It's not "impossible" to decarbonize the production of aluminum without driving up cost, as the page claims. Once green electricity and hydrogen is cheaper than CO2-emitting energy, this will be possible and even profitable.
The problem with decarbonizing is not being solved with green electricity and hydrogen. The (direct) emissions come from the carbon anodes that are consumed in the process. They are usually made from petroleum coke, aka they're fossil fuels.
Alternatives are being developed, but have a somewhat troubled history. Alcoa announced in the early 2000s that they are only months away from deploying inert anode technology. They're still not there (though still working on it in a project called elysis).
From what I understand, it is cheaper. However, that's just with new power. Most of the power available now is from existing installations, making existing fossil fuel power cheaper than any new installations.
I've yet to hear about renewables that put out consistent power cheaper than gas.
You'll see the occasional article about new solar that's claimed to be cheaper than gas, but they never have sufficient storage to actually stand on their own two feet.
The last one that was posted here on HN was claimed to be cheaper than gas, but it only had enough storage to output 1/2 of its daytime production overnight. In other words, it was cheaper than gas... as long as you've got redundant gas plants to provide power overnight.
Always read the fine print when folks are pitching "cheap renewables".
Alternatives are being developed, but have a somewhat troubled history. Alcoa announced in the early 2000s that they are only months away from deploying inert anode technology. They're still not there (though still working on it in a project called elysis).