Preliminary reactor physics assessment of the HTR module with 14% enriched UCO fuel

Abstract

The high temperature reactor (HTR) Module (Lohnert, 1990) is a graphite-moderated, helium cooled pebble bed design that has been extensively used as a reference template for the former South African (Matzner, 2004) and current Chinese (Zhang et al., 2009) HTR programs. This design utilizes spherical fuel elements packed into a dynamic pebble bed, consisting of tri-structural isotropic (TRISO) coated uranium oxide (UO2) 500μm fuel kernels with a U-235 enrichment of 7.8% and a heavy metal loading of 7g per pebble. This fuel type was previously qualified for use in Germany for pebble bed HTRs, as well as undergoing re-qualification in South Africa for the PBMR project. It is also the fuel type being tested for use in the high temperature reactor (HTR-PM) under construction in China.In the United States, however, a different TRISO fuel form is the subject of a qualification program. The U.S. experience with HTRs has been focused upon the batch-fueled prismatic reactor in which TRISO particles are embedded in cylindrical compacts and stacked inside the graphite blocks which comprise the core. Under this type of operating regime, a smaller TRISO with a different composition and enrichment performs better than the fuel historically used in PBRs. Fuel kernels and compacting techniques more suited to prismatic core duty are currently being developed and qualified under the U.S. Department of Energy's Advanced Gas Reactor (AGR) fuel development program and in support of the Next Generation Nuclear Plant project.Interest in the pebble bed concept remains high, however, and a study was undertaken by the authors to assess the viability of using AGR fuel in a pebble bed reactor. Using the German HTR Module as the reference plant, key neutronic and thermal-hydraulic parameters were compared between the nominal design and one fueled with the fuel that is the focus of the AGR program.Seven cases, varying the total heavy metal loading and number of fuel passes, were analyzed, and recommendations are made for achieving a feasible UCO-fueled HTR Module design from a reactor physics point of view. The PEBBED-THERMIX code (Gougar et al., 2010b), developed specifically for the analysis of pebble bed HTRs, was used to compare the coupled neutronic and thermal fluid performance of the two fuel particle designs. For seven variations on the fuel loading scheme, comparisons were made of steady state core multiplication factor (k-eff) at a specified burnup, control rod worths, power and flux profiles, fuel and moderator temperatures, and power peaking factors. Also analyzed were the maximum fuel temperatures during a depressurized loss of forced cooling (DLOFC) event, as well as the reactivity behavior for a water/steam ingress scenario.The first three cases compared the HTR-Module fuel (Case 1) with fuel of the same UO2 TRISO particle design with enrichment increased to 14% (Case 2) and UCO TRISO fuel as described above. For these three cases the sphere heavy metal loading and average number of passes through the core remained constant at 7g and 15 passes. The analysis of the normal operation (steady state) equilibrium results for these three cases showed that the dominant contributor to the observed variances between the HTR Module UO2 and UCO cores is the increase in the U-235 enrichment to 14%, and not the additional moderation effects of the oxygen to carbon exchange.In addition to the steady-state analyses, two important design basis accidents were also included in this study. The first was an extreme loss of forced cooling accident caused by a large double-ended guillotine break. This event is the limiting case for the fuel temperatures. The second event was the ingress of water (in the form of steam) into the core, because of a hypothetical steam generator tube rupture. The use of UCO fuel at 14% enriched (Case 3) lead to a small increase of 48°C in the DLOFC peak fuel temperature to 1533°C. If typical uncertainty margins between 4% and 7% are taken into account, the maximum fuel temperatures are below 1660°C. However, it was also shown that only 4% of the 360,000 fuel spheres in the core have maximum temperatures above 1400°C. It was also confirmed that the DLOFC fuel temperature data for Cases 2 and 3 are essentially identical, confirming that the change in enrichment is the dominant driving factor in the differences observed between Cases 1 and 3.It was also found that the 7g heavy metal UCO fueled core is more reactive than the 7g heavy metal UO2 fueled core for a steam ingress event (1.29% for Case 1 and 1.56% for Case 3 at 660kg steam). However, the control rod shutdown worths for a full SCRAM were also compared for these two cases (5.16% and 4.30%, respectively), and it was concluded that an acceptable shutdown margin exists for both cases. It therefore seems feasible, from a water ingress point of view, to operate an HTR Module core design with UCO fuel enriched to 14% and loaded to 7g heavy metal. By comparing the results for Cases 1–3, it was shown that the higher enrichment plays the dominant role in the UCO fueled core's reactivity behavior, and not the change to UCO fuel kernels.The analysis of these two accidents, together with the acceptable results obtained from the steady-state and the control rod worth analysis, provided a preliminary insight into the behavior of the UCO-fueled HTR Module design. It is important to note that a significant analysis gap still exist in evaluating the control rod worths and water ingress effects at other core conditions, for example hot and cold shutdown conditions. A final statement on the feasibility of this fuel can only be made at that point.The fuel performance of this core design was beyond the scope of this study, and care should be taken not to equate acceptable neutronics and thermal fluid behavior with acceptable levels of fission product release rates, since many more factors are involved in this aspect of fuel design. Nonetheless, this preliminary analysis suggests that the fuel being qualified by DOE for deployment in prismatic HTRs can also be used in pebble bed reactors.