• hip prosthesis;
  • metal on metal;
  • wear tribochemical reaction layers


  1. Top of page
  2. Abstract
  6. Acknowledgements

Metal-on-metal (MoM) bearings are at the forefront in hip resurfacing arthroplasty. Because of their good wear characteristics and design flexibility, MoM bearings are gaining wider acceptance with market share reaching nearly 10% worldwide. However, concerns remain regarding potential detrimental effects of metal particulates and ion release. Growing evidence is emerging that the local cell response is related to the amount of debris generated by these bearing couples. Thus, an urgent clinical need exists to delineate the mechanisms of debris generation to further reduce wear and its adverse effects. In this study, we investigated the microstructural and chemical composition of the tribochemical reaction layers forming at the contacting surfaces of metallic bearings during sliding motion. Using X-ray photoelectron spectroscopy and transmission electron microscopy with coupled energy dispersive X-ray and electron energy loss spectroscopy, we found that the tribolayers are nanocrystalline in structure, and that they incorporate organic material stemming from the synovial fluid. This process, which has been termed “mechanical mixing,” changes the bearing surface of the uppermost 50 to 200 nm from pure metallic to an organic composite material. It hinders direct metal contact (thus preventing adhesion) and limits wear. This novel finding of a mechanically mixed zone of nanocrystalline metal and organic constituents provides the basis for understanding particle release and may help in identifying new strategies to reduce MoM wear. © 2009 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 28:436–443, 2010

Replacement of the arthritic or traumatized hip joint is a routinely performed procedure. Because of an aging population and the extension of the procedure to younger patients, technological and surgical aspects of joint replacement strategies are continually reviewed and advanced. Hip resurfacing arthroplasty has regained significant popularity in recent years, combining the preservation of bone stock and a reduced risk of dislocation with a contemporary, low-wearing metal-on-metal (MoM) joint articulation.1 Hip resurfacing has increased the popularity of the MoM articulation. For example, hip resurfacing accounts already for 7.5% of all hip replacements in Australia.2 Thus, the MoM market share of 9.1% worldwide will be growing3 (and 2008 internal estimation of Zimmer GmbH, Winterthur). A cause for concern with MoM joints, however, has been systemic metal ion release. Despite today's very low-wear rates ranging from 0.5 to 2.5 µm/year,4–6 increased ion levels in serum compared with other established bearing combinations are observed.7 Metal ion release, which can form metal/protein complexes,8 and the generation of nanoscopic wear debris,9–11 raise concerns regarding particle induced osteolysis, perivascular lymphocytic tissue responses, and metal hypersensitivity.12–15

Considerable progress has been made in understanding and controlling manufacturing variables such as alloy composition, bearing diameter, design and clearance tolerances, and surface finish. Further wear reductions will only be possible if underlying wear mechanisms are better understood. In vitro16 and retrieval17 studies found that the governing wear mechanisms are not adhesion and abrasion as in other bearings, but predominantly tribochemical reactions (TCR) and surface fatigue. TCRs occur when the surfaces of two contacting metal bearings react with the interfacial medium (e.g., synovial fluid), resulting in the alternating formation and removal of chemical reaction products at the surfaces.18 The observed nanometer-sized wear debris must stem from the uppermost tribochemically transformed zone; otherwise, small wear rates would be impossible. Indeed, using transmission electron microscopy (TEM) the top surface layer can be seen to recrystallize to nanometer grain sizes.19 The interplay between lubricant and the nanocrystalline surface layer is not well defined. An investigation into this interaction is critical because TCRs affect the composition of the layer and determine its mechanical and chemical properties (and thus stability). Our purpose was to provide a better understanding of TCRs in MoM joints by virtue of chemical and microstructural analyses of retrieved MoM bearing couples.


