Residual stresses and fatigue strength properties of ceramic/metal joints
Okabe Nagatoshi; Takahasi Manabu; Zhu Xia; Kagawa Kenichiro; Maruyama Miho
Abstract:Cyclic tensile fatigue tests were conducted for Si3N4/S45C joint specimens with copper interlayer at room temperature in order to clarify fatigue strength properties of ceramic/metal joint, considering the influence of residual stress in ceramics induced during joining process. Not only the joining process residual stress but also the residual stress behavior due to cyclic tensile loading was analyzed by using the three-dimensional elastic-plastic finite-element method. All experimental data obtained in this study were tried to arrange as the relationship between the normalized testing stress and the effective loading time by using the unified estimation method with considering effective volume and effective loading time. However, even if using the unified estimation method, the normalized strengths in the cyclic fatigue tests were clearly lower than normalized tensile strength at fracture probability of 50%. The difference of both normalized strengths was caused by the increase of the residual stress redistributed during cyclic fatigue tests. The redistributed residual stress was due to the reversible plastic deformation made in the interlayer of copper, but tended to saturate after second cycles. Then, if modifying not only the effective loading time but also the time-strength by using the unified estimation method against the superposing stress of the testing stress and the redistributing residual stress in cyclic fatigue tests, the cyclic tensile fatigue strengths agreed well with the normalized tensile strength for ceramic/metal joints as well as for monolithic ceramics. All cyclic fatigue strengths containing tensile fracture strength can be estimated well as the normalized strength with scatter properties of Weibull distribution. Key Words:Ceramic/metal joint, fatigue strength, residual stresses, Finite element method, Unified estimation method, Effective volume, Effective hold time