Acta Biomaterialia 8 (2012) 852–859 Contents lists available at SciVerse ScienceDirect Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat Tribo-electrochemical characterization of metallic biomaterials for total joint replacement N. Diomidis a,⇑, S. Mischler a, N.S. More b,1, Manish Roy c a Tribology and Interface Chemistry Group, Swiss Federal Institute of Technology of Lausanne, Lausanne, Switzerland Department of Metallurgical and Materials Engineering, Visvesvaraya National Institute of Technology, Nagpur, India c Defence Metallurgical Research Laboratory, Hyderabad, India b a r t i c l e i n f o Article history: Received 8 June 2011 Received in revised form 21 September 2011 Accepted 26 September 2011 Available online 29 September 2011 Keywords: Biotribocorrosion Fretting corrosion Biomaterials Artiﬁcial implants a b s t r a c t Knee and hip joint replacement implants involve a sliding contact between the femoral component and the tibial or acetabular component immersed in body ﬂuids, thus making the metallic parts susceptible to tribocorrosion. Micro-motions occur at points of ﬁxation leading to debris and ion release by fretting corrosion. b-Titanium alloys are potential biomaterials for joint prostheses due to their biocompatibility and compatibility with the mechanical properties of bone. The biotribocorrosion behavior of Ti–29Nb–13Ta– 4.6Zr was studied in Hank’s balanced salt solution at open circuit potential and at an applied potential in the passive region. Reciprocating sliding tribocorrosion tests were carried out against technical grade ultra high molecular weight polyethylene, while fretting corrosion tests were carried out against alumina. The wear of the alloy is insigniﬁcant when sliding against polyethylene. However, depassivation does take place, but the tested alloy showed an ability to recover its passive state during sliding. The abrasivity of the alloy depends on the electrochemical conditions of the contact, while the wear of polyethylene proceeds through third body formation and material transfer. Under fretting corrosion conditions recovery of the passive state was also achieved. In a fretting contact wear of the alloy proceeds through plastic deformation of the bulk material and wear resistance depends on the electrochemical conditions. Ó 2011 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. 1. Introduction Human joints operate by low-friction articular cartilage bearing surfaces, which are conforming and self-regenerating [1–3]. When natural joints are severely damaged, e.g. due to osteoarthritis, they are often replaced by artiﬁcial implants. In total joint replacement the implant components are generally made of metal–metal, metal–polymer, ceramic–ceramic or ceramic–polymer couples. Metal on polyethylene is a very common material coupling in total joint replacement . Knee and hip replacement joints involve a sliding contact at the articulation between the femoral component and the tibial or acetabular component during motion of the human body [5–7]. As a result the metallic components of the artiﬁcial joint are susceptible to sliding tribocorrosion (see Fig. 1). Tribocorrosion is the irreversible transformation of a material due to the simultaneous action of corrosion and wear taking place in a sliding tribological contact. It involves numerous synergy effects between mechanical and electrochemical phenomena, usually leading to an acceleration of material loss [8–10]. ⇑ Corresponding author. Tel.: +41 216932952. 1 E-mail address: [email protected]ﬂ.ch (N. Diomidis). Present address: Nuclear Power Corporation of India Ltd, Mumbai, India. Fretting corrosion is a particular form of tribocorrosion involving a small amplitude relative displacement or vibration, usually between surfaces that are meant to be ﬁxed to each other . In the particular case of orthopedic implants micro-motions are known to occur at points of ﬁxation , while corrosion is caused by the body ﬂuids, which contain various inorganic and organic ions and molecules  (see Fig. 1). Fretting corrosion has been identiﬁed at the stem/neck and neck/head contacts of modular implants, at the stem/bone and stem cement interfaces of cemented and uncemented implants, and at the screw/plate junction of ﬁxation plates . In contrast to sliding, during fretting a considerable part of the displacement may be accommodated in the contact by elastic deformation and thus the elastic properties of biomaterials can affect the implants behavior and functionality. Furthermore, due to the closed geometry of the contact electrolyte replenishment is difﬁcult and the electrochemical conditions might differ from those of the bulk. Furthermore, debris is easily trapped and thus the behavior of the third