[subnaught]

Some background and context from someone tangentially related to the field:

1. The overall idea here is to take an intermittent energy source (e.g. solar power) and "store" it as chemical fuel, in this case hydrogen and oxygen. This is what plants do, and we can also view fossil fuels as resulting from the "storage" of millions of years of solar energy. Note also that you get the water back when you burn the hydrogen, so there is no net consumption of water, it's just a carrier.

2. While you can split water without a catalyst, most of the energy gets wasted as heat, so this is not a great way to go if you're trying to do energy storage.

3. Efficient catalysts exist for this reaction, but they are based on rare and expensive metals, typically Pd, Pt, and Ir. As a result, there has been a search for catalysts involving "first-row" metals such as Fe, Co, Ni, etc.

4. There are variety of metrics for an electrocatalyst (efficiency, stability, cost, etc), but it's a fair bet that if this were significantly better than state-of-the-art, it would be in Science or Nature rather than PNAS.

[dnautics]

Beyond the expensive of the metals another problem is duty cycles. Most transition metal catalysis is oxygen sensitive, and it seems like for some reason the first step of splitting water is creating oxygen. Plants go through great lengths to separate oxygen synthesis (photosystem II) from electron consumption. Most hydrogen production in lower organisms (like e coli) occurs entirely in anoxic conditions. Engineered systems for generating hydrogen via algae typically are temporally segregated (harvest light during day, produce hydrogen at night) which defeats the purpose and is also chemically steppy (carbohydrate intermediates).

[novaleaf]

> and it seems like for some reason the first step of splitting water is creating oxygen

good laugh on that :)

[logicallee]

can you compare storing an intermittent energy source by means of a chemical fuel, hydrogen and oxygen, as compared with in a battery? Just compare and contrast every aspect that matters. Just to be clear, this is (in effect) a battery, right? So what are its characteristics as compared with, say, lithium ion batteries. (I am particularly interested in weight and in number of duty cycles, which sounds like it's "unlimited" as opposed to lithium ion which is really not that many cycles, right?) I'm also far outside the field, just interested. Thanks!

[epistasis]

"Cycles" don't really make any sense here, but you can get some estimates of costs and efficiencies here:

https://en.wikipedia.org/wiki/Power_to_gas

Lithium ion batteries for grid storage typically use an NMC chemistry, and standard warranties are for 10 years of daily cycling.