Hacker News Reader

Comment by jpk | Hacker News Reader

My understanding is that building the reactors are significantly more expensive. So until the price of Uranium increases enough to justify the total cost of the Thorium reactors, we won't see many of them outside research settings.

The other issue is building a new reactor of any type right now is a giant pain in the ass, which doesn't help either.

View on HN

Thorium reactors should be quite a bit cheaper. Light-water reactors operate at 160 atmospheres of pressure. They need a very strong reactor vessel, which currently can only be forged by a single facility in Japan. They need a large containment dome because if a pipe breaks, that high-pressure water will flash into steam with a thousand times the volume. They use lots of redundant emergency active cooling mechanisms. Some of them have giant slabs of ice inside the containment dome, which reduce the volume of steam if it's released but have to be constantly refrigerated, to keep them frozen in close proximity to a nuclear reactor core.

LWRs are resupplied every year with expensive fabricated fuel rods, which are non-standard and can only be purchased from the company that sold the reactor. The fuel rods are complicated because they have to withstand a thousand-degree temperature gradient, and the fuel pellets are prone to cracking from the production of xenon gas. Xenon and other reaction products prevent the use of more than one percent or so of the energy potential of the nuclear fuel, another reason the rods are frequently replaced, and the reason we have so much nasty nuclear waste.

The nuclear industry makes most of its revenue from selling those fuel rods.

LFTRs operate at atmospheric pressure. No super-strong steel, no containment dome, no ice. It's a liquid fuel, so no proprietary fuel rods. Xenon just bubbles out of it.

The fuel has a strong "negative coefficient," meaning the reaction slows down as it gets hotter. If it nevertheless gets too hot, a salt plug melts and all the fuel drains into a passive cooling tank. No need for all those active cooling systems.

On top of that, LFTRs operate at higher temperature, so the turbine is more efficient, and the waste heat can be used to desalinate seawater. They don't require water cooling. As a bonus, marketable reaction products can be separated from the liquid fuel (http://flibe-energy.com/products/).

Misguided government regulation is the main problem, but Sorenson's company plans to get around that by selling to the military first.

"Misguided government regulation is the main problem, but Sorenson's company plans to get around that by selling to the military first."

Sorry but is this subtle irony? My apologies if it's not, I don't know anything about this field. It just seems weird that the military wouldn't be subject to the same regulations that the govt. sets? Or in the US are the military able to ignore some of the regulations around this?

The NRC has no authority over nuclear aircraft carriers and submarines. The military's stuff is classified, and it does its own regulation.

The latter, I'm sure. When you are a war economy, the military rules supreme.

In the video he states that the cost would be 30-50% lower because the reactors are very simple. The cost for uranium based fuel is also significantly more expensive (watch the vide). He states that the reason people don't build these is because they don't know about them, or they are stuck in the existing structure of legacy reactor thinking.

The reason that anything new is a pain in the ass is simply because of bureaucracy.

It might be that you could make a thorium reactor much safer than current light-water reactors for cheaper, but the laws regulating our nuclear power were written under the assumption that all reactors were light-water reactors - so in practice you would end up with all the fancy 100% reliable control systems needed to keep the positive feedback loops in a traditional reactor from blowing themselves up as well as the special metallurgy needed by Thorium reactors.

That is the case, and it's a big problem. NRC regulations for example require "fuel integrity," which doesn't exist with liquid fuel.

Sorenson's company plans to market initially to the military, which has need of compact energy sources for remote bases and isn't constrained by the NRC.

CANDU, one of the best uranium designs, has the same problem.

It's sort of like inkjet printers. Either it's cheaper up front and more expensive in the long run, or it costs a fortune up front but the ink is cheap.

To be completely honest, though, one of the biggest obstacles to widespread CANDU implementation in the past was that it isn't a breeder.

Reply on news.ycombinator.com