It has been an attractive idea for some time to reduce the long-term potential hazard of the waste by chemical removal of the actinides and subsequent transmutation in a neutron flux. The overall incentive for actinide partitioning is not very great. The reduction of the ingestion hazard after recycling equilibrium has been reached will be only modest, and the technical effort will be enormous. The technology for actinide partitioning is not available as yet, and considerable development will be required to make it available. Moreover, it has to be considered that part of the actinides are transferred from the waste to the fuel cycle on recycling, where they may create an even greater hazard than in the waste.

The overall effect of actinide partitioning depends not only on the degree of chemical separation but also on the efficiency of transmutation. At present transmutation would have to be performed by recycling the separated actinides to LWRs, where it will be less effective than in a LMFBR. The reduction of the potential hazard achieved by actinide removal will decrease with repeated recycling of these actinides as a result of the buildup of the higher actinides and will eventually attain an equilibrium value. Figure 11.17 is a plot of equilibrium hazard index reduction factors in LWR uranium waste versus age of the waste for 99.5 and 99.9 percent chemical separation efficiency. Between 100 and 50,000 years, reduction factors are found of not more than 5 and 30, respectively [С2]. Any actinide separation higher than 99.9 percent makes it necessary to consider 99 Tc as well and seems out of reach, as presently nothing even close to 99.9 percent is technically feasible.

Figure 11.17 reflects the effect that actinide partitioning and transmutation has on the actin­ide hazard index of only the HLW itself. If the total quantity of actinides accumulated in the HLW and in the fuel cycle is considered, the same equilibrium reduction factor will eventually be at­tained provided that a constant nuclear power level is assumed, but it will take a very long time. In the fuel-cycle study performed for the American Physical Society [P2], an example with recycling the actinides to a LMFBR has been calculated that is shown in Fig. 11.18.

It should also be obvious that actinide reduction in HLW is reasonable only if an equivalent reduction of actinides in non-high-level waste, such as refabrication waste, can also be achieved. Also, 129 I must be considered in a long-term hazard balance.

Chemical separation. Current concepts for high-efficiency separation of actinides call for improved plutonium recovery, coextraction of uranium and neptunium with subsequent partitioning by valence control, and extraction of amercium and curium from the HAW stream. There are a number of major problems to be solved before a technically feasible process will be available.

Actinide losses to undissolved residues of fuel and to solids generated in the process have to be eliminated. To improve the recovery of plutonium, inextractable forms have to be identified and means have to be found to recover them.

For the recovery of americium and curium from the waste stream, cation-exchange and extraction processes appear most promising. The outstanding problem is a highly effective separation of actinides from lanthanides. The latter would be harmful upon transmutation in thermal reactors because of the high-neutron-capture cross sections of some of them. An actinide/lanthanide fraction would probably have to be separated first from the other fission products and waste components and then the actinides would have to be recovered with high purity. Also, by taking into account that substantial additional waste streams would have to be managed without significantly increasing the overall waste quantity, it is obvious that the recovery of americium and curium will be the most difficult task in waste partitioning [B5].

Transmutation. Recycling actinides to the LWRs will decrease the average material neutron multiplication factor by only 0.8 percent, provided that they are of high purity [C2]. Recycling to LMFBRs, however, will be preferred. There will be less neutron capture in impurities, such as lanthanides, and the average fission-to-capture ratio of the actinides should be higher in a fast spectrum than in a thermal one.

Recycling of actinide waste will increase radiation problems associated with processing of fuel. After a few cycles, for example, 552 Cf builds up to the strongest neutron source and reaches 1012 njs per MT of heavy metal at 150 days after discharge.

Figure 11.19 is a schematic flow sheet for actinide recycling.