Historical review of research on brittle crack propagation arresting technology for large welded steel structures developed in Japan with the application of Kca parameters

In response to the concern of increased risk of brittle fracture accompanying the recent enlargement of container ships, experimental research is being conducted to investigate brittle crack propagation arrest properties in Japan. The objective is to obtain the required toughness of the material to arrest brittle crack propagation in a 100-mm thick plate, which is considered to be the maximum thickness used in such applications. The use of K-ca as a method for determining arrest toughness is a main difference with respect to methodologies employed in Europe and the United States. In this review, we compare the approaches for determining brittle crack propagation arrest properties that are used in Japan with those used in Europe and the United States. Moreover, we review recent research trends, particularly with respect to the background and development of K-ca parameters. With regard to the industrial application techniques concerning arrestability of brittle crack propagation in steel plates, studies in the ship and storage tank research fields date back to after World War II, while some attention is also seen for nuclear power and line pipes. These research procedures were initially established in Europe and the United States, but was first adopted by Japan. However, soon after, Japan and the time when the research fell downward due to progress of steel manufacturing technology and defect management technology. Since then, research has actively resumed, and original contributions are being realised. The background of this work in Japan, and the creation of the K-ca concept will be explained herein. Further, the background of research on brittle crack propagation arrest properties in very large container ships, determination philosophy for deriving demand values, and ultrawide brittle crack propagation tests in the study of 75-mm thick material and their results are described. In both of the scenarios considered, i.e. one in which cracks are generated from the top of the hatch side coaming and arrested on the upper deck, and the other wherein brittle cracks occur at the upper deck end and are arrested in the hatch side coaming, the required K-ca was found to be 6000 N/mm(3/2).