Safety Analysis of the MIT Nuclear Reactor for Conversion to LEU Fuel Wilson and Dunn GTRI Program, Nuclear Engineering Division Argonne National Laboratory 9700 South Cass Ave, Argonne, IL 60439 USA Newton, Jr. and L. Hu MIT Nuclear Reactor Laboratory 138 Albany St, Cambridge, MA 02139 USA ABSTRACT The Massachusetts Institute of Technology Reactor, MITR, is a compact research reactor licensed for 6 MW operation using highly enriched uranium (HEU) fuel. As a part of HEU minimization in the Global Threat Reduction Initiative, a series of safety analyses have been carried out to support conversion of MITR to low enriched uranium (LEU) fuel. Conversion analyses using an ORIGEN-MCNP coupling code, MCODE, have been benchmarked to experimental data from historical (1975-1976), and a series of recent (2007-2009) MITR cores. These models have been used in an MITR fuel management system to introduce proposed LEU fuel with subsequent management of fuel movements in order to minimize power peaking. Specific three dimensional power shapes for each individual core state, element, and stripe within a plate were generated. Limiting power distributions were found to occur on plates of depleted elements when reshuffled adjacent to the heavy water reflector surrounding the core. Since six control blades separate the hexagonal MITR core from the reflector, limiting power distributions were found at the end of cycle due to control blade withdraw. Analyses estimate that 30% fewer elements will be needed to maintain operations at 7 MW for LEU compared with current HEU operations at 6 MW. These results have been used in steady-state thermal-hydraulic analysis with the result that the limiting safety system setting power based on the onset of nucleate boiling is MW. 1. Introduction The Massachusetts Institute of Technology Reactor (MITR) is a research reactor in Cambridge, Massachusetts designed primarily for experiments using neutron beam and in-core irradiation facilities. It delivers a neutron flux comparable to current LWR power reactors in a compact 6 MW core using highly enriched uranium (HEU) fuel. The submitted manuscript has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory ( Argonne ). Argonne, a Department of Energy Office of Science laboratory, is operated under Contract No. DE-AC02-06CH11357. The Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government. Work supported by US Department of Energy, Office of Global Threat Reduction, National Nuclear Security Administration (NNSA). In the framework of non-proliferation policies, the international community presently aims to minimize the amount of nuclear material available that could be used for nuclear weapons. In this geopolitical context, most research and test reactors both domestic and international have started a program of conversion to the use of Low Enriched Uranium (LEU) fuel. International efforts conceived during Atoms for Peace in 1955  began as a concerted program for Reduced Enrichment Research and Test Reactors (RERTR) starting in 1978 following the plan of Lanning (of MITR), et. al. [2,3]. Fuels allowing higher uranium densities have enabled many reactors to convert to LEU and maintain performance [4,5]. Currently, a new type of LEU fuel based on a high-density alloy of uranium and molybdenum (UMo) is undergoing qualification testing, and is expected to allow the conversion of domestic high performance reactors like the MITR-II reactor . The MITR core has a hexagonal design that contains twenty-seven fuel positions in three radial rings (A, B, and C), as shown in Figure 1. The core is light water moderated and cooled and is surrounded by a D2O reflector. Boron impregnated stainless steel control blades are located at the periphery of the core on each of the sides of the hexagon and have a total upward travel of 53 cm. In addition, fixed absorbers can be installed in the upper axial region of the core in a hexagonal configuration between the A and B rings as well as in three radial arms extending to the edge of the core. MITR reactor fuel loading is very flexible, and typically at least three of these positions (two in the A-ring) are filled with either an in-core experimental facility or a solid aluminum dummy element to reduce power peaking. The remaining positions are filled with rhomboid-shaped MITR-II fuel elements. Each fuel element can be rotated, flipped, stored and returned to the core as needed over the course of 2-3 years. 