Ok so aluminum is still a better candidate than iron powder since it doesn't use hydrogen. I know I'm in the hydrogen sublemmy arguing against the use of hydrogen here, but green hydrogen uses a ton of electricity and it's an extra step that costs some energy efficiency, so why not cut out the need for green hydrogen and use electricity directly to smelt aluminum oxide into aluminum metal?

Aluminum carries 31MJ/kg of energy vs irons 7MJ/kg.

They also don't address how the iron oxide is reduced back it iron metal. I'm assuming they use DRI for that (Direct Reduced Iron) where you pelletized the oxide and pass hot hydrogen over it in an oven. You still need to sinter the pellets to allow for optimal gas contact around each pellet along with internal gas passage, which means you do need to grind it down after you reduce it back to metal. I don't know where your numbers are coming from.

Where did you find info on reducing a iron oxide powder without sintering directly? Is that a fluidized bed process? AFAIK there's either DRI pellets, or you do a vortex lance method the Chinese pioneered but that produces liquid iron and liquid slag.