Available online at www.sciencedirect.com

www.elsevier.com/locate/jmbbm

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 1

http://dx.doi.org/101751-6161/& 2014 El

nCorrespondingþ351 253510175x517

E-mail address: s

Research Paper

Biotribological behavior of Ag–ZrCxN1�x coatingsagainst UHMWPE for joint prostheses devices

S. Calderon V.a,b,n, J.C. Sanchez-Lopezc, A. Cavaleirob, S. Carvalhoa,b

aGRF-CFUM—Physics Department, University of Minho, 4800-058 Guimarães, PortugalbSEG-CEMUC Mechanical Engineering Department, University of Coimbra, 3030-788 Coimbra, PortugalcInstituto de Ciencia de Materiales de Sevilla (CSIC-US), Avda. Américo Vespucio 49, 41092 Sevilla, Spain

a r t i c l e i n f o

Article history:

Received 13 August 2014

Received in revised form

23 September 2014

Accepted 29 September 2014

Available online 8 October 2014

Keywords:

ZrCN

UHMWPE

Sputtering

Tribology

Hardness

Amorphous carbon

.1016/j.jmbbm.2014.09.028sevier Ltd. All rights rese

author at: Universidade465; fax: þ351 253510461ecave44@gmail.com (S. C

a b s t r a c t

This study aims to evaluate the structural, mechanical and tribological properties of

zirconium carbonitrides (ZrCxN1�x) coatings with embedded silver nanoparticles, produced

with the intention of achieving a material with enhanced multi-functional properties,

including mechanical strength, corrosion resistance, tribological performance and anti-

bacterial behavior suitable for their use in joint prostheses. The coatings were deposited by

direct current (DC) reactive magnetron sputtering onto 316 L stainless steel, changing the

silver content from 0 to 20 at% by modifying the current density applied to the targets.

Different nitrogen and acetylene gas fluxes were used as reactive gases. The coatings

revealed different mixtures of crystalline ZrCxN1�x, silver nanoparticles and amorphous

carbon phases. The hardness of the films was found to be mainly controlled by the ratio

between the hard (ZrCxN1�x) and soft (Ag and amorphous carbon) phases in the films,

fluctuating between 7.4 and 20.4 GPa. The coefficient of friction, measured against ultra-

high molecular weight polyethylene (UHMWPE) in Hank’s balanced salt solution with

10 g L�1 albumin, is governed by the surface roughness and hardness. The UHMWPE wear

rates were in the same order of magnitude (between 1.4 and 2.0� 10�6 mm3 N�1 m�1),

justified by the effect of the protective layer of albumin formed during the tests. The small

differences were due to the hydrophobic/hydrophilic character of the surface, as well as to

the silver content.

& 2014 Elsevier Ltd. All rights reserved.

rved.

do Minho, Dept. Física, Campus de Azurém, 4800-058 Guimarães, Portugal. Tel.:.alderon V.).

1. Introduction

The modification and functionalization of material surfaces

have been appealing topics of investigation in the material

science and engineering fields, since the surface is the main

material portion to interact with the environment. In parti-

cular, medical devices, such as surgical tools and prosthesis

devices, are particularly interesting in this field due to the

need of surface multi-functionality in order to withstand

mechanical loading and maintain the biocompatible character,

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 184

including chemical stability and low tendency to microbialcolonization. As a result, multifunctional coatings are beingdeveloped in order to improve the current materials used in themedical devices field. The chemical inertness of the materialshas been usually achieved by coating the devices with very lowreactive ceramic material such as diamond-like-carbon materi-als (Hauert et al., 2013), nitrides (Subramanian et al., 2011) oroxides (Shi et al., 2000), while the antimicrobial capabilities havebeen achieved by introducing antimicrobial agents, such assilver (Betts et al., 2005; El-Kady et al., 2011; Lok et al., 2007; Kellyet al., 2010; Cao et al., 2011; Dowling et al., 2003). The tribologicalproperties, on the other hand, are strongly dependent on thecombination of the materials and lubricant used in the system.For knee and hip joint prosthesis, for instance, 316 L stainlesssteel, Co–Cr–Mo alloys and titanium alloys are usually coupledwith ultra-high molecular weight polyethylene (UHMWPE)because of its excellent properties of bio-compatibility, chemi-cal stability, effective impact load damping, low friction coeffi-cient and superior fracture toughness (Dong et al., 2007; Turellet al., 2005). However, it has been evidenced that the wear ofUHMWPE may be the limiting factor that compromises thelong-term performance of joint prosthesis (Goswami andAlhassan, 2008) by generating polymer debris particles, beinghighly susceptible to osteolysis, a leading cause of failure forsuch devices (Buford and Goswami, 2004). Thus, specific atten-tion has been placed on the causes of UHMWPE wear.

