Tribological improvements of plasma immersion implanted CoCr alloys

Tribological improvements of plasma immersion implanted CoCr alloys

Surface & Coatings Technology 204 (2010) 2928–2932 Contents lists available at ScienceDirect Surface & Coatings Technology j o u r n a l h o m e p a...

891KB Sizes 0 Downloads 3 Views

Surface & Coatings Technology 204 (2010) 2928–2932

Contents lists available at ScienceDirect

Surface & Coatings Technology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s u r f c o a t

Tribological improvements of plasma immersion implanted CoCr alloys J.A. García a,⁎, C. Díaz a, S. Mändl b, J. Lutz b,c, R. Martínez a, R.J. Rodríguez a a b c

AIN Centro de Ingeniería Avanzada de Superficies, 31191 Cordovilla-Pamplona, Spain Leibniz-Institut für Oberflächenmodifizierung, 04303 Leipzig, Germany Translational Centre for Regenerative Medicine, Universität Leipzig, Germany

a r t i c l e

i n f o

Available online 27 March 2010 Keywords: Ion implantation Plasma immersion Biocompatibility Tribology

a b s t r a c t CoCr alloys are widely used in some specific hip and knee prosthesis. In particular, they are used in advanced metal on metal (MoM) prostheses. However, while these new implants have an intermediate wear rate between low wear ceramic–metal and high wear polymer–metal prostheses, a high risk of metalosis produced by wear debris, originated in the metal–metal sliding contact, is still present. Oxygen plasma immersion ion implantation (PI3) provides a practical method to improve wear resistance of CoCr alloys, thus reducing the amount of wear particles. This paper describes the tribological and wear resistance improvements of CoCr alloy implanted by PI3. Friction and wear tests were performed to investigate the modifications obtained by oxygen implantation. Glow Discharge Optical Emission Spectrometry (GD-OES) analysis was employed to evaluate the oxide layer thickness and compositional changes of the metallic alloying elements produced by the oxygen treatment. Moreover, SEM analysis was carried out for detecting wear particles after ball on disk tests in a biological solution simulating the human body fluid. The reported results indicate that PI3 improves the wear resistance of CoCr alloy to be used in human body implants. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Cobalt based alloys are one the most employed biomaterials for articulating hip and knee prosthesis due to their very suitable combination of mechanical and tribological properties as well as their good corrosion resistance [1]. However, new concepts in the prosthesis design and evolution in surgical techniques have biased many investigations for improving lifetime and performance of Co alloys. Concerning failure causes, different investigations have found the following problems: i) migration of toxic metal ions due to corrosive failure; ii) debris generated by abrasive wear; iii) adverse micro- and nanotopography; iv) low bioactivity; and v) non-hydrophilic surface [2–6]. Advanced surface treatments are very suitable techniques to improve materials for biomedical applications, and due to its specific characteristics like non-delamination risk, ion implantation is one of the most promising methods for improving prosthesis surfaces [7–9]. Recent publications have found that nitrogen implantation is a good solution for improving wear resistance in CoCr alloys in dry conditions, reducing debris, and thus reducing particle-induced osteolysis (the major cause of a loosening after hip joint replacement) [10]. However Öztürk et al. reported a considerable increase in migration of toxic metal ions after nitrogen ion implantation [2]. Similar results have been obtained for fretting wear experiments in simulated body fluid after nitrogen PIII [11]. Looking for solutions for the migrations effects Díaz et

⁎ Corresponding author. Tel.: + 34 948 42 11101; fax: + 34 948 42 11100. E-mail address: [email protected] (J.A. García). 0257-8972/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2010.03.050

al. studied oxygen implantations and reported a dramatic reduction in concentration of toxic metals [12], coupled with moderate increases in wear resistance. The aim of this investigation is the optimisation of tribological properties of CoCr alloys by means of oxygen plasma immersion ion implantation, by obtaining a combination of excellent barrier behaviour of oxygen layer preventing ion metal migrations with a good wear resistance.