  1. Top of page
  2. Abstract
  6. Acknowledgements


This chemical and microstructural investigation of TCR layers is based on the same MoM retrieval collection described in earlier studies.17, 19, 20 Briefly, the collection consists of 42 retrieved McKee-Farrar prostheses from five manufacturers worn by 14 male and 28 female patients. All prostheses were implanted and retrieved by a single surgeon.21 The average patient age at implantation was 61 (range: 38–82) years, and the prostheses lasted for 13.6 (range: 1.3–22) years. None of the components (acetabular cup or femoral stem) was removed for excessive wear. The implants came in femoral head diameters ranging from 35 to 42 mm. All implants were made of low carbon-cast cobalt alloys according to ASTM F75/ISO 5832-4 with about 26 wt-% Cr and 5–6 wt-% Mo. Differences regarding the elemental composition among manufacturers were summarized previoulsy.17 At the time of removal, all prosthesis were carefully rinsed to remove blood and subsequently sterilized and packed.

Light Microscopy

All samples were inspected for the presence of macroscopically visible TCR layers using a stereo light microscope (Wild Microscope M420, Leica, Glattbrugg, Switzerland). The presence and location of layers on the bearing surfaces were mapped for both heads and cups. Additional surface details, in particular the characteristics of the layers, were obtained with a reflective light microscope in the bright-field mode (Axiotech Vario 100, Carl Zeiss, Oberkochen, Germany). Here, a relatively thin 5×-objective allowed unrestrained views of the cups' inner bearing surfaces.

Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDS)

The articulating areas of the cups and heads of several samples were investigated by means of field-emission SEM (Hitachi S-4100, Kyoto, Japan). The surfaces remained uncoated to allow later chemical analyses. Secondary (SE) and backscattered electron (BSE) images were recorded from 2–10 keV, where the low voltage contributed to more topographic detail in both imaging modes. Energy-dispersive X-ray analysis (EDS) was used to evaluate differences in chemical composition between areas. EDS (Model 6816, Oxford Instruments, High Wycombe, UK) was performed at 5 and 10 keV. The system allows quantitative chemical analyses with an accuracy of about 1 wt.% for elements with an atomic number >14. Lighter elements are detected qualitatively.

Based on the light microscopy and SEM results, five representative MoM couplings were chosen to undergo detailed analyses using SEM and further sophisticated techniques as described below. Manufacturer origin and demographic details of those implants are listed in Table 1.

Table 1. Demographic Details of the XPS/TEM Subgroup
 Couple ACouple BCouple CCouple DCouple E
  1. XPS = photoelectron spectroscopy; TEM = transmission electron microscopy.

Alloy trade nameCoballoyVinertiaVinertiaZimaloyZimaloy
ManufacturerDow, UKHowmedica, USAHowmedica, USAZimmer, USAZimmer, USA
Head diameter (mm)3540354237
Cup inclination (°)4048503837
Implantation siderightleftleftleftright
Age at surgery (years)7351665852
Time in situ (years)7.517.91913.612.1

Photoelectron Spectroscopy (XPS)

To locate the areas of interest, light microscopy and integrated scanning X-ray induced secondary electron imaging (SXI) were used prior to XPS analysis. The distributions of chemical elements on and beneath the surface were resolved by means of XPS analyses (PHI Quantum 2000, Physical Electronics Inc., Eden Prairie, MN). Samples were exposed to a monochromatized X-ray beam (Al Kα = 1486.6 eV) with 20, 50, or 100 µm lateral resolution. Low-energy electrons and argon ions were used simultaneously to compensate for electrical charging of insulating surface areas during analysis. Emitted photoelectrons were analyzed with a hemispherical electron energy analyzer equipped with a channel plate and a position sensitive detector. The electron take-off angle was 45°. The analyzer was operated in the constant pass energy mode of 117.40 or 58.7 eV, giving a total energy resolution of 1.70 or 1.04 eV, respectively. The residual background pressure inside the spectrometer was better than 2 × 10−9 mbar during analysis. The binding energy scale was calibrated for the Au-4f electrons at 84.0 eV. Elemental concentrations are given in atomic percent (normalized to a total of 100 at%) using the photoelectron peak areas after Shirley background subtraction (Muli-Pack, Version 6.0, Physical Electronics Inc.) and the built-in PHI sensitivity factors for the calculation. Next to the expected alloy elements, special attention was directed toward the occurrence of carbon (C), oxygen (O), nitrogen (N), phosphorus (P), sulphur (S), sodium (Na), magnesium (Mg), and calcium (Ca). The detection limit was 1 at%. The elemental concentrations are presented as a function of distance from the surface as obtained by acquiring sputter depth profiles. The latter were acquired by material removal using 4 kV argon-ion etching in between consecutive analysis of the elemental concentrations. The sputter rate is material dependent and was determined to be 20 nm/min for SiO2.


