that not "generating" oxygen?
-- Vernon, Jan 23 2016


Just get a tree.
-- not_morrison_rm, Jan 23 2016


//how is that not "generating" oxygen?//

Well, yes, I guess. But I think what you're aiming at is better described as "purifying oxygen from air". No matter.
-- MaxwellBuchanan, Jan 23 2016


Re semantics I agree with Max. You are using the device to purify oxygen out of the impure mix that is air.

Re iron I like the idea of using plain iron. Hemoglobin works lots of jiggery pokery to side step the fact that you cant get iron that hot in an organism. But heat /cool in a machine would work fine. You could do it with induction.

The trick here is to maximally reuse the heat, ideally keeping the iron hot enough that you can push it forward and back thru oxidize / deoxidize. If waste heat could somehow be used to drive the evacuation of the chamber to receive the oxygen that would be slick.
-- bungston, Jan 26 2016


//Hemoglobin works lots of jiggery pokery//

Yes and no. At high concentrations, oxygen binds to haemoglobin; and at low concentrations, it's released. So it works a bit like a sponge transferring water from a puddle to a dry paper towel.

However, there is quite a lot of cleverness to ensure that the dissociation curve is very steep, for efficient transfer.
-- MaxwellBuchanan, Jan 26 2016


I suppose compared to something like the Krebs cycle hemoglobin is pretty straightforward.
-- bungston, Jan 27 2016


//compared to something like the Krebs cycle hemoglobin is pretty straightforward.//

There's no such thing as the Krebs cycle, not in isolation anyhow. Especially when, for convenience, it gets separated from the electron transport chain. Succinate dehydrogenase is a functional component of complex II in the electron transport chain so it all gets a bit interconnected, and a lot of it will happily run backwards. This morning's discovery is that humans would probably be a bit more heart-attack proof if aconitase wasn't so fussy about H2O2.
-- bs0u0155, Jan 27 2016



random, halfbakery