Effects of Precipitates and Albumin in Simulated Body Fluids on Pin-on-Disk Wear Behavior of Biomedical Co-Cr-Mo Alloys

Abstract

Pin-on-disk wear tests using Co-Cr-Mo (CCM) alloy pins and disks were conducted in 0.14 M NaCl solutions with and without albumin. To clarify the effect of precipitates in the CCM alloys on the alloy wear behavior, as-received (having precipitates) and solution-treated (having no precipitates) CCM alloys were used as specimens. Friction coefficients during wear testing were measured. After the wear testing, the mass loss of pins and disks, wear tracks on the disks, and the wear debris were examined using scanning electron microscopy (SEM) and laser microscopy (LM). The concentration of metallic ions in solution was analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). In the solution without albumin, the friction coefficient increased with increasing sliding time and discontinuous wear grooves with attached and detached sections were observed, indicating that adhesive wear was dominant. On the other hand, in the solution with albumin, the friction coefficient was constant independent of the sliding distance and continuous wear grooves were observed, indicating that abrasive wear was dominant. High amounts of mass loss were detected from the as-received alloys in the albumin solution, and it is theorized that the wear debris (including precipitates) enhanced the third-body wear. The concentration of Cr ions in solution was lower than the expected values based on the alloy composition. In the solution with albumin, it was thought that the Cr ions bonded with the albumin, increasing the amount of available Cr ions in solution.

1. Introduction

Biomedical Co-28Cr-6Mo (CCM) alloy is used in wear parts of artificial joints because of its excellent mechanical properties, wear resistance, and corrosion resistance. Metal-on-metal (MoM) type artificial joints using CCM alloy in a ball and cup configuration have minimal risk of material fracture and high design variability.¹⁾ On the other hand, such joints do have relevant concerns regarding the formation of fine wear debris¹⁾ and allergic reactions induced by elution of metallic ions²⁾, compared to metal-on-polymer (MoP) type artificial joints. For these reasons, the use of MoM type artificial hip joints declined sharply in England and Wales since 2008.³⁾ Because MoM type artificial joints possess superior durability and strength compared to MoP type artificial joints, it is believed that MoM type artificial joints are more advantageous for long-term use. In order to realize wider use of MoM type artificial joints, improved wear resistance (through microstructural control of the CCM alloy) and a better understanding of the wear behavior (such as formation of wear debris and elution of metallic ions in biological media) are necessary.

The American Society for Testing and Materials (ASTM) F75 standard permits the addition of carbon and nitrogen to biomedical CCM alloys at 0.35 and 0.25 mass%, respectively.⁴⁾ The addition of carbon and nitrogen in CCM alloys suppress the formation of the intermetallic σ-phase, which negatively impact the mechanical properties of the CCM alloy,^5,6) and are related to the precipitation strengthening by carbides and nitrides.⁷⁾ In our previous studies, pin-on-disk wear tests involving CCM alloy pins and Al₂O₃ disks in 1 mass% lactic acid and Kokubo solutions were used to study the wear behavior of these systems, with a focus on the effects of precipitates present in the CCM alloys.⁸⁾ The mass loss and ion elution of as-cast CCM alloys with precipitates are much higher than those of solution-treated precipitate-free CCM alloys. The likely explanation here is that three-body abrasion was promoted by the wear debris (including precipitates). Chiba et al. also reported that these precipitates promoted abrasive wear based on pin-on-disk wear tests using CCM alloy pins and disks in Hanks' solution.⁹⁾

The environments in which artificial joints are used include both inorganic ions and proteins. Albumin is the most abundant protein in plasma with a typical concentration of 5 g/100 mL in blood. Albumin has a high affinity to a very wide range of materials, including metal ions and amino acids. Albumin is also known to maintain the pH and osmotic pressure of plasma.¹⁰⁾

Many previous studies have explained the wear behavior of CCM alloys containing precipitates in simulated body fluids (SBF) with albumin.^11–18) However, the effect of precipitates in CCM alloys on the wear behavior in solutions with albumin are not entirely clear; in fact, both increase^11–13,18) and decrease¹⁵⁾ in the mass loss after wear tests have been reported. In addition, these previous reports all involved precipitates, and there are no reports on the wear behavior of precipitate-free CCM alloys, and only a few report on the ion elution of CCM alloys during wear tests in solutions containing albumin.

