Effects of plasma nitriding ion beam flux density and time on the properties of CoCrMo alloy

Effects of plasma nitriding ion beam flux density and time on the properties of CoCrMo alloy

Vacuum 119 (2015) 214e222 Contents lists available at ScienceDirect Vacuum journal homepage: www.elsevier.com/locate/vacuum Effects of plasma nitri...

3MB Sizes 0 Downloads 5 Views

Vacuum 119 (2015) 214e222

Contents lists available at ScienceDirect

Vacuum journal homepage: www.elsevier.com/locate/vacuum

Effects of plasma nitriding ion beam flux density and time on the properties of CoCrMo alloy D.C. Ba, L. Xu, Q. Wang* School of Mechanical Engineering & Automation, Northeastern University, No 3 Wenhua Road, Shenyang 110819, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 December 2014 Received in revised form 23 May 2015 Accepted 23 May 2015 Available online 3 June 2015

The effects of plasma nitriding ion beam flux density and deposition time on structural and tribological properties of medical forged CoCrMo alloy were investigated. Plasma nitrided processes under pure nitrogen were conducted at various ion beam flux densities for 2, 4, 6 and 8 h. The active species included in the nitrogen plasma, phase compositions, microstructures and surface microhardness were characterized by means of OES, XRD, SEM and microhardness test techniques. A milliammeter was used to detect the ion beam flux density. Dry wear tests were performed using a ball-on-disc tribotester. The OES data suggested that excited neutral molecules N*2 and Nþ 2 played important roles in DC plasma nitriding. The experimental analyses confirmed that plasma nitriding process has shown promise in producing thicker, harder, better hydrophilic and more wear resistance layers than the conventional CoCrMo alloys. As compared with untreated CoCrMo alloy, the nitrided CoCrMo alloy showed lower wear rates and wear scar widths, and the specimen nitrided at 5.8 mA/cm2-8 h exhibited the lowest wear rate and better drysliding wear resistance. The adhesive wear was the main mechanism for untreated CoCrMo alloy while the wear mechanisms of nitrided specimens were fatigue wear, abrasive wear and slight adhesive wear. Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.

Keywords: Medical grade forged CoCrMo alloy Plasma nitriding Ion beam flux density Time Wettability Wear

1. Introduction Implant materials are usually metallic materials like austenitic stainless steels, Co-based alloys and titanium alloys, which are used to perform the functions of living tissues in the human body, and they are in continuous contact with body fluids [1]. Therefore, these metallic materials, especially the ones used in repairing and substituting for hard tissues, should have perfect mechanical properties, superior wear resistance and high corrosion resistance as well as favorable biocompatibility in physiologic environments [2e5]. Among the metallic materials, CoCrMo alloys are widely used in orthopedic medical implants because of their excellent properties imparted by a thin oxide layer formed spontaneously on the alloy surface [6]. However, owing to the aggressive biological effects of hip or knee prostheses in long-term applications, the CoCrMo alloy is prone to wear and fretting corrosion, and the protective passive oxide layer is not sufficient for the parts that work in contact. The wear particles and corrosion products, accumulated at tissues surrounding the implant, may migrate to distant

* Corresponding author. Tel./fax: þ86 024 83680450. E-mail address: [email protected] (Q. Wang). http://dx.doi.org/10.1016/j.vacuum.2015.05.032 0042-207X/Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.

parts of the body and lead to changes of biological process and negative body reactions, which restricts the lifetimes and usage of CoCrMo alloy. Due to the fact that wear and corrosion behavior occurs on the surface of CoCrMo alloy, surface treatments have been widely used to compensate the defects and improve comprehensive properties in the recent years. Plasma nitriding is a plasma-activated thermochemical method for improving the performance of components with respect to hardness, wear and corrosion resistance [7]. It has always been attractive to such requirements as energy and cost saving because of its relatively lower temperature, shorter treatment period and better applicability compared with the conventional nitriding process [8]. Çelik et al. [9] investigated the effects of plasma nitriding treatment of the CoCrMo alloy at various temperatures, and they found that the mechanical and tribological properties of nitrided specimens were better than that of the untreated specimens. Furthermore, the similar experimental processes were conducted by Wang et al. [6], and the nitrided CoCrMo alloy also showed superior wear resistance. Despite extensive researches [10e17], the studies mainly focused on the influence of treatment temperatures and time, the mechanisms of plasma nitriding have not been clearly understood yet, especially the effects of all kinds of active species like excited neutral molecules N*2 or atoms N*, excited