To validate the XPS findings and to gain additional information about the subsurface microstructure, head and cup sections were further investigated by means of TEM (EM400, Phillips, Eindhoven, The Netherlands; and Tecnai F20ST, FEI, Eindhoven, The Netherlands) with the use of EDS and electron energy loss spectroscopy (EELS). A custom preparation technique was employed using two parallel cuts of the contacting areas with a thickness of 500 µm, which were glued together with a two-component adhesive. To minimize the gap between the contacting surfaces, the convex head was glued to the concave cup. The as-prepared samples were then fixed in a brass cylinder (3 mm diam.) using a slotted pipe (2.5 mm diam.) and a suitable adhesive. A heat treatment at room temperature for 30 min and at 150°C for 2 h resulted in a sufficient bonding strength of the composite setup. After drying, the compounds were cut into 400 µm-thin slices using a corundum wheel on a low speed saw. After conventional wet grinding to a thickness of 100 µm, specimens were further thinned from both sides by means of a dimple grinder (Model 656, Gatan GmbH, Munich, Germany) and an ion mill (PIPS 691, Gatan GmbH). TEM investigations were performed under an accelerating voltage of 120 kV after a sample thickness of less than 100 nm had been reached. In a recent study with worn high-nitrogen steel samples, this technique did not introduce artefacts into the surface.22. EDS line scans were performed on these cross-sectional samples to plot the elemental composition from surface to depth. EELS was used to verify the results of EDS measurements. For this technique an electron spectrometer is required, which measures the energy of randomly deflected electrons of the electron beam; energy loss can be associated with a specific element. EELS mapping was used to determine the relative local distribution of cobalt (Co), chromium (Cr), and C on the section of interest.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Light Microscopy

All 42 samples, heads and cups, displayed TCR layers. Most were visible with the unaided eye and were found in or adjacent to the articulating areas.

SEM and EDS Analyses

Large quantities of TCR layers were identified on nearly every component. By SEM, thick layers appeared fragmented; thin ones appeared smooth. Quite often the thick layers were scratched. EDS analyses revealed high carbon contents (Fig. 1). In addition, traces of N, O, P, S, chlorine (Cl), Na, Mg, potassium (K), and Ca were found occasionally. In areas without TCR layer, carbon and trace elements were not detected.

thumbnail image

Figure 1. SEM BSE-picture (top) and EDS spectra (bottom) within the primary articulating surface area of a retrieved MoM head depicting the high C content within tribochemical reactions layers (square symbol). For comparison the area adjacent to the TCR layer (triangle) shows no indication of C.

Download figure to PowerPoint

XPS Analyses

As was shown earlier,23 thick denatured protein layers might stick rigidly onto the passive layer of cobalt alloys. Because chromium-oxide forms in a moist environment, this is expected and was verified using XPS (data not shown). Here, the focus is on TCR layers with an appearance similar to those in Figure 1. These layers seem level with metallic-like surface areas in the immediate surroundings and are predominantly found within the contact area. Figure 2 displays a light microscopy image with the corresponding SXI overview from such a location. Due to differences in light reflection, the two areas can easily be distinguished from each other. The SXI overview image suggests a difference in chemical composition between these two areas and were further investigated using XPS (“Point 1” located on the metallic looking surface and “Point 2” on the layer; Fig. 2).

thumbnail image

Figure 2. The primary articulating surface section of a retrieved MoM head being about 1 × 1 mm in size. Left: light microscopy image showing a tribochemical reaction layer which discriminates from the bright looking metal surface. Right: SXI area pattern of the same region highlighting the differences in chemical composition. “Point 1” and “Point 2” indicate locations for XPS measurements.

