Effect of angular displacement amplitude on the torsional fretting corrosion behavior of CoCrMo alloy in different synovial fluid

Abstract

The T–θ curves, friction torque curves, impedance and morphology were measured to research the effect of angular displacement amplitude on the torsional fretting corrosion behavior of the CoCrMo alloy in different synovial fluids. The results showed that the torsional fretting process could be divided into a running-in stage, a fluctuate stage and a steady-state stage both in NS and FBS. When the friction process was stable, the twisting fretting process of CrCoMo alloy was in the partial slip stage at θ = 2° and in the gross slip stage at θ = 6°, and the friction energy dissipation increased with the increase of θ. The results of electrochemical impedance spectroscopy showed that the impedance of the CoCrMo alloy decreased with the increase of θ, and when θ was larger, the decrease of impedance was more obvious. Also, the decrease of impedance in FBS was more obvious than that in NS. The surface roughness of CoCrMo after wear was enlarged with the increase of θ. The wear zone was extended to the central parts from the edge of the contact area, and wear was also severe in the central parts at a large value of θ. In NS, there was an obvious curved furrow at the edge of wear zone, and the roughness was much larger than that in FBS.

---

1. Introduction

As important components of biological material, artificial joint materials have been developed rapidly in recent years. Commonly used artificial joint materials are metallic materials, ceramic materials¹ and polymer materials.² Compared with other materials, metallic materials have always been the most common materials for artificial joints due to their high strength, high toughness, and so on.³ At present, clinical applications of combined artificial hip joints are becoming more and more frequent, and the CoCrMo alloys⁴ have been widely used for their good biocompatibility and high friction resistance.

Fretting corrosion will be produced by the combined action of fretting wear and corrosion into body fluid in each link interface of the artificial joint material, such as the link interface between the joint head and the joint handle, the bone screw and the bone plate, the joint and the bone cement, and so on.^5–14 The CoCrMo–Ti₆Al₄V alloy hip joint was the most typical component implant with CoCrMo alloy as the joint head and Ti₆Al₄V alloy as the joint handle, while the study of J. S. Kawalec et al.⁵ and Viceconti et al.⁶ showed that the matching pair of these two alloys was sensitive to fretting corrosion. Ocran et al.⁷ studied the corrosion and fretting corrosion of medical grade CoCrMo alloy in a clinically relevant simulated body fluid environment with a variety of electrochemical characterization techniques and tribological experiments. Swaminathan et al.⁸ studied the effects of potential and frequency on the fretting corrosion of Ti₆Al₄V and CoCrMo surfaces. In the majority of the potential range tested, the fretting corrosion behavior of CoCrMo/Ti₆Al₄V and CoCrMo/CoCrMo was similar and was dominated by the CoCrMo surface. Contu et al.⁹ studied the corrosion behavior of a CoCrMo implant alloy during fretting in bovine serum. The result showed that the corrosion current of CrCoMo alloy was significantly lower in serum both at pH 4.0 and 7.0. In addition, the biological electrolyte increases the repassivation rate at neutral pH. Swaminathan et al.¹⁰ also studied the fretting corrosion of CoCrMo and Ti₆Al₄V interfaces. To demonstrate the capabilities of the new system and validate the proposed model, experiments were performed to understand the effect of the applied normal load on the fretting corrosion performance of Ti₆Al₄V/Ti₆Al₄V, CoCrMo/Ti₆Al₄V, and CoCrMo/CoCrMo material couples under potentiostatic conditions with a fixed starting surface roughness. The results of this study showed that fretting corrosion is affected by the material couples, the normal load and the motion conditions at the interface. In addition, Geringer et al.^11–13 summarized the friction/fretting corrosion mechanisms of implants, and proposed current trends and outlooks for implants. Meanwhile, Brown et al.⁵ (1995) found that fretting corrosion can accelerate the crevice corrosion of modular hip tapers. Human tissues will be adversely affected when the wear debris and corrosion products accumulate to reach a certain amount in the body, and a large amount of wear debris and corrosion products will lead to the loss of implants in the clinic.¹⁴