D.C. Ba et al. / Vacuum 119 (2015) 214e222 þ molecule ions Nþ 2 or atom ions N , photons and electrons on the evolution of the nitrided layer. Previous investigation on nitriding of AISI 316 stainless steel concluded that the excited neutral molecules N*2 governed the nitriding process, and the density of N*2 had significant effect on nitriding kinetics [18]. Relative description can also be found in the literature about ion nitriding process [19]. In the present study, medical forged CoCrMo alloy has been plasma nitrided at various treatment conditions, and the reactive species included in the nitrogen plasma has also been identified. The aim of this study is to investigate the effect of the reactive species, ion beam flux density as well as deposition time on the nitriding behavior of medical grade forged CoCrMo alloy. The ion current was measured by a milliammeter, and the reactive species were discerned by optical emission spectroscopy. The structural and compositional characterization of the plasma nitriding layers is investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM). The hardness and surface wettability as well as tribological properties are respectively investigated by microhardness tester, optical dynamic/static contact angle meter and ball-on-disc tribotester.

2. Experimental procedures Medical grade forged CoCrMo alloy (ISO 5832-12) was used in the experiments with the nominal composition shown in Table 1. The tensile strength, elastic modulus and elongation are 700 MPa, 230 GPa and 3.0%, respectively. Disc-like samples (25 mm diameter  5 mm thickness) were prepared from the cylindrical bars of the CoCrMo alloys by linear cutting machine. All samples were separately immersed in acetone and dilute hydrochloric acid for 15 min to remove surface contamination and oxidizing layer, and were subsequently ground by abrasive papers of 320, 500, 800, 1000 and 2000 grits. Before plasma nitriding, the specimens were ultrasonically washed in alcohol and distilled water for 10 min, then dried with nitrogen gas and inserted into the vacuum chamber immediately. The samples were placed into a holder which is the cathode of the nitriding equipment, as shown in Fig. 1. As can be seen, the vacuum chamber constitutes of a quartz tube (80 mm inner diameter and 1200 mm length) which can be evacuated down to 1  104 Pa prior to the introduction of working gas, and it is surrounded by a resistance furnace of 500 mm long to reach a maximum temperature of 1200  C. The temperature during nitridation is recorded by a thermocouple inserted in the sample. Plasma nitriding process was carried out in a DC power unit with maximum working voltage of 1000 V. The typical operating parameters applied during plasma nitriding process are listed in Table 2. Prior to the plasma nitriding process, the specimens were subjected to cleaning by argon sputtering for 30 min under a voltage of 600 V and pressure of 150 Pa, and then were performed in gas of pure nitrogen. The furnace temperature was increased at a rate of 10  C/min to the setting temperature, and the specimens were allowed to cooled in the furnace in working gas after the nitriding process was completed. During DC plasma nitriding process, the plasma emission spectra was monitored by EPP2000 spectrometer ranging from 190 to 1100 nm. X-Ray diffraction was carried out on a PW3040/60 diffractometer using the CuKa1 (l ¼ 1.5406 Å) radiation source. The Table 1 Chemical composition of forged CoCrMo alloy. Co/wt.%

Cr/%

Mo/%

Mn/%

Ni/%

Si/%

C/%

Bal.

