Hacker News new | past | comments | ask | show | jobs | submit login
Hysata electrolyzers produce cheapest hydrogen at 95% efficiency (hysata.com)
34 points by belter 3 days ago | hide | past | favorite | 26 comments

The important part is not just efficiency, it's also capital cost of the electrolyzer system. It can make sense to reduce efficiency (say, by increasing current density) if that reduces capital cost per kW. I don't see the cost/kW here, although it claims "low capex". Low capex is particularly important when driving the electrolyzers with sources with intermittency.

Exactly, capital cost (which is a function both of upfront cost and the lifetime) is the dominant cost of electrolyzed hydrogen right now.

Right now in California a bunch of organizations are prematurely adopting hydrogen tech that is not ready to be deployed, for buses and trains, despite failed pilots of these all over the globe. Decision makers are using the wrong parameters to make investments that last decades, and it's extremely disheartening.

Pretty much all hydrogen hype out there feeds into these bad decisions but the inappropriateness of the metrics is somehow being ignored.

You can view the hype as trying to drive up interest in projects to move electrolyzers (and other hydrogen technology) down their experience curves.

For intermittent sources it's also important to be able to cycle on/off (or throttle to partial power) frequently. In reading about green hydrogen projects I saw it mentioned that traditional alkaline electrolyzers could not safely run in an intermittent fashion, although proton exchange membrane electrolyzers could. Unfortunately, the source that mentioned this did not say why alkaline units were unsafe to run this way, and my searches for more information were not successful.

My guess is something like "it relies on a certain outflow rate of gases to prevent hydrogen/oxygen mixing" but I'm not sure why that is a particular issue for traditional alkaline electrolyzers, or how difficult it is to mitigate. The alkaline units otherwise require only common materials, have been manufactured forever, and seem like the natural thing to build more of for coupling to intermittent renewables.

Inability to cycle rapidly can be dealt with by including batteries, and/or by having a bank of electrolyzers that are not all operating at the same time.

I hear people talking about intermittent operation of electrolyzers, but this sounds to me like the stupidest thing to do:

* Intermittent operation means you need to either build a huge buffer tank, or overprovision your transport infrastructure compared to your average production rate

* Your customers need to deal with a less reliable supplier

* Intermittent operation means shorter maintenance intervals, more stress on your electrolysis units and higher demand on your BoP

* When you've put down a huge capex, the marginal cost of not running the plant is quite high, so it makes sense to run 24/7 even at high energy prices

It's talked about because the sources are intermittent. There will be large price swings in the grid of the future; hydrogen can act as a source of dispatchable demand. For dispatchable demand you want a reasonable application with minimal capex per watt (and also low fixed opex), and hydrogen fits that bill.

What other application do you think has as low a capex per watt as electrolysis? Maybe heating thermal batteries? But that's not useful for long term smoothing. Data centers have something like 100x the capex/watt of hydrogen.

(2023): "...Currently, the production cost of renewable hydrogen (using renewable energy) is at least twice that of hydrogen produced from fossil fuels. Hysata says its technology will slash costs and produce hydrogen “well below” a competitive target price of $2 per kilogram (approx. US$1.50/kg)..." - https://arena.gov.au/blog/hysata-to-build-next-generation-hy...

Edit: someone more knowledgable responded with more accurate numbers about energy conversion- read their comment.

This still seems like it would only make sense when there is an energy surplus (e.g. excess solar generation) and it is useful to store energy in this form rather than a battery.

You are right that this makes sense to store an energy surplus. The reason you see such a large gap between "41.5 kWh/kg" and "33.6 kWh/kg" is that the latter value refers to what is called the lower heating value (LHV) [1]. The higher heating value (HHV) of hydrogen is 39.4 kWh/kg, and that's where the 95% efficiency claim comes from: 39.4 / 41.5 = 94.9%.

This has a good explanation of higher and lower heating values for chemical fuels: https://en.wikipedia.org/wiki/Heat_of_combustion#Higher_heat...

[1] https://en.wikipedia.org/wiki/Energy_density#In_chemical_rea...

Isn't hydrogen difficult to handle and store, easily leaks out, extremely flammable/explosive, etc? When I was a kid, hydrogen cars where the future. Doesn't seem that way now.

Hydrogen will always be needed to produce chemicals, as in the Haber-Bosch process that renders modern agricultural productivity possible:


Decarbonizing primary steel production could consume vast quantities of electrolytic hydrogen for reducing iron oxide ores to metallic iron.

Underground storage of hydrogen for subsequent combustion in gas turbines is also one of the most promising ways to achieve seasonal electricity storage, for e.g. powering a country through the winter with excess renewable electricity generated in the spring. There's currently a large scale pilot program for this concept under construction in Utah:


Hydrogen would also be useful for large scale production of fuels from biomass.


Is this a big deal? This report mentions electrolyzers hitting 80-95% efficiency back in 2004: https://www.nrel.gov/docs/fy04osti/36705.pdf

The big H2 inefficiencies are still the compression, transportation, and conversion back to electricity.

There have been %90+ efficient atmospheric co2 direct to propane synthesis in labs for a couple years now. Trying to transport hydrogen seems insane given that we already have direct propane fuel cells that are highly efficient too. And an infrastructure for holding and transporting propane if it isn't available on site.

Interesting, do you have any links to share, it seems like direct-propane synthesis (3-carbon chain) is less likely than methane. Is it some kind of sweet spot, or does direct butane work too?

Something like direct butanol synthesis and fuel cells would be the ultimate I think, a liquid fuel that could be burned as jet fuel as well as in a fuel cell.


Edit: my time frame is off, I'm bad at estimating elapsed time in general. So we are probably a ways off from commercialization.

Correct me if I am wrong. My understanding is that hydrogen is an efficient way of getting energy only if it is used as a larger scale storage instead of batteries. Is it?

The round trip efficiency of hydrogen and electricity is much less than batteries, even at large scale.

What hydrogen has going for it is extremely low capex per unit of energy storage capacity, perhaps $1/kWh for a system that stores compressed hydrogen in solution mined salt caverns.

in itself and at the moment with current technology yes.

but hydrogen is used to make other things that make it very easy to store as well.

Ammonia is the leading product at the moment.

I wonder how the unit cost of production compares to gray hydrogen in areas where methane is cheap.

How does this end up at the top of HN with one comment and only 14 points?

14 points in only 20 minutes is a huge rate of vote accumulation. I spend a lot of time watching the new queue, and this behavior seems about right to me.

I don't know how voting works fully, but I suspect the karma of the voters has an impact, or maybe the karma of the submitter too?

Hydrogen rises?


The machinations of the Wizards of Dang are a mystery to all.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact