If true, you're right, it may be possible for H2 to become unbeholden to carbon. For now, let me reword to say that classic techniques for producing H2 have always been beholden to carbon. It's not yet a pejorative mischaracterization to say it's always been that way. I did gloss over the possibility of this changing, however. That feels like a comfortable bet, but you are right that it changing is probably not strictly impossible.
As for the mining, the same applies here as it applies to computer chips and even aluminum ion antiperspirants, it's all terrible for the environment. We should be careful here, too, however. This kind of mining is a different kind of destructive. It's primarily destructive to waterways and not the atmosphere/climate story. And like you I have no idea how to compare them. Meanwhile, toxic mining also seems necessary for perhaps all highway-safe automobiles (EVs, H2Vs, internal combustives), though I'm also not prepared to speak on the varying extents of toxic mining necessary for each and would not be surprised if you found that data and could support the idea that EVs are meaningfully worse. Our conversation needs that info, too.
Agreed.
Bear in mind that I have no particular dog in this fight. I'm not sure that I'm fully onboard that the effects of CO2 are anywhere near as dire as some people predict, but I am in full agreement that CO2 is a greenhouse gas and that we should be trying to find alternative solutions to burning hydrocarbons. Especially because even beyond CO2, there are other pollutants.
But as I've said before, to some extent you need to come at these things from a cost-benefit analysis, rather than picking the technology you want to win and assuming it will just scale. I've said before that I'm in data storage, and although there are huge predictions that HDDs will go the way of the dodo to be replaced by SSD, they don't factor in the capital intensiveness of the NAND industry and that it's only capable of economically producing a small portion of the world's total bit storage capability. It doesn't even matter that it's superior in performance--you simply cannot economically produce nearly enough bits to replace it wholesale. And as the world's bit demand goes up exponentially, NAND flash would have to grow at a much faster exponential to overtake HDD entirely, and I don't see that happening for decades.
I personally worry that battery EVs are similar. They can be far superior in performance, and in pollution [as we improve our electricity generation]. But if the world's ability to economically mine lithium, or cobalt, or other metals used in the batteries doesn't come close to the world's annual automotive demand, it doesn't matter one bit if they're "better". You can't build and sell something if you don't have raw materials. Now, maybe that worry is unfounded. Maybe we can easily mine more than enough, and improve battery recycling to re-use what's already been mined, etc. I don't know because it's not my field of expertise, unlike data storage.
The advantage of H2 is that hydrogen is plentiful. The disadvantage [as you point out] is that it's not plentiful in H2 form, so to get it we must separate it from another source, which can be difficult. Oh, and that whole "exploding" thing. But if we can find a way to economically extract hydrogen in a non-polluting way, that doesn't increase carbon footprint, and we can figure out how to do it without causing explosions, it may tip the cost-benefit scale from battery EVs to hydrogen fuel cell EVs. But again, that is also not my field of expertise, and I'm not going to say I know which one is going to win as I don't have the requisite knowledge. Hence why I'm postulating, not declaring.