27.7

6.0

0.62

0.12

0.76

0.10

215

Fig. 1. Schematic diagram of plasma nitriding equipment.

analysis of each specimen was run in grazing incidence X-ray diffraction (GIXRD) between 30 and 70 , with step interval 0.1 and incidence angle of 5 . The surface microstructure analysis was carried out on Ultra Plus scanning electron microscopy, and the measurement of the modified diffusion layers thickness was obtained from Axiovert 200 MAT metallographic microscope. The roughness of Ra was measured by using TR200 roughness tester. The microhardness was measured by 401MVD microhardness tester, with the applied loads from 100 gf to 500 gf and the loading time of 10 s. Surface wettability was evaluated by the contact angle. The sessile drop method was carried out for the contact angle measurement on SL200 optical dynamic/static contact angle meter under room temperature in the air. The amount of distilled water used for the measurement was 0.5 mL. The wear tests were carried out at room temperature on a ball-on-disc tribotester at the constant load of 5 N under ambient environment condition without a lubricant. During the whole experiment, a 3 mm diameter SiN ceramic ball was used to form wear tracks on the surface of the selected samples. The disc rotated at 168 rpm and a total number of 1800 s was set for each test. The wear rate can be calculated by equation described in literature [6]. In order to calculate the wear volume, the substrates weight was recorded before and after the wear experiments. Wear scar widths after wear tests were examined using Axiovert 200 MAT metallographic microscope. In view of the scatter of the microhardness and contact angle data, the microhardness and contact angle values were averaged over at least four measurements at different points under the same condition. 3. Results and discussion Fig. 2 shows the nitrogen optical emission spectra ignited by DC power supply at 550 V, 200 Pa pressure. It is found that the spectral lines include a relatively strong excited neutral molecules N*2 (268.2 nm, 315.9 nm, 337.1 nm, 357.1 nm, 380.5 nm, 662.3 nm and 775.3 nm in wavelength, respectively) and a relatively weak excited molecule ions Nþ 2 (391.4 nm) [7,18]. However, there is no spectral evidence for excited neutral atoms N* and excited ions Nþ. Baldwin et al. [18] suggested that with the plasma nitriding treatment on stainless steels, the nitriding reaction was attributed to the excited neutral molecules N*2, and similar research can also be found in literature [19]. Although previous studies were mainly focused on the role of the excited neutral molecules N*2, in this work, the effect of the relatively weak excited molecule ions Nþ 2 on the nitriding of medical forged CoCrMo alloy also should be noted, in addition to the excited neutral molecules N*2. The discharge nitrogen molecule Nþ 2 is the most active specie, and it can be easily attracted by the negative biased substrate and react with the elements of alloy surface [20].

216

D.C. Ba et al. / Vacuum 119 (2015) 214e222

Table 2 Different nitriding processes of forged CoCrMo specimens. Sample no.

Temperature ( C)

Pressure (Pa)

Time (h)

Voltage (V)

Mean surface roughness (mm)

Mean friction coefficient

1 2 3 4 5 6 7 8

Untreated 800 800 800 800 800 800 800

0 200 200 200 200 200 200 200

0 4 4 4 4 2 6 8

0 350 450 550 650 650 650 650

0.09 0.13 0.16 0.22 0.28 0.26 0.35 0.39

0.30 0.22 0.26 0.33 0.41 0.35 0.46 0.52

Fig. 2. Nitrogen optical emission spectra (OES).

In the process of plasma nitriding forged CoCrMo alloy, ion bombardment energy plays a decisive role in film quality [5]. In order to provide even better analysis of ion bombardment effect on the film quality, the concept of ion beam flux density J was introduced in this work. Fig. 3 displays the trend of ion beam flux density and OES intensity of two spectral features as a function of treatment voltage. As can be seen, the ion beam flux density data and the intensity of the molecular-ion spectral line (N*2 337.1 nm and Nþ 2 391.4 nm) in the nitrogen plasma both increase significantly with the elevation of treatment voltage, and the minimum and maximum ion beam flux density values have been measured as 1.2 mA/cm2 at 350 V and 5.8 mA/cm2 at 650 V. Apparently, the operating voltage can be able to significantly influence the ion

Fig. 3. Ion beam flux density and intensities of spectral lines from N*2 and Nþ 2 at different operating voltages.

