Today, the oxygen in the atmosphere is about 1018 kg. The estimated carbon biomass of the earth is about 4×1015. Less than 0.5%.

Estimated fossil fuels are around 5 times as much as existing carbon biomass. If these estimates are appropriate, then probably before the oxygen was released it was not locked up in carbon compounds.

Oxygen could have been extracted from inorganic stuff and cycled through the biomass.

Which reduced minerals were left behind? Not silicates and carbonates. Little magnetite crystals?

Maybe there was more SiO4 before (things like olivine), and living things tended to produce silicates with less oxygen (things like kaolin)? There isn't a whole lot of silica dissolved in seawater at any one time, but there's a whole lot of silica available and it could add up eventually. Silicates with more oxygen weathered away, and silicates with less oxygen were deposited?

Maybe there's so much free oxygen today because some H2 got produced, and it rose right out of the atmosphere and got swept away. So there's less water and more oxygen than there used to be.

Maybe the biomass estimates, and the fossil fuel estimates, don't actually tell us about the carbon, and there is lots of reduced carbon elsewhere.

I'm speculating, does anybody have good answers?


6 Answers 6


the great oxygenation event was caused by the evolution of photosynthesis, photosynthesis turns CO2 and water in to sugar and oxygen. so the oxygen came from CO2 and water.

keep in mind CO2 levels were many many times higher (~20X) than today. Hydrogen was not released it was combined with carbon and oxygen to make sugars

enter image description here

  • 3
    $\begingroup$ This is not a complete answer. What happens when the sugar is metabolized with oxygen to produce wager and CO2 again? That would give no net accumulation of oxygen. $\endgroup$ Commented Apr 30, 2018 at 12:50
  • 2
    $\begingroup$ These are not plants, they are algae they are not respiring, respiration did not evolve (or at least did not become widespread) until AFTER oxygen was made abundant by this reaction. $\endgroup$
    – John
    Commented Apr 30, 2018 at 17:43
  • $\begingroup$ Well, even cyanobacteria respire in some form, they have to in order to live. The point is that something must be happening to that fixed carbon, of whatever form, to stop it recombining with the oxygen, or otherwise oxidising. $\endgroup$ Commented May 1, 2018 at 13:33
  • 2
    $\begingroup$ Anaerobic respiration, no free oxygen involved. The fixed carbon is sugar, which does not need to be oxidized to be metabolized, there needs to be large amounts of free oxygen before spontaneous oxidation is a consideration. Oxygen levels would have just kept rising had aerobic respiration not evolved. $\endgroup$
    – John
    Commented May 2, 2018 at 1:00
  • $\begingroup$ @AndrewJonDodds You are right that if sugar (and organic matter) were metabolized, oxygen would be consumed again. However, with time, large quantities of organic matter get trapped in sediments and doesn't get metabolized again, leaving the oxygen in the atmosphere. Fossil fuels are just a tiny part of all that organic matter. $\endgroup$
    – Pere
    Commented Jul 13, 2018 at 18:20

I'd like to add to the other answers.

Yes, it is correct that photosynthesis by cyanobacteria caused the rise of atmospheric oxygen. Some important points to talk about:

Estimated fossil fuels are around 5 times as much as existing carbon biomass. If these estimates are appropriate, then probably before the oxygen was released it was not locked up in carbon compounds.

Fossil fuels are irrelevant to the discussion. Fossil fuels only started forming in the Phanerozoic, about 500 million years ago. The rise of oxygen occurred more than 2 billion years ago.

Oxygen could have been extracted from inorganic stuff and cycled through the biomass. Which reduced minerals were left behind? Not silicates and carbonates. Little magnetite crystals?

Nothing, actually. The minerals found on the surface of the Earth formed as a response to the presence of atmospheric oxygen. This means that oxidised minerals on the surface only appeared after we had atmospheric oxygen. This means that something else had to be reduced. For example, sedimentary detrital pyrite was a thing! Not possible today with the presence of oxygen.

Maybe there was more SiO4 before (things like olivine), and living things tended to produce silicates with less oxygen (things like kaolin)?

Not accurate. Both olivine and kaolinite are stable over a very wide oxidation states. Doing the charge-balance math shows that the silica exists as Si4+ in all silicates. If anything, olivine is a reduced formed (it has Fe2+) and is not stable on the surface as it oxidises ("rusts").

Silicates with more oxygen weathered away, and silicates with less oxygen were deposited?

Not really.

The key point is that the oxygen came from photosynthesis, and reduction of H2O and CO2 to form organic matter. Carbon and hydrogen form the organic stuff, and the leftover O2 is emitted to the atmosphere. But do we have enough carbon? You say:

was there that much CO2 before? Oxygen is supposedly the most common element in the earth's crust, about 466 parts per thousand by weight.

But, most of the oxygen is locked up in minerals and is completely irrelevant for this discussion as it does not participate in the reactions. Carbon is being constantly emitted by volcanism. It is then consumed by either mineral carbonation reactions (such as formation of olivine to magnesite: magnesium carbonate) on the surface, or by formation of organic matter. Therefore, CO2 doesn't have the chance to accumulate in the atmosphere. However, oxygen isn't consumed by reactions, so it accumulates.

In the beginning, oxygen didn't actually accumulate. It also had a sink: dissolved Fe2+ in the oceans. Oxidation of that iron caused formation of insoluble Fe3+ oxides and hydroxides, which we see today as banded iron formations. Only once the iron was consumed, the oxygen could start accumulating in the atmosphere.

  • 3
    $\begingroup$ "In the beginning, oxygen didn't actually accumulate. It also had a sink: dissolved Fe2+ in the oceans." In the beginning, there was no free oxygen. It was all bound to something. There were things in the oceans like Fe2+, because there was not enough oxygen to oxidize them. After the Great Oxygenation Event, there was free oxygen. About 10^18 kg. Oxygen weighs 16 grams per mole, so that's around 10^20 moles. 10^20 moles of oxygen that used to be attached to something is now free. What happened to whatever it was attached to? Where is it now? $\endgroup$
    – J Thomas
    Commented Apr 29, 2018 at 13:51
  • 2
    $\begingroup$ That's my key point. No free oxygen -> free oxygen. Something got reduced and stayed reduced, because before the oxygen was not free and now it is. What was it that used to be connected to oxygen, that stopped being connected to oxygen so the oxygen can be free? Unless the numbers are wrong (and they might be) there wasn't enough carbon to have an atmosphere with 20% CO2. The oxygen had to come from somewhere else and get converted to CO2 before it could be turned to O2. Then the reduced carbon could get converted to CO2 again and turned to O2 again. $\endgroup$
    – J Thomas
    Commented Apr 29, 2018 at 13:56
  • 1
    $\begingroup$ The carbon is off by a hundredfold or more, and 2 oxygens versus 1 carbon is a 2-fold difference. The carbon that I have listed is not enough. There needs to be a lot more carbon somewhere else that's reduced now, or else the oxygen came from somewhere else. As of now, the most plausible theory I've seen is that it came from water and the hydrogen has gone into space. Second best is that there are a whole lot of reduced minerals today that are under the sea floor, or underground, sealed under clay, or turned into clay, that are reduced more than they used to be. $\endgroup$
    – J Thomas
    Commented Apr 30, 2018 at 1:26
  • 2
    $\begingroup$ @JThomas well, there is plenty of pyrite in marine sediments. I'm not sure about mass balance considerations, but it could be the reduced stuff you're looking for. This is the only reduced solid mineral I can think of, other than hydrocarbons. Also remember that a lot of the stuff on the seafloor gets subducted back into the mantle so doing mass balance calculations based on current day biomass and fossil fuel mass is missing out on a lot of carbon. $\endgroup$
    – Gimelist
    Commented Apr 30, 2018 at 5:57
  • 1
    $\begingroup$ Thank you! So there could have been much more sulfate in the oceans in the past, which got reduced and stored away where the released oxygen couldn't get to it. Lots of sulfur available for sulfur photosynthesis. So we could get iron sulfide or elemental sulfur deposited on the ocean bottom, and the part that didn't get metabolized and sent back up would get subducted and removed. The removed sulfur wouldn't return until volcanism or hydrothermal vents bring it back. Thank you! sciencedirect.com/science/article/pii/0016703784900899 $\endgroup$
    – J Thomas
    Commented Apr 30, 2018 at 11:20

Actually quite interesting.

Oxygen is made by the splitting of water, either by photosynthesis or by UV-based photo-dissociation in the upper atmosphere.