Download figure to PowerPoint

Point 1

The first two nanometers show a combination of C, O, N, and P. Further below, Cr and O prevail for about 5 nm. The Cr-signal shows mostly chromium-oxide with a binding energy of 576.8 eV. Then, the concentrations of these elements gradually decrease, while the levels of Co, metallic Cr (574.2 eV), and Mo increase toward the expected concentration (Fig. 3, top chart). This XPS profile represents the expected chromium-oxide passivated cobalt alloy surface with some organic C-O-N contamination on top of it.

thumbnail image

Figure 3. XPS charts showing the elemental distribution of Points 1 and 2 (Fig. 2) in atomic concentrations versus depth from the surface. Note the difference in scale between charts. Point 1 (top chart): a thin contamination layer (containing C) is followed by an O-rich Cr layer with about 8 nm in thickness before the base material is reached. Point 2 (bottom chart): C prevails down to a depth of 120 nm before the elements of the base material take over.

Download figure to PowerPoint

Point 2

A completely different picture emerges from the chemical analysis of Point 2, which is only 200 to 300 microns away from Point 1. Again, at the surface C-O-N contamination prevails. Subjacent to this layer, C is the most prominent element for the next 120 nm. The C binding energy peaks at 285 eV throughout depth without indication of carbides (281–283 eV). Some C[BOND]O bonding near the surface is indicated. At about 50 nm depth, the alloy's base elements [Co, Cr, Mo] begin to appear and rise to their expected concentration levels, while the C concentration slowly starts to decrease. Throughout, the Cr-signal is 90% metallic, 10% oxidic. Hence, the presence of a distinct passivation layer could not be observed. Atomic concentrations versus depth are displayed in Figure 3, bottom chart.

TEM Analyses

In agreement with previous investigations,19 directly below the contact surface, the microstructure is nanocrystalline (Fig. 4). Except for some face-cubic-centered (fcc) crystals, most crystals have a hexagonal-closed-packed (hcp) lattice structure and stem from strain induced phase transformation. Figure 5 displays the TEM micrograph of a location similar to Point 2 in Figure 2 in the EDS mode, which does not provide the same level of contrast as the standard bright-field mode depicted in Figure 4 (i.e., blurring the nanocrystalline microstructure of the material). However, the EDS mode allows for elemental identification. The EDS profile along the dotted line is depicted in Figure 5. Because of the wedge shape of the TEM specimen (required for its preparation), the base material elements Co, Cr, and Mo show a steady increase with distance from the surface because more signal is reaching the detector. Nevertheless, within the first nanometers of the surface, the gradient is obviously steeper, indictaing a thin zone of reduced metal content. Interestingly, the C and O lines show a distinctly different behavior with a steep increase directly at the surface. In particular, the C distribution seems to fit the XPS results suggesting that C prevails at the surface and then decreases to a depth of 150 nm. These results were repeatedly verified on samples taken from three other retrievals, whereby the thickness of the carbon-rich layer varied from 50 to 200 nm. Furthermore, EELS mapping confirmed the presence of high amounts of C below the surface (Fig. 6). The specific structure suggests clusters of C within the cobalt-alloy substrate.

thumbnail image

Figure 4. TEM micrograph showing a nanocrystalline subsurface microstructure. Left: bright-field image. Right: dark-field image showing the positions of the strain induced hexagonally closed packed CoCrMo nanocrystals.

Download figure to PowerPoint

thumbnail image

Figure 5. High-resolution TEM picture of the subsurface microstructure directly at the worn surface using the EDS mode (nanocrystals do not show in this mode). The dotted line depicts the measurement pathway for the EDS profiles shown below (firstt profile: total detected signal). The dashed line marks the 150 nm depth line.

Download figure to PowerPoint

thumbnail image

Figure 6. EELS elemental subsurface distribution at a location similar to that of Figure 2. Bright areas indicate high amounts of element-specific material, whereas dark areas denote their absence. Close to the surface, increasing amounts of C can be found, while the main alloy constituents, Co and Cr are decreased. Note the distribution of C, which suggests the occurrence of local clusters indicating mechanical mixing with the metal. These C clusters show as white spots within the C-map.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  6. Acknowledgements

Considering the environmental conditions of the specific tribosystem of the artificial hip joint, all four major wear mechanisms (abrasion, adhesion, surface fatigue, and tribochemical reactions) can apply.18 Typically, MoM hip joints operate in boundary or mixed lubrication mode,24 depending on the head diameter and clearance tolerances. Hence, TCR layers are expected for MoM joints and have been described.25–29 TCR layers were recognized as “deposits” and/or “precipitates,” which belies their importance in the tribosystem. In this study we demonstrated that TCR layers do not simply adsorb onto the bearing surface; TCRs also modify the cobalt-alloy substrate, transforming subsurface layers from purely metallic to composite-like.