Several authors have reported that the applied load (Donget al., 2007; Rostoker and Galante, 1979; Kim et al., 2011) andthe counterpart material properties, like hardness (Bufordand Goswami, 2004), roughness (Turell et al., 2005; Goswamiand Alhassan, 2008; Buford and Goswami, 2004; Cho et al.,2004) and ability to resist particle formation, are the mainfactors affecting UHMWPE wear. The environment in whichthe materials interact has also been identified to play acrucial role in the final tribological behavior. Studies aboutthe influence of the different components of the simulatedbody fluid have demonstrated that the adsorption of proteinsstrongly affects the wear and coefficient of friction (COF) ofthe system, depending on the ability of the surface to adsorbsuch molecules (Myant and Cann, 2014). Albumin absorptionreduces the wear rate of the material due to the ability of theadsorbed layer to protect the surface, reducing the materialtransfer and direct contact between the surfaces up to someextent (Hall and Unsworth, 1997; Gispert et al., 2006, 2007;Alves et al., 2014). However, contradictory results have beenreported regarding the benefits of the albumin to the frictionproperties; some reports attribute an increase of COF due tothe mutual interactions between the molecules of albumin(Fang et al., 2009) while others (Hall and Unsworth, 1997;Gispert et al., 2006. 2007) state that the protective character ofalbumin reduces the friction. Thus, surface properties suchhydrophobicity and roughness, known to affect directly theprotein adsorption (Gispert et al., 2006; Wang et al., 2012),should not be overlooked in regard to their influence on thetribological behavior.

As a result, a material capable of showing a broad scope ofall the mentioned properties may be of great importance forevolving the medical devices currently in use. Transitionmetal carbonitrides (MeCN) materials may be good candidatesfor that purpose. Among the large MeCN family of materials,

ZrCxN1�x has been proven to have tunable propertiesdepending on the non-metallic elements contents, alteringthe hardness, friction coefficient, and biocompatibility in awide range (Balaceanu et al., 2010; Chen and Liu, 2005; Braicet al., 2011; Grigore et al., 2010; Hollstein et al., 2001; Gu andChen, 2006; Silva et al., 2010; Larijani et al., 2010). However, inorder to provide the system with antimicrobial capabilities itis necessary to add an antimicrobial agent, such as silver,widely used and proven to be efficient. Nonetheless, adding afourth element to the ternary system ZrCxN1�x may alter theproperties, highlighting the importance of determining theproper amount of the constituents (Zr, C, N and Ag) toguarantee an appropriate tribological performance.

Consequently, the aim of this work is to evaluate theinfluence of the chemical, structural and morphological proper-ties of Ag–ZrCxN1�x films on the tribological performanceagainst UHMWPE as counterpart in biological medium. Thisanalysis will help to assess the viability of this system to beused in the field of prosthetic devices. The Ag–ZrCxN1�x filmswould work as a multifunctional coating in which the silverserves as an antibacterial material, the coating itself as a barrierfor the delivery of ions from the substrate (SS316L), and themixture of hard and soft phases as a composite material thatenhance the mechanical and tribological performance withoutcompromising the UHMWPE.

2. Materials and methods

2.1. Coating production

Direct current (DC) unbalanced reactive magnetron sputteringwas used to deposit Ag–ZrCxN1�x films. The sputtering reactorconsists of two magnetrons equipped with pure Zr (99.2 at%,100�200�6mm3) and Ag (99.9 at% 100�200�6mm3) targetslocated in opposite positions in a closed magnetic field config-uration and a rotational substrate holder located at 70mm fromthe targets. A Joule effect resistor positioned at 80mm from thesubstrates is used to regulate the temperature of the chamberand a gas inlet is located in one side of the reactor, homogeniz-ing the gases throughout a gas pipe around the chamber.

Table 1 shows the deposition parameters and variablesmonitored during the process. The chamber base pressurewas set between 2.4�10�4 and 4.2�10�4 Pa and the deposi-tions were carried out using two different acetylene (1.2 and2.4 sccm) and nitrogen fluxes (3 and 9 sccm), attaining work-ing pressures between 0.60 and 0.69 Pa. The zirconium targetcurrent density was maintained at 10 mA cm�2, while thesilver target current density was varied between 0.25 and0.75 mA cm�2 to tune the silver content in the films. The biasvoltage (�50 V), substrate rotation speed (8 rpm) and cham-ber temperature (100 1C) were monitored and kept constantfor all the depositions.

Mirror polished samples of 316 L stainless steel and silicon(100) were used as substrates. The 316 L substrates werepolished using emery paper (from 240 to 2400 grit) and thenmirror polished with a diamond solution to obtain surfaceroughness lower than 8 nm. The samples were cleaned bymeans of an ultrasonic bath in distilled water, ethanol andacetone during 10 min, to enhance the film adhesion. An

Tab

le1–Dep

osition

param

etersan

dco

atings

pro

perties

.