2. HEU and LEU Fuel Designs Each rhomboid-shaped fuel element contains fifteen aluminum-clad fuel plates using HEU (93% enriched) in an aluminide cermet matrix with a fuel thickness of mm and a fueled length of 57 cm. The cladding of each fuel plate is mm thick after machining mm x mm square fins along the length of the plate to increase heat transfer to the coolant. The fuel has an overall density of g/cm3, with a total loading of 508 g 235U in each element. An LEU core has been redesigned so that the 15-plate HEU element is replaced by an 18-plate LEU element (see Figure 2) with a mm U-10wt% Mo 17 g/cm3 monolithic (fuel Figure 1. Layout of the MIT Reactor core. Fuel meat width mm Side plate flat-to-flat mm Element end flat-to-flat mm Channel width mm Figure 2. Schematic of MITR LEU fuel element drawn with 18 finned plates. 60 Finned fuel plates (18) formed from a single piece of a uranium alloy casting) foil in mm clad with mm aluminum fins. Heat transfer area provided by the addition of plates is offset by decreasing the water gap between plates from mm to mm in order that the element and core may remain otherwise unchanged . 3. Neutronic Modeling Benchmarking of Model to Measured Data A number of neutronic models have been made for the MIT reactor. The Monte Carlo code MCNP has been used for many evaluations of HEU and LEU core and experiment designs. The basic reactor design and fuel structure has also been input into the MCNP-ORIGEN linkage code MCODE for fuel management and burnup evaluations . Capabilities to study core and experimental loadings are very important for the MITR reactor since fuel loading is very adaptable. MITR, which does not have a fixed core reloading pattern, uses flexible fuel management on an ongoing basis. In addition to the symmetric fuel element which can be rotated and/or flipped for efficient fuel utilization, MITR loads a variable core configuration in which an adjustable number of in-core experimental, dummy and fueled elements are arranged among the 27 core loading positions. Since there is no fixed core reloading, it is important to have versatile fuel management capabilities integrated with neutronic modeling to evaluate potential LEU, and analogous HEU, cores series as well as for ongoing fuel management. As part of the LEU conversion of MITR, neutronic models have been developed and benchmarked, with good agreement to experimental data at both beginning of cycle (BOC) and end of cycle (EOC). Historical and modern core configurations modeled are listed in Table 1. Table 2 lists modeled reactivity within 1% of predicted critical states across a wide variety of Table 1. MITR-II configurations of start-up and recent modern cores. fresh startup core physics testing in 1975-76, and twelve recent cores (BOC to EOC) operated in 2008-2009. Nearly fresh historical cores were not modeled at the EOC core state since they were either nearly fresh, or due to multiple configuration changes while the core fueling was being operated (such as the control system being changed one blade at a time during Core 2 from Cd to borated stainless steel). These model benchmarks, and others, showing agreement across many core configurations, and control blade states are documented in . Subsequent to benchmarking of the MCODE model, safety analyses and other information appropriate for conversion were prepared [10-13]. Integrated Fuel Management and Nodal Optimization To allow studies such as these, the graphical user interface (GUI) has been designed and built into the MCODE model in a version called MCODE-FM , which includes the ability to model all aspects of fuel management at the MIT reactor, including fuel flipping, rotating and storage for later use. A critical search algorithm was also utilized for criticality via control blade motion during depletion. The number of axial burnup nodes as well as fuel plate grouping can also be varied, depending on resolution and computational time needed, Table 3. As a part of nodal optimization of reactor analysis, automated nodalization of the MCODE-MCNP models has been fully implemented by Horelik  to allow the user to divide plates, or set of plates, into any combination of axial or lateral (stripe) independently depleting