The TCR layer consists of organic, ceramic, and metallic constituents that are well mixed. At first glance these findings are unexpected, but the specific composition of the TCR layers explains the success of self-mating cobalt–chromium alloy joints in the human body: direct metal–metal contact never occurs in the presence of a TCR layer—even without fluid film separation. Thus, adhesion, which could lead to catastrophic seizure of the contacting surfaces, is prevented. Indeed, no signs of adhesion were identified on 84 articulating surfaces of this retrieval collection.17 Obviously, TCR layers are essential to keeping wear rates low.

Tribochemical reactions depend on the mechanical and chemical interaction between body and counterbody, the interfacial medium, and the environment. According to classical theory, reaction layers are generated within or adjacent to the contacting areas and require mechanical action. Friction between the contacting bodies causes an increase in temperature and a rise of the inner energy of the uppermost layers of the deformed materials in contact. Both features enhance the surface reactivity, and oxidized islands are generated.30–33 These oxide layers flake off the surface after reaching a critical thickness. Now, freshly activated, bare metal is presented to the interfacial medium causing metal–ion release. The interfacial medium is likely involved in the generation (reformation) of TCR layers. For example, proteins can stick to the activated surfaces, forming deposits. This may slow the repassivation process, yet a chromium–oxide layer is still generated.34 The specific bonding mechanism is not well understood, but can be attributed to the high number of free Co and Cr ions close to the surface, which easily form metal/protein complexes.8 In turn, these complexes are adsorbed onto the metallic surfaces.35 These protein layers adhere rigidly to the surfaces23 and are typically found on passive metal films.17

The subsurface carbon must stem from these or other environmental carbon sources. At Point 2 in Figure 3, within the first 100 nm of the TCR layer, the nonmetallic elements were 89% C, 7% N, and 4% O. This is similar (though not equal) to albumin, a major protein constituent of synovial fluid. Human albumin contains 63% C, 17% N, 19% O, and <1% S (neglecting hydrogen).36 XPS and EELS readings suggested the presence of carbon clusters, not dissolved carbon (furthermore, the measured C content is far too high to be attributed to carbides). However, it is still unclear how carbon clusters can extend up to 200 nm into the bulk, given the thermodynamic conditions of the hip joint. Although locally elevated temperatures between 60 and 80°C are conceivable,16 no thermally driven diffusion process can be postulated that would account for driving organic matter into a metallic solid solution within a time frame of years. Similarly, a mechanically driven diffusion process37 is implausible under mild sliding wear conditions: the essential impact energy for this process is not present in total hip joints. Therefore, another mechanism must be operant, capable of blending organic material with a metal substrate. Based on recent molecular dynamics (MD) simulations, such a mechanism was investigated by Rigney et al. and termed “mechanical mixing.”38–42

Putting two different metallic materials A and B into contact, MD simulations revealed the formation of vortices in the vicinity of the interface during sliding conditions. The convective material transport is most pronounced in regions with high vorticity. Interestingly, the material transport is not restricted by the interface A/B, but material exchange between both bodies can take place. Such a mechanism is capable of mixing materials over a number of atomic distances and has been experimentally validated for several tribosystems.43–45 In the case of MoM hip replacements, the tribosystem is very complex, and the computer simulation of all aspects (e.g., organic constituents of the synovial fluid; materials with strain induced phase transformation) is currently impossible. However, the same principles apply, suggesting that areas with oxide layers and/or adsorbed proteins are cluster-wise incorporated into the convective material transport. This, in concert with the external shear stresses due to friction, facilitates the transformation of the uppermost subsurface layers into a nanocrystalline microstructure of cobalt–chromium alloy. The nanocrystals are known to rotate under mechanical shear stresses,46 which would then support the mixing process even further.

