I only have two small issues with a wholesale switch to nuclear power:
1) We need, and don't have, a good long-term (~centuries) method of sequestering about a half-dozen long lived fission products that are hazardous to life. (As planetary emperor, I'd say sequester them in low-melting glass, make concrete with the glass, and lower it into geologic subduction zones on the ocean floor. But we are not doing that.)
2) We have this Git-R-Done mentality that is likely to result in hasty permitting, shoddy construction, and lax enforcement. This has only gotten worse in recent decades with the rise of the global corporatocracy. It will lead to countless additional industrial accidents, like TMI or the SL-1 event.
Both of these problems are essentially political in nature, and not technical. But as I continue to watch the political arena, I become more and more cynical about the possibility of any useful change occurring before the wheels simply come off of our "civilization".
I'm thinking that if we had enough coal (and hoping that we don't), we earthlings would simply continue to burn it until the atmosphere became rich enough in CO2 that we could no longer breathe it.
Aside from e.g. Tc-99, I'm not sure much needs to be done about fission products. The major issue is the total toxicity, and once it's dropped below the level of the original uranium what's the point?
We've only had ONE SL-1 event and ONE TMI Unit 2 in 60 years. Designs and other things have been changed so that the same type of accident will not happen again, and even if we could expect one every 3 decades, it wouldn't be anything to worry about compared to the alternative.
WRT isotopes, I was thinking more Sr-90 and Cs-137 and such. Things that are actively accumulated by living things. Iodine is _very strongly_ accumulated ... most of its isotopes are either quite temporary or nearly stable and very low-level, however.
Once we have TPTB acting out of panic and issue (2) gets some traction, there's no telling where this crud ends up. I'm really talking about political dynamics more than engineering here, and it's pretty hard to track anything on a time horizon longer than a decade, particularly with the gathering storm clouds of change we can see on the horizon.
The half-life of Sr-90 is 28.8 years, Cs-137 is 30.1 years. If you can isolate these for as little as 1000 years (not very difficult), their concentration is reduced by a factor of over 1 billion.
I'm not worried about Git-R-Done on the back end. There is nothing that needs to be done about SNF for many years, and if we e.g. pyroprocess our stocks of PWR fuel to reclaim the Pu for starting IFRs or LFTRs, we'll get the fission products in some kind of readily-sequestered salt. Once soaked up in a zeolite and packed in stainless steel, it can go back into the concrete dry-storage casks for a century without anyone needing to worry about it much.
The real threat is failures of PWRs and BWRs. Those can dump stuff like I-131 where people get substantial exposure in a short time. Unpressurized reactors like lead- and sodium-cooled technologies don't have those problems, and I suspect that Git-R-Done will be forced toward those rather than things which require very large forged pressure vessels.
I hope you are right when it comes to the Git-R-Done stage. There seems to have been a good bit of resistance, well really anything with the word "nuclear" tends to elicit political hysteria, and resistance to processing of spent fuel. So it just piles up emitting evil-looking Cerenkov radiation in these pools in our rusting fleet of PWRs. Probably enough of it to keep us in electricity for a century without extracting any more actinides from earth's crust.
I'm seeing a lot of political ads on TV recently, 'tis the season, you know, and not only is the political arena devoid of science and engineering, it also lacks basic common sense. And these are the guys who are supposed to "lead", hah.
If David LeBlanc's claim of 0.8 tons of U/Th per GW-yr is reasonable (and translates from LFTR to IFR), and the 60,000 tons of SNF stored in the USA is about 95% uranium and higher actinides, we've got something around 70,000 GW-yr of electricity available from it. The USA's average consumption is around 450 GW, so it represents 150+ years at our current rate if burned in IFRs. Multiply by 5 to include depleted uranium from enrichment.
The idea of an IFR that needs nothing but the spent fuel from an existing reactor at the same site may sell to the public. The cooling pool will eventually empty, the dry casks hold rods which have been decaying down for up to 40 years; processing this material into an IFR makeup fuel stream and a fission-product stream makes no more mass (I suspect that the bulk goes up) and presents no new issues. The stainless-steel cans of absorbed salt can go straight back into the empty casks.
The problem starts to be one of pace: if an IFR can get almost 20 times as much energy out of the spent rod as the PWR got from the new one, it will take several hundred years to go through the SNF inventory. We'd have to add multiples of our current generating capacity to work through it in a human lifetime, or separate some of it into at least 3 streams (e.g. reclaimed uranium, Pu+Am+Cm, and waste material) just for the sake of being rid of it. Fluoride volatilization in molten salt followed by electrolysis can do this, but I wonder about the non-technical issues.
OTOH, maybe Git-R-Done will demand this be done in 10 years so the Pu+Am+Cm can be used to start all the new reactors. We'd still get rid of the SNF. I just hope that nobody's crazy enough to try to do this with a massive Purex plant.