I'm looking for a way to bypass the bidirectional diffusive gas exchange that happens in human alveoli, to allow aquanauts to breathe a standard Nitrogen-Oxygen air mix at very high pressures (up to 300 atmospheres) found in ambient pressure deep submarine habitats. Being able to use ambient pressure habitats at this depth would be great from a structural engineering standpoint, but nitrogen narcosis would be absolutely deadly so I'm wondering if anyone knows a way of stopping nitrogen from diffusing into the bloodstream.

I'm disregarding High Pressure Nervous Syndrome (HPNS) here, on the assumption that since whales have found a workaround, a similar sort of neurochemical alteration would also be viable in humans.

edit: A quick clarification on nitrogen narcosis vs HPNS.

Nitrogen narcosis is just one of a number of different gas narcoses, any inert gas will cause narcotic effects past a certain pressure as it diffuses into the brain and (as best eh boffins can tell) physically interfering with the process of chemical signaling in the brain. Thus replacing nitrogen with another inert gas (ie: helium) is a no-go.

High Pressure Nervous Syndrome on the other hand is not a function of breathing gas (there is some confusion about this, as the Wikipedia article refers to it as occurring with helium based breathing mixes, however subsequent research has revealed that it's actually something that occurs due to pressure distorting biochemical enzymes in the body, particularly in the nervous system. The best guess in the field of cetacean neurology right now (from what I can tell) is that deep diving whales use neurochemical transmitters that are less susceptible to this pressure induced distortion in function.

You can also assume that the relative concentration of atmospheric O2 and C02 in the high pressure habitat has been altered to create partial pressures equivalent to those at sea level - that is nitrogen makes up a larger overall percentage of the local breathing mix.

  • $\begingroup$ And yes, that's supposed to say impermeable, not impereable. $\endgroup$
    – Algebraist
    Commented May 20, 2019 at 2:39
  • $\begingroup$ Are you trying to bioengineer humans or are you looking for tech? $\endgroup$
    – Willk
    Commented May 20, 2019 at 2:40
  • $\begingroup$ The ideal system is something that minimizes the amount of alternation that has to be done to the baseline human. Within that, it can be either a biological mechanism, some sort of dry nanotech, or even a cybernetic implant. $\endgroup$
    – Algebraist
    Commented May 20, 2019 at 2:44
  • $\begingroup$ Welcome to Worldbuilding.SE Algebraist. Thanks for jumping right in with an interesting question. Please check out our tour and help center. $\endgroup$
    – Cyn
    Commented May 20, 2019 at 4:38
  • 1
    $\begingroup$ Good question, but just to clarify, Whales haven't figured out a workaround, they just hold their breath for a very long time,, its basically free diving, humans have managed below 200 meters free diving, and the risks from nitrogen aren't there as the diver hasn't inhaled pressurized air, normal sea level air, so maybe a suit that can hold back the pressure and allow sea level air to be breathed is the way to go $\endgroup$ Commented May 20, 2019 at 11:49

4 Answers 4


I'm gonna go out on a limb here and say no (and get away with it because you haven't asked for hard science!)

$N_2$ is smaller and lighter than $CO_2$, and neither are polar. There's no trivial way to make a membrane that would let the larger molecule through and reject the smaller. Technology does exist to filter nitrogen (used in oxygen concentrators) and carbon dioxide (as used in rebreathers) out of the air, but this wouldn't help if you're using the result as a breathing gas because you've either massively increased the partial pressures of the rest of the components of your breathing gas (which is dangerous, see below) or you've filled it with some other inert gas to make up the balance, thus negating your original design goal (and no alternative gas is perfect, all have costs and risks associated with them).

Your best bet is to make an alternate gas exchange mechanism that doesn't need to use ambient atmospheric pressure to get gas into and out of blood, ie. a specialised artificial lung (ECMO). You'd probably have to dissolve your required breathing gases into some kind of fluid (much like a liquid breathing approach) and then let the gasses diffuse across into the bloodstream across a simpler membrane.

This would be a non-trivial bit of engineering and medicine, if you wanted it to be compact, reliable and safe in underwater environments. I note that this approach was used in Peter Watts' Starfish and his other Rifters books, with a surgically implanted lung replacement. He went a step further and made it extract oxygen from water via electrolysis, too (read the link for more details and related work). If you didn't want that, you could still use it as an underwater breathing system... just take a suitable carbon dioxide scrubber and oxygen source that works with the breathing fluid, and run it as a sort of liquid-loop rebreather.