In this process, there is still a hidden friction between the implants, in particular, the fretting friction in the torsional process.^15–20 Although some scholars^21–23 studied the torsional fretting friction behavior of different metal ball heads against a UHMWPE socket, there were few reports on the torsional fretting behavior of the CoCrMo alloy. In addition, most of the torsional tests did not take into account the effects of the liquid environment. This paper will discuss the torsional fretting corrosion mechanism of the CoCrMo alloy under different angular displacement amplitudes in different medium environments. The experiments can provide a certain theoretical basis and experimental support for the torsional fretting corrosion behavior of the CoCrMo alloy in an actual implant, and the results can reveal the electrochemical behavior, fretting behavior and damage mechanism during the process of torsional fretting corrosion.

2. Experimental procedure

A Ti₆Al₄V alloy ball and a CrCoMo alloy plane were used as the friction couple to discuss the effect of different angular displacement amplitudes (θ) and synovial fluids on the torsional fretting corrosion mechanism of the CoCrMo alloy. The chemical composition and physical mechanical properties of the Ti₆Al₄V alloy ball and the CoCrMo alloy plane are shown in Tables 1 and 2. The diameter of the Ti₆Al₄V alloy ball with a mirror surface is 5 mm, and the size of the CoCrMo plane with a surface of R_a = 0.04 μm (roughness) is Φ 22 mm × 7 mm. A metallographic mosaic test machine was used to package the CoCrMo alloy plane sample, as shown in Fig. 1. Normal saline (NS, 0.9% NaCl)²⁴ and fetal bovine serum (FBS, 25% volume of calf serum)²⁵ were respectively selected as the test media.

Table 1 Chemical composition of Ti₆Al₄V alloy and CoCrMo alloy (in wt%)

Ti6Al4V	Element	Ti	Al	V	Fe	C	O	N	H
	Percentage	89.729	6.05	3.95	0.11	0.014	0.14	0.006	0.001
CrCoMo	Element	Co	Cr	Mo	Ni	Fe	Mn	Si	C	N
	Percentage	63.25	28.10	5.90	0.88	0.65	0.63	0.22	0.25	0.12

---

Table 2 Physical mechanical properties of Ti₆Al₄V alloy and CoCrMo alloy

Material	Tensile strength	Yield strength	Elongation ratio	Elastic modulus	Poisson ratio	Hardness
Ti6Al4V	950 MPa	860 MPa	0.12	110 GPa	0.3	348 HV
CoCrMo	970 MPa	660 MPa	0.18	200 GPa	0.3	321 HV

---

Fig. 1 Sample used to investigate tribology corrosion.

The torsional fretting corrosion test was carried out in the UMT experiment concerted with a high precision low speed reciprocating rotary table, as shown in Fig. 2. The experimental apparatus includes four parts: a positioning and loading system, a twist drive system, a twist control system and a data acquisition system. The test parameters were set after the samples were installed. The applied pressure was 100 N, the twist angle displacement amplitude (θ) was respectively set to 2°, 4° and 6°, the fretting frequency was 1 Hz, the process of torsional fretting was reciprocating motion and the cycle time was 4800 s. A CS type electrochemical workstation was used to measure the electrochemical impedance spectroscopy in the torsional fretting test, a Ag/AgCl electrode was used as the reference electrode, and a 10 × 10 × 0.1 mm Pt electrode (99.99%) was used as the auxiliary electrode. The impedance frequency range was 10⁵ to 10⁻² Hz, and the sine wave amplitude of AC impedance was 10 mV.^26,27 Each experiment consisted of three groups of parallel experiments.

Fig. 2 Schematic of torsional frictional tester, 1 upper sample, 2 lower sample, 3 rotary table, 4 support device, 5 vertical actuating device, 6 horizontal actuating device, 7 3-D force sensor, 8 torque sensor, 9 stepping motor, I. Axis of upper specimen, II. Axis of motor rotation.