beam flux density J. Although the ionization ability of nitrogen and ion beam flux density increases with increasing treatment voltage, it should be noted that extremely high treatment voltage may possess a negative effect on plasma-surface nitriding interactions and even lead to the failed operation because of the occurrence of excessive sputtering on the sample surface. Owing to the complexity of plasma-surface interactions and to the nature of the excited neutral molecules N*2 and excited molecule ions Nþ 2 involved in these nitriding interactions, the appropriate treatment voltage should be selected to maintain the dynamic equilibrium between the sputtering of the sample surface and the reaction of the gas species on the sample surface [8]. The XRD patterns of plasma nitrided specimens at various operating ion beam flux densities and time periods compared with untreated substrate material are presented in Figs. 4 and 5, respectively. As to the untreated specimen shown in Figs. 4 and 5, typical diffraction peaks with relatively high intensity identified as a (111), a (200), ε (100), and ε (101) are obviously demonstrated, and there is a higher concentration of the a phase or the a/ε combination. The a phase (in some papers also known as g) with face centered cubic crystal structure is a non-equilibrium phase formed during cooling, while ε phase with hexagonal closed packaged structure is an equilibrium phase, and it distributes as thin plates in the CoCrMo alloy. The patterns indicated distinctive phases including s, CrN and Cr2N, which formed during the plasma nitriding process of forged CoCrMo alloy. Fig. 4 shows the XRD evolution after nitriding at 800  C-4 h at different ion beam flux densities. It is determined that s phase forms at the nearly same angle where the ε phase disappears after nitriding treatments. According to the diffraction patterns of untreated and nitrided specimens, the peaks of a phase and s phase have been both shifted to higher 2q values after plasma nitriding, and the amount of shifting increases with the operating voltage and ion beam flux

D.C. Ba et al. / Vacuum 119 (2015) 214e222

217

Fig. 4. XRD patterns of untreated and nitrided forged CoCrMo alloys at different ion beam flux densities for 4 h.

Fig. 5. XRD patterns of untreated and nitrided forged CoCrMo alloys at different operating times.

density. The peak shifting of a phase and s phase was probably due to a gradient in nitrogen near the sample surface, overlapping of adjacent reflections from phases or residual stresses, resulting in lattice distortion [1,8,21]. At low ion beam flux density J (1.2 mA/ cm2, 350 V), the nitrided specimen structure mainly consists of a and s cobalt parent phases. When the ion beam flux density reaches 2.6 mA/cm2, the CrN phase peak is appeared at about 62.5 . As the ion beam flux density continues to rise, besides CrN and a and s cobalt parent phases, a new peak of Cr2N phase at 55 can be also observed. Meanwhile, the effects of the nitriding time on specimens are shown in Fig. 5. Similar to the XRD evolution in Fig. 4, the shifting amount of a phase and s phase, and the diffraction peak intensities of CrN and Cr2N have also increased with increasing treatment time. In addition, it should be noted that the amount of s phase at 46.5 increases significantly with the elevation of ion beam flux density J and time. However, the amount of s phase at about 41.5 decreases with nitriding time at 5.8 mA/cm2, and this phase almost disappears when the samples are nitrided above 6 h, as shown in Fig. 5. Previously, in the studies [9,15], it was observed that s phase disappeared when the forged CoCrMo alloy was nitrided at 800  C. However, as seen in Figs. 4 and 5, the significant amounts of s phase at about 46.5 still can be all detected in the nitrided samples at 800  C. From the Fig. 5, it can be seen that the peak intensity of s

phase decreases while the CrN and Cr2N phases increases with increasing nitriding time. In the subsequent experiments, the experimental results confirm that the s phase of nitrided specimen completely disappears with increasing process temperature up to 900  C [5]. Since the s phase is a cobalt rich phase, the formation of this phase is related to the concentration of cobalt or Co/Cr ratio increase. It has been observed that the intensities of CrN and Cr2N phases evidently increase with the elevation of ion beam flux density and time, which result in decomposition of chromium from the s phase. Therefore, chromium element may bond with active species (N*2 and Nþ 2 ) and form CrN and Cr2N phases. In addition to these, cobalt nitride phases and molybdenum nitride phases cannot occur in the nitriding layers owing to the weaker affinity to nitrogen active species. The SEM micrographs of the unnitrided and plasma nitrided forged CoCrMo alloys are given in Fig. 6. According to the Fig. 6(a), the unnitrided specimen exhibits a relatively smooth and featureless surface while the nitrided specimen surface becomes rougher due to the nitrogen ion Nþ 2 bombardment. From the surface roughness of all the specimens presented in Table 2, it is clearly seen that the surface roughness of the unnitrided sample was about 0.09 mm and the nitrided samples were measured between 0.13 and 0.39 mm. In addition, it was also observed that the surface roughness increases with the ion beam flux density and time elevation, as