In this study, pin-on-disk wear tests using CCM alloy pins and disks were conducted in 0.14 M NaCl solutions with and without albumin. As-received CCM alloys having precipitates and solution-treated (ST) CCM alloys having no precipitates were used for the pins and disks. The purpose of this study is to clarify the effects of precipitates in CCM alloys and albumin in solution on the pin-on-disk wear behavior of these alloys.

2. Experimental Procedures

2.1 Specimens

Commercial CCM bars (Sulzer Ltd., ϕ44 mm, as-received) were used in this study. The chemical composition of the as-received CCM alloy is shown in Table 1. To dissolve the precipitates in the as-received CCM alloys, the alloys were sealed in silica ampoules with Ar gas and solution-treated at 1523 K for 1.8 ks, followed by quenching with water (ST).

Table 1 Chemical composition of as-received CCM alloy used in this study (mass%).

Co	Cr	Mo	Si	Mn	Fe	Ni	C	N
Bal.	27.4	6.09	0.70	0.66	0.49	0.37	0.09	0.18

The microstructures of the as-received and ST CCM alloys were determined using electron backscatter diffraction (EBSD, JSM-7800F, JEOL) to measure the grain sizes of these alloys. The hardness of the alloys were measured using a micro Vickers hardness tester (HM-211, Mitsutoyo) with a 1 kgf load. The grain size and hardness of the as-received and ST CCM alloys are shown in Table 2. The grain size of the CCM alloys increased after solution treatment, while the hardness decreased.

Table 2 Grain size and Vickers hardness of as-received and ST CCM alloys.

Alloy	Grain size, d/μm	Vickers hardness, HV
As-received	9.1	363
ST	48.8	322

The precipitates were electrolytically extracted from the as-received and ST CCM alloys in a 10% sulfuric acid solution⁵⁾, and the precipitate phases were identified using X-ray diffraction (XRD, D8 ADVANCE, Bruker AXS K.K.). XRD patterns of these precipitates are shown in Fig. 1. M₂₃X₆-type (M: Metallic elements, X: carbon and nitrogen), η-phase (M₆X-M₁₂X-type), and M₂X-type precipitates were detected from the as-received CCM alloy. No precipitates were detected from the ST CCM alloy, indicating complete dissolution of precipitates after solution treatment.

Fig. 1

XRD patterns of precipitates electrolytically extracted from as-received and ST CCM alloys.

2.2 Pin-on-disk wear tests

Pin-on-disk wear tests were conducted using CCM alloy pins and disks using a conventional pin-on-disk apparatus (FPR-2100, RHSCA). The CCM alloy pins were 5 mm in diameter and 25 mm in length, and the contact surface of the pins was worked to a spherical radius (SR) of 100 mm to avoid stress concentration at the edge of the pins. The CCM alloy disks were 44 mm in diameter and 5 mm thick. Both the contact surfaces of the pins and disks were mirror-polished (Ra < 0.05 μm) and ultrasonically cleaned in ethanol and then ultrapure water for 300 s each. 0.14 M NaCl solutions with and without albumin (20 g∙L⁻¹) were used as the test solution. The pH of these solutions was adjusted to 7.4 using 10 mM NaOH solution. 70 mL of the 0.14 M NaCl solution at room temperature was used for all experiments. The disks rotated at a rate of 25 mm∙s⁻¹ (13.3 rpm) at a sliding diameter of 36 mm. The rotation or sliding speed was set to the same value of a previously reported hip joint simulator.¹³⁾ The pin load was set to 1.0 kgf, corresponding to 144 MPa of contact stress as calculated using the Hertzian contact theory.¹⁹⁾ During wear testing, the friction coefficient between the pin and disk was measured every 300 s.