The wear surface roughness and 3D morphology of the CrCoMo alloy plane were respectively measured by a JB-4C precision roughness tester and a Micro-XAM three-dimensional surface shape analyzer. Meanwhile the optical and micro morphology of the wear surface of the CrCoMo alloy plane were respectively observed using an optical microscope and a S-3000 scanning electron microscope (SEM).

3. Results

3.1 T–θ curves and time-varying friction torque curves

The values of friction torque value (T) and angular displacement amplitude (θ) gathered by the data acquisition system after the torsional fretting corrosion test are shown in Fig. 3. The data acquisition rate of the system was 20 times per second, and all the experimental data under different cycle times were processed by Matlab 7.0 to form the T–θ curves.

Fig. 3 Curves of friction torque and angular amplitude against the number of cycles.

The T–θ curves of torsional fretting under different conditions are shown in Fig. 4. As shown in Fig. 4a, the T–θ curve shows an almost horizontal line after 10 cycles. With the increase in cycles, the wear of the surface film and the direct contact of the two alloys results in the gradual expanse of the T–θ curve. According to the study of Zhang,^28,29 the relative motion state will be in the partial slip regime, mixed fretting regime, and gross slip regime when the T–θ curve shows a line, an ellipse and a parallelogram, respectively. In the process of the increase in cycles from 10 to 100, the shape of the T–θ curve switches from a line to a parallelogram, moreover, the relative motion state has changed to a gross slip regime from a partial slip regime. After 1000 cycles, the shape of the T–θ curve changes to an ellipse. The value of friction energy dissipation keeps a balanced state in the terminal stage of torsional fretting, which is equal to the area inside the loop of the T–θ curve. The shape of the T–θ curve in Fig. 4b is a parallelogram in the experimental preliminary stage, and the curve maintains the parallelogram shape throughout the total fretting process, which means that the relative motion state is in the gross slip regime. Furthermore, the size of parallelogram doesn’t vary obviously in the terminal stage, which means that the friction energy dissipation possesses a stable value. Moreover, as shown in Fig. 4a and b, the amplitude of the angular displacement is bigger, so the friction energy dissipation in a stable state is larger.

Fig. 4 T–θ curve of torsional fretting under different conditions.

From Fig. 4c (FBS, θ = 2°), it can be observed that the T–θ curve in FBS presents a similar trend to that in NS under the same displacement amplitude. The T–θ curve experiences the switch from a line to a parallelogram, then to an ellipse, and reaches a relatively stable state eventually. In the terminal period, the friction energy dissipation starts to be stable. Fig. 4d (FBS, θ = 6°) shows that the T–θ curve nearly forms a line in the initial stage, which differs from the experimental result of Fig. 4b.

Fig. 5 shows the friction torque time-varying curves under different test conditions. The initial torque value is smaller, and then gradually increased until the final torque values tended to be stable, as shown in Fig. 5a (NS, θ = 2°), which is consistent with the T–θ curve under this condition. The friction torque curve can be divided into three stages, the run-in stage (two body friction stage), the fluctuate stage (two body friction tending to three body friction), and the steady-state stage (the formation of the third body and the amount of solution erosion achieving a dynamic balance). Fig. 5b (NS, θ = 6°) shows that the value of friction torque increases rapidly at the beginning of the experiment, and the speed of increase is much faster than that of Fig. 5a. This phenomenon further indicates that the large θ can make the surface film quickly torn and damaged. After the value of friction torque reaches its peak, there is a process of gradual decline tending to a stable value. The torque change process in Fig. 5b can also be divided into the run-in stage, fluctuate stage and steady-state stage. Similar to Fig. 5a and b, Fig. 5c (FBS, θ = 2°) and Fig. 5d (FBS, θ = 6°) show that the friction torque time-varying curves also include the run-in stage, fluctuate stage and steady-state stage. As shown in Fig. 5, the value of the friction torque in NS is slightly higher than that in FBS for the same value of θ; meanwhile, the variation trend of the friction torque is closely related to the value of θ.

Fig. 5 Friction torque change over time under different conditions.

