Computational analysis of microstructure of ultra high molecular weight polyethylene for total joint replacement

Ultra high molecular weight polyethylene (UHMWPE, or ultra high), a frequently used material in orthopedic joint replacements, is often the cause of joint failure due to wear, fatigue, or fracture. These mechanical failures have been related to ultra high's strength and stiffness, and ultimately to the underlying microstructure, in previous experimental studies. Ultra high's semicrystalline microstructure consists of about 50% crystalline lamellae and 50% amorphous regions. Through common processing treatments, lamellar percentage and size can be altered, producing a range of mechanical responses. However, in the orthopedic field the basic material properties of the two microstructural phases are not typically studied independently, and their manipulation is not computationally optimized to produce desired mechanical properties. Therefore, the purpose of this study is to: (1) develop a 2D linear elastic finite element model of actual ultra high microstructure and fit the mechanical properties of the microstructural phases to experimental data and (2) systematically alter the dimensions of lamellae in the model to begin to explore optimizing the bulk stiffness while decreasing localized stress. The results show that a 2D finite element model can be built from a scanning electron micrograph of real ultra high lamellar microstructure, and that linear elastic constants can be fit to experimental results from those same ultra high formulations. Upon altering idealized lamellae dimensions, we found that bulk stiffness decreases as the width and length of lamellae increase. We also found that maximum localized Von Mises stress increases as the width of the lamellae decrease and as the length and aspect ratio of the lamellae increase. Our approach of combining finite element modeling based on scanning electron micrographs with experimental results from those same ultra high formulations and then using the models to computationally alter microstructural dimensions and properties could advance our understanding of how microstructure affects bulk mechanical properties. This advanced understanding could allow for the engineering of next-generation ultra high microstructures to optimize mechanical behavior and increase device longevity.