ML24319A262
| ML24319A262 | |
| Person / Time | |
|---|---|
| Issue date: | 11/21/2024 |
| From: | Hossein Esmaili NRC/RES/DSA/FSCB |
| To: | Scott Krepel Office of Nuclear Reactor Regulation |
| References | |
| Download: ML24319A262 (1) | |
Text
MEMORANDUM TO:
Scott Krepel, Branch Chief Division of Safety Systems Office of Nuclear Reactor Regulation Nuclear Methods and Fuel Analysis FROM:
Hossein Esmaili, Branch Chief Division of Systems Analysis Office of Nuclear Regulatory Research Fuel and Source Term Code Development
SUBJECT:
EVALUATION OF RE-CRITICALITY WITHIN THE LOWER PLENUM DUE TO FUEL FRAGMENTATION, RELOCATION AND DISPERSAL To support development of regulatory guidance for addressing high burnup fuel dispersal and its consequences, staff from the Fuel and Source Term Code Development Branch (FSCB) have performed evaluations and analyses to demonstrate that the risk of re-criticality within the reactor coolant system due to fuel fragmentation, relocation and dispersal (FFRD) is negligible.
For this task, FSCB staff created a simplified model of the system using the Criticality Safety Analysis Sequence (CSAS) from Oak Ridge National Laboratorys SCALE neutronics code package to perform the criticality computations, with SCALE/ORIGAMI being used to determine the isotopic compositions of the fuel fragments. The analysis of the system determined that, even under conservative particle packing fractions and fluid conditions, the effective neutron multiplication factor remained substantially less than one. Therefore, re-criticality is not considered to be of concern following such a scenario.
Contact:
Andrew Bielen, RES/DSA 301-415-2366 November 21, 2024 Signed by Esmaili, Hossein on 11/21/24
S. Krepel Discussion The potential for FFRD during certain transient scenarios has introduced additional considerations that must be accounted for to demonstrate that the reactor system and fuel remains in a safe state following upset conditions. One such concern is that of local re-criticality:
if sufficient fuel mass was to collect within an ex-core region of the reactor coolant system (RCS), combined with sufficient water such that advantageous conditions for a nuclear chain reaction were achieved, then this subsystem could potentially become critical. This is an undesirable configuration because there could be significant additional fission energy generation within a portion of the system that is not designed to sustain such a condition, and resulting local temperature conditions could result in unexpected failure of components relied upon to maintain coolable geometry.
There are several locations within the RCS which might be collection points for fuel fragments released due to FFRD. This work focused upon the lower plenum of the reactor vessel as the most limiting location from a re-criticality standpoint. This is because the potential combination of fuel mass, range of potential fuel/water relative packing fractions, and leakage considerations (i.e., the surface-to-volume ratio of the ex-core fueled region, presence of steel adjacent to the fuel pile to minimize neutron leakage) make this region particularly susceptible to re-criticality.
Given the uncertainty in the specific geometry, packing fraction, and temperature conditions that might be realized in the fuel pile following FFRD, a range of fuel/water packing fractions, pile depths, and temperature conditions were considered to fully bound possible post-transient configurations.
System Model The computations were completed using a simplified Westinghouse 4-Loop lower plenum model1 in CSAS-Shift run in continuous energy mode, with isotopic composition of the fuel fragments determined using SCALE/ORIGAMI[1]. Shift is SCALEs modern, massively parallel Monte Carlo particle transport code. The ORIGAMI depletion was completed using 8% enriched UO2 which was burned to 55 GWd/MTU prior to the assumed LBLOCA and subsequent FFRD, with no cooling period. This burnup level is the threshold at which FFRD becomes a concern.
From a reactivity standpoint, fuel burned higher than this will have a lower fissile content and thus be less reactive. Therefore, it is conservative to evaluate all fuel at this burnup level. An actual core would have a mix of burned fuel in the pile, depending on the burnups of the fuel assemblies in the core prior to transient initiation. OBIWAN was used to pull out the burnt fuels specific atomic densities per isotope to be used directly within the criticality computations.
These atomic compositions account for solid fuel components, while noble gases are removed to reflect that they would not be present in the fuel pile. In particular, no reactivity credit for absorption in Xenon-135 is taken within the burned fuel composition.
Particle Model The particulate fuel within the model was represented using a homogenous mixture of non-borated water and fuel at specified packing fractions as opposed to random fuel particles to reduce computational time. This water-fuel composite approach was justified by comparing unit cell cases of a 2mm spherical particle suspended in a 2mm cube of water with a 2mm cube of 1 Other lower plenum designs could result in minor changes in the calculated keff. However, as noted below, the calculated keff is much less than 1 for even very large dispersed fuel masses, so the conclusions in this analysis are valid for all existing PWR and BWR designs.