3.2 Electrochemical properties

Fig. 6 shows the Nyquist plots (Fig. 6a) and Bode plots (Fig. 6b) of the CoCrMo alloy in NS under different angular displacement amplitudes, in which the symbols represent the actual measured value and the solid lines indicate the fitted values. According to the study of Geng,³⁰ the classical R(CR) circuit is used to fit the electrochemical impedance data using Z-View software. The curves show a high consistency between the actual measured value and the fitted values both in the high frequency region and the intermediate frequency region, but not in the low frequency region due to the non ideal capacitor properties. The samples have a better corrosion resistance before friction, but the impedance values obviously decrease over the process of friction, while the magnitude of the decrease is at a relatively lower level than the increase of θ. The phase angle peak before friction is extended to the low frequency region from the medium frequency region (see Fig. 6b) which demonstrates the existence of an oxide film or a passivation film on the surface of the alloy. However, the left shift of the peak area in the process of friction reveals evidence for the destruction of the passive film, and the left shift of peak area increases when increasing θ, making specimen scraping serious and magnifying the potential difference with the surrounding non-friction region, which makes the impedance values decrease further.

Fig. 6 (a) Nyquist plots and (b) Bode plots of the CoCrMo alloy under different angular displacements in NS.

The Nyquist plots (Fig. 7a) and Bode plots (Fig. 7b) of the CoCrMo alloy in FBS with different angular displacement amplitudes are displayed in Fig. 7. The measured values are in good agreement with the fitted values over the whole range of scanning and the impedance of the friction is decreased with the increase of θ in comparison to that before friction. From Fig. 7b, with the increase of θ, the phase angle decreases and the peak area becomes narrower and shifts to the left, which indicates that the surface integrity of the specimen is destroyed. The exposure of metal in the abrasion area makes the region have low potential and low impedance, and this response allows the sustained decline of impedance as the amplitude of the angular displacement increases.³¹

Fig. 7 (a) Nyquist plots and (b) Bode plots of the CoCrMo alloy under different angular displacements in FBS.

3.3 Roughness and 3D morphology of surface

Fig. 8 shows the surface profile of the wear area of the CoCrMo alloy. The shapes under different conditions are almost W type, and the concave part of the wear area deepens gradually with the increase of θ. In addition, the concave part is gradually extended to the wear center, which makes the width of the wear region become larger. The variation in roughness of the wear region of the CoCrMo alloy under different angular displacement amplitudes is displayed in Fig. 9. The result shows that the increase of θ could aggravate the surface roughness of the sample, and the surface roughness in NS is larger than that in FBS under the same value of θ.

Fig. 8 Surface profile of the wear region of CoCrMo samples.

Fig. 9 Variation of contact surface roughness of CoCrMo alloys vs. different angular displacements in NS and FBS.

The three-dimensional morphology of the wear region of CoCrMo alloy under different media and different angular displacement amplitudes can be seen in Fig. 10. From Fig. 10a (FBS, θ = 2°), we can see slight damage in the middle of the wear region. In the process of a normal load, work hardening exists in the intermediate zone, and stress concentration may form easily at the edge. In such case, the surface may rupture under friction to form a trench, accompanied by many asperities. The plastic deformation will be strengthened at larger θ values (Fig. 10b, FBS, θ = 4°) to deepen the groove. If θ becomes even bigger (Fig. 10c, FBS, θ = 6°), serious wear and tear of the trench region will expand to the middle part. Fig. 10d (NS, θ = 6°) shows the generation of a furrow in the periphery of the severe wear zones. The damage in the middle part is lighter than that at the edge part because the external line speed of the Ti₆Al₄V alloy ball is greater than the internal line speed. The wear of the CoCrMo alloy in NS is more intense than that in FBS at the same value of θ.

Fig. 10 3D morphology of the wear region of the CoCrMo alloy under (a) θ = 2° in FBS, (b) θ = 4° in FBS, (c) θ = 6° in FBS and (d) θ = 6° in NS.