body is critical . The commonly used metallic biomaterials for orthopedic applications (stainless steel, titanium and CoCrMo alloys) owe their high corrosion resistance to the spontaneous formation of a passive surface oxide layer which forms the interface between the alloy and the environment [16,17]. The properties of surface ﬁlms dictate the results of chemical and mechanical interactions at the 1742-7061/$ - see front matter Ó 2011 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2011.09.034 N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 Fig. 1. A schematical representation of a total hip joint replacement prosthesis. The types of motion and surface degradation mechanisms of the implant metallic components are shown. interface. Both sliding tribocorrosion and fretting corrosion of passive metals lead to local damage or removal of the passive ﬁlm, as well as detachment of metal particles, leading to mechanical wear. As the counterbody moves on the depassivated surface area re-oxidizes, a process involving a charge transfer reaction at the interface which yields dissolved metal ions and a solid oxide. As a result in both tribocorrosion situations passive ﬁlm degradation induces wear-accelerated corrosion. The repeated removal of oxide ﬁlms produces particles and ions, which can result in adverse biological reactions and can lead to mechanical failure of the device [18–21]. Therefore, an approach combining both electrochemistry and tribology is a necessary means to study these complex phenomena and to assess the biocompatibility of candidate metallic materials. In recent years different electrochemical techniques have been combined with tribology  and methodologies have been developed  that allow the study of tribo-electrochemical systems. Two types of electrochemical tests are commonly combined with tribology: chronopotentiometry and potentiostatic polarization. During chronopotentiometric measurements the evolution of the open circuit potential (Eoc) of the sample is monitored before, during and after a mechanical perturbation. The open circuit 853 potential is the difference in electrical potential spontaneously established between the metal and the solution. Upon disruption of the passive surface ﬁlm the underlying metal is uncovered and the Eoc drops to more cathodic values, indicating the increased oxidation tendency of the exposed metal. In such a case the measured Eoc is a mixed potential resulting from galvanic coupling between passive and depassivated areas in the sample surface . On the other hand, during potentiostatic polarization tests one can control the potential and thus study the response of the material under different oxidizing conditions, such as in the case of local tissue inﬂammation when oxidizing agents will be present in the environment. The potential of the sample is externally imposed and the evolution of the current is monitored during sliding or fretting. In this case disruption of the passive surface ﬁlms leads to an increase in the measured anodic current originating from the accelerated oxidation reactions taking place at the uncovered metal surface . The scope of this work is to evaluate the combination of electrochemistry and tribology in the study of the wear behavior of biomedical alloys. For this, laboratory set-ups are used to test a newly developed b-titanium alloy in different tribological contacts in physiological solution. The selected counterparts were UHMWPE and alumina in order to simulate the ball/acetabular cup sliding contact and the femoral stem/ball fretting contact, respectively. 2. Materials and methods 2.1. Characteristics of the alloy In this work a titanium alloy with b microstructure (see Fig. 2), Ti–29Nb–13Ta–4.6Zr, was studied [26,27]. b-Alloys exhibit a modulus of elasticity considerably lower than commonly used Ti–6Al– 4V and Co–28Cr–6Mo alloys and thus decrease the risk of stress shielding of the femur . The added beneﬁt of using Nb, Zr and Ta as alloying elements is that they tend to form dense surface oxides, increasing the stability of the passive layer . The alloy was prepared by melting in a vacuum arc electric furnace (non-consumable) and casting in the form of pancakes, followed by forging. A vacuum arc electric furnace in which the charge can be melted in a water cooled copper crucible under vacuum (103 mbar) was employed. Prior to melting the furnace chamber was purged with argon gas once and then again ﬁlled with the gas up to 532 mbar. The charge was melted by arcing emitted from a tungsten bit brazed to a copper stringer rod suspended above the charge. A d.c. potential of up to 30–32 V and current of 1000 A was applied between the tungsten cathode and charge material, which formed the anode. A stirring coil around Fig. 2. (a) Optical micrograph and (b) X-ray diffraction spectrum of the Ti–29Nb–13Ta–4.6Zr alloy. 