[_ph_]

The article names it, but then does not dive deeper into it: the hydrogen chain has huge energy losses. First, electrolysis unfortunately has not a high efficiency - I have read numbers around 70%, but then the compression of the produced hydrogen takes a lot of energy, as compressing gases heats them up. Then the hydrogen has to be transported and stored, requiring further energy. Finally, you have to use the hydrogen, there the efficiency varies between 60% (fuel cell) or as proposed in the article with a combustion engine as low as 30%.

At the same time, there are less and less applications which cannot be powered by electricity directly, having only a fraction of losses compared to the equivalent hydrogen chain. So as long we do not have more than 100% reneweable electricity generation, we should be very concerned about the efficiency of our energy usage.

[mojomark]

Did you read the full article? It's in there - he developed a new Titanium electrode to improve the efficiency ond reduce the footprint of the electrolysis hardware and he also developed a new hydride (with a weak H2 bond to support release of H2 using waste heat) for greatly improved energy density. The specific values are in the article; maybe someone nicer than me will do more work to pull that for you.

[_ph_]

I rechecked the article, and I could not find any specific values in there. Care to point them out to me? I could only find some hand-wavy statements about a somewhat improved electrolysis device, but no numbers about its efficiency. Nor what those "hydrides" supposed to be.

[sebazzz]

How does that compare to using batteries? Batteries are re-usable of course, but also degrade over time and need to be replaced.

[_ph_]

As do fuel cells and H2 tanks - H2 is quite aggressive to the materials. But most importantly, the charge-discharge cycle of a battery is about 90% efficient, which beats hydrogen easily. And with quickly dropping battery prices they even become feasible for large-scale storage, see the Australian battery built by Tesla.

[ekun]

It mentions this in the article Li-ion go have a life cycle of roughly 1,000 recharges whereas these have 10,000.

[parametrek]

Depends on the li-ion. There are chemistries out there right now that are good for 20,000 cycles like LTO. And a substantial amount of li-ion research is extending cycle life.

It doesn't matter how good hydrogen becomes. Battery tech will continue to improve faster than hydrogen tech. H2 will never catch up.

[lokopodium]

Hopefully solar cells will become cheap enough quickly enough. It would be awesome to have a small hydrogen plant at home.

[_ph_]

What would you use the hydrogen plant for? For the same amount of solar energy, a Tesla is going about 3x further than a Mirai, and a heat pump would provide much more heat your house.

[marcosdumay]

For energy storage for the night.

Open fuel cells storage can be much cheaper than battery based storage. You can store energy for a long period, paying by the maximum power, not storage capacity.

That said, molten salt cells look way more promising than hydrogen.

[lokopodium]

Mirai is one of the first production HFC vehicles, well below what's feasible. Total solar energy is not really a constraint, refueling time and amount of stored energy are.

[_ph_]

Total solar energy is a constraint, as someone has to pay for the cells and to have space to set them up. If you want to power your own car, don't you think it makes a difference whether you need 1x or 3x the area covered with solar cells?