Here's the fact that reorganizes the whole storage conversation: lithium-ion is excellent at the wrong problem. It's the best technology in the world for delivering a lot of power for a few hours — exactly what a phone, a car, or a grid's evening peak needs. But its cost is dominated by the cathode and anode materials, so the longer you want it to discharge, the more of those expensive materials you buy. Doubling the storage duration roughly doubles the cost. That's fine at four hours and brutal at four days.
Why does the grid need four days? Because a system running mostly on solar and wind has to ride through stretches when neither shows up — a still, cloudy week in winter. Filling that gap with lithium would be ruinously expensive, so the field is hunting for chemistries whose cost barely rises with duration. The mechanism that makes those chemistries cheap is that they store energy in abundant materials, and accept lower efficiency or power in exchange.
Take metal-air. A grant like A*STAR's US10964982B2, "Rechargeable metal-air battery cell, a battery stack and method of manufacturing the same," describes the family. The mechanism in plain terms: one electrode is a cheap, abundant metal, and the other "electrode" is effectively oxygen pulled from the air — so you're not paying for a heavy, expensive second electrode material at all. That's how the cost-per-hour-stored drops; the trade-off is typically lower round-trip efficiency and slower response than lithium. For multi-day storage, that's an acceptable trade, because the alternative is not storing the energy at all.
Thermal storage attacks the same problem from a different angle, storing energy as heat in cheap material — the approach in Rondo Energy's recent grant US12644396B2. Different physics, same economic logic: make the energy-holding medium so cheap that adding hours is almost free, and put the cost into the conversion equipment you size once. Whether the carrier is a metal, a brick, or a tank of flowing electrolyte, the long-duration design principle is identical.
The accurate framing, which the hype usually mangles: this is not lithium versus its replacement. Lithium will keep owning short-duration storage; it's genuinely the best tool there. Long-duration is a different market with different winners, and a grid built on renewables needs both. Treating it as a horse race misreads the engineering — they're complements, sized for different jobs.
So when you read that some new battery will "kill lithium," check the duration it's built for. If it's a multi-day chemistry, it isn't competing with lithium at all — it's filling the gap lithium was never economic for. The patents make the distinction concrete: cheap abundant materials, lower efficiency, longer duration. That's the trade, and it's the right one for the problem.