3.4 Morphology

Fig. 11 shows the optical morphologies of the CoCrMo alloy surface under different angular displacement amplitudes in different solutions. As shown in Fig. 11a (NS, θ = 2°) and d (FBS, θ = 2°), the wear of the peripheral part is serious while that of the central area is slight under the smaller value of θ (2°), which reflects the fretting process run in the partial slip regime. In addition, the wear in NS is more serious. As shown in Fig. 11b (NS, θ = 4°) and e (FBS, θ = 4°), the wear of the central region under greater angular displacement amplitudes (θ = 4°) is increased, which indicates that the surface in NS has closed to a gross slip regime, but the wear is slight in FBS although there are traces of slip on the surface of the CrCoMo alloy. In Fig. 11c (NS, θ = 6°) and f (FBS, θ = 6°), the wear region of the CrCoMo alloy under the largest value of θ (6°) is extended to the entire contact area, and the torsional fretting between the two alloys is running in the gross slip zone. The optical morphology properties of the wear zone are consistent with the motion state reflected in the T–θ curves.

Fig. 11 Optical morphology of the wear region of CoCrMo samples.

4. Discussion

CrCoMo alloys and Ti₆Al₄V alloys have been commonly used in joint replacement surgery, and they will produce mutual friction in the human body. The form of friction is very complicated and changeable, and torsional fretting is one of the most common forms. Although a lot of research^{15–18,22,23} on the torsional fretting of different artificial joint prosthesis has been done, these studies don’t take into account the impact of the human body environment. In addition, the process of torsional fretting can affect the electrochemical properties of the CrCoMo alloy in the simulated body fluid. According to the fitted curve of the impedance diagram, the polarization resistance diagram of the CoCrMo alloy at different angular displacement amplitudes is displayed in Fig. 12. We can see that the polarization resistance of the CoCrMo alloy is high in these two media, and decreases significantly with torsional motion of small θ values. However, the decrease of polarization resistance in NS is not as obvious as that in FBS. The root cause of this is that the surface film is very thin and easy to abrade, which results in the non-significant change of the electrochemical surface state with the increase of θ, and the lubrication and protection effect of FBS may protect the abrasion area from serious damage at small θ, but a large θ value (6°) could result in strong friction, causing a further decline of the polarization resistance by the complete contact of the metal matrix with the solution. According to the measurement of electrochemical impedance spectroscopy, the torsional fretting can greatly reduce the impedance value of the CrCoMo alloy, thus reducing the corrosion resistance of the alloy. The corrosion resistance of the CrCoMo alloy decreases with the increase of the angular displacement amplitude. In addition, the effect of torsional fretting on the corrosion speed of the CrCoMo alloy in FBS solution was stronger than that in NS solution.

Fig. 12 Polarization resistance of the CoCrMo alloy under different angular displacements in (a) NS and (b) FBS, respectively.

