Hacker News Reader

Comment by pjc50 | Hacker News Reader

> unless government policies mandate it or restrict batteries or blue hydrogen

Yes, but I think this the most likely outcome. Natural gas is only cheap in certain areas, and the past few years have made everyone very, very aware of the geopolitics involved in getting hold of it. While global warming is not going away, and I question the extent to which CCS actually happens with blue hydrogen.

Batteries are capital equipment in the same way as electrolysers are. They're great at short term storage, but medium-term is still a bit more of an issue. "Restrict batteries" is obviously not on the table except for stupid retail corner cases where utilities have captured the regulator.

There's a potential market for lots of green H2 in Haber nitrogen, metals refining, and synthetic jet fuel etc, but only if the cheap CO2 emitting option is priced out or banned, or H2 electrolysers get comparable capital prices to battery storage.

View on HN

With CO2 emitting option is priced out or banned, direct hydrogen production using high temperature sulfur–iodine cycle without H2 electrolysers could be economic option. The heat could be supplied from high temperature nuclear reactor.

Back to the corrosion discussion from the original article: when handling sulphuric acid at 800C, what kind of piping material do you need?

Glass coated steel, silicon carbide.

"Material selection, simulation, and experimentation of sulfuric acid decomposition in a pilot-scale sulfur-iodine thermochemical cycle for hydrogen production"

https://www.sciencedirect.com/science/article/abs/pii/S00162...

> With CO2 emitting option is priced out or banned

GP was talking about injecting the CO2 back into the well, not releasing it to the environment. There are even standards for specific injection wells used for long term storage (EPA Class VI).

Processes that involve heating sulfuric acid vapor to decomposition don't sound terribly practical. If you thought seawater corrosion was challenging...

When would that ever make sense over solar?

How to make hydrogen production cheaper and easier: include an atomic reactor component.

Huh?

I’d be interested in hearing about some scenario where this actually costs less, given the cost of building anything nuclear in 2026.

I agree, the experience building nuclear reactors is mixed bag. Some builds failed, like Flamanville 3, Hinkley Point C, Vogtle 3. Some builds succeeded: Barakah nuclear power plant, Fuqing 5,6. It really depends on maturity of supply line and political support.

The real question is: how do we produce hydrogen from the coming massive overbuilding of cheap-but-variable solar. Nuclear reactors are a whole different animal: even if we build them "cheaply" they're not going to approach the costs of overbuilt solar, so those nuclear watts will be better used for other purposes.

Coupling PV to electrolyzers and efficient hydrogen production using solar energy is still open research problem. Batteries will be probably needed.

"The low efficiency of PV-electrolyzer systems can be attributed to several factors: intrinsic losses in both the PV and electrolyzer units, energy consumption by balance-of-system components (e.g., inverters, thermal management), and, most critically, ineffective electrical coupling. Although some researchers advocate for direct coupling as a cost-effective solution, variable solar input remains a major challenge. Fluctuations in solar irradiance can cause the power delivered to fall outside the acceptable operating range of electrolyzers, leading to frequent shut-downs and start-ups. These cycling events can accelerate degradation, particularly in PEM electrolyzers, and also affect the purity and yield of hydrogen"

"Recent studies also highlight the integration of battery energy storage systems (BESS) into large-scale PV-CSP hybrid plants as a strategic enhancement. With anticipated declines in battery costs, this integrated approach may become increasingly viable in the near future."

https://link.springer.com/article/10.1007/s44373-025-00080-4

Reply on news.ycombinator.com