Agcu

rren

tden

sity

[mAcm

�2]

C2H

2

[sccm]

N2

[sccm]

Thicknes

s[lm]

Dep

osition

rate

[lm

h�1]

Rou

ghnes

s[nm]

Zr

[%]

N [%]

C [%]

Ag

[%]

O [%]

CþN Zr

H[G

Pa]

E[G

Pa]

Con

tact

angle

[deg

rees

]

UHMW

PEwea

r[�

10�6

mm

3N

�1m

�1]

SS31

6L–

––

––

870

––

––

––

––

8675

1.49

70.21

ZrC

N/

Ag

ZrC

N1

01.2

32.3

1.9

27718

4933

150

31.0

20.47

2.0

2607

963

79

2.08

70.07

ZrC

N2

0.25

1.2

31.8

1.7

2473

4624

186

60.9

18.672.2

2227

1898

72

1.38

70.02

ZrC

N3

0.75

1.2

32.0

2.1

31713

3820

1620

61.0

12.271.1

1647

1195

73

2.04

70.04

ZrC

N/

Ag/a-

(C,N

)

ZrC

N4

0.5

2.4

31.6

1.5

18710

3518

299

91.4

13.471.2

2067

2010

471

1.69

70.03

ZrC

N5

0.25

1.2

91.1

1.0

1174

3239

187

41.8

11.570.8

1737

1410

173

1.64

70.16

ZrC

N6

0.5

2.4

91.5

1.4

1576

2130

2713

92.8

7.47

0.8

1237

996

71

1.92

70.05

Stan

darddev

iationsofth

eco

mposition

arebe

twee

n0.5an

d1.2at%.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 1 85

in situ etching process, immediately before the coatingsproduction, was carried out using argon atmosphere and apulsed power supply. Thereafter, a 200 nm Zr interlayer wasdeposited before the multifunctional Ag–ZrCxN1�x layer tofurther enhance the adhesion between the film and thesubstrate.

2.2. Structural and chemical characterization

The structure was studied by X-ray diffraction in a PANaly-tical XPert PRO micro-diffractometer with Cu Kα (1.540598 Å)radiation in a grazing angle mode at 11 incidence angle.Thicknesses were evaluated by the Calotest method utilizinga rotation sphere 20 mm in diameter at 1000 rpm for 90 s inorder to obtain the desired wear.

The surface morphology and cross section morphology ofthe films were assessed using a NanoSEM—FEI Nova 200scanning electron microscope, with a secondary electrondetector. Energy dispersive X-ray spectroscopy was utilizedto determine changes on the composition after the tribologi-cal test. The acceleration voltage was kept at 10 keV for bothimages and chemical composition analysis.

The surface roughness of the bare and coated steel wasmeasured with a contact stylus profilometer, using 10 mm ofdistance at 0.1 mm s�1 of scanning speed. Ra and Rq valueswere obtained as an average of 5 measurements in differentzones of the films.

Raman spectra measurements (200–2000 cm�1) were car-ried out to the samples using a LabRAM Horiba Jobin Yvonspectrometer equipped with a CCD (charge–coupled device)detector and a He–Ne laser (532 nm) at 5 mW before and afterthe tribological test, in order to determine alterations on thechemical structure of the films.

The coating chemical composition was analyzed by aCameca SX 50 electron probe micro-analyzer (EPMA), usingstatistical analysis of 5 randomly selected spots in eachsample. The phase composition was then calculated consid-ering that the ZrCxN1�x phase possesses all the Zr atomsavailable in the films (disregarding oxygen contamination)and, therefore, carbon-containing phases are formed by thecarbon and nitrogen in excess ([C]þ[N]–[Zr]) measured byEPMA. The number of atoms in the silver state correspondsexactly to the measured atomic percentage. Nonetheless, itmust be stressed that the presence of residual oxygen in thesystem indicates oxides formation, most probably of Zr, aspreviously shown (Calderon V et al., 2013), and therefore, thephase estimation should only be taken as an approximation.

2.3. Mechanical characterization

The hardness and Young modulus were evaluated by aMicroMaterials Nanotest system equipped with a Berkovichindenter at a maximum load of 5 mN. The indenter wasmaintained 5 s at maximum load and 30 s at lower load forthe thermal drift. The loading and unloading rates weremaintained equal at 0.167 mN s�1. The values reported arean average of 16 points measured.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 186

2.4. Tribological characterization

The tribological properties were evaluated by reciprocatingfriction tests using 10 mm ultrahigh molecular weight poly-ethylene (UHMWPE CHIRULEN 1020) balls in Hank's balancedsalt solution (HBSS) with 10 g L�1 of albumin from bovineserum (BSA) at 37 1C in a CSM ball-on-disk tribometer. Thetest parameters were set to 10 N of applied load; 20 mm s�1 oflinear speed; 10 mm track length; 80,000 cycles of duration at1 Hz of acquisition rate. Under these conditions, the corre-sponding initial Hertzian contact pressure was calculated in40 MPa.

