In Nuclear Power, Size Matters

The LFTR uses thorium dissolved in molten floride salt. It is proven tech, since the US government
built one back in the late 60s and ran it for 5 years -- with 1.5 years at full power...

The devil is in the details.

While it is indeed possible to build an LFTR, that old bugger called economics tends to come and mess things up.

First of all you need a larger amount of fissile materials since the molten salt transports it out of the core. and around the entire primary loop. Secondly, as with sodium, you need to have a secondary loop to make things safe. Then there's the hydrolysis that can occur at low temperatures, which means you have to keep the salt molten. If the reactor has problems, that may involve drawing power from the grid. The reprocessing technologies kinda work, but are unproven at large scale, and nobody has an idea what the cost will be for a large reactor. They also imply building reprocessing tech for every single plant, which increases capital costs.

Then there is the startup material. Natural uranium is not good enough, so you either need to breed U-233 in a different reactor ( proliferation concern ) , use highly enriched U-235 ( proliferation concern, expensive ) , or startup on plutonium. Now plutonium in a thermal spectrum leads to accumulation of Curium, which is a troublesome waste product that cannot be efficiently destroyed in a thermal reactor.

Add in that while Thorium and Uranium dissolves easily in fluoride salts, plutonium and the other actinides do not. In fact, even at high temperatures with a completely pure salt, the solubility of Pu fluorides is just a few percent. The molten salt reactor experiments got around these issues by using a very exotic salt. Beryllium and Lithium fluorides, with the lithium enriched in Li-7. Now, beryllium is highly toxic, expensive and difficult to work with. It's such a pain that the US and UK considered developing new nuclear warheads that did not use it, even though it is the best lightweight neutron reflector there is. Enriched lithium-7 is a different problem in itself, and even if 99% pure, you will get quite a bit of tritium when it is exposed to neutrons. Perhaps not more than in a CANDU reactor, but all tritium control systems ever designed are made for water coolant.

Then is the issue of in-core materials. The molten salt reactor developed by the US dealt with damage to in-core materials by replacing the graphite core materials frequently. Not only is this expensive, but it's not very fun to handle radioactively contaminated graphite. It is hard to reprocess since it forms organic compounds and is difficult to dissolve in nitric acid. Pyro-processing by electro-refining and similar is also poorly suited for graphite. This is one of the reasons why the pebble bed reactors are usually seen as "once through". Nobody has come up with a practical way to deal with the graphite. Since the material will be in direct contact with the fuel salt, it will likely adsorb quite a bit of contaminants.

Plateout on heat exchangers is another issue. The noble metals have poor solubility in fluoride salts, so unless a very potent ( i.e expensive ) reprocessing system is able to get rid of them quickly, they will plate out on the cold parts of the reactor, which is usually the heat exchangers. A suggested solution is to use graphite-based heat exchangers, which has its own spectrum of development issues and research needs.

I'm not saying molten salt reactors can never become a good idea. I'm just saying that in comparison to the number of issues that need to be resolved to make them practical for a power plant, they are extremely hyped.

Source: http://rss.slashdot.org/~r/Slashdot/slashdotScience/~3/1OGasBLRoJI/in-nuclear-power-size-matters

light field camera world series game 1 exotic animals exotic animals scott hall lra lra