FAILURE BEHAVIOR OF BWR LOWER HEAD DUE TO REACTION WITH MOLTEN METALLIC MATERIALS

抄録

Decommissioning work of Tokyo Electric Power Company (TEPCO) Fukushima Daiichi Nuclear Power Station (1F) is in progress. Internal investigations, sim-tests, and accident analyses are underway to understand the state of reactor pressure vessels (RPV) and primary containment vessels after the accident by TEPCO. During the severe accident (SA), various materials, including stainless steel control rods, B₄C, Zr-alloys channel box materials, and fuel rods, melt in the core region and fall into the bottom of the RPV. According to a possible scenario for 1F unit 2 based on the internal investigation, there is a possibility that most of the fuel oxides could not melt completely, and a metal melt pool could be formed. The metal pool could react with structural materials such as the Control rod drive (CRD), consisting mainly of a stainless steel housing and an Inconel stub tube, and cause the failure of the RPV. The reaction between the metal pool and the structural materials is important in understanding the RPV failure behavior. In this study, the ELSA (Experiment on Late In-vessel Severe Accident Phenomena)-1 test, which focuses on the damage caused by the eutectic melting of the liquid metal pool and CRD structures, was conducted.

A test sample simulating the CRD structure at the lower head was fabricated and loaded with Fe-Zr alloy as the simulated metal corium. The sample was gradually heated up to about 1400°C using the LIESAN (Large-scale Equipment for Investigation of Severe Accidents in Nuclear reactors) test facility, and in-situ observation was performed using a video camera.

The test results showed that the CRD structural material reacted with the metal debris and melted and failed at about 1050-1250°C, which was lower than the melting point of the CRD itself. It was also observed that the molten material flowed into the CRD, suggesting that the CRD structure was preferentially damaged during SA. From the cross-section observation of the sample and phase diagrams on its materials, it was found that the housing and the stub tube of the CRD melted and failed due to the Zr-Fe and Zr-Ni eutectic reactions with Zr in the molten metals, respectively. In the experiment, it took a certain amount of time from the start of the CRD reaction with the metal pool until it melted and broke, indicating that the reaction speed was also an important factor in the failure of RPV.