By removing ambient pressure from the equation you not only deal with gas toxicity issues, but also with some gas expansion issues when you change depth. There's much less risk of decompression sickness when there's no dissolved gasses to form bubbles in your body, for example. You'd still have to deal with keeping the lungs, sinuses and eustachian tubes pressurised (Watts' system flooded the breathing passages with saline when the artificial lung was operating), so you still have to be careful about squeezes and overexpansion injuries.

nitrogen narcosis would be absolutely deadly

Many things become deadly at high enough pressures. Acute oxygen toxicity will cause seizures, long term exposure to high pressure oxygen will damage the lungs (amongst other things). You'll need to be super careful about other contaminants which may go from being an irritation on the surface to being fatal at depth. You've also got other serious, non-biological issues, such as the fact that there's lots of extra oxygen in the air that will make fires in your habitats really exciting .

I'm disregarding High Pressure Nervous Syndrome here, since whales have found a workaround

Whales, and indeed all other diving sea mammals, hold their breath. This sharply limits how long their dives can be. It also limits the maximum amount of any one gas that can diffuse into their bodies. They also have adaptations to better fill their blood and muscles with oxygen pre-dive and limit gas transfer from the lungs at depth, reducing the ability of undesirable gasses to diffuse into their blood and then cause toxic or narcotic effects, or risk of the bends. It isn't at all clear that they have "found a workaround" in the sense you mean, because they may well simply not expose their nervous systems to the types and amounts of dissolved gasses that deep-sea divers do.

Remaining at depth and continuously breathing from some other air source will result in the "inert" parts of your breathing gas dissolving into your blood stream, something that does not happen to whales. Disregard diving mammals when considering long-term underwater habitation; they don't do it and aren't adapted for it. The longest dive by a mammal is a little over 2 hours, by something with quite different physiology to humans. I'm pretty certain that if you have whales and walruses underwater breathing equipment, you'd find that they develop a whole raft of pressure-related illnesses in due course, just like humans do.

Remaining under high pressure for extended periods of time has a whole new set of issues which are poorly understood. Have a read up on the risks posed to saturation divers for examples of this sort of problem. Changing the atmosphere composition is unlikely to fix all these issues.

  • $\begingroup$ The idea is in fact to remove gas toxicity and gas expansion issues, by preventing Nitrogen from getting into the bloodstream. I'd also considered the ECMO approach, but was hoping for something that requires a little less bulk or active machinery. $\endgroup$
    – Algebraist
    Commented May 20, 2019 at 13:54
  • 1
    $\begingroup$ @Algebraist yeah, I guessed as much. But you're just out of luck; there's no membrane that will do what you want, and anything using ambient pressure will end up needing to use problematic inert gasses to keep the partial pressures acceptable. I'm fairly certain you're stuck with ECMO-type stuff. $\endgroup$ Commented May 20, 2019 at 13:56

Depending on how deep you go, Nitrogen is not your only problem. Once you get below about 60m of depth, oxygen toxicity also kicks in and CO2 poisoning is also something divers really have to worry about. The way they deal with all of this is special air mixes that contain percentages of inert gas to reduce the amount of oxygen and nitrogen they breathe in, and letting their exhalations escape completely meaning that the CO2 can't build up in an enclosed system.

So; at-pressure environments deep in the ocean are possible, but you're going to be bringing in some inert gases and trying to keep your environment as closed as possible. But, how do you get the balance right in the first place?

Let's start with the nitrogen; nitrogen filtration is a thing, and it's possible to purify and extract nitrogen from the atmosphere already. We can do the same thing with Oxygen to some degree, and we even understand the theory of chemically freeing oxygen from CO2. It's not easy, but it's possible.

So; in its simplest form, the approach you need is;

1) bring down a tank of inert gas.
2) start filtering out and capturing nitrogen, but
3) start back-filling atmospheric pressure with inert gas.

You also need to do the same thing with oxygen if you can, for emergencies. Keep monitoring your atmospheric levels, and when your oxygen partial pressure is at around 0.2 ATM, and your Nitrogen is at around 0.6 ATM, you've achieved your balance.

Over the long term, your environment needs plants to survive of course, so you should have large domes of crops naturally converting CO2 to O2 for you. Also, plant lots of legumes like beans. These take the nitrogen out of the atmosphere and 'fix' it into the soil, which is needed for good crop management. (This is why you can't just strip all the nitrogen out of the atmosphere.)

Put simply though, all the equipment you need on an industrial scale are already available for both the capture, release and monitoring of constituent gases in your atmosphere. Over the long term you want to also try and develop as sustainable an eco-system as possible in your domes, as this will ease the wear and tear on your equipment. This is actually quite achievable with current technology; the only word of caution is that the earth's ecology is a complex balance of interaction which you cannot hope to perfectly replicate in your underwater environment; you're still going to need to refresh aspects of your environment, including atmosphere, from time to time.

