Here is a more extensive look at fuel reprocessing and refabrication in EBR-II.
Some of the footage is actually the same as that in the IFR film we recently showed, although that is less obvious because this is in black-and-white. Close comparison with another EBR-II film shows that the material used for coating the glass moulds for casting the fuel pins was at some point changed from graphite to thorium oxide, or perhaps vice versa.
The narration emphasizes an important economic aspect related to the closed nuclear fuel cycle. The longer it takes for fuel removed from the reactor to be cooled, reprocessed, made into new fuel, and reloaded into the reactor, the larger the total inventory of fuel required per megawatt of capacity. Locating the reactor and the fuel facility together shortens this time, and incidentally avoids the inconvenience of shipping the spent and fresh fuel. Reprocessing methods which do not require waiting months for decay of the short-lived fission products can further shorten the time, although this means that all operations have to be carried on behind heavy shielding.
Early on at EBR-II, the fuel remained in the core 135 days, was cooled for 15 days, and could be reprocessed in under 30 days. Thus the out-of-reactor fuel inventory could be reduced in principle to 1/5 of the quantity in the reactor. This is a big difference from concepts involving oxide fuel and away-from-reactor PUREX reprocessing, in which the out-of-reactor inventory is often half or more of that in the reactor. Later on, the metal fuel pin performance was improved to the point that fuel could remain in the core close to two years. Although a somewhat larger cooling time would have to be allowed, this still implies a very small out-of-reactor inventory. So, even if the quantity of fuel required in the core per megawatt of output is greater than with oxide fuel, the overall requirement might actually be less.
Given that, in general, a metal-fuel fast-neutron reactor has a larger breeding gain (the plutonium which it produces per megawatt-day of operation, over and above that required for feeding back into the reactor) than an oxide-fuel fast reactor, the overall concept appears very favorable for supplying fissile material in a fully-nuclearized energy economy.