854 N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 the copper crucible stirred the melt pool and homogenized the melt composition. Each pancake was melted four times to produce a homogeneous composition. The pancake was then forged at a temperature of 900 °C in a forging press. The composition of the alloy measured by inductively coupled plasma optical emission spectrophotometry is 27.2 wt.% Nb, 11.5 wt.% Ta, 4.8 wt.% Zr, 0.05 wt.% S, 0.05 wt.% C, and 230 ppm H2. The Ti–29Nb–13Ta– 4.6Zr alloy has a hardness of 249 MPa and an elastic modulus of 55 GPa. Cylindrical specimens were cut with dimensions of 20 mm diameter and 5 mm thickness. Sample preparation was done by progressively grinding with emery papers of 240, 400 and 500 grit, followed by ﬁne polishing with diamond paste (6 lm) and ﬁnal polishing with alumina powder (3 lm), leading to an Ra of 0.05 lm. The specimens were then washed in deionized water and ultrasonically cleaned in acetone and ethanol for 10 min each. 2.2. Experimental conditions All tests were done in Hank’s balanced salt solution (HBSS), consisting of NaCl (8.00 g l1), KCl (0.40 g l1), CaCl2 (0.18 g l1), NaHCO3 (0.35 g l1), NaHPO42H2O (0.48 g l1), and MgCl26H2O (0.10 g l1) at 37 ± 1 °C. A Wenking LB 95 L Auto Range laboratory Potentiostat served to control the potential of the titanium specimen disk (working electrode). A platinum wire and a mercury sulfate electrode (MSE) were connected to the potentiostat as counter and reference electrodes, respectively. Experiments were performed at open circuit potential (Eoc) and at an applied potential of +2 V vs. MSE corresponding to the passive region (Epass), as shown in Fig. 3. For tests at open circuit potential the samples were immersed in the electrolyte 1800 s before the initiation of sliding. For tests at a passive applied potential the samples were immersed in the solution at rest potential for 300 s and then the anodic potential was applied for 1800 s prior to the initiation of sliding. Reciprocating sliding tribocorrosion tests (Fig. 4) were carried out with a sliding tribo-electrochemical apparatus, details of which are given in a separate publication . Pins with a hemispherical end of 12 mm diameter and 38 mm length were made up of technical grade UHMWPE and were used as counterbodies. The applied load was 6.5 N, at a frequency of 1 Hz, with a stroke length of 5 mm for 3600 s. This resulted in an initial maximum Hertzian contact pressure of 23 MPa. The wear of the hemispherical tip of the polymer pin was determined by measuring the ﬂat end formed during rubbing on optical micrographs and geometrically calculating the corresponding volume of the worn spherical cup. Fretting corrosion tests (Fig. 4) were carried out with a fretting tribo-electrochemical apparatus described in detail elsewhere . Alumina balls of 10 mm diameter (SWIP AGBrügg, G10 AFBMA ﬁnish) were used as counterbodies. Fretting corrosion experiments were carried out at an applied normal load of 10 N, with a displacement of 100 lm and a frequency of 1 Hz was applied for 3600 s. This resulted in an initial maximum Hertzian contact pressure of 400 MPa. The vertical position of the counterbody was monitored using a Keyence LC2420 laser distance meter at a resolution of 0.01 lm. The mean frictional coefﬁcient (l) was calculated by dividing the tangential force by the normal force when the ball was in the middle of the stroke. Laser proﬁlometer (UBM Telefokus UBC14) surface scans were performed at a resolution of 300 points mm1 and the wear volumes were calculated according to Fouvry et al. . Scanning electron microscopy (Philips XL30FEG SEM) and optical microscopy were used to characterize the morphology of the wear tracks. 3. Results 3.1. Sliding tribocorrosion The evolution of the coefﬁcient of friction and the potential with time during sliding tribocorrosion experiments on Ti–29Nb–13Ta– 4.6Zr at open circuit potential are shown in Fig. 5. The coefﬁcient of friction goes through a run-in period of about 700 cycles and then a steady-state value of around 0.3 is achieved. Before the initiation of sliding the measured open circuit potential reﬂects the presence of a passive ﬁlm on the alloy surface in contact with the electrolyte Fig. 3. Electrochemical potentio-dynamic polarization of Ti–29Nb–13Ta–4.6Zr in HBSS. Fig. 4. Schematic diagram of the tribological test systems. Fig. 5. Evolution of the coefﬁcient of friction and the open circuit potential during sliding tribocorrosion experiments at Eoc. 855 N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 Table 1 The wear volume of UHMWPE pins after sliding tribocorrosion experiments on Ti– 29Nb–13Ta–4.6Zr at open circuit potential and at a passive applied potential. Electrochemical condition Wear volume (104 mm3) Eoc Epass 4.08 ± 1.12 1.62 ± 0.16 transfer ﬁlm is detected at open circuit potential (Fig. 7), indicating that polyethylene transfer is not responsible for the measured electrochemical response. The wear of the UHMWPE pin after tribocorrosion tests at different potentials is shown in Table 1. 