The specific UHMWPE wear rate was calculated as thevolume of the spherical cap corresponding to the worn areameasured on the UHMWPE, divided by the applied load andthe total sliding distance. This wear rate value is believed tobe overestimated since the surface of the UHMWPE is com-posed of plastically deformed microscopic crests, which arethe main parts of the material that were submitted to wear.However, this estimation of the specific wear is in a very goodagreement with the tendency achieved by the measurementsof the material weight losses, carried out before and aftertribological tests, although the measured values with thisprocedure were of the same order of magnitude of thestatistical errors (�10�4 g).

2.5. Contact angles

The contact angles of the samples were measured usingultra-pure water and an automated contact angle measure-ment apparatus (OCA 15 Plus; Dataphysics, Germany).

3. Results and discussion

The thickness of the films was maintained between 1.1 and2.3 mm. Table 1 shows the deposition conditions, chemicalcomposition and some measured properties of the coatings.The composition analysis revealed a successful introduction

Fig. 1 – (a) Coatings X-ray diffractograms, ZrC0.5N0.5 diffraction(ICDD-00-087-0719) are shown for comparison and (b) coatings R

of carbon and nitrogen in the coatings with the change offluxes, as well as a wide range of silver content varying from 0to 20 at%. Table 1 also reports the ratio between the sum ofcarbon and nitrogen, and zirconium ((CþN)/Zr), allowing togroup the samples depending on the amount of CþN excesswith respect to a hypothetical stoichiometric ZrCxN1�x in thecoatings. The first group consists of three samples depositedwith the same acetylene and nitrogen fluxes varying only theAg current density on the target (ZrCN1, ZrCN2 and ZrCN3).These samples have almost no excess of carbon and nitrogen(i.e. (CþN)/ZrE1). The second group formed by samplesZrCN4, ZrCN5 and ZrCN6 possesses an increasing carbonand nitrogen excess, allowing the presence of amorphousC–C and/or C–N phases in the coatings. The set of sampleswill be referred hereafter as ZrCN/Ag or ZrCN/Ag/a-(C,N) forcoatings with and without amorphous carbon-based phases,respectively. With these two groups it will be possible tobetter understand the influence of the silver and amorphouscarbon phases on the ZrCxN1�x mechanical and tribologicalproperties.

3.1. Structural and morphological characterization

The evaluation of the coating structure by X-ray diffractionand Raman spectroscopy revealed a mixture of crystallineand amorphous phases. Fig. 1a and b show the evolution ofthe crystalline phases and the Raman shift as a function ofthe (CþN)/Zr ratio of the samples. XRD diffraction pattern isin agreement with the presence of a face-centered cubic (FCC)ZrCxN1�x phase. The degree of crystallinity of the ZrCxN1�x

phase is strongly dependent on the gas fluxes and the silvercurrent density, showing lower grain sizes as silver andamorphous carbon phases are increased. In the bottom partof the graph, the ZrCN/Ag samples reveal a polycrystallineorientation and broadening of the ZrCxN1�x (200) peak due tothe superposition of the diffraction plane (1 1 1) correspond-ing to an FCC Ag structure, clearly evidenced in sample ZrCN3with 20 at% of silver. This effect has been described in severalworks (Han et al., 2003; Basnyat et al., 2007; Gulbiński and

pattern (ICDD-03-065-8779) and Ag diffraction patternaman spectroscopy as a function of the (CþN)/Zr ratio.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 1 87

Suszko, 2006) and attributed to the immiscibility of silver withthe ZrCxN1�x phase, segregating to the grain boundaries andhindering the crystal growth (Musil, 2000). On the other hand,for ZrCN/Ag/a-(C,N) samples, the ZrCxN1�x crystalline phaseexhibits a reduction of the grain sizes, as the (CþN)/Zr rises,down to a level that the carbonitride phase practicallydisappears (ZrCN6). Most probably this phase will have avery low grain size (lower than 1–2 nm) and, hence, unde-tectable by XRD.

Fig. 1b shows the Raman spectra of the deposited films inthe range of 200–2000 cm�1. The presence of disorderedcarbon-based phases is manifested by two signals locatedclose to 1335 cm�1 and 1570 cm�1, corresponding to the D(disordered) and G (graphitic) bands of disordered carbon,respectively (Braic et al., 2011). The intensity of the Ramanband at 1570 cm�1 was found to exhibit a linear correlationwith the (CþN)/Zr ratio measured by EPMA, indicating that arough calculation of phase percentages could be carried outthrough the EPMA composition, described in the experimen-tal section. The repartition of the atoms into the differentpossible phases is plotted in Fig. 2. This graph is useful tohave a global overview of the phase composition necessary toestablish correlations with the coating functional properties.