218

D.C. Ba et al. / Vacuum 119 (2015) 214e222

Fig. 6. SEM images of forged CoCrMo specimens (a) untreated (b) 2.6 mA/cm2, 4 h (c) 5.8 mA/cm2, 8 h.

depicted in Table 2. As can be seen in Fig. 6(b) and (c), the surfaces of nitrided specimens are covered by particles with different sizes. At higher ion beam flux density and longer treatment time, compact nanoparticles are distributed on the coating surface while some surface defects like micro-cracks and micro-porous occur at lower ion beam flux density and shorter treatment time owing to the insufficient ion bombardment energy. Moreover, as ion beam flux density and time increasing, the small size particles are inclined to aggregate and form some larger size particles, which can also lead to the higher surface roughness. The cross-sectional micrographs of the CoCrMo sample plasma nitrided at 2.6 mA/cm2 for 4 h are shown in Fig. 7(a). According to the pattern, the nitrided layer is composed by a double layer structure. The outer layer is a uniform and continues layer which formed on top of surface. The inner layer is a diffusion layer which formed on the vicinity of the substrate. The formation of diffusion layer is due to nitrogen atoms taking interstitial places among cobalt parent structure and forming surface layers by expending the lattice. From XRD analysis, it is believed that s, CrN and Cr2N phases form the outer layer of the nitrided coating. As the ion beam flux density and time increases, the diffused layer decreases or even dissolves and the possibility of CrN and Cr2N formation increases. The variation of nitrided layer thickness as functions of ion beam flux density and time periods is given in Fig. 7(b). It is found that the nitrided layer thickness increases on the ion beam flux density increase at fixed nitridation time of 4 h or at treatment time increase at fixed ion beam flux density of 5.8 mA/cm2, and the coating thickness linearly increases with the process parameter growth. The layers thickness of nitrided samples had been measured from 6.1 mm to 16.8 mm and the maximum and minimum values were obtained from the nitrided samples at 1.2 mA/cm2-4 h and at 5.8 mA/cm2-8 h, respectively. At lower nitriding ion beam flux density and shorter treatment time, due to the lower energy and density of N*2 and weaker reaction between

Nþ 2 and alloy surface, the forming of nitrided layer depends mainly on lattice consumption induced by nitrogen atoms taking interstitial places among cobalt parent structure. As the treatment ion beam flux density and time increases, due to faster nitrogen atoms diffusion rate and more intensive nitrogen atoms and ions movement, the intensity CrN or Cr2N layer formation and the nitrided layer thickness increase [22]. Fig. 8 shows the surface microhardness values of unnitrided and nitrided specimens as a function of the applied load from 100 gf to 500 gf. As shown in Fig. 8, the surface microhardness value of untreated specimen was measured as about 320 HV, and all the plasma nitrided specimens exhibited an important increase in the surface microhardness compared to unnitrided CoCrMo specimen at different applied loads. It is a fact that the surface microhardness is monotonically dependent on the ion beam flux densities and time periods, and the hardness value for the nitrided CoCrMo alloy increases monotonically with the increase of ion beam flux density or time. From Fig. 8(a) and (b), the maximum value of 920 HV0.2 measured from the surface nitrided at 5.8 mA/cm2-8 h is increased more than two times in reference to that of unnitrided specimen. The improvement of the microhardness is believed to be mainly due to the thickness of case depths and the formation of s, CrN and Cr2N phases. The s phase hardness is similar to that of quenched FeeC (950 HV) or higher, while the Cr2N and CrN hardness values are 2175 HV0.05 and 2740 HV0.05, respectively [23,24]. The higher hardness of hard phases (CrN and Cr2N) increases the surface hardness of the nitriding forged CoCrMo samples. At lower ion energy beam density and shorter process time, the nitrided specimen structure mainly consists of s cobalt parent phase, the thickness of case depth is relatively small, and the increasing range of hardness is also small. At higher ion beam flux density and longer process time, with the formation of CrN or Cr2N phase and the thicker case depth, the hardness of nitrided sample exhibit an important increase.