All retrievals were first generation McKee-Farrar type MoM components from various manufacturers. They were made of cast cobalt–chromium alloy according to ASTM F75/ISO 5832-4. Today, MoM bearings are typically manufactured from wrought (forged) low or high carbon CoCrMo alloys (acc. to ASTM F1537 and ISO 5832/12). This is a limitation of the study; however, similar microstructural surface changes were observed for wrought low19 and high carbon16 cobalt–chromium alloys after in vitro testing. Furthermore, microstructural surface changes were found in other tribosystems with austenitic stainless steels sliding against each other in boundary or mixed lubrication mode.22 These reports suggest that our findings likely apply to current MoM bearings, and thus provide a clear direction for investigating these bearings. Recently, a mechanically mixed zone of nanocrystalline metal and organic constituents was documented for a modern, retrieved hip resurfacing implant.47

The mechanism is similar to the action of antiwear additives in high-performance engine lubricants. These additives form surface films that protect the underlying material.48 Further work is required to determine if current MoM devices exhibit the protective nanocrystalline TCR layers and could benefit from strategies to stabilize them. To make MoM bearings more durable and further reduce their wear, the generation of nanocrystalline TCR layers might be enhanced. Strategies should be employed to stabilize these layers.

In conclusion, TCR layers are found frequently on MoM bearing surfaces. These layers are generated through mechanical mixing with organic carbon stemming from the synovial fluid and are a nanocrystalline composite of metallic, ceramic, and organic material. One strategy to lower wear rates of these bearings is to promote the formation and stability of TCR layers.


  1. Top of page
  2. Abstract
  6. Acknowledgements

This study was sponsored by the Rush Arthritis and Orthopaedics Institute, the AO Research Institute, and the Institute for Materials Science and Engineering (UDE). The authors thank Prof. Dr. Dudzinski (TU Wroclaw, Poland) and Ms. Gleising for TEM support and Drs. Sudfeld and Spasova for help regarding HR-TEM and EDS, which was conducted at the Markus Institute of Prof. Dr. Farle (UDE).


