LOCA ANL Cladding Burst Tests

Rod Design Specifications

Of the series of LOCA burst tests that were conducted at ANL, three were isolated for BISON simulation. These rodlets were not irradiated and are tested out of a hot cell, known from now as OCL. The geometry of the cladding used in this experiment can be seen in the Table 1

Operating Conditions and Irradiation History

The OCL series simulated consisted of three rods identified as OCL5, OCL8 and OCL11. All three of this rodlets were made from non-irradiated archival Zircaloy-2 cladding. These rodlets were loaded with zirconia pellets to simulate fuel conductivity. These pellets had an outer diameter of 4.78mm which allowed for a loose fitting 100 cold radial gap (Monteleone, 2003) (Snell, 2005) (Billone et al., 2008).

The testing apparatus, which can be seen below in Figure 1, holds the specimen in a vertical configuration while it is heated with four radiant bulbs that provide even heating for 270 mm axially. The cladding interior is pressurized with helium to 8.28 MPa and held near there during the test with a high pressure inlet and upper and lower pressure transducers. The testing apparatus was designed such that the atmosphere external to the cladding could be controlled. The testing configurations were an argon environment, a high temperature steam environment and a cold water quench.

For experiments OCL5 and OCL8 the temperature was controlled via the radiant lights and started with a rise to about 570 K followed by a hold time of about 25 minutes. At the end of the hold the temperature was raised by 5 K per second until burst, at which time the rodlet was cooled back to room temperature at 3 K per second. Both OCL5 and OCL8 were in a argon atmosphere. Experiment OCL11 was heated in the same manner but after burst the rod was heated to about 1470 K and held of 5 minutes before being cooled at 3 K per second. This experiment was conducted in a stream environment.

A graphical depiction of the temperature profile can be seen below in Figure 2. The plot was left incomplete intentionally as the final cooling could be slow or a fast quench.

Figure 1: Diagram of the ANL LOCA burst testing apparatus c/o (Billone et al., 2008)

Figure 2: Experiment temperature profile.

Geometry and Mesh

Cladding was simulated with a mesh of the following characteristics:

• 2 higher order elements radially

• 257 higher order elements axially

These numbers came from the "coarse" option in the FuelPinMeshGenerator. The "medium" option was also simulated but did not affect the results so coarse was used to conserve runtime.

Material and Behavioral Models

The following material and behavioral models were used in these simulations:

• ZryThermal: Thermal properties for Zircalloy from MATPRO

• ZryCreepLOCAUpdate and ZryElasticityTensor: Calculates the Erbacher secondary thermal creep under loss-of-coolant accident conditions, the Limback-Andersson primary thermal creep, and the Hoope irradiation creep for Zircaloy cladding

• ZryThermalExpansionMATPROEigenstrain: thermal expansion of Zircaloy with the MATPRO model

• ZrPhase: relative amounts of the and material phases present in the zircaloy

• ZryCladdingFailure: Models the failure of Zircaloy-4 cladding due to burst under LOCA conditions, combined failure criterion used

The fuel was modeled as a simple isotropic material, using the following material and behavioral models:

• HeatConductionMaterial: isotropic thermal properties

• ComputeFiniteStrainElasticStress and UO2ElasticityTensor: elastic mechanical behavior

• ComputeThermalExpansionEigenstrain: thermal expansion with a constant instanteous thermal expansion coefficient

Details and references for all of these models listed above can be found on the linked BISON documentation pages.

Boundary and Operating Conditions

BISON inputs for the ANL experiments fell into two basic categories; unirradiated cladding burst in an argon environment and unirradiated cladding burst in a steam environment. The BISON LOCA simulations were all driven with Dirichlet boundary conditions for the cladding interior pressure and cladding exterior temperature profile. The cladding interior was held at 8.23 MPa with helium while the cladding surface was heated following the profile mentioned before (Figure 2). Initial runs had a flat axial temperature profile as the report claimed that the heat was provided in that manner. Unfortunately this resulted in a heavily ballooned cladding that did not actually burst. It was decided that a slightly peaked axial profile needed to be added to force a defect. It is very likely that the heat was not uniformly delivered, so this is an acceptable assumption. The cladding exterior pressure was held at 101300 Pa and was either inert atmosphere for the argon tests or modeled as slow flowing steam in the case of OCL11.