Porous coated Ti-6Al-4V has a fatigue strength that is approximately onethird that of the uncoated alloy. The interfacial geometry between the porous coating and the implant substrate is notch-like, leading to stress concentrations which have been implicated as one of the main causes for its reduced fatigue strength. The purpose of this study is to determine the effect of interfacial geometry on the fatigue strength of porous coated Ti-6Al-4V. The work is composed of two integrated approaches to the problem:​ numerical and experimental.

In the numerical approach, the interface between the porous coating and implant is modeled using two-dimensional finite element analysis. This analysis identifies geometric parameters that affect the value of the stress concentration factor, K_t. Based on this analysis a geometry is suggested that can reduce K_t. The purpose of the experimental approach is to verify the effectiveness of the numerical model in predicting fatigue strength. Specimens are tested based on the suggested improved geometry. Two specimen types are tested:​ one determined in the numerical analysis to have a low value of K_t and hence an improved fatigue strength;​ and one designed to have a value of K_t similar to that of a porous coated implant. The numerical analysis would be effective in predicting fatigue strength if the specimen type determined to have a low value of K_t has a higher fatigue strength than the specimen designed to have a similar K_t to a porous coated implant.

If the model is determined to be effective through the experimental analysis, then the model can be used to answer additional questions. An important additional question is the effect of shear loading on interfacial stress. Bending fatigue has been studied extensively while shear fatigue, though it has been implicated in clinical findings of loosened porous coating particles, has not. Shear loading of the interface is thus addressed with respect to the effect of changes in interfacial geometry on interfacial stress.

It was found that the numerical model is effective in predicting bending fatigue strength improvement of porous coated Ti-6Al-4V due to changes in interfacial geometry. In addition, the model correctly predicted the region of crack initiation for the two specimen types. A doubling of fatigue strength was found for the low K_t specimen type compared to the high K_t specimen type and porous coated fatigue strength values of the literature.