| Description |
The longevity of metal-on-polyethylene prosthetic hip implants, in which a CoCrMo femoral head articulates with a polyethylene acetabular liner, is often limited by polyethylene wear and osteolysis caused by polyethylene wear particles. Current approaches to reduce polyethylene wear include improving the mechanical properties of the polyethylene acetabular liner, and/or manufacturing ultra-smooth articulating surfaces. In contrast, we show that adding a patterned microtexture of concave "dimples" to a smooth CoCrMo surface significantly reduces polyethylene wear by promoting the formation of an elastohydrodynamic lubrication film, which reduces contact between the CoCrMo and polyethylene bearing surfaces. Using pin-on-disc (PoD) experiments in which a polyethylene pin articulates with a CoCrMo disc, we gravimetrically quantify polyethylene wear and demonstrate that a patterned microtexture on the CoCrMo disc significantly reduces polyethylene wear compared to a smooth, nontextured CoCrMo disc. We quantify wear of different polyethylene materials currently used in commercial prosthetic hip implants, articulating with several patterned microtexture geometries. We correlate polyethylene wear with surface topography measurements and conclude that the patterned microtexture creates microhydrodynamic bearings and an elastohydrodynamic lubricant film, which reduces contact between the articulating surfaces, thus reducing wear. We also perform electrochemical measurements to show that the microtexture does not negatively impact iv the corrosion resistance of CoCrMo. Furthermore, we use PoD experiments to measure friction between a polyethylene pin and microtextured CoCrMo discs, covering a wide range of operating conditions including sliding velocity and contact pressure. We determine how the lubrication regime changes as a function of operating conditions and show that the patterned microtexture accelerates the transition from boundary to elastohydrodynamic lubrication. Additionally, we illustrate that the patterned microtexture could enable tailoring the hip implant to specific patient needs based on activity level, gender, and age. Finally, we use machine learning methods to analyze published PoD polyethylene wear datasets and implement and cross-validate several model-based and instance-based data-driven models, and quantify their prediction error with respect to the published experiments. The data-driven models enable predicting polyethylene wear of PoD experiments based on its operating parameters, and they reveal the relative contribution of individual PoD operating parameters to the resulting polyethylene wear, thus potentially reducing the need for time consuming experiments. |
