I've been in a debate with a conspiracy nut for years. He is never going to change his fantastic ideas, but sometimes I learn something from him. I have a table for the Olkiluoto reactors in Finland. Sorry, only in Finnish. They weigh the rods in the Finnish reactors. The table is in Finnish, but I calculated about 3% weight loss in the year that they added and removed the same number of rods, 2005. It does not happen every year, that the number matches. Is 3% explainable by the Einstein's equation, for a standard light water reactor? the Finnish table http://i9.photobucket.com/albums/a57/Tero1111/artontaulukko.jpg
The amount of uranium is in a fuel rod is certainly less than 20% and probalby less than 10%. On top of that the fission products do not leave the fuel rods so there would be no change there. Finally the amount of mass lost in fission is approximately 0.08%. So the only way that you could get a 3% loss would be from something not directly as a result of fission such as corrosion and wear of the outside of the rods - which is quite possible.
How much energy were these reactors producing? E=MC^2 is a really simple equation to use, if say you had a 500 megawatt (thermal) reactor operating for 1 year that would be 1.6*10^16 watts total or joules of energy, divide it by the speed of light squared and you get only 175 grams of mater consumed! What is likely happening with these control rods to account for most of their mass loss is that fission fragments (neutrons, atoms traveling near the speed of light, sub-atomic particles) are "evaporating" matter from the rods into surroundings and beyond (neutrinos bleeding off and estimated 4.5% of nuclear fission energy into space, as the neutrinos are virtually unstoppable because they react with matter so very very rarely.)
Good explanation, neutrinos. These Olkiluoto reactors are 2500 MW, for 1 and 2 together. I think the rods were counted for the two. 220-240 rods are changed each year. http://en.wikipedia.org/wiki/Olkiluoto#Units_1_and_2 seems to be only 1760 MW for 1 and 2 together.
Some of the weight loss appears to be noble gases Kr , Xe http://files.asme.org/Divisions/NED/31871.pdf
Neutrons have weight, and they are scattered from the uranium to any medium and the control rods. Another source of weight loss. Some neutrons end up as part of the vessel itself. Read more: http://wiki.answers.com/Q/What_is_the_purpose_of_using_heavy_water_in_nuclear_reactor#ixzz2GoueTjdS
Almost none of the fission products end up outside of the rods. One of the prime considerations in reactor design is keeping fission products out of the moderator. Less than 3 neutrons are produced from fission maybe 1 neutron per fission makes it out of the rods that is about .5% of the fuel and if the fuel is 10% of the mass of the rod then that is a 0.05% mass loss. edited to add: Maybe it would be better to say 2 neutrons make it out of the rods - that is still only a 0.1% mass loss in the rods.
Loss of fission product is something that the control rod sheathing is designed to protect against. However, I'm sure there is some loss, but this is likely well-detailed and known to the people who actually load/un-load the reactor rods. 3% loss seems rather high to me at first glance, but that could be the case, not actually certain what the loss would be (other than it is not loss due to mass conversion which is negligible).
Loss of mass by neutrino emission: Only 4.5% of energy Loss of mass as energy: certainly less then 0.01% considering that a even at 3 gigawatts continuous operation for a year is only ~0.6 kg Loss of mass by neutron emission: Even 3 neutrons out of 100% U235 by mass would be 1.3% and the fuel rod is mostly other atoms and isotopes Fission Fragments: this is all we got left to explain the mass loss... almost Fuel rod stripping: it is possible though unlikely (or really wrong) that the surface of the fuel rod is being stripped away by the coolant.
Note that there is no real time frame for the 3% loss. They simply take out the oldest rods and replace with new ones, similar ones. So it is 1-4 years.
Actually that is the most likely cause. The major contributor of radiation in a shutdown reactor is the from the activated corrosion and wear products - primarily cobalt 60.
It turns out most of the neutrons are needed. Some are lost in the light water, but if you lose too much, weak U does not keep reacting. http://wiki.answers.com/Q/What_is_the_purpose_of_using_heavy_water_in_nuclear_reactor So neutrons not likely to be the 1% yearly loss.