Perhaps one can quickly get a rough order of magnitude...
Assume that you get the energy out of the reaction from the neutrons. Two popular reactions are DT with 14 MeV and DD with 2.5 MeV (2 sig figs...)
1 MeV = 1.6E-13 Joule... so each 14 MeV neut is worth 22.4E-13 Joule
A typical fission plant (e.g. San Onofre) is around 1000 MWe (electric power out). Typical thermal efficiency is around 30%, so you've got to extract roughly 3000 MW from your reaction.
1 year is roughly 8800 hours or 31.5E6 seconds, so in a year, you're going to need to generate 31.5E6*3E9 joules, or, roughly 100E15 joules. How many neutrons is that.. assuming those high energy 14 MeV neuts, 100E15/(22.4E-13), or 5E28 neutrons. Converting to moles (5E28/6E23) is approximately 1E5 moles...
Now, let's assume that we're talking about using some metal as an absorber, and that 1% of the metal atoms get turned radioactive by neutrons (that is, we need 1E7 moles of metal). (I don't know if this is reasonable.. someone correct me please).
Taking Iron, for example, 1 mole of iron weighs about 56 grams, so 1E7 moles is 5E8 grams, or 5E5 kilos, or 500 tonnes...
Clearly, anything that generates gigawatts of electrical power is going to generate copious radioactive waste. There have been some clever approaches proposed for separating out radioactive daughter products from the general mass (reducing the amount of total waste that has hot components). For instance, if the absorber were Aluminum (I don't know that this would work), you transmute some to Magnesium, which could conceivably be separated chemically, leaving non-radioactive aluminum.
Mind you, this requires separation technology that is very efficient, and works in a high radiation environment, which is no easy feat.
Also, the fusion reactor vessel itself is likely to become radioactive by activation. True, of relatively short life compared to Pu, but still a problem. Even if you only produce a small amount (percentage wise) of very hot, short lived isotopes (like Sr90 or Co60), because of the large volumes involved in any gigawatt scale machine, you're still talking about significant masses.
Created on Wednesday, March 21, 2001 7:39 PM EDT by James Lux