There are many factors that affect the torsional fretting behavior of artificial joint prosthesis, including the loading force, the corrosion environment and the angular displacement amplitude. This paper mainly discusses the influence of the angular displacement amplitude and the corrosion environment on the torsional fretting behavior of CrCoMo alloy. In FBS, the initial T–θ curve shape (Fig. 4a) is straight, and the friction torque is small and relatively stable. At this stage, the torsional fretting of the grinding pair is coordinated by the elastic deformation of the contact surface. The center of the contact area is adhesive and micro slip occurs at the contact edge, and thus fretting is running in the partial slip regime. With the increase in the number of cycles from 100 to 1000 under θ = 2° in NS, the shape of the T–θ curve changes from a parallelogram to an ellipse, which shows that the relative motion state changes from the gross slip regime to the mixed fretting regime. This result is contrary to Yi’s research.³² In Yi’s research, the initial T–θ curve was linear under small displacement amplitude, and with the increase of cycles, the T–θ curve gradually opened up into an elliptic shape, which was maintained to the end of the test. With the increase in the number of cycles, the relative motion state was not maintained as the gross slip regime in this paper. The T–θ curve gradually expands into an elliptic form which is maintained to the end of the experiment, which is due to the deformation of the material changing from the contact interface to the elastic deformation. The central region without wear and the boundary region with serious damage, shown in Fig. 11a, can be very good evidence to prove the above conclusion. A comparison between Fig. 4b and d indicates that FBS has better lubricating properties than NS, and the two alloys possess better corrosion resistance in FBS than in NS. These properties prompt the appearance of a dense passive film on the surface of the CrCoMo alloy, and the passive film will make the value of friction torque tiny. With the increase of the number of cycles, the T–θ curve expands gradually and to present a parallelogram shape, and eventually, the size of the shape becomes stable, the torsional fretting reaches a balanced state and the friction energy dissipation trends to be stable. A comparison between Fig. 4c and d indicates that the value of the friction energy dissipation is much larger under larger θ values. The damage rate of the surface film of the CrCoMo alloy can be accelerated under conditions of large θ in NS, which make the friction pair keep the gross slip regime at the beginning of the experiment. The wear rate of the CoCrMo surface film can be decelerated by the better lubrication performance of FBS. In addition, no matter in which fluid, the friction energy dissipation will increase with the increase of θ. The shape of the T–θ curve at θ = 2° in NS has the same trend of that in FBS, which shows that the influence of the different solutions under the same small displacement amplitude is very small with respect to the T–θ curve. In NS, the initial T–θ curve at θ = 6° has been turned into a parallel quadrilateral, which indicates that the torsional fretting was in the gross slip regime at the initial time. This is due to the increase of the angular displacement amplitude that can increase the contact area of the friction pair, rapidly destroy the surface of the matrix, and accelerate the torsional fretting corrosion speed. This conclusion has been also verified in ref. 15–18, 22 and 23. According to the friction torque change at different times under different conditions, the torsional fretting process consists of three stages. Cai¹⁷ indicated that the frictional torque had a close relationship with the imposed angular displacement amplitudes, namely that the frictional torque increased with the increased angular displacement amplitude. In addition, no matter what kind of solution, the friction torque under a large angular displacement amplitude experiences a rapid increase, but then falls in the run-in stage. The friction torque is related to the frictional force and the size of the contact zone. At the beginning of the test, the large angular displacement amplitude increases the contact surface area of the friction pair, and the friction torque increases rapidly. Subsequently, a large amount of elastic plastic deformation changes to plastic deformation, so the friction torque has a certain reduction.

Fig. 13 shows the SEM morphologies of the CoCrMo alloy surface under different angular displacement amplitudes and different solutions. Under smaller values of θ (Fig. 13a and d), the wear region is concentrated in the periphery of the contact zone, which characterizes typical partial slip; meanwhile, wear debris accumulation can observed around the wear scar. There is a certain number of stripping pits in the wear scar in FBS, and a massive peeling phenomenon may be seen around the wear scar. Under small displacement amplitudes, debris accumulation easily occurs around the wear area. As the displacement amplitude increases, the liquid flow is increased, and the wear debris is discharged from the contact area, thus the metal substrate is exposed and directly contacted with the grinding pair. The friction area gradually becomes larger with the increase in the contact area. With the further increase of the angular displacement amplitude, the wear of the whole contact area (especially the boundary region) is more serious. When θ = 6°, as shown in Fig. 13c (NS, θ = 6°) and f (FBS, θ = 6°), the wear area can be divided into three parts: the outermost region, the outer region and the inner region. Spall and debris can be found in the outermost region. Deep and massive furrows and the accumulation of corrosion products can be found in the outer region, which has a more serious wear and reflects the mechanism of abrasive wear. According to the early research on torsional fretting corrosion behavior,^16,17 the presence of wear debris and corrosion products in the inner region shows that the mechanism of material damage is mainly adhesion wear and corrosion wear.

Fig. 13 SEM photos of CoCrMo wear scars (A, outermost region, B, outer region and C, inner region).

5. Conclusions

(1) Friction torque can be classified into a running-in stage, a fluctuate stage and a steady-state stage. Twisting fretting is under partial slip condition at small angle displacement amplitudes (θ = 2°), and under gross slip regime at large angular displacement amplitudes (θ = 6°) at the end of the test.

(2) In different solutions, the impedance of the CoCrMo alloy decreases obviously with the increase of θ, and the degree of the decline in impedance is consistent with the increasing trend of θ. The decrease of impedance in FBS was more obvious than that in NS at the same θ value.

