[throwaway90446]

So basically, they are better on every possible front, and the technology has existed for a while?

Yet, LWR is still the dominant design. So, what's the catch? There's clearly something preferable about Uranium LWR that causes their dominance, no?

[badloginagain]

When nuclear research was first being developed, it was for the specific purpose of building the bomb. When the war ended, and the industry switched to civilian applications, they decided to go with Uranium because they had already worked out many of the issues.

Still, the US Airforce wanted a nuclear-powered bomber shortly after the Navy showed their nuclear sub and before ICBM's deprecated "end-of-days" long-range bombers. A water-contained reactor wouldn't work in an airplane, so they funded the development of the MSR (molten-salt-reactor).

Unfortunately, when they shuddered the project after the implementation of practical ICBMs, industry politics discredited MSR's in favour of the reactor-types we have now.

After fukushima, a grassroots campaign has been undertaken to resume research of MSR's as a replacement of fossil fuels for scalable carbon-free energy creation. If you are interested, you can get a great overview here: https://www.youtube.com/watch?v=P9M__yYbsZ4

[r00fus]

> Unfortunately, when they shuddered the project after the implementation of practical ICBMs, industry politics discredited MSR's in favour of the reactor-types we have now.

So essentially what you're seeing is the result of decades of regulatory capture of the NRC by the nuclear industry.

[angersock]

I think the missing factor is that there are massive regulatory hurdles in the way of testing new designs, no?

[fennecfoxen]

... well, yeah: placing massive hurdles in the way of potential competitors and their technologies is about half of the premise of regulatory capture...

[agumonkey]

It's amazing how the state of things may look like a local maximum depending on the root of the research path.

[Animats]

In light-water reactors, the highly radioactive part of the system is very simple. It's a core of metal fuel rods and support structure, some control rods, and water. All the plumbing complexity is external, and it's on water, which isn't hard to handle.

Most of the alternative reactor designs have more going on in the radioactive part of the system, or more difficult working fluids. This usually leads to trouble. Sodium-cooled reactors have sodium fires. Helium-cooled reactors have helium leaks. Pebble bed reactors have pebble jams. (The one in Germany is so jammed it can't be decommissioned.) Molten salt reactors have to pump radioactive molten salt around and run it through a chemical processing plant. In some designs that salt is a fluorine compound. Now you have all the headaches of operating a radioactive chemical plant.

Most power utilities don't want to operate a radioactive chemical plant.

[pjc50]

I wondered what had happened to pebble beds. http://en.wikipedia.org/wiki/AVR_reactor

Most power utilities don't want to operate a radioactive chemical plant

This is a brilliant phrase, thankyou.

[daeken]

I have little more than a layman's understanding of the nuclear power landscape (I've read a decent bit and watched several talks, but that's about where it ends), but it seems to me that there are at least two reasons: 1) LWR reactors have been built for a long time, so they're very well understood (known risks are better than unknown risks) and the processes are already in place to bring them up. 2) Even though they're dominant, very few LWRs are built. There simply aren't many reactors being built of any kind, so it's unsurprising that the most common type (traditionally) is what we keep building.

[taliesinb]

> LWR reactors have been built for a long time, so they're very well understood (known risks are better than unknown risks)

After Fukushima, I don't think most people would agree the risks are well understood, at least as reflected in the actions of the nuclear industry and its regulators.

The NRC (and hence most of the world's regulators) uses a "design basis" approach to establish what emergencies reactors should be able to respond to safely without the release of radioactive materials. The design basis is supposed to quantify the known risks.

In 2011 we saw just how inadequate the design basis framework was. It failed to predict risks such as multiple systems failing simultaneously, emergency generators being flooded, the plant being cut off from external help, multiple meltdowns happening simultaneously, valves getting stuck open, and a litany of other things that actually happened, resulting in the level 7 accident we saw.

While the exact set of events that happened at Fukushima Daiichi is unique, similar accidents could happen in the US, or indeed anywhere (e.g. many reactors are downstream from major dams and could theoretically experience catastrophic flooding).

And just as the NRC's computer models deemed Fukushima impossible before the accident, the NRC has largely ignored the recommendations of its own near-term task force in how to improve the regulatory situation in the US after the accident.

It is up to us to demand that the nuclear power industry, which is wielding technology of massive destructive power at the behest of its shareholders, transform itself into the transparent, accountable industry we deserve.

[everyone]

LWR was chosen during the cold war because they wanted to make as much weapons grade fuel as possible.

[aqme28]

Also because IIRC it made more sense on submarines.

[ecopoesis]

LWR is what the Navy used, and the civilian designs were just based on that.