The MIRTE (Matériaux Interaction Réflexion Toutes Epaisseurs) a program (Reference 1) has been carried out from December 2008 to June 2010 at the CEA (Commissariat à l’Energie Atomique et aux Energies alternatives) Valduc Center on the Apparatus B assembly (see Figure 1). The purpose of the MIRTE program was to measure integral reactivity characteristics of various structural materials that are typically used in nuclear facilities. The intended use of the data is to validate computer codes and associated nuclear cross section data that are used for criticality safety and reactor physics applications.
Phase 1 of the MIRTE program consisted of 43 sub-critical approaches extrapolated to critical conditions using the neutron amplification method. Three types of experimental setups have been built (see Figure 2):
(1) Interacting configurations with large screens, which consisted of two arrays separated either by water or screens with a thickness varying from 5 cm to 30 cm composed of iron, nickel, zirconium, aluminum, lead, copper, concrete with varying water contents (3%, 6%, 9%), or an empty aluminum box,
(2) Interacting configurations with thin plates, involving four arrays separated by water or cruciform plates with a thickness lower than 2 cm composed of copper, nickel, iron, titanium.
(3) Reflected configurations, which consist of one array, reflected on all four lateral sides by water, aluminum or borated glass walls.
The 43 experiments, which involved 4.738 wt.% enriched UO2 rod arrays, included 19 experiments with tested screens, nine reference experiments in which the screens were replaced by water or an aluminum box filled with air, and 15 reproducibility experiments with or without material.
Only eight experiments (comprised of four repeatability or reproducibility experiments—three with titanium and one with only water) are documented in this report. Repeatability experiments were appropriately combined into a total of four experiments (two with titanium and two reference experiments). These four
experiments were evaluated and determined to represent acceptable benchmark experiments.
In order to easily identify the different experiments, they were named according to the number of arrays involved in the configurations (4A for titanium experiments), the material to be tested, and its thickness (in millimeters). A reference experiment was identified by adding an “R” at the beginning of the experiment identifier and, concerning reproducibility experiments, the kind of reproducibility was added at the end of the identifier as follows:
(1) Rv for repeatability experiments, which consisted of a new sub-critical approach after water draining without any change in the configuration. This kind of experiment will give information about the uncertainties on the water height measurement and the extrapolation method.
(2) Rb(S1) for reproducibility experiments S1 type, which consisted of a new sub-critical approach after water draining and removal of the experimental device (support pedestal and lattices) from the experimental tank without any change in the configuration, allowing the estimation of the uncertainty on the rods positioning due to the gap between grid holes and rods.
(3) Rb(S2) for reproducibility experiments S2 type, which consisted of a new sub-critical approach after water draining, removal of the experimental device, moving and repositioning of lattices baskets, which highlights the uncertainty in lattice positioning.
(4) Rb(S3) for reproducibility experiments S3 type, which consisted of a new sub-critical approach after water draining, removal of the experimental device (support pedestal and lattices), moving and repositioning of lattices baskets and use of a new rods sample (rods sample chosen among 1261 rods). The uncertainty in rod sampling is assessed.
(5) Rb(S4) for reproducibility experiments S4 type, which consisted of a new sub-critical approach after a complete dismantling of the configuration, to evaluate the uncertainty in the screen positioning.
A number of the critical assemblies performed in the Apparatus B facility at Valduc have been evaluated as ICSBEP benchmark experiments.