Sunita P. Ho Department of Preventive and Restorative Dental Sciences, University of California San Francisco San Francisco, CA 94143 Paul F. Joseph* Department of Mechanical Engineering Clemson University Clemson, SC 29634 Michael J. Drews School of Materials Science and Engineering Clemson University Clemson, SC 29634 Thomas Boland Martine LaBerge Department of Bioengineering Clemson University Clemson, SC 29634 Experimental and Numerical Modeling of Variable Friction Between Nanoregions in Conventional and Crosslinked UHMWPE Recently, highly crosslinked UHMWPE components have been promoted for their high abrasive wear resistance over conventional UHMWPE (PE) in total joint replacement (TJR) prostheses to minimize osteolysis and consequent implant loosening. This study was aimed at investigating the role of friction gradients induced by localized coefficients of friction at both crystalline and amorphous nanoregions in PE, and crystalline and crosslinked nanoregions in crosslinked UHMWPE (XPE), in submicron wear debris gen- eration. An abrasive wear study performed on both XPE and PE using atomic force microscopy (AFM) illustrated that the onset of plastic deformation for XPE occurred at a normal load that was approximately 3 times higher when compared to PE. Coefficients of friction d of 0.2, 0.35, and 0.61, experimentally derived using AFM, were used as representative d for crystalline, amorphous, and crosslinked nanoregions, respectively, in a numerical Hertzian model. An increase in (0.20.02, 0.350.01 and 0.60.04) was observed with a decrease in crystallinity and storage modulus at 22°C. Using the Hertzian contact model, it was observed that variability in friction between nanoregions contributed to higher magnitude stresses for XPE (0.2 to 0.61; maximum eff =2.8com- pared to PE (0.2 to 0.35; maximum eff =1.1over a negligible thickness of the interfa- cial zone (IZ) between nanoregions. The experimentally observed increase in abrasive wear resistance of XPE could be attributed to an increase in the thickness of the interfa- cial zone between nanoregions with changing gradually from crystalline to crosslinked nanoregions, a situation that may not be observed with PE. This would cause a decrease in the friction gradient and resulting stresses thereby agreeing with the observed experi- mental higher abrasive wear resistance for XPE. However, in both PE and XPE, the presence of stress concentrations over a period of time could lead to irreversible damage of the material eventually generating submicron wear debris. Hence, semicrystalline, inhomogenous UHMWPE with several nanoregions (amorphous and crystalline) would be at a disadvantage for bearing application in terms of abrasive wear resistance com- pared to UHMWPE with relatively lower number of nanoregions and crosslinked nanoregions. DOI: 10.1115/1.1645530 Introduction Total joint replacement TJRprostheses consist of metallic and nonmetallic components such as mirror-finish cast cobalt- chromium Co-Cralloy or ceramic articulating on a metallic, ceramic or a polymeric ultra high molecular weight polyethylene UHMWPEcounter components 1,2. UHMWPE has been the principal polymeric material used in artificial joints since the early 1960’s to replace damaged articular cartilage in injured and dis- eased synovial joints 1,2. Despite the many advantages and wide use of UHMWPE, the generation of UHMWPE submicron or nanoscale wear debris, in TJR prosthesis has been of a primary concern for the past four decades 3,4. The submicron UHMWPE wear particles have been observed clinically to accumulate in the periprosthetic tissues. This accumulation causes an inflammatory reaction, leading to a cellular release of cytokines inducing bone resorption, and consequently loss of fixation of the device and ultimately device failure 5,6. The combined effort of in-vitro joint simulation and computa- tional modeling studies has contributed to elucidating the cause of different wear mechanisms in UHMWPE at a macroscopic scale 7,8. Wang and collaborators suggested that large scale deforma- tions, texture evolution, fracture and surface rupture within the surface region of UHMWPE components over repeated cyclic motion could cause macroscopic failure in the material eventually leading to total failure of the implant 9. Although these studies provided an insight to the wear processes in TJR prostheses, they do not explain the initiating cause of the submicron wear debris generation, a nanoscale phenomenon. Additionally, in the numeri- cal models, the coefficient of friction at the tribological contact, was either eliminated or assumed a constant. In the studies focused on identifying the initiating cause of failure, there is now a transition occurring in the field of materials analysis from continuum macro scale to discrete asperity-to- asperity nanoscale contact 10which could provide new insight to the basic problem of submicron UHMWPE submicron wear debris. It is well documented to date that when two engineering surfaces come into contact, plastic deformation may occur at the *Corresponding author: Department of Mechanical Engineering, Clemson Uni- versity, Box 340921, Clemson, SC 29634-021. Office: 235 Fluor Daniel Engineering Innovation Building. Phone: 864656-0545, Fax: 864656-4435, e-mail: paul.joseph@ces.clemson.edu Contributed by the Bioengineering Division for publication in the JOURNAL OF BIOMECHANICAL ENGINEERING. Manuscript received by the Bioengineering Divi- sion February 25, 2003; revision received October 14, 2003. Associate Editor: M. Toner. Copyright © 2004 by ASME Journal of Biomechanical Engineering FEBRUARY 2004, Vol. 126 Õ 111