D.C. Ba et al. / Vacuum 119 (2015) 214e222

219

Fig. 8. Microhardness of nitrided of forged CoCrMo alloys as functions of ion beam flux density and process time. Fig. 7. (a) Cross section image of nitrided CoCrMo alloy at 2.6 mA/cm2, 4 h (b) layer thickness of nitrided CoCrMo alloys at different treatment parameters.

The interaction between metal surface and cell/tissue which depends strongly on the surface behavior is important for the biocompatibility of metallic implants [25]. As an important factor to characterize surface performances, surface wettability, which depends on intrinsic factors of surface chemistry (such as elements and functional groups) and extrinsic factor of surface roughness, is generally evaluated by the contact angle. The variation of contact angles with untreated and nitrided specimens at different process parameters are shown in Fig. 9. As shown in Fig. 9(a) and (b), the contact angle of the untreated specimen is about 83 , and the value of contact angle significantly decreases after nitriding treatment which means the occurrence of a higher hydrophilic surface on the CoCrMo alloy. Some research shows that the hydrophilic surface is beneficial to the cell growth. From the change curve of contact angles, it is easily observed that the values of contact angle both decrease with increasing nitriding ion beam flux density and time, and the surface wettability is inclined to hydrophilicity. For examples, the value of contact angle is about 39 at 1.2 mA/cm2-4 h, and then the contact angle decreases to about 8 at 5.8 mA/cm2-8 h. Previous investigations confirmed that the contact angle of a material is determined by the surface chemistry, like elements and functional groups, and surface roughness [25,26]. After nitriding treatments, the surfaces of nitrided specimens show hydrophilic properties because of the larger proportion of polar component of surface free energy induced by nitrides structures [27]. Wenzel's

equation properly predicts the effect of surface roughness on contact angle, and it shows that the increase of surface roughness may reduce the contact angle if the contact angle of untreated sample is lower than 90 , but the roughness will increase the contact angle if the contact angle is higher than 90 [25]. Obviously, as shown in Table 2, the surface roughness increases with the elevation of ion beam flux density and time. Thus, it is not surprising to find that the specimen nitrided at 5.8 mA/cm2-8 h has the lowest contact angle of 8 . Nevertheless, drastic difference could not be expected due to the insignificant surface roughness difference among nitrided specimens. Fig. 10 represents the friction coefficient of untreated and nitrided specimens with time under dry conditions. As can be seen, the friction and wear process for both untreated and as-treated specimens consists of run-in-period behavior and stability behavior. In the initial stages of the sliding, the friction coefficient increased to maximum and then decreased to the steady values after about 200 s. From Fig. 10 and Table 2, the mean friction coefficient value of untreated specimen was about 0.30. In the case of nitrided specimens, the layers thickness and surface roughness may have great effect on the friction coefficient values. At lower operating voltages, the specimens show lower friction coefficient values, while the samples treated at higher operating voltages show higher friction coefficient values than the untreated samples. Despite the fact that the nitrided specimens at the lower operating voltages and shorter treatment time show lower case depths and surface roughness than that at the higher voltages and longer treatment time. It is observed that the friction coefficient values

220

D.C. Ba et al. / Vacuum 119 (2015) 214e222

Fig. 11. Wear rate and wear scar widths of untreated and nitrided forged CoCrMo alloys.

Fig. 9. Contact angles of plasma nitrided CoCrMo alloys versus ion beam flux density (a) and process time (b).

generally increase with operating voltages and time. However, a significant variation cannot be found in the mean friction coefficient values of the samples nitrided under different working parameters. Fig. 11 presents the relation between wear rates and wear scar widths of the specimens prepared at different process parameters.