S. Krepel mixed composition folding the water and burnt fuel into one material. The primary concern in smearing the individual particle component of the particle model into a homogeneous mixture was the potential that not explicitly modeling spatial self-shielding would have a large impact on keff. However, the difference from the particle-based model to the homogenous mixture model was at most 20 pcm, with differences becoming less apparent at higher volume fractions of fuel.
Given that a simple BCC packing fraction offers around 68%, it was therefore determined that a mixture-based fuel-water composite was justified for this model. Further details are seen in Table I of the Results discussion.
Variations on Volume Fractions within the Unit Cell composition.
Table I Unit Cell vs Mixture Justification Volume Fraction (H2O)
Keff Unit Cell Keff Mixture Difference (pcm) 0.75 0.012477 +/-
0.000138 0.012272 +/-
0.000214 20 0.50 0.109336 +/-
0.000119 0.109149 +/-
0.000276 18 0.25 0.111785 +/-
0.000244 0.111665 +/-
0.000155 12 Plenum Model The simplified model for the Westinghouse 4-Loop lower plenum was made using a hemispherical steel shell structure to represent the lower plenum, which was then filled with a fuel-water mixture, with non-borated water placed on top. The steel containment is 1.93 meters tall with a diameter of approximately 4 meters and a thickness of 10 centimeters. Non-borated water was used to offer a worst possible scenario and provide more conclusive support that FFRD would not cause concerns of re-criticality within the lower plenum. Vacuum conditions were assumed outside of the plenum, as leaked neutrons would not contribute to criticality.
Additionally, to guarantee the system was as conservative as possible, a temperature study was performed to confirm that the worst-case scenario would be at room temperature (20 C). This provided additional assurance that conservatism was prevalent throughout the computations.
S. Krepel Figure I: CSAS-Shift model of lower plenum/fuel mixture (grey-steel / blue-non-borated water /
green-Fuel/water mixture)
Results The simulation results demonstrated that criticality within the lower plenum was not possible given the assumptions made within the model. Variations to the volume fraction, and pile height within the lower plenum were made to determine limiting factors for criticality within the model.
Table II VF Variation with Bed Depth of 25 cm Volume Fraction (H2O)
Keff Fuel Mass MTU 0.2 0.714194
+/- 0.000119 2.6 0.3 0.739842 +/-
0.000197 2.3 0.4 0.782071 +/-
0.000191 2.1 0.5 0.807745 +/-
0.000220 1.8 Table III Bed Depth Variation with VF of 32% H2O Pile Depth (cm)
Keff Fuel Mass MTU 20 0.725009 +/-
0.000229 1.9 30 0.773866 +/-
0.000261 3.5 40 0.800106 +/-
0.000338 5.2 50 0.820710 +/-
0.000286 7.0
S. Krepel Table IV Temperature Study confirmations Temp C Keff Fuel Mass MTU 20 0.792961 +/-
0.000312 2.2 60 0.772808 +/-
0.000159 2.2 100 0.765358 +/-
0.000214 2.2 135 0.757625 +/-
0.000273 2.2 As shown throughout the tables above, all investigated combinations of bed depth, packing fraction, and system temperature indicate substantial margin to re-criticality. The most prevalent determining factor for k-eff to reach a value of one is the volume fraction of water to fuel within the mixture, but given the density and mass of fuel fragments, volume fractions of water high enough to reach criticality, such as 75 % water, are highly improbable in the system. Increases in the depth of the pile also reach unrealistic conditions rather quickly, as a bed depth of 50 cm and packing fraction of 68% fuel (32% water) equates to a heavy metal dispersal mass of 7 MTU. This is substantially higher than the most conservative published dispersed fuel mass in this reactor system available (e.g., ML21197A069, ML23116A214). Given these physical limitations as well as the limiting conditions of the model such as 55 GWd burnup, these studies imply that re-criticality within the lower plenum region of the reactor vessel is not possible based on our current understanding of FFRD.
S. Krepel
SUBJECT:
COMPUTATIONS TO DETERMINE POTENTIAL RE-CRITICALITY IN THE CASEOF FFRD WITH PILING WITHIN THE LOWER PLENUM DATED: NOVEMBER 21, 2024 DISTRIBUTION:
A. Bielen, RES J. Corson, RES S. Muller, RES H. Esmaili, RES C. Van Wert, NRR S. Krepel, NRR K. Webber, RES ADAMS Accession Number: Memo; ML24319A262 OFFICIAL RECORD COPY OFFICE RES/DSA RES/DSA RES/DSA RES/DSA NAME JCorson SMuller ABielen HEsmaili DATE 11/18/2024 11/18/2024 11/18/2024 11/21/2024