After wear testing for 259.2 ks (sliding distance of 6480 m), the pins and disks were ultrasonically cleaned in ethanol and ultrapure water, and then weighed to evaluate the mass loss. The wear tracks on the disk surfaces were observed using a profilometer (Dektak³ ST, ALVAC), as well as scanning electron microscopy (SEM, XL30FEG, Philips), and laser microscopy (LM, VK-X250, Keyence).

The solution after wear testing was collected and centrifuged. The concentration of metallic ions in the supernatant was evaluated using inductively coupled plasma-atomic emission spectroscopy (ICP-AES, ICPS-8100, Shimadzu). The wear debris in the solution was filtered using a polytetrafluoroethylene (PTFE) membrane filter (0.2 μm, J020A025A, Advantec) and observed using SEM.

3. Results

3.1 Mass loss, surface morphology of disks, and wear debris

The mass loss values of pins and disks after wear testing are shown in Fig. 2. The summation of the mass loss from the pin and disk is defined as the total mass loss. In the solution without albumin, the total mass loss of the as-received CCM alloy was nearly identical to that of the ST CCM alloy. On the other hand, in the solution with albumin, the total mass loss of the as-received CCM alloy was higher than that of the ST CCM alloy.

Fig. 2

Mass loss of as-received and ST CCM alloy pins and disks after wear testing in 0.14 M NaCl solutions with and without albumin.

Figure 3 shows friction coefficients between CCM alloy pins and disks during wear testing in solutions with and without albumin. In the solution without albumin, the friction coefficients of both the as-received and ST CCM alloys were constant until 50 ks (1250 m), after which they increased to 0.6–0.8, and were constant after 100 ks. In the solution with albumin, the friction coefficient was almost constant with a value of approximately 0.2–0.4, independent of the sliding time and presence of precipitates in the CCM alloy. Based on these results, the albumin in solution suppressed the increase of the friction coefficient during wear testing.

Fig. 3

Friction coefficient during the wear testing of (a,c) as-received and (b,d) ST CCM alloys in 0.14 M NaCl solutions (a,b) with and (c,d) without albumin.

Figures 4 and 5 show SEM and LM images, respectively, of the wear tracks on the disks after wear testing. Based on these images, the wear direction is from top to bottom. Meanwhile, the scales of the depth direction in the LM images are different under different conditions. In the solution without albumin, discontinuous wear grooves were observed from the surface of the disks, originating from the black parts in the image. In addition to the discontinuous wear grooves, attached and detached sections were observed by LM. On the other hand, continuous wear grooves parallel to the wear direction were observed in the solution with albumin. The width and depth of the wear grooves on the as-received CCM alloy disks were approximate 50 μm and 5 μm, respectively, and on the ST CCM alloy disks these were several μm and 1 μm, respectively.

Fig. 4

SEM images of the surface of the (a,c) as-received and (b,d) ST CCM alloy disks after wear testing in 0.14 M NaCl solutions (a,b) with and (c,d) without albumin.

Fig. 5

LM images of the surface of the (a,c) as-received and (b,d) ST CCM alloy disks after wear testing in 0.14 M NaCl solutions (a,b) with and (c,d) without albumin.

SEM images of wear debris formed during wear testing are shown in Fig. 6. In the solution with albumin, wear debris several μm in diameter were present. From the TEM analysis of these debris (not shown), the wear debris appeared to be agglomerations of smaller particles (on the order of 100 nm in diameter). In the solution without albumin, no agglomeration was observed and the size of the wear debris was approximately 100 nm.

Fig. 6

SEM images of wear debris obtained from the (a,c) as-received and (b,d) ST CCM alloys after wear testing in 0.14 M NaCl solutions (a,b) with and (c,d) without albumin.

3.2 Elution of metallic ions

The total amount of ions in solution after wear testing are shown in Fig. 7. The total amount of ions is the summation of the amounts of Co, Cr, Mo, Si, Mn, and Ni ions present in a given solution. The total amount of ions in the solution with albumin using as-received CCM alloy was highest, and that using ST CCM alloy was minimum. In the solution without albumin, the total amount of ions using as-received CCM alloys was comparable with the case of using ST CCM alloys. These results are in good agreement with the total mass loss data shown in Fig. 2.