  • $\begingroup$ If you've got a decent power source, you don't need a bunch of vegetables to do atmospheric processing for you. At the extreme end, you can do what the space station does, but really it would be easier to bring down regular supplies (or a continuous feed) of air from the surface, have a spare gas tankage and some hefty carbon dioxide filters. $\endgroup$ Commented May 20, 2019 at 12:03

You could do this with a cross between a SCUBA system and an oxygen concentrator.


Oxygen concentrators typically use pressure swing adsorption (PSA) technology and are used very widely for oxygen provision in healthcare applications, especially where liquid or pressurized oxygen is too dangerous or inconvenient, such as in homes or in portable clinics. For other purposes there are also concentrators based on membrane technology. An oxygen concentrator takes in air and removes nitrogen from it, leaving an oxygen enriched gas for use by people requiring medical oxygen due to low oxygen levels in their blood.1 Oxygen concentrators are also used to provide an economical source of oxygen in industrial processes, where they are also known as oxygen gas generators or oxygen generation plants. Oxygen concentrators utilize a molecular sieve to adsorb gases and operate on the principle of rapid pressure swing adsorption of atmospheric nitrogen onto zeolite minerals and then venting the nitrogen. This type of adsorption system is therefore functionally a nitrogen scrubber leaving the other atmospheric gases to pass through. This leaves oxygen as the primary gas remaining. PSA technology is a reliable and economical technique for small to mid-scale oxygen generation, with cryogenic separation more suitable at higher volumes and external delivery generally more suitable for small volumes.2

But a problem for your humans is that even if they purge the N2 they will still need helium to breathe. At that pressure if you get rid of nitrogen you will have remaining high pressure oxygen which will rapidly burn the lungs. You need to also cut back the oxygen to minimal percentage and introduce an inert gas (helium) to make up the missing pressure so they can inhale.

Use conjunctiva as your respiratory membrane.

subconjunctival hemorrhage

Your deep dwellers need to be protected from high concentrations of O2 and N2. The interface we have with the atmosphere (lungs) is way more than they need to exchange gases at those concentrations, and the lungs present no barrier to concentrated nitrogen gas equilibrating with the blood. Lung delicate lungs will be burned by high concentration oxygen.

They need the lungs out of the look. Your deep dwellers keep their fetal circulation. Their blood bypasses the lungs.

fetal circulation

In the fetus, instead of gas exchange in the lung, gas exchange happens in the placenta. With high concentration oxygen, gas exchange can happen over a small gas exposed area. They use the eyes.

The eyes are exposed to ambient air. Your deep dwellers run high hemoglobin and blood is exposed more widely in the eye, turning the sclera red in the manner of a conjunctival hemorrhage. Those are extravasated blood, like a bruise, but they stay bright red because the cornea is permeable to oxygen.

That takes care of oxygenation. Nitrogen entry is limited but will still gradually equilibrate. That slight excess of nitrogen, as well as CO2 produced by respiration are taken care of internally.

Excess nitrogen is fixed into urea by commensal nitrogen fixers in the gut. Urea is used by the body or excreted.

CO2 is the trickiest because we make a lot. How to ditch it without breathing? I propose it be sequestered in alkaline fluids in the stomach and cleared via burping. This will have the additional benefit of allowing phonation, since without respiration speech would not be possible. Your deep dwellers will necessarily be terse, choosing their words economically.


Should all else fail, you can turn to a biochemistry solution for this one.
Cells have transport/channel proteins in their walls, that let only specific molecular or atomic ions pass.

The ability to let only specific particles pass is related to the energetic structure of the binding part of a channel protein, which is 'gauged' onto the energetic structure of the target transport particle. Anything that doesn't match, simply cannot pass.

While I believe this could be bioenginered into some kind of breathing mask (handwaving up the speed of the process a bit), protein bindings are not without fail. Hemoglobin, which the body uses to primarily transport $\rm O_2$, can also bind to $\rm CO$, causing poisoning. But some future tech might be able to handle this.

  • $\begingroup$ I'm a bit concerned that at the pressures involved, the Nitrogen might start binding to the Hemoglobin as well, I found a paper on it, but the chemistry is far outside my area of expertise: jbc.org/content/263/27/13515.full.pdf $\endgroup$
    – Algebraist
    Commented May 20, 2019 at 20:11
  • $\begingroup$ @Algebraist: Well Nitrogen does apparently bind (source: Wikipedia) to Hemoglobin as well. My point in mentioning hemoglobin is just 'Natures engineering is not perfect, maybe future tech could do better'. I'm kind of handwaving this, but the general idea was to use advanced proteins with very selective coupling energies. At this point you probably wouldn't call it biochemistry anymore, but nanoengineering. $\endgroup$ Commented May 20, 2019 at 21:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .