OK, I managed to track down a copy of the original research. The problem was that it was published yesterday and hasn't been indexed by most search engines yet.
Sweetman, A.K., Smith, A.J., de Jonge, D.S.W. et al. Evidence of dark oxygen production at the abyssal seafloor. Nat. Geosci. (2024).
Let me just preface this by saying that there is so very much wrong with the science in this publication and even more in the pop-sci reporting.
TL;DR: Installation of the experimental equipment appears to be the cause of the oxygen production. The purported electrolysis lacks any plausible mechanism.
The first warning is that these nodules are forming from seawater and/or sediments on the ocean floor. This happens because they are more chemically stable as solid nodules than they were as constituents of the seawater. This is the basis of why salt dissolves in water, why cars rust, and why any chemical reaction proceeds spontaneously in general. It's moving to a more stable, lower energy state. The notion that these nodules could possibly contain enough energy to split the water molecules from which they formed just violates basic principles of thermodynamics.
The next problem is that these nodules form over millions of years and if they'd been electrolyzing water for all of that time then their energy would have been depleted long ago. Think of it as dropping a load of car batteries onto the ocean floor, setting them all to crack water into hydrogen and oxygen, and expecting them to be running a million years later. Even if there was energy in those nodules at some point they should be depleted long ago. To their credit the authors do caution that the oxygen production is heterogenous spatially and temporally and should not be scaled up without further study. Sadly, they proceed to do exactly that.
OK, those are just the basic flaws that struck me yesterday. Here's what I found in the paper:
This is their graph of oxygen production over the course of the 48 hours that the experimental equipment was on bottom:
![[Image: 41561_2024_1480_Fig1_HTML.png]](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41561-024-01480-8/MediaObjects/41561_2024_1480_Fig1_HTML.png)
Note the shape of the curve in almost all cases. High initial oxygen production followed by decrease over time. In almost all cases these curves are either flat, indicating no oxygen production, or pretty much classic exponential decay curves. Those curves are pretty much textbook examples of a disturbed system returning to equilibrium. Simply put, this data should indicate that their experimental equipment disrupted the system and it needed time to return to normal. The authors do discuss several possible sources of oxygen that could have been produced by the equipment but only to discard the possibility. The patterns seen in this data are incredibly obvious and should have been interpreted as an external influence produced by the introduction of the experimental equipment unless an extremely rigorous examination ruled that out. This is experimental set-up 101 and I suspect that the authors are going to get panned for publishing these flawed interpretations.
As for electrolysis of water, that requires a minimum of 1.23 Volts, although 1.5 Volts or more is common. You can change that slightly with a variety of conditions and the paper cites a total voltage of 1.60 Volts necessary for electrolysis under the conditions at the ocean floor. If you don't have the voltage you don't get any electrolysis. It's all or nothing. The maximum voltage reported in the paper was 0.95 V. Looking through the raw data I can't find that and the highest voltage there is 0.24 V. Regardless, there isn't sufficient voltage for electrolysis. It should also be noted that these voltages were measured in the lab, not on the seafloor. The pressure is lower, temperature and availability of oxygen higher, so no big surprise that some electrochemistry is occurring. Once again, a great example of a system that has been moved out of equilibrium.
This looks like a classic example of sloppy science going viral and becoming even worse once the media finished digesting it. The editors and peer review should really have picked up on this, so failures all around.
It's a bit of a shame because there are obvious red flags in the data and some very simple experiments that could have been done to validate it. Simply send the same equipment back down for two weeks and add on a hydrogen sensor. Electrolysis produces both hydrogen and oxygen, so if you don't see any hydrogen then you aren't getting any electrolysis. Elemental hydrogen is also much easier to detect than oxygen because it is so very rare in the environment. Leaving the apparatus on bottom for longer gives the set-up time to equilibrate and lets you show what the long-term oxygen production is or isn't.
None of this means that mining polymetallic nodules from the seafloor is viable or environmentally sound, just that this particular paper has some glaring flaws.
Sweetman, A.K., Smith, A.J., de Jonge, D.S.W. et al. Evidence of dark oxygen production at the abyssal seafloor. Nat. Geosci. (2024).
Let me just preface this by saying that there is so very much wrong with the science in this publication and even more in the pop-sci reporting.
TL;DR: Installation of the experimental equipment appears to be the cause of the oxygen production. The purported electrolysis lacks any plausible mechanism.
The first warning is that these nodules are forming from seawater and/or sediments on the ocean floor. This happens because they are more chemically stable as solid nodules than they were as constituents of the seawater. This is the basis of why salt dissolves in water, why cars rust, and why any chemical reaction proceeds spontaneously in general. It's moving to a more stable, lower energy state. The notion that these nodules could possibly contain enough energy to split the water molecules from which they formed just violates basic principles of thermodynamics.
The next problem is that these nodules form over millions of years and if they'd been electrolyzing water for all of that time then their energy would have been depleted long ago. Think of it as dropping a load of car batteries onto the ocean floor, setting them all to crack water into hydrogen and oxygen, and expecting them to be running a million years later. Even if there was energy in those nodules at some point they should be depleted long ago. To their credit the authors do caution that the oxygen production is heterogenous spatially and temporally and should not be scaled up without further study. Sadly, they proceed to do exactly that.
OK, those are just the basic flaws that struck me yesterday. Here's what I found in the paper:
This is their graph of oxygen production over the course of the 48 hours that the experimental equipment was on bottom:
Note the shape of the curve in almost all cases. High initial oxygen production followed by decrease over time. In almost all cases these curves are either flat, indicating no oxygen production, or pretty much classic exponential decay curves. Those curves are pretty much textbook examples of a disturbed system returning to equilibrium. Simply put, this data should indicate that their experimental equipment disrupted the system and it needed time to return to normal. The authors do discuss several possible sources of oxygen that could have been produced by the equipment but only to discard the possibility. The patterns seen in this data are incredibly obvious and should have been interpreted as an external influence produced by the introduction of the experimental equipment unless an extremely rigorous examination ruled that out. This is experimental set-up 101 and I suspect that the authors are going to get panned for publishing these flawed interpretations.
As for electrolysis of water, that requires a minimum of 1.23 Volts, although 1.5 Volts or more is common. You can change that slightly with a variety of conditions and the paper cites a total voltage of 1.60 Volts necessary for electrolysis under the conditions at the ocean floor. If you don't have the voltage you don't get any electrolysis. It's all or nothing. The maximum voltage reported in the paper was 0.95 V. Looking through the raw data I can't find that and the highest voltage there is 0.24 V. Regardless, there isn't sufficient voltage for electrolysis. It should also be noted that these voltages were measured in the lab, not on the seafloor. The pressure is lower, temperature and availability of oxygen higher, so no big surprise that some electrochemistry is occurring. Once again, a great example of a system that has been moved out of equilibrium.
This looks like a classic example of sloppy science going viral and becoming even worse once the media finished digesting it. The editors and peer review should really have picked up on this, so failures all around.
It's a bit of a shame because there are obvious red flags in the data and some very simple experiments that could have been done to validate it. Simply send the same equipment back down for two weeks and add on a hydrogen sensor. Electrolysis produces both hydrogen and oxygen, so if you don't see any hydrogen then you aren't getting any electrolysis. Elemental hydrogen is also much easier to detect than oxygen because it is so very rare in the environment. Leaving the apparatus on bottom for longer gives the set-up time to equilibrate and lets you show what the long-term oxygen production is or isn't.
None of this means that mining polymetallic nodules from the seafloor is viable or environmentally sound, just that this particular paper has some glaring flaws.