(3) In different solutions, the roughness of the CoCrMo alloy increases with the increase in θ, and the roughness in NS is higher than that in FBS. On the 3D morphology of the CrCoMo wear scar, the wear zone was extended to the central parts from the edge of the contact area, and wear was also severe in the central parts at a large θ value.

(4) From the SEM morphology of the CoCrMo wear scar, there was an obvious curved furrow at the edge of the wear zone in NS, and the roughness was much larger than that in FBS. The wear mechanism of torsional fretting in this paper was mainly the joint action of abrasive wear, corrosion wear and adhesive wear.

Acknowledgements

This paper was supported by the Doctoral Degree Teacher Research Support Program of JiangSu Normal University (No. 14XLR024), the National Natural Science Foundation of China (No. 51305177), and the National Natural Science Foundation of JiangSu Province (No. BK20130229).

References

1. J. Kusaka, K. Takashima and D. Yamane, et al., Fundamental study for all-ceramic artificial hip joint, Wear, 1999, 225(4), 734–742.
2. S. Ge, X. Kang and Y. Zhao, One-year biodegradation study of UHMWPE as artificial joint materials: variation of chemical structure and effect on friction and wear behavior, Wear, 2011, 271(9–10), 2354–2363.
3. T. F. Zhang, B. Liu and B. J. Wu, et al., The stability of DLC film on nitrided CoCrMo alloy in phosphate buffer solution, Appl. Surf. Sci., 2014, 308(30), 100–105.
4. M. Mori, K. Yamanaka and S. Sato, et al., Texture evolution and mechanical anisotropy of biomedical hot-rolled Co–Cr–Mo alloy, J. Mech. Behav. Biomed. Mater., 2015, 51, 205–214.
5. J. S. Kawalec, S. A. Brown and J. H. Payer, et al., Mixed-metal fretting corrosion of Ti₆Al₄V and wrought cobalt alloy, J. Biomed. Mater. Res., 1995, 29(7), 867–873.
6. M. Viceconti, O. Ruggeri and A. Toni, et al., Design-related fretting wear in modular neck hip prosthesis, J. Biomed. Mater. Res., 1996, 30(2), 181–186.
7. E. K. Ocran, L. E. Guenther and J. M. Brandt, et al., Corrosion and Fretting Corrosion Studies of Medical Grade CoCrMo Alloy in a Clinically Relevant Simulated Body Fluid Environment, Metall. Mater. Trans. A, 2015, 46(6), 2696–2709.
8. V. Swaminathan and J. L. Gilbert, Potential and frequency effects on fretting corrosion of Ti₆Al₄V and CoCrMo surfaces, J. Biomed. Mater. Res., Part A, 2013, 101A(9), 2602–2612.
9. F. Contu, B. Elsener and H. Böhni, Corrosion behaviour of CoCrMo implant alloy during fretting in bovine serum, Corros. Sci., 2005, 47(8), 1863–1875.
10. V. Swaminathan and J. L. Gilbert, Fretting corrosion of CoCrMo and Ti₆Al₄V interfaces, Biomaterials, 2012, 33(22), 5487–5503.
11. J. Geringer, K. Kim and J. Pellier, et al., 3-Fretting corrosion processes and wear mechanisms in medical implants, Bio-Tribocorrosion in Biomaterials and Medical Implants, 2013, pp. 45–73.
12. J. Geringer and D. D. Macdonald, Friction/fretting-corrosion mechanisms: current trends and outlooks for implants, Mater. Lett., 2014, 134(7), 152–157.
13. J. Geringer, J. Pellier and M. L. Taylor, et al., Fretting corrosion with proteins: the role of organic coating on the synergistic mechanisms, Thin Solid Films, 2013, 528(2), 123–129.
14. M. Bryant, M. Ward and R. Farrar, et al., Characterisation of the surface topography, tomography and chemistry of fretting corrosion product found on retrieved polished femoral stems, J. Mech. Behav. Biomed. Mater., 2014, 32(4), 321–334.
