DOE OSTI2012
This report presents a neutronics analysis of tight-pitch D 2 O-cooled PWRs loaded with MOX fuel and focuses essentially on the Pu breeding potential of such reactors as well as on an important safety parameter, the void coefficient, which has to be negative. It is well known that fast reactors have a better neutron economy and are better suited than thermal reactors to breed fissile material from neutron capture in fertile material. Such fast reactors (e.g. sodium-cooled reactors) usually rely on technologies that are very different from those of existing water-cooled reactors and are probably more expensive. This report investigates another possibility to obtain a fast neutron reactor while still relying mostly on a PWR technology by: (1) Tightening the lattice pitch to reduce the water-to-fuel volume ratio compared to that of a standard PWR. Water-to-fuel volume ratios of between 0.45 and 1 have been considered in this study while a value of about 2 is typical of standard PWRs, (2) Using D 2 O instead of H 2 O as a coolant. Indeed, because of its different neutron physics properties, the use of D 2 O hardens the neutron spectrum to an extent impossible with H 2 O when used in a tight-pitch lattice. The neutron spectra thus obtained are not as fast as those in sodium-cooled reactor but they can still be characterized as fast compared to that of standard PWR neutron spectra. In the phase space investigated in this study we did not find any configurations that would have, at the same time, a positive Pu mass balance (more Pu at the end than at the beginning of the irradiation) and a negative void coefficient. At this stage, the use of radial blankets has only been briefly addressed whereas the impact of axial blankets has been well defined. For example, with a D 2 O-to-fuel volume ratio of 0.45 and a core driver height of about 60 cm, the fissile Pu mass balance between the fresh fuel and the irradiated fuel (50 GWd/t) would be about -7.5% (i.e. there are 7.5% fewer fissile Pu isotopes at the end than at the beginning of the irradiation) and the void coefficient would be negative. The addition of 1 cm of U-238 blanket at the top and bottom of the fuel would bring the fissile Pu mass balance from -7.5% to -6.5% but would also impact the void coefficient in the wrong way. In fact, it turns out that the void coefficient is so sensitive to the presence of axial blanket that it limits its size to only a few cm for driver fuel height of about 50-60 cm. For reference, the fissile Pu mass balance is about -35% in a standard PWR MOX fuel such as those used in France. In order to reduce the fissile Pu deficit and potentially reach a true breeding regime (i.e. a positive Pu mass balance), it would be necessary to make extensive use of radial blankets, both internal and external. Even though this was not addressed in detail here, it is reasonable to believe that at least as much U-238 blanket subassemblies as MOX driver fuel subassemblies would be necessary to breed enough Pu to compensate for the Pu deficit in the driver fuel. Hence, whereas a relatively simple D 2 O-cooled PWR core design makes it possible to obtain a near-breeder core, it may be necessary to more than double the mass of heavy metal in the core as well as the mass of heavy metal to reprocess per unit of energy produced in order to breed the few percents of Pu missing to reach a true breeding regime. It may be interesting to quantify these aspects further in the future.
11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