Fig. 7

Total amount of ions in the 0.14 M NaCl solutions with and without albumin after wear testing of as-received and ST CCM alloys.

The amounts of each ion in solution after wear testing are shown in Fig. 8. The amounts of Co and Mo ions were on the order of the total mass loss. The amount of Ni ions was less than 7 μg, which is sufficiently low value. In the solution without albumin, the amounts of Cr and Mn ions are low.

Fig. 8

Amount of ions in the 0.14 M NaCl solutions with and without albumin after wear testing of as-received and ST CCM alloys.

4. Discussion

4.1 Effects of precipitates in CCM alloys and albumin in lubricant on wear behavior

In the solution without albumin, the friction coefficient increased with increasing sliding time and discontinuous wear grooves were observed on the surface of the disks, and adhesive wear was likely the dominant force present. Total mass loss of the as-received CCM alloys was almost the same as with the ST CCM alloys, indicating that the precipitates in CCM alloys do not affect the wear behavior in solutions without albumin.

Previous studies have reported that adsorbed albumin on CCM alloys acts as a lubricant and suppresses adhesive wear, thus, resulting in the dominant wear mechanism of abrasive wear.^11–18) They also reported that wear debris agglomerated and accumulated around the wear track.^11,12) In this study, a constant friction coefficient and continuous wear grooves were observed in the solution with albumin, indicating that abrasive wear was the dominant wear mechanism.¹⁶⁾ In addition, wear debris several μm in diameter were present, made up of agglomerated particles with diameters on the order of 100 nm (Fig. 6); these agglomerated particles are believed to cause three-body abrasion in these systems. In the case of the as-received CCM alloys, it was expected that the precipitates in the CCM alloys were included in the wear debris during wear testing. The precipitates in the as-received CCM alloys were M₂₃X₆-type, η-phase, and M₂X-type. Since the hardness of these precipitates was higher than that of the metallic matrix, it was expected that the precipitates in the wear debris accelerated the three-body wear and caused deeper and wider wear grooves to appear, compared to the ST CCM alloys. As a results, this case had the greatest total mass loss in this study.

4.2 Elution of metallic ions

CCM alloys shows high corrosion resistance due to the formation of a passive film mainly composed of Cr oxide. Muñoz and Mischler¹⁵⁾ studied the effect of albumin on the corrosion of CCM alloys using anodic polarization tests, and clarified that albumin decreased the corrosion resistance of CCM alloys in a 0.14 M NaCl solution. The protein in solution acted as a cathodic inhibitor by reducing the cathodic reaction rate, and it may affect the mass transport conditions to reduce the accessibility of the surface for oxygen.^15,20,21) During wear testing, the passive film was broken by the wear damage, and ion elution occurred at the bare surface, resulting in complicated ion elution phenomena. In addition, increasing the surface area by the formation of wear debris also affects the ion elution behavior.

The ion ratio (R_i) of Co, Cr, and Mo (the main components of CCM alloys) can be defined using eq. (1).   

\[{\rm Ion \ ratio}\ (R_{\rm i}) = M_{\text{i-sol}} / M_{\text{i-alloy}}\](1)

where M_i-sol and M_i-alloy are the mass percentages of the i ions in the solution and that of the i elements in the alloy, respectively. Calculated results from this study are shown in Table 3. When the ion ratio equals unity, the element in question exists as ions in the solution as expected from the alloy composition. The ion ratio of Cr (R_Cr) was less than unity, and less than 0.1 in the solution without albumin. Meanwhile, the pH of the solution was approximately 7, and was unaffected by the wear testing. In previous studies of static immersion of CCM alloys in simulated body fluids, Co ion release was faster than other elements.^22–24) Karimi et al. reported that the formation of Cr oxide was thermodynamically more favorable than Co oxides/hydroxides.²⁴⁾ The Cr dissolution rate was slower than the Co dissolution rate. Since the albumin reduces the accessibility of the surface for oxygen and suppresses the formation of a passivation film,^15,20,21) it is believed that the amount of Cr ions in the solution with albumin was higher than in the solution without albumin.