In the case of photosynthesis, the reduced-carbon products must be buried in order for a net accumulation of oxygen to happen. In this case the oxygen effectively comes from water:

(1) 2H2O + vis. light. -> 2H2 + O2 (Water to hydrogen and oxygen) (2) 2H2 + CO2 -> H2CO + H2O (Carbon dioxide and Hydrogen to carbohydrate and water. Carbohydrate is buried)

And from UV-dissociation, we get

(1) 2H2O + uv -> 2H2 + O2 (Water to hydrogen and oxygen, hydrogen is lost to space)

The oceans weight over 10^21kg, most of which is oxygen, so there is plenty of mass there.


Methane Photodissociation

The reaction is effectively (in an oxygen free atmosphere):

(3) CH4 + 2H2O + uv -> CO2 + 4H2 (lost to space)

Note that without hydrogen loss to space, there is a limit on the amount of oxygenation that can happen, because the reduced carbon generated in photosynthesis has to be buried geologically. Estimates for the amount of buried reduced carbon are hard to find (carbonates don't count for this purpose).

So the ultimate source of oxygen in the atmosphere would appear to be water, mostly.

  • $\begingroup$ I got that much. was there that much CO2 before? Oxygen is supposedly the most common element in the earth's crust, about 466 parts per thousand by weight. Carbon is much less, estimates range from 200 parts per million to 1,800 ppm. Most of the oxygen and carbon is already locked up, for example in the form of carbonate rock. Say we took all the reduced carbon in the world and oxidised it. Would that be enough to remove the oxygen from the amosphere and give us 20% CO2 in the atmosphere? $\endgroup$
    – J Thomas
    Commented Apr 27, 2018 at 19:02
  • $\begingroup$ @JThomas Oxygen is very abundant in the Earth's crust, but that's not where the Oxygen from the Oxygenation event came from. It came, primarily, from photosynthesis, it's often said it came from CO2, as Andrew Dodds points out, the CO2 is actually used and turned into sugar and the Oxygen is released from the H2O. howplantswork.com/2009/02/16/plants-dont-convert-co2-into-o2 The abundant Oxygen in the crust is largely static. $\endgroup$
    – userLTK
    Commented Apr 28, 2018 at 5:27

What I have learned from this is that things were very different before the Great Oxygenation Event; more than just the same elements rearranged.

If all of the free oxygen came from water (and the hydrogen escaped to outer space), that isn't very different. That's around 10^18 liters of water lost from the oceans. The surface of the earth is about 5x10^14 square meters, one liter covers 10^-3 square meters, so sea level would have been somewhere vaguely around 2 meters higher. There's a whole lot of water in the world.

People say the atmosphere used to have 20% CO2 and that's where the oxygen came from. I didn't believe it because there is probably not that much reduced carbon in the world today. But it's possible that as the carbon was reduced, it was deposited on the sea floor and then it got subducted deep underground. It may still be there, waiting to get out.

Maybe there were other oxidized compounds that got reduced. The obvious candidates are nitrates and sulfates.

Today, bacteria mostly reduce nitrates when there is not enough free oxygen. In a series of steps they reduce it all the way to N2, getting energy at each step.

Other bacteria convert N2 to ammonia, using considerable energy. Often the ones that do this are photosynthetic. They have N2 and not enough nitrogen compounds to meet their needs, so they make what they need. O2 damages the enzyme.

Possibly there used to be a lot more oxidized nitrogen dissolved in the ocean, and the balance shifted, resulting in both increased N2 and increased O2.

Maybe there used to be a whole lot more sulfur in the ocean, in the form of sulfates etc. Most of the sulfur got reduced and deposited in the form of elemental sulfur or iron sulfide etc. Then it got subducted away, removing the evidence.

What was the pH of the ancient ocean? Various organic compounds can buffer pH differently according to concentration. H2CO2 (formic acid) acts different from H2 + CO2. Did the ocean used to be more acid? I don't know, and there were a lot of different buffers. Maybe H2CO3 (carbonic acid) was always the most important, if there was much more CO2 dissolved in the oceans than any other organic acid.

Just as this could happen for sulfur or iron, it could happen with any other good electron acceptor that happened to be dissolved in the ocean. Precipitate it out, and then subduction removes most of the evidence. But only things that had oxygen and then lost it, would contribute to free O2.

It was very different, and I can't even assume that the amounts of the various elements present were like today.