3.2. Fretting corrosion Fig. 6. Evolution of the coefﬁcient of friction and the anodic current, during tribocorrosion experiments at a passive applied potential. The dotted line indicates the background current in the absence of sliding, calculated according to Mischler et al. . . Immediately upon the initiation of sliding the open circuit potential decreases, indicating a depassivation of the surface induced by mechanical removal of the passive ﬁlm. For a certain period of time during sliding the open circuit potential remains around that low value, indicating that a depassivated state prevails on a part of the surface. During this period the potential exhibits slight variations of about 75 mV resulting from local depassivation–repassivation phenomena. After about 700 s of sliding the open circuit potential suddenly increases until it reaches values similar to that measured before the initiation of sliding. This is due to regrowth of the passive ﬁlm in the wear track, indicating that at open circuit the alloy has the ability to recover its passive state while mechanical perturbation is taking place. The evolution of the coefﬁcient of friction and the anodic current (i) with time during sliding tribocorrosion experiments on Ti–29Nb–13Ta–4.6Zr at a passive applied potential are shown in Fig. 6. At Epass a steady-state coefﬁcient of friction value of around 0.3 is achieved after a run-in period. An increased anodic current is measured at the onset of sliding as a result of depassivation. During the course of the experiment a considerably increased current is measured during running in. Then it gradually decreases until the measured current is of the same order as that expected without sliding. This indicates that the material has the ability to regain its passive state during sliding at Epass. After sliding tribocorrosion experiments both the alloy and the UHMWPE counterbody were examined for wear. No measurable wear was found on the metallic alloy samples by non-contact proﬁlometry. A surface microstructure characteristic of polyethylene transfer covering a part of the wear track is found at Epass, but no Acquisition of transient values during fretting allows the plotting of fretting logs, i.e. three-dimensional graphical representations of the time evolution of the frictional force–displacement loops. Such diagrams are shown in Fig. 8 for fretting corrosion experiments carried out at open circuit and a passive potential. The fretting log diagrams have an open trapezoidal shape, indicating that fretting follows a gross slip regime. Both elastic deformation and slip at the interface contribute to accommodation of the imposed displacement. Elastic accommodation of 26 and 32 lm of the imposed 100 lm occurs at Eoc and Epass, respectively. The evolution of the coefﬁcient of friction and potential with time during fretting corrosion experiments on Ti–29Nb–13Ta– 4.6Zr at open circuit potential are shown in Fig. 9. A steady-state coefﬁcient of friction value of 0.65 is measured during fretting at Eoc. Regarding the evolution of the open circuit potential, a similar response to that measured under sliding tribocorrosion conditions is revealed. The open circuit potential decreases upon the initiation of fretting. However, the potential drop (60 mV) is smaller than the potential drop measured upon the initiation of sliding (200 mV), since the depassivated area under fretting is smaller than under sliding due to the considerably smaller imposed displacement (100 lm vs. 5 mm). After only a few hundred cycles it rises again to reach values similar to those measured before the onset of fretting. Thus passive state recovery is achieved during fretting corrosion experiments at Eoc. The evolution of the coefﬁcient of friction, the anodic current and the vertical position of the counterbody with time during fretting corrosion experiments on Ti–29Nb–13Ta–4.6Zr at a passive applied potential are shown in Fig. 10. After a run-in period a steady-state coefﬁcient of friction value of around 0.6 is measured during fretting at Epass. For fretting corrosion tests at Epass the anodic current rises immediately on the initiation of fretting, indicating depassivation in the wear track. After about 2500 s of fretting the excess current decreases, indicating passive state recovery in the wear track. Thus passive state recovery is achieved during fretting Fig. 7. Scanning electron micrographs of the wear tracks on Ti–29Nb–13Ta–4.6Zr after tribocorrosion testing at (a) Eoc and (b) Epass. 