I have to say, that I'm hoping that we will eventually get to the point where we have reasonably efficient processes (probably via clever catalysis or bioreactors) to produce synthetic fuels as a drop-in replacement for hydrocarbons, with the feedstocks being carbon dioxide and water (plus energy). Then the problem turns into one of providing the energy, which can be from renewables, or nuclear, or fusion if it ever works, or whatever. From a greenhouse gas point of view, that is then neutral, the fuel is simply a high-density energy carrier.
Things like hydrogen have real problems: hydrogen likes to quantum tunnel through things, so a tank that will happily hold natural gas may leak hydrogen like a sieve, and hydrogen is not only explosive over a wide range of concentrations, but is odourless and burns withy an invisible (to the human eye) flame. Those are significant safety barriers to overcome. Fuel cells still need fuel (really they can be seen as slow combustion that provides electricity over heat).
If we can find a path to easily synthesise hydrocarbons, for example, propane and butane for gas burning appliances, and longer chains such as heptane and octane for liquid-burning engines, then existing supply lines and technologies can be used. Technologies which have already had well over a century of refinement and efficiency gains. We don't reinvent the wheel if we don't need to.
edit- a quick google show that such chemical reactions driven by sunlight are currently at about 1%, producing methane and ethylene. That's a start. For comparison, photosynthesis in plants is about 11% efficient.
Electrochemical redox reactions producing hydrocarbons from CO2 and water look to be about 13% efficient, which is promising.
Both of these are from research papers published in 2019, so this is very much cutting-edge stuff at the moment.
Biting the hand that feeds IT