Table 3 Ion ratio (R_i) of Co, Cr, and Mo (after wear testing) of as-received and ST CCM alloys in 0.14 M NaCl solutions with and without albumin.

Element	With albumin		Without albumin
	As-received	ST	As-received	ST
RCo	1.21	1.13	1.31	1.33
RCr	0.56	0.61	0.03	0.07
RMo	0.97	1.02	2.14	1.79

According to the Pourbaix diagrams of Cr,²⁵⁾ Cr(III) compounds consisting of Cr₂O₃∙nH₂O, CrOOH, and Cr(OH)₃ are stable in neutral media. Cr ions eluted by the CCM alloys during wear testing were precipitated as Cr(III) compounds, and the amount of Cr ions in solution decreased as a result. Amorphous CrOOH is stable in room temperature NaCl solutions compared to other Cr(III) compounds.²⁶⁾ The needle-like products with high Cr contents observed on the filter after filtrating the solution after wear testing, therefore, are possibly amorphous CrOOH. In the solution with albumin, the Cr ions were bonded with albumin.²⁷⁾ Therefore, it is likely that the amount of Cr(III) precipitates was decreased, thus, increasing the ion ratio of Cr in the solution with albumin.

According to the Pourbaix diagrams of Mn,²⁸⁾ Mn²⁺ is stable in neutral media, suggesting elution of Mn ions in solution during wear testing. However, low amounts of Mn ions were present in the solution without albumin, for reasons that are not currently understood.

Several authors have observed that the increase of grain boundaries of high carbon CCM alloys can accelerate their anodic reactions.^12,21,22) In this study, the total amount of ions from the ST CCM alloys was slightly higher than from the as-received CCM alloys in the solution without albumin, observations that are in good agreement with our total mass loss data. This indicated that increasing the surface area by the formation of wear debris strongly affected the elution behavior of the CCM alloys during wear testing.

5. Conclusion

Pin-on-disk wear tests using biomedical Co-Cr-Mo alloy pins and disks were conducted in 0.14 M NaCl solutions with and without albumin, and the effects of the resulting precipitates on mass loss and metallic ion elution were investigated. The following results were obtained;

• (1)   In the solution without albumin, adhesive wear was dominant as the friction coefficient increased with increasing sliding time. Discontinuous wear grooves with detached and attached parts were observed on the worn surface of the disk after wear testing.
• (2)   In the solution with albumin, abrasive wear was dominant because the friction coefficient was constant, independent of the sliding time, and continuous wear grooves were observed on the worn surface of the disk after wear testing. Wear debris agglomerated due to the albumin present in solution. The precipitates in the as-received CCM alloys were incorporated into the wear debris, and accelerated the three-body wear to yield a greater total mass loss.
• (3)   The amount of Cr ions in solution after wear testing was lower than the expected value from the alloy's composition.

Acknowledgments

This study was financially supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (No. 26709049).