regions. This was used in a detailed study to explore the effect of depletion mesh discretization on simulation accuracy. A wide range of Table 2. Reactivity of historical startup and recent depleted MITR cores modeled at measured control positions. MITR HEU Core BOC keff EOC keff Deviation from Critical BOC Deviation from Critical EOC 1 - - 2 - - 4 - - 179 180 181 182 183 184 185 186 187 188 189 190 Average Std Dev. at 1- Table 3. Discretization of depletion and power regions to generate whole core 3-D power shapes in MITR HEU and LEU 24 element cores. Regions Geometry Nodal Optimization of Depletion Current LEU Safety Analyses  Depletion Power Shape Plate Division Each plate discrete 1 to 6 6 groups 18 Fuel Axial Division Continuous 6 to 32 6 18 Fuel Plate Lateral Division Continuous 1 to 4 1 4 Total per (LEU) Core - 100 ~ 10000 864 31104 depletion discretizations were analyzed. For safety analysis, reasonable accuracy was observed  with 6-8 axial depletion regions, since for a range of very similar shapes, it is the overall hot stripe (axially-averaged) that has been found to be an important consideration for thermal hydraulic margins . Section 4 will present overall hot stripe power distributions. Accounting for Transverse Heat Conduction using Lateral Power Stripe Size The lateral division of the stripe along the length of the fuel is important for analysis of reactor operation. As shown in Figure 3, the power distribution transverse to the direction of coolant flow must be sufficiently detailed to represent the edge of plate power peaking which occurs due to fuel self-shielding. Selection of the stripe size has important implications for the licensable reactor power. For the all-fresh LEU core, the 16 stripes has an approximately 20% higher hot stripe peak heat flux than the 4 stripe case. Since no mixing between stripes is conservatively, and realistically , assumed, using the 16-stripe case would reduce allowable limiting safety system settings (LSSS) power. However, if lateral heat conduction is taken into account, edge power peaking is mitigated since significant heat can be conducted into the unfueled sides of each plate, and removed by the coolant in the unfueled portion of the channel (next to the side plate). As shown in Figure 4 , although the 16-stripe peak heat flux was nearly 20% greater than the 4-stripe at the edge, the four cm stripe case without lateral heat conduction was conservative (lower peak clad temperature) versus sixteen cm stripes with lateral heat conduction. Prior analyses have reached similar conclusions . Figure 3. Power distribution transverse to coolant flow plotted for various stripe discretization of the hot plate (also containing the hot stripe) of an all-fresh LEU U-10Mo core. Stripe Peak Clad Temperature -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 40 45 50 55 60 65 70 75 80 85 90 95 100 105 16 Stripes, Lateral Conduction 16 Stripes, No Lateral Conduction 8 Stripes, Lateral Conduction 8 Stripes, No Lateral Conduction 4 Stripes, No Lateral Conduction 1 Stripe, No Lateral Conduction Inlet Temperature Figure 4. Peak clad temperature results from RELAP5-3D for various stripe discretization of coolant channel, both with and without lateral heat conduction . 4. LEU Core Performance Limiting LEU Power Distributions As was observed in the nodal studies discussed in Section , the depletion state of outer plates is important for MIT due to the flexible loading and rotation/flipping of elements. Limiting power is found in MITR, as shown in Figures 5-6, in the middle of life, when plates are first loaded into the C-ring alongside the heavy water reflector. The peak stripe factor of times the core average power is an increase of 63% compared to power in the same fuel plate when loaded fresh into the B-ring. In Figure 5 both the peak stripe and the stripe with peak spot for all cores were found in the same plate next to the reflector in Core 187. Since six control blades separate the hexagonal MITR core from the reflector, limiting power distributions were found at EOC due to control blade withdraw. MITR is an up-flow reactor, and so the inner ring peak stripe which was more top-peaked than the other stripes in Figure 5 was also analyzed. The most limiting stripe found was the peak stripe of Figure 5. Peak Clad Temperature ( C) Stripe Figure 5. Peak power locations of LEU fresh and depleted Cores 178-190 at BOC and EOC (plate orietation indicated with parrallel lines). 