856 N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 Fig. 8. Fretting log diagrams of Ti–29Nb–13Ta–4.6Zr at (a) open circuit potential and (b) a passive applied potential. value when indentation stops and passivity is recovered, indicating a close correlation between the mechanical, electrochemical and wear phenomena in the contact. After the fretting corrosion experiments both the alloy and the alumina counterbody were examined for wear. No measureable wear of the alumina was found, with only a few debris particles from the metal being found adhering to the ball. In contrast, wear does take place on the titanium alloy samples at both open circuit potential and at a passive potential. At both potentials a similar microstructure of the wear track largely characterized by plastic deformation and mixing of the third body is obtained (Fig. 11). A large number of debris particles were found surrounding the wear track. The wear of the alloy after fretting corrosion tests at different potentials is shown in Table 2. 4. Discussion Fig. 9. Evolution of the coefﬁcient of friction and the potential during fretting corrosion experiments at open circuit potential. 4.1. Inﬂuence of electrochemical conditions on overall wear at Epass. From Fig. 10 it can be seen that the excess current during fretting is measured only while the counterbody moves downwards, indenting the test sample. The excess current decreases to almost zero at the end of indentation of the sample. As a result, recovery of the passive state on the sample surface during fretting is accompanied by zero wear. Interestingly, the run-in period of the coefﬁcient of friction during which large variations are measured coincides with the duration of indentation of the sample by the counterbody. The coefﬁcient of friction reaches a steady-state When sliding is imposed on a Ti–29Nb–13Ta–4.6Zr/UHMWPE contact wear of the polymer occurs but no measurable wear of the alloy is found. The wear of the polymer differs depending on the imposed potential, indicating that the abrasivity of the alloy is dependent on the oxidative conditions prevailing in the contact zone. The wear volume of the polymer pin at Eoc is considerably larger that the wear volume measured after sliding tests at Epass. Additionally, the scanning electron microscopic examination indicated the presence of a polyethylene transfer ﬁlm at Epass but not at Eoc. As a result a three-body contact exists at Epass, in contrast to Fig. 10. Evolution of the coefﬁcient of friction, the anodic current and the vertical position of the counterbody during fretting corrosion experiments in Ti–29Nb–13Ta–4.6Zr at a passive applied potential. The dotted line indicates the background current in the absence of fretting, calculated according to Mischler et al. . N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 Fig. 11. Representative scanning electron micrograph of the wear track after fretting corrosion of Ti–29Nb–13Ta–4.6Zr at Epass. Table 2 The wear volume of the Ti–29Nb–13Ta–4.6Zr samples after fretting corrosion experiments at open circuit potential and at a passive applied potential. Electrochemical conditions Wear volume (104 mm3) Eoc Epass 9.87 ± 1.02 4.35 ± 0.02 857 depth of the material lost due to chemical wear is 43 nm. In reality this depth is expected to be even smaller, since the contact area and thus the wear track width increase in size with wear of the polyethylene pin. This conﬁrms that the material lost due to wear-accelerated corrosion is indeed very small and comparable with the initial surface roughness, as was seen from microstructural examination of the surface. Applying a similar calculation to the fretting corrosion data of Fig. 10 a 9 106 mm3 volume of metal is lost due to wearaccelerated corrosion. This represents only about 2% of the total worn volume, or an increase in the depth of the wear track of about 0.1 nm. This indicates that the mechanical character of wear predominates during fretting, in accordance with the smaller wear track area and higher contact pressure compared with sliding tribocorrosion tests. The above mentioned calculation highlights the different prevailing material removal mechanisms depending on the tribological contact. Under sliding conditions an electrochemical oxidation degradation mechanism predominates, accelerated by mechanical removal of the protective corrosion products. Under fretting mechanical removal of bulk material is the dominant degradation mechanism. 