What is a molten salt reactor? (opens in new tab)

Very nice. Could you add one about the waste cycle? Like, it feels true in one sense (maybe there's less fuel waste that matters, or so the claims go) but way false in another (molten salt is nasty as hell and requires a lot of maintenance which means a lot of irradiated waste).

I've been hearing for years now how fantastic Thorium is, but with every new article, I wonder: so where are all these new fantastic Thorium plants?

If they still don't exist, then surely that suggests it doesn't work, doesn't it?

> The way modern society is going, a LOT of power is going to be required

What do you mean by "modern society"? According to the IEA, "energy demand per capita is likely to remain at current levels for the next 25 years. In other [words], peak per capita energy consumption may have already arrived".[0]

Most "modern societies" have less than replacement fertility levels, so energy use is only increasing because the population is living longer, and because of immigration. To give some aggregate numbers for a set of modern societies: "The EU-27’s population is projected to increase from 446.8 million in 2019 and peak to 449.3 million in 2026 (+0.6 %), then gradually decrease to 441.2 million in 2050".[1]

[0] https://www.businessinsider.com/bernstein-says-peak-energy-p...

[1] https://ec.europa.eu/eurostat/statistics-explained/index.php...

Electric power demand will increase by what it takes to keep our cars charged up, and to drive industrial production of hydrogen for steel and other industrial processes, ammonia for fertilizer, synthetic fuels (including ammonia and hydrogen) for bigger vehicles, carbon capture from the atmosphere, water desalination, and myriad other uses.

Build-out of renewables will increasingly displace fossil-fuel generation, then (as that is exceeded) bank energy in storage systems (batteries and pumped hydro, short-term, other tech for longer term). We will need a lot of new transmission lines to trade power between current users, current generators, and storage.

None of it requires new physics, although improvements in catalysts will improve round-trip efficiency and cost of storage systems and synthesis, and new chemistries will improve cost and efficiency of new photovoltaic build-out.

There is no place for nukes in that world. They cost way too much to compete on a fair playing field.

I've been working on molten salt reactor claddings for a few years now. MSRs are very safe.

First, there are two main groups of molten salt systems. One group uses generic salt to transport heat. This usually provides efficiency gains over water based systems. These are the systems that might leak sodium and it's no big deal.

The 2nd kind of system uses a salt with the fissile material dissolved in it. This is what the article is about. Leaks in these systems are not an option. The molten salt in this case slowly attacks any metal it comes in contact with. The tanks, tubes, etc. are designed with a large safety margin and constantly monitored. The salt chemistry is also monitored for signs of the materials the storage vessel is made of.

Storing and piping molten salt proved too unreliable for the Crescent Dunes solar collector. Swapping out the simple and free heat source with a much more complex nuclear reactor doesn’t seem like a good idea.

https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Pr...

I don't know if they are safe or not but one thing that disturbs me about neo-fission propaganda is how they so nonchalantly dismiss what seem to me to be threats worth discussing. For example, MSR fans will say that they operate at standard pressure, and therefore can't rupture. Also, gaseous fission products are removed continuously. OK but what about the gaseous fission products if the removal system ceases to function? Then what about the pressure? Same question about water. MSRs are safer because they contain no water, which can't flash to steam and explode. OK, but what about water intrusion on the surprisingly wet planet Earth? What if the reactor is flooded? Would a flooded reactor stop the passive safety plug from melting?

I always had the same suspicions about the pebble bed reactor boosters. That was a design that was always sold as being inherently safe and self-moderating, with the huge, huge asterisk that if any oxygen or water ever gets anywhere near the fuel the whole thing will explode and then catch on fire, which seems like a pretty severe drawback. Any time you see a claim that a reactor vessel "does not contain" either water or oxygen, just mentally translate that to "is not designed to contain" or "does not initially contain" before reading further.

> in practice every such system has leaked liquid sodium

Just to clarify, reactors cooled by liquid sodium are not molten salt reactors but rather belong to the class of reactors often called "metal-cooled" (sodium is a metal, not a salt, see).

"in which"

I am personally not willing to put all the eggs into one basket, because I don't share your certainty about future course of things.

We have some lessons from the Energiewende in Germany, which turned out much more expensive than promised. German Green minister Jürgen Trittin assured the German public in 2004 that support for renewable energies is going to cost them "one ice cream scoop per month" [0]. Which was so wildly off-mark that you can only laugh about it bitterly in 2022.

Now it is possible that molten salt reactors are a waste of money, but I would still prefer having more options open for the future. Especially your idea that storage will be cheap needs to be tested in reality first. Currently, storing of electric energy is darn expensive.

[0] https://de.wikiquote.org/wiki/J%C3%BCrgen_Trittin

Edit: instant downvote instead of counterarguments. Energy storage is still pretty expensive regardless of the # of votes (positive or negative) that this comment attracts.

Some energy markets are already hitting saturation periods for renewables. This creates diminishing returns, where more and more of the energy generated from an intermittent source isn't actually used. We were promised that storage would be made cheap enough to capture this excess energy and release it during hours of under-production. But so far nothing has delivered storage at competitive costs.

Furthermore, renewable generation varies a lot by region. Solar's cost per watt is way different in California or Hawaii than it is in Massachusetts. This is both in terms of less sunlight (inclination of the Earth, plus weather) but also land costs. This could be ameliorated by long-distance transmission, but that has its own problems even within small states [1]. And we'd have to increase the net-cost of deployments to match.

There's none of these asterisks and hand-waving with nuclear. Heat water, spin a turbine. Energy where you want it, when you want it.

1. https://www.vox.com/videos/22685707/climate-change-clean-ene...

It is worth pursuing solutions like Moltex though because eventually we need to find a replacement for the gas power plants which will be backing up wind and solar.

Pumped hydro, cryogenic storage, thermal storage, batteries, green hydrogen etc may tackle a lot of that but we simply don’t know what solution will end of working best.

Wind and solar has to be primary focus for the next 10-15 years but once we reach 60-70% of the grid being renewable you cannot really go much further. Solutions which can deliver power on demand will be crucial. If MSR guys can get stuff going within 10 years then they may be part of the solution.

History does not end in 20 years though. Humanity will continue to exist for hundreds, thousands of years, probably more.

I agree with you that we don't need nuclear power to solve the climate change problem. But once we are done with that, we will continue to live. Research in new types of nuclear power will yield benefits, even if on a timeline that is beyond the target date for net zero emissions (2050 or so).

I came across this topic when I read about a proposed MSR powered ship. I’ve yet to see a solution for powering ships via renewables. Ships also require a relatively small reactor (since they are only powering the ship) compared to a land based reactor which will power a region. So this could be the use case needed to bring such tech to fruition.

There are countries where neither solar nor wind are a viable options.

Or water cooled reactors!