If it isn't clear what the pH of seawater was back then, I can't even be sure what the maximum amount of dissolved stuff would be. When I tried to guess at the maximum amount of dissolved iron, it turns out it varies with pH, with the amount of other dissolved chemicals, strongly with the amount of organic carbon compounds (because living things in today's ocean can't get enough iron and they scavenge it aggressively), etc. But that's today's ocean, where dissolved iron is quickly oxidized.

I expect there's room for a whole lot of uncertainty about all that. But experts can still know some things about it.

  • $\begingroup$ H2SO4 is acid. Just sulfate on its own is not acid. Also, you are missing deep earth process. Subduction and volcanic eruptions cycle carbon hydrogen and oxygen in and out of the deep earth over geological timescales. $\endgroup$
    – Gimelist
    Commented May 4, 2018 at 22:55
  • $\begingroup$ Yes, I missed about H2SO4. Convert it to S, 2 O2 and H2, and pH does not change. pH could change by buffering with organic acids. For example CO2 + H2 creates formic acid, which buffers water in a way that CO2 and H2 do not. But I was wrong. Subduction removes things from the ocean bottom. Volcanoes and mid-ocean ridges add stuff from the mantle. So the ocean floors are recycled every hundred million years or so. If we could assume the mantle gets thoroughly mixed, then it would continually reveal stuff close to the mantle average. But it does not mix that well. $\endgroup$
    – J Thomas
    Commented May 5, 2018 at 10:27
  • $\begingroup$ Before, there was no free oxygen. Suddenly (geologically speaking) there was a lot. Where did it come from? Was it attached to things in the mantle, and suddenly it came out, and now the mantle has slightly less oxygen than before? Did it get released from chemicals in the ocean, and the de-oxygenized chemicals suddenly got shoved into the mantle, so the mantle has less oxygen? Was it attached to hydrogen, and suddenly the hydrogen got blown away? $\endgroup$
    – J Thomas
    Commented May 5, 2018 at 10:54

Someone (I've lost track of whom. Tl;dr.) is seriously underestimating the amount of reduced carbon in sediments.

Having examined (and logged, and submitted to oil companies and government, hundreds of thousands of descriptions of drill cuttings, I'd estimate (very crudely) the proportion of elemental carbon in fine-grained sediments ("mudrocks" if you work for BP ; "silt and shale" for the rest of the world) at around 1%. That's of the total rock volume. Wherever you look in mudrocks, you see myriads of tiny (silt and finer grade) flakes of black elemental carbon. Some of them are palynomorphs (a whole industry in itself), but most are amorphous carbon to poorly crystalline graphite.

Once carbon gets to graphite near the Earth's surface, it tends to stay there unless it gets finely dispersed in a wildfire - which is going to be generating charred organic matter of it's own.

I don't see any problem with the Earth's atmosphere depositing around 1m total thickness of carbon dust over the approx 3 Gyr since the GOE started.

Now, doing the same to terraform Venus ... that's going to produce about 90m of carbon dust on the surface. Still less of a challenge than making an atmosphere for Mars.


The oxygen that we now see as $\ce{O2}$, or at least a large part of it, likely existed primordially in rocks.

When Earth differentiated its rock andviron, more was involved than just a physicsl separation of metal from rock. Under the GPa-level pressures generated by Eath's self-gravity, seemingly stable compounds undergo changes in their chemistry. One such change of properties is iron oxides can decompose to release oxygen[1], forming the iron or iron-rich materials that were then dense enough to sink beneath the mantle rock.

[A]bove ~60 GPa, iron oxides (particularly η-Fe2O3) start to decompose, producing oxygen. Based on estimates of the amount of BIFs subducted into the Earth’s mantle, the amount of oxygen produced by the formation of Fe5O7 alone can be as high as 8 to 10 times the mass of oxygen in the modern atmosphere.

Such reactiobs continue to occur in the mantle today.

The oxygen released by such reactions does not necessarily appear directly as $\text{O}_2$; it might instead formed oxygenated compounds (such as atmospheric carbon dioxide or sulfate monerals), from which elemental oxygen can eventually be extracted through organic processes.

Cited Reference

  1. E. Bykova, L. Dubrovinsky, N. Dubrovinskaia, M. Bykov, C. McCammon, S.V. Ovsyannikov, H.-P. Liermann, I. Kupenko, A.I. Chumakov, R. Rüffer, M. Hanfland, V. Prakapenka (2016). "Structural complexity of simple Fe2O3 oxide at high pressures and temperatures". Nature Communications 7, 10661; doi: 10.1038/NCOMMS10661.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.