Compared with the untreated specimen, the wear rates and wear scar widths of all nitrided specimens have obviously decreased, which show the better wear resistance under ambient environment conditions without a lubricant. Under various process parameters, the wear rates and wear scar widths reduce basically with the increase in nitriding ion beam flux densities and time. The sample nitrided at the highest ion beam flux density and longest time exhibits the lowest wear rate and wear scar width, and the wear resistance increases more than two times as compared to that of the untreated specimen. Moreover, it should be noted that the wear rates of specimens nitrided at 5.8 mA/cm2-6 h and 5.8 mA/ cm2-8 h did not change significantly. In Fig. 12, the wear tracks of the untreated and nitrided specimens at various process parameters are given. As can be seen in Fig. 12(a), a lot of abrasive debris and plowing grooves occur in the wear track of the untreated sample, and the scratching is wide and deep. Excessive adhesion and plastic deformation are evidently observed on the worn surface at initial friction stage, which confirm that adhesive wear is the dominating wear mechanism of the untreated sample under dry condition. The wear mechanisms are different with the untreated and treated CoCrMo alloys during different friction stages, and the adhesive wear may transform to abrasive wear if nitrided layer breaks down and creates debris with the increase of friction time and sliding distance. Fig. 12(b) and (c) show the wear scars of the nitrided specimens at different process parameters. While the ploughings and fatigue spallings occur on

Fig. 10. Friction coefficient of untreated and nitrided forged CoCrMo alloys under dry wear conditions.

D.C. Ba et al. / Vacuum 119 (2015) 214e222

221

Fig. 12. Micrographs of wear tracks on the forged CoCrMo alloys under dry conditions (a) untreated (b) 2.6 mA/cm2, 4 h (c) 5.8 mA/cm2, 8 h.

the wear tracks of the nitrided specimens at lower nitriding ion beam flux density and shorter time, shallowing scratching and fewer abrasive debris are observed on the worn surface of the specimen nitrided at higher nitriding ion beam flux density and longer time. The thick nitrided layer effectively prevents the ball to reach to the substrate and decreases the amount of plastic deformation. As a result, the main wear mechanisms for the nitrided specimens are fatigue wear, abrasive wear and slight adhesive wear. Evidently, the wear resistance improvement of the nitrided specimens is related to the CrN and Cr2N phases formed in the nitrided layer and the thick nitrided layer. It is known that the CrN and Cr2N are the hard phases with good adhesion. As shown in Figs. 4 and 5, with the nitriding ion beam flux density and time increase, the contents of CrN and Cr2N phases also increase. In addition, the nitride layer thickness and surface microhardness are also increased after nitriding treatments. The strengthening of the surface may lead to the reduction of the surface wear in friction and wear. For example, the specimen nitrided at lower ion beam flux density and shorter time with thinner nitrided layer thickness may easily reach to the SiN ceramic ball, and lead to the poor wear resistance. Meanwhile, high hardness, enough thickness and good adhesion of the nitrided layer induced by higher ion beam flux density and longer time may effectively hinder the contact between the friction ball and substrate. However, it is worth to note that the wear resistance does not further improve with the increasing of the ion beam flux density and time due to the increasing of Cr2N phase content. The hardness and wear resistance of Cr2N phase are less than those of the CrN phase [6]. Thus, a reasonable process parameter should be selected during nitriding treatment. 4. Conclusions Untreated forged CoCrMo alloys show very weak biocompatibility and unacceptable tribological behavior. Plasma nitriding

process produces thick, hard, highly hydrophilic and wear resistance layers on CoCrMo alloy for biomedical applications. Therefore, DC plasma nitriding processes on forged CoCrMo alloys were carried out at various ion beam flux densities for different durations of 2e8 h, and the effects of the ion beam flux density as well as deposition time periods on the nitriding behavior were investigated. The OES data suggested that the excited neutral molecules N*2 and excited molecule ions Nþ 2 played important roles in DC plasma nitriding. The intensities of N*2 and Nþ 2 , as well as ion beam flux density, increased with increasing operating voltages. At lower nitriding ion beam flux density and shorter time, the nitrided layer structure was composed of s and CrN phases; at higher nitriding ion beam flux density and longer time, s and CrN, as well as Cr2N phases, appeared at the surface of nitrided specimens. The intensities of CrN and Cr2N phases increased with increasing nitriding ion beam flux density and time; however, the intensity of s phase increased with increasing ion beam flux density while decreased with time increasing. With increasing nitriding ion beam flux density and time, layer thickness, surface roughness, microhardness and wear resistance increased while water contact angle decreased, and surface became highly hydrophilic. Due to the effect of surface roughness on the wettability, the CoCrMo alloy nitrided at 5.8 mA/cm2-8 h had the lowest contact angle of 8 . As compared with untreated CoCrMo alloy, all CoCrMo alloy after plasma nitriding treatments had lower wear rates and wear scar widths and the specimen nitrided at 5.8 mA/cm2-8 h, exhibited the lowest wear rate and better dry-sliding wear resistance. The wear mechanism for untreated specimen was mainly adhesive wear while the nitrided specimens changed the wear mechanism to fatigue wear, abrasive wear and slight adhesive wear because of breakage of nitrided layers. From the experimental results shown in this paper, it should be recognized that rising both plasma nitriding ion beam flux density and time lead to increasing of structural and tribological properties of CoCrMo alloy.