01020304050607080900 102030405060Distance from Bottom of Fuel (cm)Heat Flux (W/cm2 of foil) with Peak SpotPeak StripeInner Ring Peak Stripe Figure 6. Peak power locations of LEU fresh and depleted cores at BOC and EOC, for peak location, peak stripe (C-ring), and inner ring (A/B-rings) peak stripe. A-2A-1A-3B-1B-2B-3B-4C-1B-5B-6B-7B-8B-9C -3C-4C-5C-6C-7C-8C-9C-10C-11C-12C-13C-14 C-15C-2Blade 2Bla de1Bla de 5Bla de6Bla de4Bla de3Hot SpotHot StripeFresh, BOC179,180,183-85,189-90 EOC183-5,189-90BOC&EOC 181-2BOC&EOC 188Fresh BOC 179-180,185-6BOC 187 EOC179,180,187BOC&EOC186BOC&EOC 181-2BOC 187-8 EOC179,187BOC183-4,189-90 EOC180,183-6,189-90EOC188Core Structural Experimental Performance and Operation Limits Analyses with these tools show that HEU experimental performance will be maintained with the use of monolithic U10Mo LEU fuel and an upgrade in operating power from 6 MW to 7 MW, as listed in Table 4. MITR LEU core neutron flux can exceed cm-2 s-1 for in-core (> MeV), and 5x1013 cm-2 s-1 for ex-core experiments (< eV). Analyses also estimate that 30% fewer elements will be needed to maintain operations at 7 MW for LEU compared with current HEU operations at 6 MW. These results have been used in subsequent steady-state thermal-hydraulic analysis with the result that the limiting safety system setting power based on the onset of nucleate boiling is MW . Table 4. LEU performance at 7 MW vs. HEU counterpart core at 6 MW. LEU Core In-Core Irradiation (A2) Twelve-inch Beam Port Two-inch Pneumatic Facility Fission Converter Window Below-core Thermal Beam Facility Energy > MeV < eV < eV < eV < eV LEU Fresh 22 Element 109% 103% 103% 102% 102% LEU 185 BOC 107% 103% 104% 104% 103% LEU 185 EOC 107% 101% 104% 104% 101% LEU 186 BOC 107% 103% 106% 106% 103% LEU 186 EOC 108% 103% 106% 107% 104% LEU 187 BOC 108% 103% 106% 105% 103% LEU 187 EOC 107% 102% 105% 106% 103% LEU 188 BOC 107% 103% 105% 104% 103% LEU 188 EOC 106% 101% 103% 103% 100% LEU 189 BOC 107% 104% 106% 105% 104% LEU 189 EOC 106% 102% 104% 104% 102% LEU 190 BOC 107% 103% 105% 106% 103% LEU 190 EOC 106% 101% 103% 104% 101% 5. Conclusions and Future Work MCODE has been used to determine power peaking in proposed LEU cores. These power distributions were used to determine the thermal-hydraulic safety limits. In this determination, engineering hot channel factors (EHCFs) were modelled as distributions with both measurement and calculation uncertainties quantified in thermal hydraulics limits analysis using statistical propagation. In addition, the interior coolant channel (full channel) is assumed more limiting than all possible end channel configurations. Uncertainties in the end channel significantly impact whether interior or end channels are limiting, and so further analysis will be required as manufacturing uncertainties in the channel gap become available for the LEU fuel element. Alternate LEU designs are also being investigated to improve fabricability. Current HEU plates, and LEU designs to date, have been finned with 10 mil deep grooves along the length of the coolant channel to increase heat transfer. However, plate fabrication may be improved by decreasing the groove depth to allow increased cladding thickness (and tolerances) under the fins. In order to offset the loss of heat transfer in a design with less fin area, thinner foils are being studied in order to optimize power peaking. These would be incorporated in outer plates of elements since, in the designs to date, the highest power peaking has been found in the plate of depleted elements loaded adjacent to the heavy water reflector. Acknowledgements The authors would like to acknowledge of the many contributions talented students and staff at MITR and ANL to this effort. The contributions of Nick Horelik are especially acknowledged for his detailed efforts in MCODE-FM development which were utilized in this work. References  A. Glaser, About the Enrichment Limit for Research Reactor Conversion: Why 20%?, Proc. RERTR 2005, Boston, MA, Nov. 6-10, 2005.  D. Lanning et. al., Proposed National Plan for Development of High-Uranium-Density Research and Test Reactor Fuel to Accommodate Use of LEU, RSS-TM-7, Argonne National Laboratory (1977).  