Slow Reactivity Insertion in Subcritical Reactors. Application to ADS

Abstract

As there is a need to deal with the problem of radioactive waste of nuclear plants, prototypes of subcritical reactors are being developed at the moment. They are known as ADS and they have as their main goal, the actinides transmutation. The neutron source, which keeps the reactions in a subcritical reactor, is very intense, that is why a high energy proton accelerator is required. For many applications, it is important to design methods to measure the intensity of the source. ADS are expected to work in a subcritical range between 0.92< keff <0.97 and because of that, control rods have a limited use and many designers are considering not to include them. They can be substituted by reflector or combustible elements displacements. In order to calibrate in reactivity these displacements, one can either take advantage of the pulsing structure of the neutron source in microseconds time scale, or can design methods for the minute time scale wherein the source is constant, that is the objective we want to achieve here. If the reactivity control is made with the reflector, in between the extreme states of open reflector and closed reflector, both keff are calculated and a reactivity ramp is estimated appropriate to the reflector movement velocity. This calibration should be made empirically by measuring the response to the ramp with neutron detectors. In the response to the ramp, the specific parameters are: the source intensity, the initial keff and the ramp slope, which defines the objective to be reached in the calibration process. Measurements are available from experimental subcritical reactors in the operation range of the ADS, which allow getting ready this calibration method based on the reactivity ramps. However, in order to fit the measurements to a theoretical model, the point kinetics equations need to be solved for this case. With the Prompt-Jump approximation and the consideration of slow ramp compared to the time constants of delayed neutrons, the integration of these equations and the non-linear fit of the measurements to the resulting method are detailed. The static approximation is enough when the reactor is in a typical subcritical state. However, when close to a critical state, the reactivity is no longer a ramp but a parabola which lead us to estimate the neutron source by means of a non-linear fit of the experimental values and assuming a known reactivity initial value.