Technique for Introducing Internal Defects with Arbitrary Sizes and Locations in Metals via Additive Manufacturing and Evaluation of Fatigue Properties

Mechanical component failure is usually caused by metal fatigue originating from small defects in metallic materials. Thus, it is important to precisely capture the fatigue properties of materials containing small defects. Fatigue tests of materials with artificial surface defects introduced by drilling have been conducted. Using the resulting data, an equation for predicting the material fatigue limit has been proposed on the basis of the √area parameter model, and its effectiveness has been confirmed for various materials. However, for additive manufactured (AM) materials that contain internal defects resulting in failure, controlling the size of the defect where the fracture originates is extremely difficult. Therefore, verification of the predictive ability of the √area parameter model for AM materials is impossible, in contrast with other materials that fail because of surface defects. In this context, developing a technique to intentionally introduce internal defects with arbitrary sizes at arbitrary locations can provide insights that help predict the fatigue limit of AM materials. This study aimed to establish a technology for quantitatively evaluating the effect of internal defects on the fatigue properties of AM materials by introducing internal defects with arbitrary sizes at arbitrary locations via AM. Specimens with different defect sizes and locations were prepared. Prior to the fatigue tests, the defect sizes and locations were measured non-destructively via X-ray computed tomography (CT). The fatigue tests were conducted in air at room temperature. All the specimens failed because of the intentionally introduced internal defects, and the fatigue lives became shorter with increasing defect sizes, except for the specimens with defects adjacent to the surface. In those cases, fatigue cracks easily reached the surface; therefore, the fatigue lives were speculated to be shorter than those of the specimens with the same defect sizes. Moreover, the defect sizes determined from the fracture surfaces by scanning electron microscopy were nearly consistent with those determined by X-ray CT.