4.3. Recovery of the passive state Eoc, at which an alloy–polymer two-body contact predominates. The above mentioned results reveal that the presence of the UHMWPE transfer ﬁlm, which protects the polyethylene pin from further wear, is inﬂuenced by the electrochemical conditions in the contact. In the case of the fretting corrosion experiments a different situation prevails. This is not surprising since alumina is much harder than the metal while UHMWPE is softer. Under fretting the alumina counterbody shows no indication of wear, while measurable wear of the Ti alloy is found. In this case also the electrochemical conditions inﬂuence the total wear volume. A smaller amount of wear is measured after tests done at a passive potential than at Eoc. This could be explained by the different structure and chemical composition, as well as the greater thickness of the passive layer at higher potentials [29,34,35]. However, after a certain period of fretting wear of the alloy stops. 4.2. Wear-accelerated corrosion Even though no wear of the metallic alloy can be measured, the electrochemical parameters (potential and current) measured during sliding reveal that depassivation of the surface does take place. Depassivation will lead to material loss since the uncovered metallic surface will spontaneously re-oxidize, releasing metal ions, indicating synergy between mechanical and electrochemical phenomena. In the case of tests at an applied potential the excess current due to sliding can be measured (see Fig. 6) and the amount of oxidized material can be calculated using Faraday’s law: V¼ ItM nFd ð1Þ where I is the excess current due to rubbing, t is the duration of rubbing, M is the atomic mass of the metal, n is the metal oxidation valence, F is the Faraday constant, and d is the density. For tests done at open circuit potential such a calculation cannot be made because no external current circulates. If the measured excess current is assumed to result from the oxidation of Ti to TiO2 a total of 1.61 104 mm3 of metallic material is lost due to wearaccelerated corrosion. Assuming that the size of the wear track can be approximated by multiplying the initial Hertzian contact diameter (740 lm) with the stroke length (5 mm), then the average When sliding tribocorrosion experiments are carried out at either open circuit or applied potential the electrochemical response of the alloy during sliding reaches values expected without sliding. This indicates that the alloy tends to regain its passive state despite mechanical perturbation. The ability of an alloy to regain its passive state while sliding is a critical property in orthopedic prostheses, since the release of metallic ions due to wear-accelerated corrosion will be limited during the lifetime of the implant. For a material to be able to recover its passive state during sliding after depassivation has taken place it is necessary that depassivation ceases, since the repassivation rate of the metal does not change considerably during the course of the experiment. In order for this to happen the stress acting on the passive surface needs to be lower than the critical value required for passive ﬁlm removal. Thus the nature of the passive ﬁlm and particularly adhesion to the substrate and the ability to resist delamination are critical properties. In the present experiments the friction coefﬁcient is stable and does not change considerably with potential. Thus the critical parameter for passive ﬁlm removal is the contact pressure. Since the metallic alloy shows very little wear the evolution of the contact pressure from the initially applied 23 MPa during the course of the experiment depends on wear of the polyethylene. For sliding tests carried out at Eoc, at which a thin passive layer is expected, depassivation takes place immediately upon initiation of sliding. On the other hand, for tests at Epass, at which a thicker passive ﬁlm is expected , apart from an initial current peak, the anodic current does not increase signiﬁcantly until a few hundred cycles have occurred. This indicates that the thickness of the passive layer is also a critical parameter for depassivation. In the case of fretting corrosion an initial maximum contact pressure of 400 MPa is applied. During the course of the test only the metallic alloy sample wears, while the alumina counterbody is unaffected. As a result, the evolution of the pressure in the contact depends on wear of the alloy. The link between the wear volume and the recovery of passivity has been demonstrated for a b-Ti–13Nb–13Zr alloy . According to Fig. 10 depassivation of the alloy stops at the moment when wear also stops. Since wear of the alloy stops the size of the wear track measured at the end of the test can be assumed to be the same as at the moment of passivity