Fig. 3 depicts the SEM images in cross-section and surfaceviews of the coatings. The cross-section images of the coat-ings show mostly a columnar growth with less definedcolumns as silver and a-(C,N) phases increase. For ZrCN/Agsamples with higher content of silver, the very small brightspots are believed to be silver nanoparticles. These whiteareas are also evident in the ZrCN/Ag/a-C coatings both insurface and cross-section images.

The surface morphology is much more compact as theamorphous phases rise in the samples ZrCN/Ag/a-(C,N),revealing a predominant effect of amorphous phases incontrolling the morphology of the coatings as compared withsilver. This effect can be explained by the ability of carbonphases to surround the ZrCxN1�x crystal in a composite-likeconfiguration, in contrast to the silver phase, which tends tosegregate and coalesce as isolated nanoparticles.

Fig. 2 – Phase atomic composition calculated by EPMA.(ZrþCþN), a-(CþN) and Ag atomic percentages are associatedto the ZrCxN1�x, amorphous carbon and silver phases,respectively.

Previous publications have highlighted the ability of silverto segregate to the surface in similar matrices, particularlywith the increment of the silver content (Escobar Galindoet al., 2013) and time (aging) (Manninen et al., 2013). However,the Ag–ZrCxN1�x samples prepared in this work did not reveala similar phenomenon, since none of the samples has shownthe presence of silver aggregates on the surface. This factshould be taken into account when comparing the tribologi-cal behavior with reported data as the self-lubricating proper-ties of such soft metallic phase cannot be regarded.

3.2. Mechanical and tribological characterization

3.2.1. HardnessThe hardness is correlated with the structural and morpholo-gical characteristics of the films. This property has beenreported to depend on the ratio between the soft and hardphases in the coatings (Yao et al., 2006). As described in thestructural characterization, the films are mainly composed ofthree phases, metallic Ag, a-(C,N) and ZrCxN1�x. The plot offilm hardness vs. ZrCxN1�x phase percentage gives a linearcorrelation (cf. Fig. 4). This hardness reduction with thediminution of the ZrCxN1�x phase percentage in the film,points out the ratio between hard and soft phases as thekey-feature controlling the hardness properties, independentlyof the relative proportion between the silver and amorphouscarbon phases. Previous works have already highlighted thenegative influence of both silver and carbon amorphousphases on the hardness properties (Han et al., 2003; Basnyatet al., 2007; Yao et al., 2006; Sánchez-López et al., 2012).However, in order to discriminate the influence of silver andamorphous carbon on the hardness separately, a multiplelinear regression was carried out using the Ag and a-(C,N)contents. These calculations found a linear regression with aconfidence level greater than 95% (po0.05) and R2¼0.994(cf. Eq. (1)).

H¼ 20:35 �0:385 Ag phase�0:204a�ðC;NÞphase½GPa� ð1ÞThis empiric mathematical correlation (Eq. (1)) revealed

that the effect of silver on hardness reduction is almost twicethe carbon phase. It must be stressed that this correlation isonly valid for the silver and a-(C,N) percentage ranges studiedin this work.

3.2.2. Coefficient of frictionFig. 5 plots the dependence of the average COF measured foreach sample versus the content of a-(C,N) phase, Rq rough-ness and hardness. In all the cases, the lowest value isobtained with the uncoated substrates pointing the impor-tance of the surface properties in determining the mechan-ical strength of a material. Among the Ag–ZrCxN1�x samples,a significant reduction in the COF is observed when a-(C,N) ispresent. This diminution could be explained in a boundarylubrication regime (that can be expected in these lubricatedtests), since the presence of this phase between surfaceasperities in close contact, where the liquid media couldnot get access or be hindered, would lead to a decrease in thefriction. It is well known the low friction values achieved withC-based coatings in dry sliding; however, a reduction offriction is also observed from 0.15 to 0.12 in the absence of

Fig. 3 – Surface and cross-section SEM images for Ag–ZrCxN1�x coatings (a) ZrCN1, (b) ZrCN2, (c) ZrCN3, (d) ZrCN4, (e) ZrCN5and (f) ZrCN6.

Fig. 4 – Films hardness evolution as a function of ZrCxN1�x

phase content with its correspondent linear regression.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 188

the a-(C,N) phase, implicating that this should not be the onlyfactor influencing the values of the COF. Fig. 5b and c depictsthe tendencies of COFs with the film roughness and hardnessvalues, respectively. The observed increase of the measuredCOFs with these properties can be understood as a dominantploughing component affecting the friction forces. Hard andrough surfaces create more interlocking points than softsmooth surfaces, and thus when rubbing the films materialagain a softer polymer, as the UHMWPE used in the pros-theses, a severe plastic deformation occurs and ploughs out agroove. This action creates a resistance to motion and there-fore, the higher the hardness and the roughness, the higherthe measured friction value is.