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Schmalzried TP, Fowble VA, Bitsch RG, et al. 2007. Total hip resurfacing. In: CallaghanJJ, RosenbergAG, RubashHE, editors. The adult hip. 2nd ed. Philadelphia, PA: Lippincott; p 969979.
  • 2
    Hing C, Back D, Shimmin A. 2007. Hip resurfacing: indications, results, and conclusions. Instr Course Lect 56: 171178.
  • 3
    Orthopedic Network News. 2007; 18
  • 4
    Rieker CB, Schön R, Köttig P. 2004. Development and validation of a second-generation metal-on-metal bearing. Laboratory studies and analysis of retrievals. J Arthrop 19 (Suppl 3): 511.
  • 5
    Semlitsch M, Willert HG. 1997. Clinical wear behaviour of UHMWPE cups paired with metal and ceramic ball heads in comparison to metal-on-metal pairings of hip joint replacements. Proc Inst Mech Eng [H] 211: 7388.
  • 6
    Chan FW, Bobyn JD, Medley JB, et al. 1999. Wear and lubrication of metal-on-metal hip implants. Clin Orthop Rel Res 369: 1024.
  • 7
    Rasquinha VJ, Ranawat CS, Weiskopf J, et al. 2006. Serum metal levels and bearing surfaces in total hip arthroplasty. J Arthroplasty 21 (Suppl 2): 4752.
  • 8
    Hallab NJ, Mikecz K, Vermes C, et al. 2001. Differential lymphocyte reactivity to serum-derived metal–protein complexes produced from cobalt-based and titanium-based implant alloy degradation. J Biomed Mater Res 56: 427436.
  • 9
    Doorn PF, Campbell PA, Worrall J, et al. 1998. Metal wear particle characterization from metal on metal total hip replacements: transmission electron microscopy study of periprosthetic tissues and isolated particles. J Biomed Mater Res 42: 103111.
  • 10
    Catelas I, Campbell PA, Bobyn JD, et al. 2006. Wear particles from metal-on-metal total hip replacements: effects of implant design and implantation time. Proc Inst Mech Eng [H] 220: 195208.
  • 11
    Brown C, Williams S, Tipper JL, et al. 2007. Characterisation of wear particles produced by metal on metal and ceramic on metal hip prostheses under standard and micro-separation simulation. J Mater Sci Mater Med 18: 819827.
  • 12
    Korovessis P, Petsinis G, Repanti M, et al. 2006. Metallosis after contemporary metal-on-metal total hip arthroplasty. Five to nine-year follow-up. J Bone Joint Surg Am 88: 11831191.
  • 13
    Miloŝev I, Trebŝe R, Kovac S, et al. 2006. Survivorship and retrieval analysis of Sikomet metal-on-metal total hip replacements at a mean of seven years. J Bone Joint Surg Am 88: 11731182.
  • 14
    Jacobs JJ, Hallab NJ. 2006. Loosening and osteolysis associated with metal-on-metal bearings: a local effect of metal hypersensitivity? Bone Joint Surg Am 88: 11711172.
  • 15
    Lachiewicz PF. 2007. Metal-on-metal hip resurfacing: a skeptic's view. Clin Orthop Relat Res 465: 8691.
  • 16
    Wimmer MA, Loos J, Nassutt R, et al. 2001. The Acting wear mechanisms on metal-on-metal hip joint bearings—in vitro results. Wear 250: 129139.
  • 17
    Wimmer MA, Sprecher C, Hauert R, et al. 2003. Tribochemical reaction on metal-on-metal hip joint bearings—a comparison between in-vitro and in-vivo results. Wear 255: 10071014.
  • 18
    Wimmer MA, Fischer A. 2007. Tribology. In: CallaghanJJ, RosenbergAG, RubashHE, editors. The adult hip. 2nd ed. Philadelphia, PA: Lippincott; p 215–226.
  • 19
    Büscher R, Täger G, Dudzinski W, et al. 2005. Subsurface microstructure of metal-on-metal hip joints and its relation to wear particles generation. J Biomed Mater Res Part B Appl Biomater 72B: 206214.
  • 20
    Sprecher CM, Schneider E, Wimmer MA. 2005. Verschleisserscheinungsformen und – mechanismen von explantierten Metall–Metall Hüftendoprothesen in Relation zur Lebensdauer. Zeitschrift Mater 47: 96100.
  • 21
    Täger KH. 1976. Untersuchungen an Oberflächen und Neogelenkkapseln getragener McKee-Farrar Endoprothesen. Arch Orthop Unfall-Chir 86: 101113.
  • 22
    Büscher R, Gleising B, Dudzinski W, et al. 2004. The effects of subsurface deformation on the sliding wear behaviour of a microtextured high-nitrogen steel surface. Wear 257: 284291.
  • 23
    Wimmer MA, Nassutt R, Sprecher C, Loos J, Täger G, Fischer A. 2006. Investigation of stick phenomena in metal-on-metal hip joints after resting periods. Proc Inst Mech Eng [H] 220: 219227.
  • 24
