Let me preface this with my personal opinion about nuclear power: I 
nukes. I was raised about 10 miles from a huge plant with two reactors. In a 1980 referendum here in Sweden I voted Yes. I think we have wasted 35-45 years being scared of nukes, for very little reason, and because of this, most reactors in the world that are currently operating tend to be on the older end.
So, these past few years, with climate change looming and renewables not generating nearly enough electricity, nuclear has seen a renaissance. But - and here it comes - I think it’s totally misguided to place our hope in nuclear power to solve climate change, even though nuclear may very well, in one form or another be the best solution in the very long perspective, i.e. centuries and not decades.
So why do I think this? The TL;DR version is that the math doesn’t add up. It simply can’t be done.
• Uranium is not renewable. We use U235 of which there is a finite and not very big amount. Confirmed estimates say there are about 4.2M tons of U235 and we currently use about 60K tons. This means that we have fuel for 70 years at the current rate of use. Should the use ten-fold… Well, let’s not ten-fold the use. Cite.
Another problem with current technology is that it won’t mitigate CO2 emissions in a very meaningful way. By 2040 the estimate is reducing carbon in the range 2 - 5 percent. Cite
• What about U238? Well. If it could be done in a good way, this would mean that we don’t have to worry about running out of fuel, as U238 is more than 99 percent of our total uranium. However, the only way we currently know how to use it is with Fast Breeder Reactors. And it doesn’t seem that it’s a viable way forward, as we’ve been experimenting with FBRs since the 1950’s and still haven’t solved all problems. BTW, one of those problems is that a byproduct of using U238 in reactors is PU239. Cite.
There are currently three (3) breeder reactors operating in the world. Two in Russia and one in India. Cite.
• Another problem is that it’s expensive and price is rising. Since 2009 cost for running nuclear plants has gone up about 33 percent, while solar has gone down 90 percent.
Globally the cost of renewables is now significantly below that of either nuclear power or gas. According to Bloomberg New Energy Finance (BNEF), wind and solar power are now the cheapest form of new electricity in most of the world. Furthermore, BNEF anticipates that it will be more expensive to operate existing coal or natural gas power plants in five years than to build new solar or wind farms.
Cite: PDF page 293
• Nuclear plants are expensive and take a lot of time to build. The poster child for modern nuclear power plants is Olkiluoto in Finland. Construction on its third reactor (a generation III reactor) started in 2005 and was projected to cost €3BN when it was completed in 2009. It became operational last year at a cost of close to €9BN, but was quickly shut down:
electricity production started on 12 March 2022. In May 2022, foreign material was found in the turbine steam reheater, and the plant was shut down for about three months of repair work. On 30 September 2022, the reactor reached its maximum output power for the first time.
I’ve been trying to find a good source that estimates if nuclear power is commercially viable without support/subsidies. There are a lot of sources out there that look into costs. Almost all of them agree that the upfront costs of construction, the security measures when running the plant, insurance fees and the disposal of waste are huge, whereas actually running the plant is cheap. Almost all ongoing projects do seem to have some kind of support from their respective countries and I can’t find a single project that is running without any kind of subsidies.
• Thorium. I’ve been hearing about Th for about 15 years. There are indeed a lot of advantages. It’s a lot more effective than uranium, it’s a lot more abundant. The reason for that may be that Th232, which is the isotope proposed for nuclear power has a half-life of 14 BN years.
But there is only one (1) reactor running right now. It’s in China and although the program launched in 2011, it’s not being scaled in any purposeful way.
Operated by the Shanghai Institute of Applied Physics (SINAP), the Wuwei reactor is designed to produce just 2 megawatts of thermal energy, which is only enough to power up to 1,000 homes. But if the experiments are a success, China hopes to build a 373-megawatt reactor by 2030, which could power hundreds of thousands of homes.
Cite.
• SMR. We’re reaching the end of this very long post and I’ve saved the Small Modular Reactor to last. As evidenced in this thread, this is where our nuclear hopes and dreams are. In the U.S. the poster child is Nuscale. They started in 2007 and still hasn’t started its first prototype. In march 2020 things were delayed due to security risks. It now looks as if the $6.1BN SMR will be completed in 2030, producing 50 MW. Cite.
• Those 50 MW should raise questions. The International Energy Agency, IEA, estimates that that we need to increase nuclear with 10 GW to 2030. My handy calculator tells me that we need to build 200 of these SMRs from now to then. There are currently two (2) up and running with another four being constructed. Six more are being licensed. The rest (32 units) are at different stages of design. This is very far from 200 and we’d need to ramp up SMR deployment to a new SMR - up and producing power - around every two weeks starting now, if we’re to reach the estimate from IEA.
I am not saying that we should stop all plans. But I think we realistically have to face the fact that nuclear power is not the solution to climate change, and that it’s even hard to estimate to what extent it can actually help. Obviously some. We should’ve done what we’re doing now 40 years ago, but public sentiments back then made it politically impossible.
.