Nickel Hydrogen Batteries by NASA (opens in new tab)

The numbers thrown out for this (~$60/kWh) are comparable in price to what LFP might cost in the next couple of years, the C rates for these are much lower (C/2).

A hydrogen fuel cell also vents and you can ignite the concentrated stream like a blow torch. At 5% the pressure of a H fuel cell, the battery probably vents less hydrogen.

How are today's batteries doing when put in the barbecue fire? The new ones only need to not do (much) worse.

> But what about the hydrogen?

Doesn’t it say the hydrogen is vented as H2O?

The cells are pressure vessels, so normal cube-square scaling laws don't apply. Instead you need to use pressure vessel scaling laws, which also account for the needed wall thickness.

Pressure vessel scaling laws say that all cylinders have the same mass efficiency, and making long thin cylinders is easier than making short squat cylinders.

As I understand it, something like a lead-acid battery using volumes of acid and volumes of reactants so a cube gives them more power with the same surface area. NiMH batteries use boundaries between states instead of acid. Therefore, you want long thin batteries of alternating materials to make them more efficient.

Or, to put it a different way, NiMH batteries require a large interior surface area, and so the square/cube law forces them to look longer and thinner as they get larger.

> 7 GWh, or about 400 million dollars

This implies a bulk cost of about $57/kWh which is slightly cheaper than present day LFP (the primary competitor for grid-scale battery storage). IMO, it needs to be no more expensive than LFP, because LFP cells can be spec'd for ~8 years and it's hard to get people to invest on timescales longer than ~10 years.

That looks suspiciously similar to a standard gas tank size.

Generally you make a battery by combining cells - they have a 1-1.5V terminal voltage. The image lower down shows a rack of them in a warehouse. I suspect you'd stack them up to a few hundred volts and plug them into an inverter.

I wonder if they require the same balancing as more delicate chemistries.

This video discusses that: https://www.youtube.com/watch?v=2zG-ZrC4BO0

But, keep in mind that Undecided is overly optimistic about pretty much every emerging technology.

The use case for this is to charge the battery during the day and discharge it during the night. It is not for seasonal energy storage. There are no batteries that can be used for seasonal storage, nothing comes within a factor of 100 of being economical. But batteries for seasonal storage requires very few cycles, one per year. This battery here can go through many thousands of cycles. If you don't use it for daily storage, you are paying for something you don't need.