In addition to the material properties, the composition of

the fluid is known to play an important role in the measure-

ment of the COF of the tribological system due to chemical

and/or physical interaction with the material surface. Albumin

has demonstrated to protect the surface of the materials

against wear; however, its role on the COF is not well under-stood. Fang et al. (2009), for instance, have reported that theabsorption of albumin on both UHMWPE and CrCoMo surfacesmay induce an increase of the COF due to the interactionbetween the molecules of albumin. Contrarily, Hall andUnsworth (1997) and others (Gispert et al., 2006, 2007) havereported that the protection of albumin reduced both wear andCOF. Despite of being a very complex process that depends onfluid pH, ionic strength and temperature (Gispert et al., 2006),as well as on the hydrophobicity of the surface, several reportsagree that the albumin adsorbs easier to hydrophobic surfacesthan hydrophilic materials (Gispert et al., 2006; Wang et al.,2012). In this case, the contact angle of each film, and thestainless steel, were evaluated in order to investigate thecorrelation between the adsorption of the albumin with thehydrophobic behavior of the coatings. The results showed thatthe only hydrophilic surface (ZrCN1) has the highest COF.However, among the other samples with similar hydrophobi-city levels (see Table 1) the COFs did not show a clearcorrelation with the contact angle. This certainly means thatthe effect of ploughing in friction should be given moreconsideration than the decrease of adhesion forces by effectof chemical bonded layers (albumin) or solid lubricant (amor-phous carbon-based phases).

3.2.3. WearThe wear of the UHMWPE counterpart was evaluated bymeasuring the permanent scar left at the end of thetribological test.

The UHMWPE wear rate did not exhibit significant differ-ences between the samples (1.4 to 2.0�10�6 mm3 N�1 m�1),hampering the analysis of the surface roughness, hardnessand COF effects on the polymer wear, probably explained dueto the formation of a protective albumin layer adsorbed onthe material surfaces. Nonetheless, the UHMWPE wear rateshows a particular trend as a function of the silver content,

Fig. 6 – Films-UHMWPE wear factor evolution as a functionof the silver content. The values in degrees represent thecontact angle of each sample.

Fig. 5 – (a) Films-UHMWPE coefficient of friction evolution as a function of a carbon amorphous phases content, (b) roughnessand (c) hardness of the coatings.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 1 89

indicating that the wear of the UHMWPE increases byincreasing the silver content in the films, as seen in Fig. 6.Recent investigations have found that silver nanoparticles onsurfaces may delay the adsorption of the albumin (Kumariet al., 2013), and therefore, the system may be less protectedby the albumin as the silver nanoparticles increase in thematerial. The hydrophobic character of the samples alsoinfluences the albumin adsorption, as explained in the pre-vious section, thus, the high wear rate exhibited by thesample ZrCN1 may also be due to low albumin adsorptionas it is indeed the single sample showing a hydrophiliccharacter. In summary, the wear of the soft UHMWPEmaterial results from the balance between abrasive wearand plastic deformation, as determined by the film hardness,microasperities and debris particles, and boundary lubrica-tion by albumin, controlled by the silver content and hydro-philic/hydrophobic character.

The wear rate of the films was not measurable in thepresent conditions, and therefore, the effect of phase com-position (ZrCxN1�x, Ag and a-(C,N) phases) on the film wearcould not be concluded.

SEM examination of the film wear tracks and UHMWPE ballscars of two representative samples (ZrCN1 and ZrCN4), belong-ing to the ZrCN/Ag and ZrCN/Ag/a-(C,N) groups respectively, are

shown in Fig. 7. Although it is frequently reported the transfer ofUHMWPE material to the counterpart, none of the samplesshowed the presence of transferred polymer material, as shownin Fig. 7, fact explained by in the wiping effect of the lubricant(Hall and Unsworth, 1997).

The wear debris are usually reported in the ranges of 0.1 to100 mm (Ge et al., 2008), still being controversial the ability ofthe submicrometer particles to generate inflammatoryresponse that results in osteolysis (Hamid Reza SeyyedHosseinzadeh, 2012). In this study the wear debris from thepolymer or films could not be detected either in the fluid oron the surface of the materials, analyzed by STEM and SEMimaging, respectively. Thus, it is believed that the size of theparticles is below 100 nm, since the roughness of the filmswere below 100 nm, and hence, the abrasive crests of thefilms only trim nanometric parts of the polymer.

Raman spectroscopy performed in both the film weartrack and the rubbed UHMWPE did not show any significantchanges after the tribological test, as shown in Fig. 8. How-ever, the UHMWPE exhibits an increasing oxygen contenttrend as the wear rate increases as shown in Fig. 9, whichmay be related to the polymer degradation, since themechanical solicitation may promote the formation of radi-cals through bond scission, inducing the UHMWPE oxidation(Kurtz, 2004).