2. Experimental Co28Cr6Mo alloy was obtained as commercial material according to ASTM F1537, respective ISO 5832-12. The studies were done on flat samples of 30 mm of diameter. They were mirror polished with a 1 μm diamond suspension and cleaned in an acetone ultrasonic bath before making the different treatments. The process temperature between 500 and 600 °C was low enough to avoid phase transitions or recrystallisation of the base material. PI3 oxygen implantations were carried out in a high vacuum chamber at a base pressure lower than 10− 4 Pa and a working pressure of 0.18 Pa. The plasma was generated by an RF plasma source operating at 40.68 MHz. For all experiments a high voltage pulse of 10 kV was used and maintained for 1 h, with a total incident dose between 3 and 6 × 1018 oxygen atoms/cm2. The fluence per pulse was obtained from reference implantations into Si at varying pulse numbers by Rutherford backscattering spectroscopy [13]. The temperature itself was measured with an IR pyrometer, calibrated against a thermocoupe. A closed oxide layer of 10–15 nm will be established at these fluences

J.A. García et al. / Surface & Coatings Technology 204 (2010) 2928–2932

with an integrated oxygen content below 1017 cm− 2. Excess oxygen is either removed by sputtering or diffusing towards the bulk [14]. A Fisherscope H100VP microindenter was used to measure the Universal Hardness (HU) [15]. The equipment software allows to control indentation depth, final load and the number of loading and unloading steps. A Vickers indenter tip was used and the maximum test load was 2 mN. For each sample, thirty indentations were done in order to obtain a good statistical average. The indentation depths were larger than the 10% of the implanted layer depth, which implies that the measured value is a mixture of the layer and the substrate hardness [16]. The friction measurements were done in a pin/ball-on-disk tribometer FALEX 320PC, immersing the samples in Hank's solution as a lubricant. Hank's solution, simulating the human body fluid [17], has a content of 0.9% NaCl, pH 7.4 and during the test its temperature was maintained at 37 °C ± 1 °C. The parameters chosen for these tests were 50 g of applied load; 45% of humidity, 5000 cycles and a linear speed of 0.1 m/s. The counterbodies were a 3 mm-diameter CoCr and UHMWPE bearing balls simulating two of the most common tribological pairs used nowadays (metal–metal and metal–PE). Moreover, an optical profilometer WYKO RST 500 was used to measure the roughness before and after the tests, after cleaning samples in an ultrasonic bath. Furthermore, in order to estimate the wear coefficient after the ball-on-disk tests, the volume loss in the wear track was measured by using the same optical profilometer [18]. Quantitative depth-profiles of different elements, such as oxygen, nitrogen, aluminium, vanadium, titanium, chromium, cobalt and molybdenum were measured by glow discharge optical emission spectroscopy (GDOES) using a JY GD Profiler HR (Jobin-Yvon Horiba, France) RF source operating at 650 Pa and 40 W. Further corrections were applied on the optical signals in order to take into account the influence of the residual hydrogen signal. A Field Emission Scanning Electron Microscopy (FE-SEM) HITACHI was used to observe the treated surface of the samples, the wear tracks and the particles collected by nanometric filtering the Hank's solution after friction and wear tests. Thus, the amount of wear debris and its chemical composition could be determined as a proxy for the changes in the biocompatibility.

3. Results 3.1. GDOES analysis Oxygen concentration profiles do not show significant differences at the studied range of temperatures. In all the cases, for the different implantation conditions (Table 1) there is a 40–50 nm modified thin layer where the oxygen concentration is about 25–30 At.% (Fig. 1). Co concentration shows a peak in the most outer surface (about 20 nm thick) for all the implantation temperatures, as it is shown in Fig. 1. These peaks, are also reported for other authors [12,19] and suggest the existence of a thin cobalt oxide layer in concordance with literature [2]. Furthermore, it is also possible to note in Fig. 1, that the Cr concentration goes up when the Co concentration decreases, suggesting the formation of some species of chromium oxides.

2929

Fig. 1. Compositional profile for the implanted sample at 520 °C.