recovery. Measuring the size of the wear track allows approximate calculation of the pressure at the moment of passivity 858 N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859 of conditions where corrosion is signiﬁcant or not. Such phenomena have a high clinical relevance, and thus the tribo-electrochemical approach could be beneﬁcially applied to conditions more representative of the in vivo situation (e.g. hip and knee simulators). Tribo-electrochemical techniques have shown that Ti–29Nb– 13Ta–4.6Zr recovers a passive surface state under both sliding and fretting contacts in a physiological solution. This is a critical property for biomedical applications, which if taken into account in alloy development and implant design could lead to decreased material loss and increased biocompatibility. Fig. 12. Schematic representation of the loading conditions and the depassivation mechanism during sliding and fretting tests. recovery. The critical pressure for depassivation in the different fretting corrosion tests is in the range 100–200 Mpa, as shown in Fig. 12. Such a pressure is considerably larger than the critical pressure for passive ﬁlm removal found under sliding, which is lower than 23 MPa. This difference in critical pressure for depassivation can be attributed to different depassivation mechanisms. Under sliding no wear or plastic deformation of the alloy takes place according to Fig. 7. Depassivation during the early stages of sliding proceeds by passive ﬁlm delamination. On the other hand, under fretting the application of a contact pressure higher than the hardness of the alloy results in plastic deformation of the metallic material and in the production of third body particles. This plastic deformation of the material in the wear track is responsible for depassivation. As fretting proceeds mechanical mixing and compaction of the surface oxides takes place. Furthermore, plastic deformation induces the formation of a nanocrystalline tribologically transformed structure in the immediate subsurface which is considerably harder than the base alloy . The presence of such a structure could be conﬁrmed by nanoindentation measurements. The phenomenon of passivity recovery during sliding and fretting discussed above appears to be a characteristic of the alloy. Other b-titanium alloys have exhibited the capability to recover a passive state under either sliding or fretting conditions [26,27]. However, only Ti–29Nb–13Ta–4.6Zr consistently recovered passivity under both sliding and fretting. The typically used Ti6Al4V does not recover passivity when tested in the same fretting tribometer under similar conditions [11,14]. 5. Conclusions A tribo-electrochemical approach has been applied to the study of the biotribocorrosion behavior of a newly developed b-titanium alloy for biomedical applications in simulated body ﬂuid in fretting and sliding contacts. The combination of electrochemistry and tribology has been demonstrated to constitute a powerful tool for the characterization of biomaterials for joint replacement. Electrochemical conditions are crucial in the wear of materials for biomedical applications. Under sliding the electrochemical conditions can inﬂuence the behavior of the third body and the build-up of polyethylene transfer ﬁlms. Under fretting electrochemical parameters can be used as an in situ measure of wear. Tribo-electrochemical tests can quantify the amount of depassivation and the abrasivity of materials. Combining tribology with electrochemistry allows monitoring of passive ﬁlm removal and regrowth, and the identiﬁcation Appendix A:. Figures with essential color discrimination Certain ﬁgures in this article, particularly Figs. 1, 3–6, 9, 10 and 12, are difﬁcult to interpret in black and white. The full color images can be found in the on-line version, at doi:10.1016/ j.actbio.2011.09.034. References  Podsiadlo P, Kuster M, Stachowiak GW. Numerical analysis of wear particles from non-arthritic and osteoarthritic human knee joints. Wear 1997;210:318–25.  Dowson D. A comparative study of the performance of metallic and ceramic femoral head components in total replacement hip joints. Wear 1995;190:171–83.  Podsiadlo P, Stachowiak GW. 3-D imaging of surface topography of wear particles found in synovial joints. Wear 1999;230:184–93.  Wilches LV, Uribe JA, Toro A. Wear of materials used for artiﬁcial joints in total hip replacements. Wear 2008;265:143–9.  Long M, Rack HJ. Titanium alloys in total joint replacement – a materials science perspective. Biomaterials 1998;19:1621–39.  Jasty M. Clinical reviews: particulate debris and failure of total hip replacements. J Appl Biomater 1993;4:273–6.  Niinomi M. Recent metallic materials for biomedical applications. Metall Mater Trans A 2002;33:477–86.  