Proc of 1st RERTR, Argonne, IL, Nov. 9-10, 1978, ANL/RERTR/TM-1, Argonne National Laboratory.  NRC, Safety Evaluation Report related to the Evaluation of Low-Enriched Uranium Silicide-Aluminum Dispersion Fuel for Use in Non-Power Reactors, NUREG-1313 (1988).  J. Snelgrove, et al., Development of Very-High-Density Fuels by the RERTR Program, Proc. RERTR 1996, Seoul, Korea, Oct. 7-10 1996.  Meyer, et. al., Progress in the Development of Very High Density Low Enrichment Research Reactor Fuels, Trans. European Research Reactor Conf. (RRFM), Prague, Czech Republic, March, 2012.  T. Newton, et. al., Completion of Feasibility Studies on Using LEU Fuel in the MIT Reactor, Proc. RERTR 2009, Beijing, China, Nov. 1-5, 2009.  Z. Xu, P. Hejzlar, and M. Kazimi, MCODE, Version - An MCNP-ORIGEN Depletion Program, Massachusetts Institute of Technology (2006).  E. Wilson, et. al., Comparison and Validation of HEU and LEU Modeling Results to HEU Experimental Benchmark Data for the Massachusetts Institute of Technology MITR Reactor, ANL/RERTR/TM-10-41, Argonne National Laboratory (2010).  Newton, Jr., L-W Hu, B. Forget, Horelik, Connaway, T. Gerrity, Plumer, Wilson, and Stevens, Fuel Management and Safety Analysis for LEU Conversion of the MIT Nuclear Reactor, Proceedings of the RERTR Conference, Santiago, Chile, October, 2011.  Wilson, Horelik, A. Bergeron, Newton, Jr., F. Dunn, L. Hu, Stevens, Neutronic Modeling of the MIT Reactor LEU Conversion, Trans. Am. Nuclear Soc. 106, 1, 849-852 (2012).  Wilson, N. Horelik, Dunn, Newton, Jr., L. Hu, and Stevens, Power distributions in Fresh and Depleted LEU and HEU Cores of the MITR Reactor, ANL-RERTR-TM-12-3 Revision 0, Argonne National Laboratory, February 2012.  Wilson, Newton, Jr., F. Dunn, L. Hu, Stevens, Conceptual Design Parameters for MITR LEU U-Mo Fuel Conversion Demonstration Experimental Irradiations, ANL-RERTR-TM-12-28, Argonne National Laboratory June 2012.  P. K. Romano, B. Forget and T. H. Newton, Jr., Development of a Graphical User Interface for In-core Fuel Management Using MCODE, Proceedings of the Conference on Advances in Nuclear Fuel Management, Hilton Head, South Carolina, April, 2009.  N. Horelik, Expanding and Optimizing Fuel Management and Data Analysis Capabilities of MCODE-FM in Support of MIT Research Reactor LEU Conversion, MS thesis, Massachusetts Institute of Technology, December, 2011.  E. Wilson, T. Newton, Jr., N. Horelik, T. Gerrity, H. Connaway, and B. Forget, Neutronic Analysis Capabilities for Conversion of the MIT Reactor to LEU Fuel, Proceedings of the RERTR Conference, Warsaw, Poland, October, 2012.  L. Hu, K. Chiang, Wilson, Dunn, Newton, Jr., and Stevens, Thermal Hydraulic Limits Analysis for LEU Fuel Conversion of the MIT Reactor, MITNRL-12-01, Massachusetts Institute of Technology, March 2012.  B. Dionne, J. Stevens, S. Kalcheva, and E. Konen, Applicability of a Hot Channel-Based Hot Stripe Approach to Model the Azimuthal Power Peaking in a BR2 Fuel Assembly , Proceedings of the RERTR Conference, Santiago, Chile, October 2011.  Cheng, Heat Conduction in a NBSR Fuel Plate - Effect on Wall Heat Flux, Brookhaven National Laboratory Memo, April 6, 2010.  T. Newton, Jr., G. Kohse, M. Ames, Y. Ostrovsky, and L. Hu E. Wilson, F. Dunn and J. Stevens, Activities in Support of Conversion of the MIT Nuclear Reactor to LEU Fuel, Proceedings of the RERTR Conference, Warsaw, Poland, October, 2012.  H. Connaway, Development of a Core Design Optimization Tool and Analysis in Support of the Planned LEU Conversion of the MIT Research Reactor (MITR-II), MS thesis, Massachusetts Institute of Technology, September, 2012.
Safety Analysis of the MIT Nuclear Reactor for …
Safety Analysis of the MIT Nuclear Reactor for Conversion to LEU Fuel E.H. Wilson and F.E. Dunn GTRI Program, Nuclear Engineering Division Argonne National Laboratory
Link to this page:
Please notify us if you found a problem with this document:
Related search queries
Massachusetts Energy Storage Initiative, STATE OF CHARGE Massachusetts Energy Storage Initiative, State of Charge, Hydrogen: A Future Energy Carrier, Schlumberger, QUALITY MANAGEMENT FOR HEALTH CARE, QUALITY MANAGEMENT FOR HEALTH CARE DELIVERY, New Construction Reference Guide, Technological Tipping Points, Realizing the Potential of Blockchain, Quantum machine learning, UNITED STATES COURT OF APPEALS FOR, UNITED STATES COURT OF APPEALS