3.2. Tribology Ultramicrohardness results showed significant increases, around 30% in the best cases in relation to the reference sample for HU in the most outer surface of all the implanted samples, as gathered in Fig. 2. Due to the very thin modified layer (40–50 nm), the indentation depths at 2 mN of the final load were larger than the 10% of the implanted layer, which implies that real thin layer hardness is higher than the measured values. The results obtained in friction tests carried out against a CoCr counterbody on samples implanted at different temperatures present significant decreases in friction coefficient with respect to the untreated sample (Table 1). While a value of 0.38 in friction coefficient is found for the untreated sample, we obtained 0.10 for the 520 °C implanted sample and 0.04 at 540 °C. When implantation temperature goes up to 570 °C, friction coefficient goes up to 0.30, which is still below the reference sample value. Moreover, the implanted samples didn't show significant variations in roughness parameters like Ra, showing a good agreement with the friction results, as usually increases in roughness are associated with increases in friction coefficient. The results of wear tests show decreases for all the implanted samples except for the lowest implantation temperature (Table 1). For the samples implanted at 520 °C and 540 °C the wear coefficient were reduced by about a factor 2 (from 6.5 to 3.1 × 10− 16 Nm− 2), as it is possible to observe in Fig. 3, where the wear tracks for the unimplanted and implanted samples at 540 °C are shown. Even more improvements were obtained for the samples implanted at high temperature, showing a decrease of about one order of magnitude. However, additional friction tests of implanted samples against a UHMWPE counterbody show an increasing of the friction coefficient.

Table 1 Tribological properties. Samples

Reference50gr 520 °C 540 °C 570 °C 610 °C

Friction coefficient

Wear rate (Nm− 1/2 ⁎ 10− 16)

Roughness (Ra) nm Before wear test

After wear test

0.38

6.4

10.47

16.65

0.10 0.04 0.30 0.21

0.4 3.1 3.5 0.7

8.37 15.11 12.44 10.23

13.40 9.95 8.00 12.70

Fig. 2. HU profile for the PI implanted sample at 520, and the unimplanted sample.

2930

J.A. García et al. / Surface & Coatings Technology 204 (2010) 2928–2932

Fig. 3. Wear tracks for the unimplanted (top) and for the implanted samples at 540 °C.

The SEM and EDS analysis of debris gathered after friction and wear ball on disk tests did not show clear differences about the composition between the untreated and treated samples. In all the cases these particles have a high concentration of Co and in some case Cr. These results are in agreement with the GDOES analysis and also with literature [2] (Fig. 4), showing a CoO thin layer in the most outer surface followed by a CoCrO layer. The size of the particles was higher in the treated samples in relation to the reference sample, probably due to changes in the wear process induced by the implantation treatment. Particle size was in the range of microns in all cases, from 100 to 600 microns in treated samples and between 1 and 20 microns in the reference sample. 3.3. GXRD analysis The results obtained in the GXRD analysis showed the presence of two additional peaks for the treated samples. No clear differences were pointed out for the four implantation conditions. Fig. 5 gathers the diffraction patterns for four different implantation temperatures. Com-

paring implanted samples with untreated samples an extra peak appears at 36.2° and a second peak is at 51°. These extra peaks correspond to Co oxide and Cr.

4. Conclusions Co28Cr8Mo alloy has been implanted at different temperatures in order to optimize the treatment parameters. The obtained tribological results point out the following aspects: o There are no significant differences in the oxygen profiles for the different studied implantation temperatures, despite the different diffusion conditions in the range of temperature of the experiments (520 °C–610 °C). Minor variations for the segregated Co and Cr content were observed. o Ultramicrohardness tests revealed an increase of more than 80% in the hardness for the first nanometers in the implanted samples. Technical problems for measuring the real hardness of extremely

J.A. García et al. / Surface & Coatings Technology 204 (2010) 2928–2932

2931

Fig. 4. Mapping of a particle from treated sample at 570 °C.

thin coatings (40–50 nm), suggest bigger increases for these treatments. o Friction coefficient in Hank's solution obtained for the implanted samples was reduced from 0.36 for the untreated sample to below

0.1 at 520 and 540 °C. This results are consistent with the non variation in roughness for the different implantation experiments. o Wear results showed important decreases in wear rate of 100% for the 540 and 570 °C, and close to a factor 10 for the implanted

Fig. 5. GXRD diagrams for the different studied samples.