4. Conclusions

Ag–ZrCxN1�x coatings were produced by DC reactive magne-tron sputtering, varying the silver content between 0 and20 at%. The coatings exhibited crystalline ZrCxN1�x phase,crystalline silver nanoparticles and amorphous carbonphases, being the latter two responsible for a reduction ofthe ZrCxN1�x grain sizes, an improvement of the films densityand a reduction of the hardness, with their increment.Nonetheless, the amorphous carbon phases revealed a largereffect controlling both features, as compared to silver.

The COFs revealed a dominant ploughingmechanism on thefriction forces dependent on the roughness and hardness ofthe films. No wear rate was measurable on the coatings and theUHMWPE wear rate did not exhibit significant differences

Fig. 7 – Scanning electron microscope images of the wear track on the films and wear mark on the UHMWPE of tworepresentative samples (a) and (b) wear track of ZrCN1, (c) wear mark on the UHMWPE used against sample ZrCN1, (d)and (e) wear track of ZrCN4, (f) wear mark on the UHMWPE used against sample ZrCN4. Optical images of the UHMWPE scarrubbed against (g) ZrCN1 and (h) ZrCN4.

Fig. 8 – Raman spectroscopy before and after the tribologicaltest for a representative sample (ZrCN2) and itscorresponding UHMWPE counterpart.

Fig. 9 – Energy dispersive X-ray analysis of oxygen on thepolymer wear mark.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 190

between the samples, attributed to the formation of a protective

layer of albumin on the material surface. The small variance of

the UHMWPE wear rate was correlated to this protective layer,

which is dependent on the hydrophobic/hydrophilic surface

character as well as on the silver content.

The results support the idea of utilizing Ag–ZrCxN1�x

coatings as a surface modification option for multifunctionalpurposes, adding functions such antimicrobial effect, corro-sion resistance and maintaining or even improving theirmechanical and tribological performance.

Acknowledgements

This research is partially sponsored by FEDER funds throughthe program COMPETE- Programa Operacional Factores deCompetitividade and by Portuguese national funds throughFCT-Fundação para a Ciência e a Tecnologia, under theprojects ANTIMICROBCOAT—PTDC/CTM/102853/2008 and inthe framework of the Strategic Projects PEST-C/FIS/UI607/2011”, PEST-C/EME/UI0285/2011 and SFRH/BD/80947/2011.This work has also been supported by the Spanish MECthrough the Consolider FUNCOAT (CSD2008-00023) projectand Junta de Andalucía (P10-TEP-67182 project).

r e f e r e n c e s

Alves, C.F.A., Oliveira, F., Carvalho, I., Piedade, A.P., Carvalho, S.,2014. Mater. Sci. Eng., C 34, 22–28.

Balaceanu, M., Petreus, T., Braic, V., Zoita, C.N., Vladescu, A.,Cotrutz, C.E., Braic, M., 2010. Surf. Coat. Technol. 204,2046–2050.

Basnyat, P., Luster, B., Kertzman, Z., Stadler, S., Kohli, P., Aouadi, S.,Xu, J., Mishra, S.R., Eryilmaz, O.L., Erdemir, A., 2007. Surf. Coat.Technol. 202, 1011–1016.

Betts, A., Dowling, D., McConnell, M., Pope, C., 2005. Mater. Des.26, 217–222.

Braic, M., Braic, V., Balaceanu, M., Zoita, C.N., Kiss, A., Vladescu, A.,Popescu, A., Ripeanu, R., 2011. Mater. Chem. Phys. 126, 818–825.

Braic, M., Braic, V., Balaceanu, M., Vladescu, A., Zoita, C.,Titorencu, I., Jinga, V., 2011. Surf. Coat. Technol. 206, 604–609.

Buford, A., Goswami, T., 2004. Mater. Des. 25, 385–393.Calderon V, S., Galindo, R.E., Benito, N., Palacio, C., Cavaleiro, A.,

Carvalho, S., 2013. J. Phys. D: Appl. Phys. 46, 325303.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 4 1 ( 2 0 1 5 ) 8 3 – 9 1 91

Cao, H., Liu, X., Meng, F., Chu, P.K., 2011. Biomaterials 32, 693–705.Chen, C.-S., Liu, C.-P., 2005. J. Non-Cryst. Solids 351, 3725–3729.Cho, H.J., Wei, W.J., Kao, H.C., Cheng, C.K., 2004. Mater. Chem.

