Oxygen Toxicity & ROS

Why oxygen is a slow poison

Leave a steel wreck in the sea and it slowly crumbles into orange flakes. That rust is oxygen quietly tearing the metal apart, the same gas you're breathing, doing chemistry. Your body runs on a carefully controlled version of that exact reaction to make energy. At high pressure the control slips, and a little of the same destructive chemistry starts happening inside your cells. That is the root of every kind of oxygen toxicity, the fast brain seizure and the slow lung burn alike. This article is the why underneath both. The one idea to carry out of it: the safe limit isn't a fixed wall, it's a balance that shifts with your day.

The sparks: reactive oxygen species

Burning oxygen for energy is a bit like grinding metal: most of the work is useful, but a few sparks fly off. Those sparks are reactive oxygen species (ROS): unstable molecules that will grab onto and damage almost anything nearby. Three matter most:

• Superoxide, the first spark, leaking off your cells' energy machinery.
• Hydrogen peroxide, what superoxide gets converted into (yes, the same stuff in the bottle under your sink).
• The hydroxyl radical, the nastiest of the three, and the one that actually shreds cell membranes, proteins, and DNA.

At normal surface pressure your cells leak only a tiny trickle of these, a small fraction of the oxygen passing through, and a cleanup crew of antioxidant enzymes mops them up as fast as they appear. The system sits in easy balance.

Why pressure tips the balance

Go deep on a rich mix and far more oxygen dissolves into your blood and tissues than your body was built for. Your cellular machinery gets flooded, and the spark rate climbs. Past a point the cleanup crew can't keep up. The leftover sparks start doing damage, and that imbalance has a name: oxidative stress. Everything else in oxygen toxicity is just where that damage lands.

Where the damage lands

In the brain, a seizure. Nerve cells are especially fragile here: they run hot, carry few spare antioxidants, and are built from the fatty molecules ROS love to attack. The sparks knock out GABA, the chemical that tells neurons to settle down, and without those brakes the brain can run away into a convulsion. That's CNS oxygen toxicity, and it can happen in minutes.

In the lungs, a slow burn. The same sparks inflame lung tissue over hours and days instead of minutes. That's pulmonary oxygen toxicity: slower, and far more forgiving.

What makes it worse: exercise and CO₂

Two things you control on a dive push the spark rate up.

Working hard burns more oxygen, which means more sparks, right when your antioxidant crew is already stretched thin by the pressure. This is why hard exertion at depth is the single biggest thing you can do to bring on a seizure, not just a little extra risk, but a real tipping point.

Carbon dioxide is the sneaky one. When CO₂ builds up, it widens the blood vessels in your brain and floods it with more blood, and more oxygen, raising the local spark rate even though your oxygen pressure (PO₂) on the computer hasn't moved. Raised CO₂ turns up as a documented factor in a meaningful share of investigated rebreather seizures, and it's almost certainly an undercount. CO₂ creeps in from skip-breathing, from dense gas that's hard to move, from a failing rebreather scrubber, and from working too hard. Managing it is part of managing oxygen.

Can't I just take antioxidants?

It's the obvious question, and the answer is no. Megadosing vitamin C and E before a dive doesn't meaningfully raise your enzymatic cleanup crew or reach the mitochondria where the sparks fly, so it won't buy you a higher oxygen limit. There's even evidence that blanket antioxidant dosing blunts useful adaptations your body makes to exercise. What does grow the cleanup crew is ordinary fitness: training nudges your own antioxidant defences upward over time. So the protective move isn't a pill, it's being fit, rested, and hydrated.

Why this matters for you

The oxygen limits you're taught aren't arbitrary. They mark roughly where the spark rate starts to catch up with the cleanup crew for an average person on an average day. A few things follow:

• Manage your CO₂. It raises the spark rate through a side door your oxygen gauge can't see.
• Treat exertion at depth as genuinely riskier than the same effort at the surface, because your margin is already thinner.
• Respect that the margin moves. Tired, dehydrated, cold, ill, or already loaded with oxygen from earlier dives, the same diver tolerates less. On those days, dive a more conservative PO₂ and take longer surface intervals.

That shifting margin is also part of why a hard, oxygen-heavy dive day leaves you so wrung out: residual oxidative stress is part of post-dive fatigue, not just tired muscles. No limit is a personal guarantee.

Keep reading

• CNS Oxygen Toxicity: when the brain runs away
• Pulmonary Oxygen Toxicity: the slow lung version
• Gas Density & CO₂ Buildup: where the CO₂ comes from

References

1. Gur I, Arieli Y, Matsliah Y. Methylphenidate and the risk of acute central nervous system oxygen toxicity: a rodent model and observational data in human divers. Diving and Hyperbaric Medicine. 2024;54(3):168–175.
2. Auten RL, Davis JM. Oxygen toxicity and reactive oxygen species: the devil is in the details. Pediatric Research. 2009;66(2):121–127.
3. Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine. 5th ed. Oxford University Press, 2015.
4. Demchenko IT, et al. Two faces of nitric oxide: implications for cellular mechanisms of oxygen toxicity. Journal of Applied Physiology. 2008.
5. Bitterman N. CNS oxygen toxicity. Undersea and Hyperbaric Medicine. 2004;31(1):63–72.
6. Pollock NW. Hazards in Rebreather Diving. 2024 (CO₂ as a contributing factor in rebreather oxygen-toxicity events).

The chemistry here is why I drill CO₂ control and PO₂ discipline in every CCR and trimix course. Ask me about training.