You decouple the transformation (charge/discharge) from the capacity (liquid volume), with the goal of making the latter "a standard pressure, watertight tank."
But I believe last time they came up here, people said the charge/discharge still needed some work.
Also, long-term storage will very likely use some different chemistry from short-term. High-temperature batteries have some very interesting trade-offs that I have no idea how will pan-out in practice. Things are mostly not settled on that area, it looks like a very interesting thing to work on.
I.e. putting energy into nuclear reactors so that we can produce U-235. Although I guess technically breeder reactors, although like-to-like is less fun fantasy than solar -> fissile.
https://www.abc.net.au/news/2023-06-23/vanadium-flow-battery...
IIRC, this particular chemistry is an Australian development, as well.
It seems like time is the bottleneck in basically all cases rather than overall capacity as well.
Batteries have no bottleneck. Just lots and lots of things one can improve a bit with some amount of work.
It sounds like a false premise to me.
It's a perfectly fine point to make, and it's one of the things worth optimizing.
Thus, lithium looks like one of the fundamental bottlenecks for grid storage. It can kinda work on high-cost small-size pilot projects, but we probably won't be able to use it for real.
Sodium on the other hand has all of the same desirable chemistry properties, but scales much better. And iron has all the cost benefits, but undesirable chemical properties. (And there are, of course, people working on C-H vs. C-OH bonds that are completely out of the box.)