    Dowson D, Hardaker C, Flett M, et al. 2004. A hip joint simulator study of the performance of metal-on-metal joints: Part II: design. J Arthroplasty 19 (Suppl 3): 124130.
  • 25
    Täger G, Euler E, Plitz W. 1997. Changes in shape of the McKee-Farrar hip endoprosthesis. Orthopäde 26: 142151.
  • 26
    Plitz W, Huber J, Refior HJ. 1997. Experimentelle Untersuchungen an Metall–Metall-Gleitpaarungen und ihre Wertigkeit hinsichtlich eines zu erwartenden in-vivo-Verhaltens. Orthopäde 26: 135141.
  • 27
    Chan FW, Bobyn JD, Medley JB, et al. 1996. Engineering issues and wear performance of metal on metal hip implants. Clin Orthop 333: 96107.
  • 28
    McKellop H, Park SH, Chiesa R, et al. 1996. In vivo wear of 3 types of metal-on-metal hip prostheses during 2 decades of use. Clin Orthop 329: 128140.
  • 29
    Semlitsch M, Streicher RM, Weber H. 1989. Verschleißverhalten von Pfannen und Kugeln aus CoCrMo-Gußlegierung bei langzeitig implantierten Ganzmetall Hüftprothesen. Orthopäde 18: 377381.
  • 30
    Lin IJ, Nadiv S, Grodzian DJM. 1975. Changes in the state of solids and mechano-chemical reactions in prolonged communition processes. Minerals Sci Eng 7: 313336.
  • 31
    Lauer JL, Fung SS. 1982. Microscopic contour changes of tribological surfaces by chemical and mechanical action. Trans ASLE 26: 430436.
  • 32
    Quinn TFJ. 1983. NASA interdisciplinary collaboration in tribology. A review of oxidational wear. NASA Contractor Report 3686.
  • 33
    Sullivan JL. 1987. The role of oxides in the protection of tribological surfaces. In: Proc. Conf. “Tribology—friction, lubrication, and wear. Fifty years on, 1.-3.7.1987.London: IMechE Publ. Ltd. p 283301.
  • 34
    Igual Muñoza A, Mischler S. 2007. Interactive effects of albumin and phosphate ions on the corrosion of CoCrMo implant alloy. J Electrochemical Soc 154: C562C570.
  • 35
    Yan Y, Neville A, Dowson D. 2006. Biotribocorrosion—an appraisal of the time dependence of wear and corrosion interactions: II. Surface analysis. J Phys D Appl Phys 39: 32063212.
  • 36
    Universal Protein Resource, UniProtKnowledgebase (UniProtKB), Section Swiss-Prot (UniProtKB/Swiss- Prot), P02768 (ALBU_HUMAN), Isoform 1 [UniParc],
  • 37
    Odunuga S, Li Y, Kraschnotchekov P, et al. 2005. Forced chemical mixing in alloys driven by plastic deformation. Phys Rev Lett 95: 045901-1045902-4.
  • 38
    Rigney DA, Hammerberg JE. 1998. Unlubricated sliding behavior of metals. MRS Bull 23: 3236.
  • 39
    Rigney DA, Hammerberg JE. 1999. Mechanical mixing and the development of nanocrystalline material during the sliding of metals. Proc TMS Fall Meet 465474.
  • 40
    Rigney DA. 2000. Transfer, mixing and associated chemical and mechanical processes during the sliding of ductile materials. Wear 245: 19.
  • 41
    Subramanian K, Wu JH, Rigney DA. 2004. The role of vorticity in the formation of tribomaterial during sliding. Mater Res Soc Symp Proc 821: 97102.
  • 42
    Karthikeyan S, Kim HJ, Rigney DA. 2005. Velocity and strain-rate profiles in materials subjected to unlubricated sliding. Phys Rev Lett 95: 14.
  • 43
    Venkataraman B, Sundararajan G. 2000. Correlation between the characteristics of the mechanically mixed layer and wear behaviour of aluminium, Al-7075 alloy and Al-MMCs. Wear 245: 2238.
  • 44
    Singh JB, Cai W, Bellon P. 2007. Dry sliding of Cu–15 wt%Ni–8 wt%Sn bronze: wear behaviour and microstructures. Wear 263: 830841.
  • 45
    Hahn M, Gleising B, Dudzinski W, et al. 2008. Electron microscopical investigationof thermally sprayed coatings of cylinder walls after motor tests. Metallography (in press).
  • 46
    Karthikeyan S, Wu JH, Rigney DA. 2004. The role of vorticity in the formation of tribomaterial during sliding. In: AndersonPM, FoeckeT, MisraA, RuddRE, editors. Proceedings Symposium on Nanoscale Materials and Modeling-Relations among Processing, Microstructure, and Mechanical Properties. Vol. 821 Warrendale, PA: MRS; p 961966.
  • 47
    Pourzal R, Theissmann R, Morlock M, et al. 2009. Micro-structural alterations within different areas of articulating surfaces of a metal-on-metal hip resurfacing system. Wear DOI: 10.1016/j.wear.2009.01.012; (in press).
  • 48
    Mosey NJ, Müser MH, Woo TK. 2005. Molecular mechanisms for the functionality of lubricant additives. Science 307: 16121615.