15. Z. Cai, S. Gao and X. Gan, et al., Torsional fretting wear behaviour of nature articular cartilage in vitro, Int. J. Surf. Sci. Eng., 2011, 5(5/6), 348–368.
16. Z. B. Cai, S. S. Gao and M. H. Zhu, et al., Tribological behavior of polymethyl methacrylate against different counter-bodies induced by torsional fretting wear, Wear, 2011, 270(3–4), 230–240.
17. Z. B. Cai, G. A. Zhang and Y. K. Zhu, et al., Torsional fretting wear of a biomedical Ti₆Al₇Nb alloy for nitrogen ion implantation in bovine serum, Tribol. Int., 2013, 59(59), 312–320.
18. K. Chen, D. Zhang and G. Zhang, et al., Research on the torsional fretting behavior of the head–neck interface of artificial hip joint, Mater. Des., 2014, 56(4), 914–922.
19. H. Quan, S. Gao and M. Zhu, et al., Comparison of the torsional fretting behavior of three porous titanium coatings for biomedical applications, Tribol. Int., 2015, 92, 29–37.
20. W. Song, W. Fei and Z. Liao, et al., Study on torsional fretting wear behavior of a ball-on-socket contact configuration simulating an artificial cervical disk, Mater. Sci. Eng., C, 2015, 55, 22–33.
21. J. Yu, Z. Cai and M. Zhu, et al., Study on torsional fretting behavior of UHMWPE, Appl. Surf. Sci., 2008, 255(2), 616–618.
22. Z. B. Cai, J. Yu and Z. R. Zhou, et al., Torsional Fretting Behaviors of UHMWPE against Different Counter-Bodies, Advanced Tribology, Springer, Berlin Heidelberg, 2009, pp. 298–299.
23. Z. B. Cai, M. X. Shen and J. Yu, et al., Friction and wear behaviour of UHMWPE against titanium alloy ball and alumina femoral head due to torsional fretting, Int. J. Surf. Sci. Eng., 2013, 7(1), 81–95.
24. D. He, T. Zhang and Y. Wu, Retracted: mechanisms responsible for synergy between fretting and corrosion for three biomaterials in saline solution, Wear, 2001, 250(1–12), 180–187.
25. F. Megret, J. P. Hugnot and A. Falconar, et al., The effects of calcium phosphate deposition upon corrosion of CoCr alloys and the potential for implant failure, J. Biomed. Mater. Res., Part A, 2005, 75A(2), 365–373.
26. B. C. Syrett, S. S. Wing and B. C. Syrett, et al., Electrochemical investigation of fretting corrosion of surgical implant materials, Corrosion, 1978, 34, 11.
27. J. Geringer, J. Pellier and M. L. Taylor, et al., Electrochemical Impedance Spectroscopy: Insights for Fretting Corrosion Experiments, Tribol. Int., 2013, 68(6), 67–76.
28. J. Zhang, F. A. Moslehy and S. L. Rice, A model for friction in quasi-steady-state sliding part I. Derivation, Wear, 1991, 149(1–2), 1–12.
29. J. Zhang, F. A. Mosleby and S. L. Rice, A model for friction in quasi-steady-state sliding part II. Numerical results and discussion, Wear, 1991, 149(1–2), 13–25.
30. H. Geng, D. K. Zhang, K. Chen and Q. L. Wang, Galvanic corrosion behaviors between Ti₆Al₄V alloys and CoCrMo alloys in saline solution, Mater. Express, 2014, 4(3), 213–220.
31. R. Dmitry, R. Rashmi and Y. Judy Chia-Chun, et al., An electrochemical investigation of TMJ implant metal alloys in an artificial joint fluid environment: the influence of pH variation, Journal of Cranio-Maxillo-Facial Surgery, 2014, 42(7), 1052–1061.
32. Y. I. Jing-Hua, X. Z. Lin and Z. B. Cai, et al., Study on torsional fretting corrosion behaviors of Ti₆Al₄V alloy in the serum solution, Funct. Mater., 2012, 7(43), 919–923.

---