Nuclear was cheaper when more of it was built [1]. Economies of scale make things cheaper. A production run of 40 steam generators is a lot cheaper than 4 steam generators.

Proponents of a primarily solar + wind grid are betting on a breakthrough in energy storage. If that breakthrough does not transpire, we'll either have to give up on stopping carbon emissions or use nuclear power.

1. https://www.researchgate.net/figure/Overnight-Construction-C...

Converting atmospheric CO2 into fuels could contribute to this effort. But bacterial and plant-based fuel production may still be more economical and produce fewer overall emissions than even a solar array and a carbon capture plant.

Converting atmospheric CO2 into hydrocarbon fuels requires hydrogen as an input, so it'd probably be easier to just store the hydrogen directly. Right now, almost all hydrogen is produced through steam reformation [1] which emits CO2. Electrolysis is inefficient and corrosion of electrodes makes it expensive and hard to scale. Capturing atmospheric CO2 is similarly difficult. Carbon Dioxide is at very low concentrations in the atmosphere so it takes a really long time to sequester meaningful amounts of it. Similar issue with biomass: it produces energy very slowly and doesn't have the scale required.

There's a reason why plans for a primarily renewable grid assume that compressed air, synthetic ammonia, giant flywheels, or something else will provide storage for orders of magnitude cheaper than batteries: because existing storage systems aren't capable of meeting the storage demands of intermittent generation. Will one of these systems deliver a storage breakthrough? Maybe. But it's not wise to bet the future of your electrical grid on a technological breakthrough that hasn't happened yet.

1. https://en.wikipedia.org/wiki/Steam_reforming