Comment by phenol | Hacker News Reader

Comment by phenol | Hacker News Reader

Parent Full thread

critically, the electricity here is not actually directly breaking apart the water into hydrogen, so you can't directly calculate an energy efficiency like you can for electrolysis. this company is using electrohydraulic fracturing to shatter ultramafic rocks and thereby increase the rate of natural geochemical processes like serpentinization or the Schikorr reaction in which rocks react with water to produce geologic hydrogen. the actual source of the free energy that drives the reaction is partially reduced Fe(II) species in the crust (olivine, wüstite, etc) that can absorb an oxygen atom from water to produce free H2 (2 FeO + H2O -> Fe2O3 + H2).

something like 3-6% of the Earth's crust is ferrous Fe(II) so in principle the quantity of hydrogen that could be produced in this manner is something ungodly like 1e15 tons. but the problem is the natural rate of formation is very slow because it's rate-limited by low rock surface areas and slow formation of new rock faces. hence shattering rocks with electricity to increase the reaction rate.

source: I worked on the 2024 us govt research effort that funded these guys

https://arpa-e.energy.gov/news-and-events/news-and-insights/...

I think GP meant "in terms of hydrogen out per joule, how does this process (on average) compare to electrolysis".

sure, and I'm saying that it's not a well-defined number because it's not a closed thermodynamic process. it's kinetically limited, so your H2 yield is a function of how much time you're willing to wait. it also obviously will depend on the type of rock, electrohydraulic fracturing parameters (pressure, voltage, pulse sequence) and so on. and in practice one of the biggest challenges in this field is actually capturing and collecting the hydrogen you stimulate, which is difficult to model and can depend on the precise geology and downhole microbiology. you simply can't compare it to electrolytic H2 on an atoms-per-joule basis.

You cannot compare it in general, but in a particular case you can very well compute it so you can determine whether you have spent well your energy, or you have wasted it because using it otherwise would have produced more hydrogen.

Even if you might not be able to compute the efficiency before the process is actually completed, it would be foolish to start it without at least some rough estimate, to see if there are chances for this to be worthwhile.

I agree with the previous poster that the article should have included some estimate of the efficiency they hope to obtain, otherwise it is impossible to say whether this is newsworthy in a positive way (it could be newsworthy in a negative way, if they had wasted resources for a negligible outcome).

The article acknowledges:

> Hydrogen ... currently takes more energy to make than it yields, and the cheapest and most common way is by reacting steam with methane, a potent greenhouse gas.

> It’s possible to make zero-carbon hydrogen by splitting water with electrolyzers powered by renewable energy. But in most cases, the process is too expensive to be economical...

and also says,

> A back-of-the-envelope calculation convinced him that the cost of stimulated geologic hydrogen could easily compete with hydrogen made from methane. “If we get the technology right,” he concludes, “this could be huge.”

It's pretty clear from reading the article that actual yields will be highly dependent on discovering ideal mining sites, but the fact that people are building startups to research and develop this kind of implies someone ran some numbers and thought it was worthwhile enough to be "newsworthy".

thank you for actually reading the article and realizing that the team of energy scientists + engineers doing this have thought more carefully about the viability of the process than a HN commenter that learned about it this morning

kind of. the real figure of merit here is the levelized cost of hydrogen (LCOH), which requires you to build a technoeconomic model to simulate the total cost and production of a hypothetical unit over its lifetime. focusing just on "energy efficiency" will be misleading (in favor of geoH2!) because, relative to a green H2 elyzer, this approach has relatively lower operational costs (OPEX) and relatively higher capital costs (CAPEX) because you don't need stoichiometric amounts of electrons but you do need to drill a bunch of boreholes. if you just looked at "energy efficiency" you'd fail to amortize the drilling capex which is probably a bigger cost driver than electricity opex.

and fwiw estimates of LCOH by this process are $0.50-1.50/kg compared to maybe $3-6/kg for green H2. but these are obviously pie-in-the-sky NOAK figures and a technoeconomic analysis is only as good as the assumptions you feed into it. is it actually cost effective? there is only one way to know for sure, and these guys are gonna find out it's the hard way.

Thanks for sharing all of this insight. Here is some more hydrogen conversation https://www.fourier.earth/

This makes me wonder how many tons of free hydrogen would it take to react with a significant percentage of Earth's oxygen to the point it would harm life.

Reply on news.ycombinator.com