Phys. 88, 9–16.Dong, G.-N., Hua, M., Li, J., Chuah, K.B., 2007. Mater. Des. 28,

2402–2416.Dowling, D., Betts, A., Pope, C., McConnell, M., Eloy, R., Arnaud, M.,

2003. Surf. Coat. Technol. 163, 637–640.El-Kady, A.M., Ali, A.F., Rizk, R.A., Ahmed, M.M., 2011. Ceram. Int.

38, 177–188.Escobar Galindo, R., Manninen, N.K., Palacio, C., Carvalho, S.,

2013. Anal. Bioanal. Chem. 405, 6259–6269.Fang, H.-W., Hsieh, M.-C., Huang, H.-T., Tsai, C.-Y., Chang, M.-H.,

2009. Colloids Surf., B 68, 171–177.Ge, S., Wang, S., Gitis, N., Vinogradov, M., Xiao, J., 2008. Wear 264,

571–578.Gispert, M.P., Serro, A.P., Colaco, R., Saramago, B., 2006. Wear 260,

149–158.Gispert, M.P., Serro, A.P., Colaco, R., do Rego, A.M.B., Alves, E.,

da Silva, R.C., Brogueira, P., Pires, E., Saramago, B., 2007. Wear262, 1337–1345.

Goswami, T., Alhassan, S., 2008. Mater. Des. 29, 289–296.Grigore, E., Ruset, C., Li, X., Dong, H., 2010. Surf. Coat. Technol.

204, 1889–1892.Gu, J.-D., Chen, P.-L., 2006. Surf. Coat. Technol. 200, 3341–3346.Gulbinski, W., Suszko, T., 2006. Surf. Coat. Technol. 201,

1469–1476.Hall, R.M., Unsworth, A., 1997. Biomaterials 18, 1017–1026.Hamid Reza Seyyed Hosseinzadeh, A.E.a.A.S.S., 2012. The bearing

surfaces in total hip arthroplasty—options. In: Fokter, D.S. (Ed.),Material Characteristics and Selection. Recent Advances inArthroplasty, INTECH.

Han, J.G., Myung, H.S., Lee, H.M., Shaginyan, L.R., 2003. Surf. Coat.Technol. 174–175, 738–743.

Hauert, R., Thorwarth, K., Thorwarth, G., 2013. Surf. Coat.Technol. 233, 119–130.

Hollstein, F., Kitta, D., Louda, P., Pacal, F., Meinhardt, J., 2001. Surf.Coat. Technol. 142–144, 1063–1068.

Kelly, P., Li, H., Benson, P., Whitehead, K., Verran, J., Arnell, R.,Iordanova, I., 2010. Surf. Coat. Technol. 205, 1606–1610.

Kim, D.-W., Lee, K.-Y., Jun, Y., Lee, S., Park, C., 2011. Int. J. Precis.Eng. Manuf. 12, 1111–1118.

P. Kumari, P. Majewski, J. Nanomater., 2013 (2013).S.M. Kurtz, The UHMWPE Handbook: Ultra-high Molecular

Weight Polyethylene in Total Joint Replacement, AccessOnline via Elsevier, 2004.

Larijani, M.M., Zanjanbar, M.B., Majdabadi, A., 2010. J. AlloysCompd. 492, 735–738.

Lok, C.N., Ho, C.M., Chen, R., He, Q.Y., Yu, W.Y., Sun, H., Tam, P.K.H.,Chiu, J.F., Che, C.M., 2007. J. Biol. Inorg. Chem. 12, 527–534.

Manninen, N.K., Ribeiro, F., Escudeiro, A., Polcar, T., Carvalho, S.,Cavaleiro, A., 2013. Surf. Coat. Technol. 232, 440–446.

Musil, J., 2000. Surf. Coat. Technol. 125, 322–330.Myant, C., Cann, P., 2014. J. Mech. Behav. Biomed. Mater. 34,

338–348.Rostoker, W., Galante, J., 1979. J. Biomed. Mater. Res. 13, 957–964.Sanchez-Lopez, J.C., Abad, M.D., Carvalho, I., Escobar Galindo, R.,

Benito, N., Ribeiro, S., Henriques, M., Cavaleiro, A., Carvalho, S.,2012. Surf. Coat. Technol. 206, 2192–2198.

Shi, W., Dong, H., Bell, T., 2000. Mater. Sci. Eng., A 291, 27–36.Silva, E., Rebelo de Figueiredo, M., Franz, R., Galindo, R. Escobar,

Palacio, C., Espinosa, A., Calderon V, S., Mitterer, C., Carvalho, S.,2010. Surf. Coat. Technol. 205, 2134–2141.

Subramanian, B., Muraleedharan, C.V., Ananthakumar, R.,Jayachandran, M., 2011. Surf. Coat. Technol. 205, 5014–5020.

Turell, M.E., Friedlaender, G.E., Wang, A., Thornhill, T.S., Bellare, A.,2005. Wear 259, 984–991.

Wang, K., Zhou, C., Hong, Y., Zhang, X., 2012. Interface Focus 2,259–277.

Yao, S.H., Su, Y.L., Kao, W.H., Cheng, K.W., 2006. Surf. Coat.Technol. 201, 2520–2526.