Design and Analysis of Sodium-cooled Fast Reactor Systems for Nuclear Waste Minimization

Ching-Sheng Lin, Purdue University

Abstract

The primary objective of this study was to develop fuel cycle options that minimize the nuclear waste generation using sodium-cooled fast reactor systems. In addition, this study aims to develop a new method that improves the accuracy in analyzing one of the developed options that the moderated target assembly was loaded in the fast reactor core. This document summarizes the objectives and lines of work focused on the design studies and performance analyses of transuranic (TRU) elements transmutation in fast reactor systems, the development of new depletion method, and the assessment of how well the design meets the objectives. Two “two-stage” fast spectrum fuel cycle options (Options 1 and 2) were proposed. The first option is a two-stage fuel cycle option of continuous recycle of Pu in sodium-cooled fast reactor (SFR) and subsequent burning of minor actinides (MAs) in an accelerator driven system (ADS). The second option is a two-stage SFR/ADS fuel cycle option with moderated target assemblies employed in SFR to reduce the amount of MAs to be sent to the second-stage ADS. Design studies were performed to develop reference designs for the SFR core, an ADS blanket, and a moderated target assembly. The fuel cycle performance was evaluated based on the mass flow data for a nuclear fleet of 100 GWe-yr electricity production. For comparison, a single-stage SFR fuel cycle with homogeneous recycling of TRU was evaluated as well (Option 3). For the fuel cycle Option 2, the less energetic neutrons in the moderated target assembly enhance the transmutation performance. It is found that the use of moderated target assemblies in SFR reduces the number of required second-stage ADS by a factor of six without deteriorating safety characteristics. However, the reduced mean free path of neutrons in moderated target assembly presents a potential challenge to the conventional homogenized method used in fast reactor neutronics analyses. Thus, a pin depletion method based on VARIANT transport solutions was developed and implemented in a computer code named VAREPD to exam the fuel inventory change inside the moderated target assembly. The verification test result suggests that VAREPD calculation can accurately retrieve the nuclide density distribution inside the moderated target assembly. The root-mean-square (RMS) error in the nuclide densities at the end of cycle is 5.5% for Np-237, 3.6% for Pu-238, 4.4% for Pu-239, 3.0% for Cm-242 and 4.1% for Cm-244. The corresponding two-sigma uncertainties of the reference Monte Carlo solution obtained with the SERPENT code are 0.6%, 1.2%, 1.8%, 1.0% and 4.4%, respectively. It is also found that both the VAREPD and the homogenized assembly calculation of REBUS-3 yield reasonably accurate assembly-averaged nuclide densities. It is concluded that the burnup calculation with homogenized assembly models provides satisfactory mass flow data for the fuel cycle analysis but the new pin depletion method would be required in the design optimization and the post irradiation examination, in which an accurate assessment of temperature and fluence distributions is important. In summary, analysis results showed that all three proposed fuel cycle options could achieve high reduction in the nuclear waste generation because of the continuous recycle of the Pu and MAs. The SFR in the Options 1 and 2 may have a potential benefit from the fuel fabrication and reprocessing points of view. On the other hand, Option 3 with homogeneous recycle of TRU in SFR may have the economic advantage over Options 1 and 2. Moreover, co-extraction of Pu and MAs would reduce the proliferation risk.

Degree

Ph.D.

Advisors

Yang, Purdue University.

Subject Area

Nuclear engineering

COinS