2932

J.A. García et al. / Surface & Coatings Technology 204 (2010) 2928–2932

sample at high temperature (610 °C). Taking into account that there are no differences in thickness with implantation temperature, one possible explanation is chemical composition and microstructure of these thin layers may be the reason for these clear differences. o GXRD analysis showed that the chemical compositions of the layers generated in the implantation are cobalt and chromium oxides. There are not clear differences for the four implantation temperatures. o The SEM and EDS analysis of the wear debris showed that these particles have Co as principal element in their composition, in agreement with literature that proposes a first thin layer of cobalt oxide. Acknowledgements The authors would like to thank the support received from: Gobierno de Navarra for the project IIQ011979.RI1 BIO-COAT. and the Science and Innovation Ministry for the project MAT-2007.66550-C02-01. J.L. further acknowledges the support from the German Federal Ministry of Education and Research (BMBF, PtJ-Bio, 0313909). References [1] J. Cawley, J.E.P. Metcalf, A.H. Jones, T.J. Bnd, D.S. Skupien, Wear 255 (2003) 999. [2] O. Öztürk, U. Türkan, A.E. Eroglu, Surf. Coat. Technol. (2006) 5687.

[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]

Y. Okaaki, E. Gotoh, Biomaterials 26 (2005) 11. K. Yoshida, M. Moita, H. Mishina, JSME Int. J. Ser. C 46 (4) (2003). T. Hanawa, Mat. Sci. Eng., C 24 (2004) 745. F.C. Giacomelli, C. Giacomelli, A. Spinelli, J. Braz. Chem. Soc. 15 (4) (2004) 541. S. Mändl, B. Rauschenbach, Surf. Coat. Technol. 156 (2002) 276. S. Mändl, Surf. Coat. Technol. 201 (2007) 6833. X. Tian, C. Gong, S. Yang, Z. Luo, R.K.-Y. Fu, P.K. Chu, IEEE Trans. Plasma Sci. 34 (2006) 1160. J. Lutz, S. Mändl, Reduced Tribocorrosion of CoCr Alloys in Simulated Body Fluid after Nitrogen PIII, PIII&D Conference, Brazil, 2009. T. Lutz, J.W. Gerlach, S. Mändl, Surf. Coat. Technol. 201 (2007) 6690. C. Díaz, J. Lutz, S. Mändl, J.A. García, R. Martínez, R.J. Rodríguez, Nucl. Instrum. Methods Phys. Res., B 267 (2009) 1630. D. Manova, S. Mändl, B. Rauschenbach, Plasma Sources Sci. Technol. 10 (2001) 423. J. Lutz, A. Lehmann, S. Mändl, Surf. Coat. Technol. 202 (2008) 3747. J.L. Loubet, J.M. Georges, G. Meille, Micr. Techn. Mater. Sci. Eng. ASTM STP 889 (1986) 72. J.A. García, R. Martínez, B. Lerga, G.G. Fuentes, R.J. Rodríguez, S. Mändl, Surf. Coat. Technol. 200 (2005) 573. A. Oyane, H.-M. Kim, T. Furuya, T. Kokubo, T. Miyazaki, T. Nakamura, J. Biomed. Mater. Res. A 65 (2003) 188. J.A. García, A. Guette, A. Medrano, C. Labrugère, M. Rico, M. Lahaye, R. Sánchez, A. Martínez, R.J. Rodríguez, Vacuum 64 (2002) 343. M. Ueda, L.A. Berni, R.M. Castro, H. Reuther, C.M. Lepienski, P.C. Soares Jr., Surf. Coat. Technol. 200 (2005) 594.