222

D.C. Ba et al. / Vacuum 119 (2015) 214e222

Acknowledgment This work was supported by the Project Sponsored by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (2014475) and Shenyang Science and Technology Plan of China (No. F12028200).

References [1] [2] [3] [4]

[5] [6] [7] [8]

€ J Bionic Eng 2014;11:303e10. Çelik A, Aslan M, Yetim AF, Bayrak O. Milosev I, Strehblow HH. Electrochim Acta 2003;48:2767e74. Lutz J, Díaz C, García JA, Blawert CS. Surf Coat Technol 2011;205:3043e9. Okur S. Structural, compositional and mechanical characterization of plasma nitrided CoCrMo alloy. The Graduate School of Engineering and Sciences of _ Izmir Institute of Technology, Master of Science in Physics; 2009. Xu L, Ba DC, Wang Q. Chin J Vac Sci Technol 2014;34:1019e23. Wang QL, Zhang L, Dong JD. J Bionic Eng 2010;7:337e44. Baldwin MJ, Fewell MP, Haydon SC. Surf Coat Technol 1998;98:1187e97. Wang Q, Ba DC, Ming Y, Xu L. Plasma Sci Technol 2014;16:960e3.

€ Alsaran A, Kaymaz I, _ Yetim AF. Surf Coat Technol 2008;202: [9] Çelik A, Bayrak O, 2433e8. [10] Hosseini SR, Ahmadi A. Vacuum 2013;87:30e9. € [11] Oztürk O, Türkan U, Eroglu AE. Surf Coat Technol 2006;200:5687e97. [12] Bazzoni A, Mischler S, Espallargas N. Tribol Lett 2013;49:157e67. € re JP, Drouet M. Surf Coat Technol 2010;204: [13] Pichon L, Okur S, Oztürk O, Rivie 2913e8. [14] Dong JD, Wang QL, Zhang L. China Surf Eng 2010;23:73e7. [15] Wang QL, Zhang L, Shen H. Surf Coat Technol 2010;205:2654e60. [16] Wei R, Booker T, Rincon C. Arps J Surf Coat Technol 2004;186:305e13. [17] Laning BR, Wei R. Surf Coat Technol 2004;186:314e9. [18] Baldwin MJ, Fewell MP, Haydon SC. Surf Coat Technol 1998;98:1187e91. [19] Michel H, Czerwiec T, Gantois M. Surf Coat Technol 1995;72:103e11. [20] Leyland A, Fancey KS, James AS, Matthews A. Surf Coat Technol 1990;41: 295e304. [21] Ferkel H, Glatzer M, Estrin Y. Scr Mater 2002;46:623e8. [22] Li GJ, Peng Q, Li C. Surf Coat Technol 2008;202:2749e54. [23] Bertrand G, Mahdjoub H, Meunier C. Surf Coat Technol 2000;126:199e209. [24] Dong H. Int Mater Rev 2010;55:65e98. [25] Lin Z, Lee GH, Liu CM. Surf Coat Technol 2010;205:391e7. [26] Zhao CS, Liu XD, Nomizu M, Nishi N. Biomaterials 2003;24:3747e55. [27] Jiang N, Zhang HJ, Bao SN, Shen YG, Zhou ZF. Physic B 2004;352:118e26.