There is one important material constraint before expanding solar PV technology to the multi-terawatt scale: silver. Right now silver-containing conductive past

There is one important material constraint before expanding solar PV technology to the multi-terawatt scale: silver. Right now silver-containing conductive pastes are the overwhelmingly popular choice for forming cell contacts. I expect copper contacts to replace silver pastes at some point. Plated copper has higher conductivity than printed silver paste and of course the raw metal is much cheaper, but it takes more effort to integrate plating into manufacturing flows. SunPower already uses copper metallization at gigawatt scale cell production levels, so I'm pretty confident that other manufacturers can follow when silver costs go up. The one important possibly-constrained battery material is cobalt as a component of lithium-ion battery cathodes. But stationary storage batteries don't need to maximize energy density and therefore don't need to use the same cobalt-bearing cathodes as those in mobile electronics and electric vehicles. The key attribute of a stationary grid-storage battery is lifetime -- encompassing both calendar life and cycle life. Long-life batteries don't need cobalt if the energy density is negotiable. FWIW you can see from my comment history that I have no qualms about the scalability or safety of nuclear power either. I think that deep decarbonization may require additional nuclear power. The cost trajectory for storage-backed renewables just looks more promising than the cost trajectory of new nuclear power right now. It appears that the least-cost decarbonization path will emphasize renewable/storage buildout until the marginal decarbonization cost rises high enough to justify more high-cost (but rock-steady) nuclear reactors.