Gagan Srivistava — 2012-13 Fellow
The proposed research involves the analysis and modeling of the mixed lubrication process in an artificial hip joint and its lifetime prediction. Wear debris generated during everyday operation accelerates wear in hip joints. About a decade ago, researchers in Japan proposed that dimples in the surface led to reduced wear in hip joints. They suspected that the debris particles fell into these dimples and away from the partial contact regions. However, their results were only experimental and models are incapable of capturing the progressive debris generation and particle dynamics, which is needed to predict the type of surfaces that are robust against contamination.
In artificial joints, engineers and scientists in collaboration with medical doctors have sought to design hip joints that operate in the full fluid (non-contact) lubrication regime as healthy human joints do; yet, the designed artificial joints usually operate in the mixed lubrication regime which creates massive amounts of wear debris, usually at sub-micron sizes. This wear debris causes the disease osteolysis, which results in the patient needing prosthesis surgery, more commonly known as the joint replacement. Recent press has focused on the wear debris in metal on metal hip joints, which has been very detrimental to patient health. To date, there has been almost no development of predictive modeling capabilities of these joints that accounts for the mixed lubrication (fluid and contact mechanics) and the resultant particle-generated wear mode of the artificial hip. Many hip lubrication models focus on either pure elastohydrodynamic modeling, pure boundary lubrication, surface wear or mixed lubrication. The only models which attempt to integrate the particle-augmented wear into the framework are finite element models (FEM) models. However, even those models simulate dry contact (ignoring fluid) between a single wear particle and a surface. While experimental screening and innovative bio-material development has advanced artificial hip joint technology, a lack of multiphysics model makes new joint designs and materials quite expensive, in terms of time and money, to screen and implement. For example, a 3-5 month test and characterization period is based on a 5 million cycle test (representing 5 years of patient use). Since hip prostheses successfully operate up to 10-15 years in patients, particle-augmented mixed lubrication (PAML) modeling simulations would provide an inexpensive life prediction method, avoid high consumable costs, and complement existing experimental approaches.
As mentioned earlier, the proposed research will use multiphysics modeling to analyze the tribological phenomena behind hip joints, in order to design an optimized textured hip joint interface that is capable of trapping wear debris in reservoirs created by surface texture. The numerical aspect of this study will involve simulation and analysis of the slurry (i.e., synovial joint fluid plus debris particles) flowfield during hip articulation using in-house code, called PAML Lite. PAML Lite has been developed for chemical mechanical polishing (another particle-fluid-wear problem) and recently extended to hip joint interfaces. This model, which is presently based on a cylindrical coordinate system, will be tailored more comprehensively to the artificial hip joint, which is a spherical coordinate system. Additionally, surface features in the form of cavities on the cup and/or head would be incorporated into the model to assess the hip’s ability to create hydrodynamic pressures that trap the debris being caused by the joint’s partial contact modes. The model will be used to vary the reservoir dimensions, pitches, and depths to find the optimal configuration for laying out the reservoirs on the surface.
The computational domain in this study will simulate the thin fluid gap between two counter-rotating walls which represent the surfaces of the hip head and cup. The results of this simulation will allow for an accurate prediction of the 3-D flow and pressure fields during day to day operations like walking, the resulting wear, and ultimately the lifetime prediction.
In the conventional empirical-based design methodology, the coefficient of friction (COF) and wear are measured after the experiments are conducted and then the design iteration loop changes the texture dimensions until optimization is achieved. This method is costly and ignores the predictive capability of tribology. The proposed model-based design methodology would require the engineer to input the activity or input conditions (i.e., the patient weight or load, speed, initial texture dimensions, substrate roughness, and hip reservoir materials) into the model. Consequently, the model would predict the resulting COF, wear levels, and joint lifetime. Whenever the source of damage is known (i.e., wear particles) and the mode of failure can be modeled (i.e., wear), the scenario for developing a lifetime prediction framework emerges as in the case of this hip problem.