TerraPower's Natrium: A Fast Neutron Reactor with Built-In Grid Level Storage

That's for their sodium test facility, which they have just started construction on.

https://news.ycombinator.com/item?id=40658852

They applied for their NRC permit on March 28, 2024.

https://www.nrc.gov/reactors/new-reactors/advanced/who-were-...

They can extract a little bit more energy out of spent fuel using a process like DUPIC [1], but they don't extract the vast energy remaining in the U-238, and they also do not transmute it into shorter-term waste like fast-neutron reactors do, so the total value is pretty low.

There are lots of CANDUs outside of Canada though! [2]

[1] https://dspace.mit.edu/bitstream/handle/1721.1/45495/4225543... (thesis)

[2] https://en.wikipedia.org/wiki/CANDU_reactor#Active_CANDU_rea...

Candu reactors are cool, but there are some reasons not a lot of people use them outside Canada.

The main one is that a certain reactivity feedback loop is positive. Meaning as things go bad, they tend to get worse. There are numerous feedback loops in a reactor, and the others are negative, so overall the reactors are safe. But inherently they are less safe than pressurized water reactors. Details here [1].

Another problem is the amount of waste. Because CANDU reactors use unenriched uranium, they extract less energy from the same amount of fuel. Put it differently, they need more fuel to get a certain amount of energy. About 10 times as much. After the fuel is used, it becomes waste, and you need to handle it somehow.

Yet another problem is that deuterium is not as good a moderator as hydrogen. It does not absorb neutrons, and because of that it results in much better neutron economy, but it takes a larger volume of heavy water to achieve the same amount of moderation. So a Candu reactor tends to be bigger. It is also much more complex, because the heavy water and the regular water move in different circuits. All in all, a Candu reactor is a bit more difficult to maintain. For example, it needs to be overhauled after 30 years in order to get a life extension of another 30 years. A PWR can go for 40 years before an overhaul. That's not a huge difference, but it's something.

Of course, the other thing is that you need the initial investment in the heavy water. Last time I read about it, that heavy water added about $1 BN to the construction cost. The idea was that over the life of the reactor, one would recoup that in the cheaper cost of the fuel, which does not need enrichment. Still, $1 BN is a big number.

As for the spent fuel. You are right, Candu reactors can use the spent fuel from PWRs. But that's just in theory. I don't think in practice they do that. Reactors are complex enough. Dealing with the uncertainty of using a different fuel that the one they've been tested with is not something operators like to do without a very good reason. So in practice Candu reactors don't use spent fuel or thorium, or even MOX fuel (a mix of uranium and plutonium).

[1] https://en.wikipedia.org/wiki/Void_coefficient