Diomidis N, Celis JP, Ponthiaux P, Wenger F. Tribocorrosion of stainless steel in sulfuric acid: identiﬁcation of corrosion-wear components and effect of contact area. Wear 2010;269:93–103.  Landolt D, Mischler S, Stemp M, Barril S. Third body effects and material ﬂuxes in tribocorrosion systems involving a sliding contact. Wear 2004;256:517–24.  Iwabuchi A, Lee JW, Uchidate M. Synergistic effect of fretting wear and sliding wear of Co-alloy and Ti-alloy in Hank’s solution. Wear 2007;263:492–500.  Barril S, Mischler S, Landolt D. Electrochemical effects on the fretting corrosion behaviour of Ti6Al4V in 0.9% sodium chloride solution. Wear 2005;259:282–91.  Windler M, Klabunde R. Titanium for hip and knee prostheses. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors. Titanium in Medicine. Berlin: Springer-Verlag; 2001. p. 703–46.  Hallab NJ, Jacobs JJ. Orthopedic implant fretting corrosion. Corros Rev 2003;21:183–213.  Hiromoto S, Mischler S. The inﬂuence of proteins on the fretting-corrosion behaviour of a Ti6Al4V alloy. Wear 2006;261:1002–11.  Diomidis N, Mischler S. Third body effects on friction and wear during the fretting of steel contacts. Tribol Int: in press. doi:10.1016/j.triboint.2011. 02.013.  Virtanen S, Milosev I, Gomez-Barrena E, Trebse R, Salo J, Konttinen YT. Special modes of corrosion under physiological and simulated physiological conditions. Acta Biomater 2008;4:468–76.  Milosev I. Metallic materials for biomedical applications: laboratory and clinical studies. Pure Appl Chem 2011;83:309–24.  Jones FH. Teeth and bones: applications of surface science to dental materials and related biomaterials. Surf Sci Rep 2001;42:75–205.  Marino CEB, Mascaro LH. EIS characterization of a Ti-dental implant in artiﬁcial saliva media: dissolution process of the oxide barrier. J Electroanal Chem 2004;568:115–20.  Hodgson AWE, Mueller Y, Forster D, Virtanen S. Electrochemical characterization of passive ﬁlms on Ti alloys under simulated biological conditions. Electrochim Acta 2002;47:1913–23.  Merritt K, Brown SA. Biological effects of corrosion products from metals. In: Fraker AC, Grifﬁn CD, editors. Corrosion and degradation of implant materials, STP 859. West Conshohocken, PA: ASTM International; 1983. p. 195–207.  Mischler S. Triboelectrochemical techniques and interpretation methods in tribocorrosion: a comparative evaluation. Tribol Int 2008;41:573–83.  Diomidis N, Celis JP, Ponthiaux P, Wenger F. A methodology for the assessment of the tribocorrosion of passivating metallic materials. Lubr Sci 2009;21:53–67. N. Diomidis et al. / Acta Biomaterialia 8 (2012) 852–859  Diomidis N, Göckan N, Ponthiaux P, Wenger F, Celis JP. Assessment of the surface state behavior of Al71Cu10Fe9Cr10 and Al3Mg2 complex metallic alloys in sliding contacts. Intermetallics 2009;17:930–7.  Stemp M, Mischler S, Landolt D. The effect of mechanical and electrochemical parameters on the tribocorrosion rate of stainless steel in sulphuric acid. Wear 2003;255:466–75.  More NS, Diomidis N, Paul SN, Roy M, Mischler S. Tribocorrosion behavior of b titanium alloys in physiological solutions containing synovial components. Mater Sci Eng C 2011;31:400–8.  Diomidis N, Mischler S, More NS, Roy M, Paul SN. Fretting-corrosion behavior of b titanium alloys in simulated synovial ﬂuid. Wear 2011;271: 1093–102.  Pilliar RM. Metallic biomaterials. In: Narayan R, editor. Biomedical materials. Berlin: Springer Verlag; 2009. p. 41–81.  Diomidis N. The surface degradation of metal joints. In: Affatato S, editor. Wear of Orthopaedic Implants and Artiﬁcial Joints. Cambridge: Woodhead Publishing, in press. 859  Stojadinović J, Bouvet D, Declercq M, Mischler S. Effect of electrode potential on the tribocorrosion of tungsten. Tribol Int 2009;42:575–83.  Barril S, Debaud N, Mischler S, Landolt D. A tribo-electrochemical apparatus for in vitro investigation of fretting-corrosion of metallic implant materials. Wear 2002;252:744–54.  Fouvry S, Kapsa Ph, Zahouani H, Vincent L. Wear analysis in fretting of hard coatings through a dissipated energy concept. Wear 1997;203(/204):393–493.  Mischler S, Barril S, Landolt D. Fretting corrosion behavior of Ti-6Al-4V/PMMA contact in simulated body ﬂuid. Tribology 2009;3:16–23.  Milosev I, Metikos-Hukovic M, Strehblow HH. Passive ﬁlm on orthopaedic TiAlV alloy formed in physiological solution investigated by X-ray photoelectron spectroscopy. Biomaterials 2000;21:2103–13.  Karthega M, Raman V, Rajendran N. Inﬂuence of potential on the electrochemical behaviour of b titanium alloys in Hank’s solution. Acta Biomaterialia 2007;3:1019–23.  Sauger E, Fouvry S, Ponsonnet L, Kapsa Ph, Martin JM, Vincent L. Tribologically transformed structure in fretting. Wear 2000;245:39–52.