[samtho]

I don’t think those extrapolations are totally fair. You’re assuming today’s technology is frozen, inflexible, and adoption is going to linear. If we have learned anything from the last decade it is that adoption curves are getting steeper.

If you recall from my original post, I said nothing about that hypothetical, abundant hydrogen being born from any particular source, but I did have solar in mind specifically. I personally think solar cells and panel arrays will get more efficient at producing power that we will eventually have a global excess. If we have grids producing renewable electricity when conditions allow, they can redirect some of the excess energy and use electrolysis to produce hydrogen creating energy reserves for when the power source is not available to them.

Hydrogen is still not super efficient because we have not even scratched the surface of optimizing it’s consumption. If we had concerted efforts towards that goal, we can get it to work for us in ways we never thought possible. I don’t think lugging around large batteries is the future, rather our fuel storage needs will be met by hydrogen tanks.

[mrpopo]

I'm not assuming linear adoption. I explicitly said that even if adoption was FOUR TIMES faster than now, the amount of renewable energy would barely reach the same amount of primary energy as we use now by 2050.

This is a optimistic scenario. Current trend is mostly linear. See:

Wind power generation, world: https://ourworldindata.org/grapher/wind-generation?tab=chart...

Solar power generation, world: https://ourworldindata.org/grapher/solar-energy-consumption?...

You can choose linear/log, it doesn't look exponential to me.

Add to this the fact that primary energy generation is increasing all over the world, so the share of renewables is increasing even less.

The target is 2050. There is no R&D that will get rid of electrolysis and Carnot cycles by then. Maybe by the end of the century if you want. But for the next 30 years, batteries will be in the lead.