Since its discovery, the phenomenon of nuclear fission is the object of extensive theoretical and experimental studies. However, we are still far from a complete understanding of the fission process. Nuclear theory can satisfactorily explain the process of neutron-induced fission at thermal neutron energies, but it meets problems at high neutron energies. However, new applications are nowadays developed involving neutron-induced fission in this energy domain. An example of such an application is accelerator-driven systems (ADS) which are dedicated to transmutation of highly radioactive nuclear waste. Conceptual studies of ADS require new nuclear data on neutron-induced reactions within a wide incident energy range. Along with structural, spallation target and other materials, data on neutron-induced fission are especially required for two nuclides, 232Th and 238U. At present, however, there are no published neutron-induced fission yield data for either 232Th or 238U at energies above 20 MeV.
In this thesis, I present measurements of fission fragment mass yields at neutron energies from 10 to 60 MeV for 232Th and 238U. The experiment was done at the Louvain-la-Neuve quasi-monoenergetic neutron beam facility. A multi-section Frisch-gridded ionization chamber was used as the fission fragment detector. The fission fragment mass yields were measured at peak neutron energies of 33, 45, and 60 MeV. In addition, data for the neutron-energy intervals 9-11, 16-18, and 24-26 MeV were also extracted from the low-energy tail. The measurement results show that the symmetric fission component increases with incident neutron energy for both uranium and thorium, but it is more enhanced for thorium. The uranium results were compared to the only existing set of experimental data for neutron energies above 20 MeV. Reasonable agreement was found. However, our data show a lower symmetric fission component. For thorium, the present data are the first above 20 MeV.
Model calculations with the TALYS code have also been done. This code is based on the multi-modal random neck-rupture model extended for higher excitation energies. We included a phenomenological model into the code and achieved a good description of our experimental results.