REFERENCES

• 1)  M. Hannemann, et al.: Orthop. Traumatol.: Surg. Res. 99 (2013) 263–271. 10.1016/j.otsr.2013.01.005
• 2)  R.A.E. Clayton, I. Beggs, D.M. Salter, J.T. Patton and D.E. Porter: J. Bone Joint Surg. Am. 90 (2008) 1988–1993. 10.2106/JBJS.G.00879
• 3)  A.J. Smith, P. Dieppe, K. Vernon, M. Porter and A.W. Blom: Lancet 379 (2012) 1199–1204. 10.1016/S0140-6736(12)60353-5
• 4)  V.A. Mayer: 2013 annual book of ASTM standards, section thirteen, medical devices and services, vol. 13.01 (ASTM International, PA, 2013).
• 5)  S. Mineta, S. Namba, T. Yoneda, K. Ueda and T. Narushima: Metall. Mater. Trans., A 41 (2010) 2129–2138. 10.1007/s11661-010-0227-1
• 6)  S. Mineta, Alfirano, S. Namba, T. Yoneda, K. Ueda and T. Narushima: Metall. Mater. Trans. A 44 (2013) 494–503. 10.1007/s11661-012-1399-7
• 7)  T. Narushima, S. Mineta, Y. Kurihara and K. Ueda: JOM 65 (2013) 489–504. 10.1007/s11837-013-0567-6
• 8)  K. Ueda, K. Nakaie, S. Namba, T. Yoneda, K. Ishimizu and T. Narushima: Mater. Trans. 54 (2013) 1281–1287. 10.2320/matertrans.ME201316
• 9)  A. Chiba, N. Nomura and Y. Ono: Acta Mater. 55 (2007) 2119–2128. 10.1016/j.actamat.2006.10.060
• 10)  S. Sugio, A. Kashima, S. Mochizuki, M. Noda and K. Kobayashi: Protein Eng. 12 (1999) 439–446. 10.1093/protein/12.6.439
• 11)  L.C. Julián and A.I. Muñoz: Tribol. Int. 44 (2011) 318–329. 10.1016/j.triboint.2010.10.033
• 12)  J.L. Tipper, P.J. Firkins, E. Ingham, J. Fisher, M.H. Stone and R. Farrar: J. Mater. Sci. Mater. Med. 10 (1999) 353–362. 10.1023/A:1026473723777
• 13)  R. Varano, J.D. Bobyn, J.B. Medley and S. Yue: Proc. IMechE, Part H: JEIM 220 (2006) 145–159.
• 14)  M.A. Wimmer, J. Loos, R. Nassutt, M. Heitkemper and A. Fischer: Wear 250 (2001) 129–139. 10.1016/S0043-1648(01)00654-8
• 15)  A.I. Muñoz and S. Mischler: J. Mater. Sci. Mater. Med. 22 (2011) 437–450. 10.1007/s10856-010-4224-0
• 16)  D. Sun, J.A. Wharton and R.J.K. Wood: Wear 267 (2009) 52–60. 10.1016/j.wear.2009.01.011
• 17)  H. Mishina and M. Kojima: Wear 265 (2008) 655–663. 10.1016/j.wear.2007.12.006
• 18)  Y. Yan, A. Neville, D. Dowdon and S. Williams: Tribo. Corros. 39 (2006) 1509–1517.
• 19)  J.A. Greenwood: Tribol. Int. 30 (1997) 235–237. 10.1016/S0301-679X(96)00051-5
• 20)  S. Hiromoto, E. Onodera, A. Chiba, K. Asami and T. Hanawa: Biomaterials 26 (2005) 4912–4923. 10.1016/j.biomaterials.2005.01.028
• 21)  A.I. Muñoz and S. Mischler: J. Electrochem. Soc. 154 (2007) C562–C570. 10.1149/1.2764238
• 22)  Y. Okazaki and E. Gotoh: Biomaterials 26 (2005) 11–21. 10.1016/j.biomaterials.2004.02.005
• 23)  H.-Y. Lin and J.D. Bumgardner: J. Orthop. Res. 22 (2004) 1231–1236. 10.1016/j.orthres.2004.04.005
• 24)  S. Karimi, T. Nickchi and A.M. Alfantaz: Appl. Surf. Sci. 258 (2012) 6087–6096. 10.1016/j.apsusc.2012.03.008
• 25)  M. Pourbaix: Atlas of Electrochemical Equilibria in Aqueous Solutions, (NACE, Houston, 1974) p. 262.
• 26)  H. Yashiro, T. Itoh and K. Tanno: Corros. Sci. 32 (1991) 125–138. 10.1016/0010-938X(91)90037-P
• 27)  J. Yang and J. Black: Biomaterials 15 (1994) 262–268. 10.1016/0142-9612(94)90049-3
• 28)  M. Pourbaix: Atlas of Electrochemical Equilibria in Aqueous Solutions, (NACE, Houston, 1974) p. 290.