Wear 306 (2013) 296–303

Contents lists available at ScienceDirect

Wear

0043-16

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/wear

The role of a tribofilm and wear debris in the tribological behaviourof nanocrystalline Ni–Co electrodeposits

C. Ma a,n, S.C. Wang a, L.P. Wang b, F.C. Walsh a,c, R.J.K. Wood a

a National Centre for Advanced Tribology at Southampton (nCATS), Engineering Sciences, University of Southampton, Highfield, Southampton SO17 1BJ, UKb Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou, PR Chinac Engineering Materials, Engineering Sciences, University of Southampton, Highfield, Southampton SO17 1BJ, UK

a r t i c l e i n f o

Article history:

Received 18 October 2012

Received in revised form

14 January 2013

Accepted 29 January 2013Available online 10 February 2013

Keywords:

Ni–Co

Tribofilm

Debris

Friction

Wear

48/$ - see front matter & 2013 Elsevier B.V. A

x.doi.org/10.1016/j.wear.2013.01.121

esponding author. Tel.: þ44 7909331442; fa

ail address: c.ma@soton.ac.uk (C. Ma).

a b s t r a c t

The tribofilm and debris generated from wear tests acting as a third body can make important

contributions to a tribo-system. The effect of tribofilm on sliding friction and wear depends strongly on

the applied sliding conditions, and it is still lack of understanding due to insufficient experimental

studies. In the present research, the electrodeposited nanocrystalline Ni–Co coatings with different

cobalt content were prepared. The worn surfaces as well as the wear debris were characterised by

surface analysis techniques after dry sliding tests against a stainless steel ball. The tribofilms containing

iron from the counterparts were formed on the worn surface of the coatings (less than 60 at.% Co),

which exhibited high coefficients of friction and wear rates. No tribofilm or iron transfer from the pin

was found on the Co-rich coatings (more than 70 at.% Co) but a dramatic friction reduction of 50% and

improved wear resistance were experienced. The wear debris contains a mixture of face-centred cubic

(fcc) metallic phase and fcc oxidised phase, irrespective of the coating composition. The oxidised debris

cannot form an efficient lubricative film to promote separation of the sliding surfaces.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Nanocrystalline nickel and Ni–Co layers are important engi-neering coatings which can be used in aerospace, automobile andgeneral industries due to their high strength, low temperaturesuperplasticity, anti-wear and anti-corrosion resistance [1–4].Electrodeposition has been considered as one of the most tech-nologically feasible and economically superior methods toproduce nanocrystalline coatings. In the past few years, thepreparation and optimisation of the coatings have attractedextensive attention. Previous studies showed that experimentalvariables, such as cobalt ions concentration, current density,additive type and bath temperature, can influence the Co contentand grain size, thus determine the microstructure and mechanicalproperties [5–7]. Electrodeposited nanocrystalline Ni coatings andnickel-rich Ni–Co coatings have a fcc structure. With furtherincrease in the cobalt content, the hexagonal close-packed (hcp)structure is formed [8,3]. It has been found that the hcp structureof cobalt-based alloys (Co–Re, Co–Mo and Co–Cr) leads to lowercoefficient of friction and wear rate than their counterparts withfcc structure [9,10]. Inman et al. [11,12] reported that thepresence of hcp structure in a Co–Cr alloy (Stellite 6) restricted

ll rights reserved.

x: þ44 2380595096.

material removal and reduced material transfer when slidingagainst body centred cubic (bcc) structured Incolony MA956.Moreover, Persson et al. [13] suggested that the excellent lowfriction properties and galling resistance of Stellite 21 wereattributed to the transformation from fcc structure to easilysheared hcp structure in the tribofilm which is enriched withCo. Wang et al. [3] also reported that the cobalt-rich alloysexhibited a reduced coefficient of friction and improved wearresistance against an AISI-52100 stainless steel ball compared tothe nickel-rich coatings. The formation of a tribofilm was thoughtto be responsible for this switch in performance. However, therehas been no further investigation of the participation of tribofilmsand wear debris in the friction and wear of Ni–Co coatings thatcould explain such behaviour.

If the generated wear debris is not simultaneously removedfrom the contact surfaces, it can form a three-body contact whichresults in physical and chemical interactions [14]. The layer ofcompacted wear debris, referred to as a tribofilm, may beinadvertently formed on the rubbing surfaces. Many studies showthat the formation of stable tribofilms leads to a significantreduction in the coefficient of friction and the wear rate byredistributing contact stresses, establishing mild wear with theoxide films acted as solid lubricants [15,16]. The properties oftribofilms can lead to remarkable differences in friction and wear.For example, for a TiAlN/VN coating [17], the presence of a thinhydrated tribofilm resulted in the low coefficient of friction (0.53)

C. Ma et al. / Wear 306 (2013) 296–303 297

at room temperature. Nevertheless, it increased to approximately1 at 300 1C due to the breaking of the –OH bonds and thetransformation of the monohydrate to an amorphous (TiAlV)NxOy

with higher inherent friction. Tribofilms also have an accelerativeeffect on the tribological behaviours [18]. Luo et al. [19] reportedthat the coefficient of friction of TiAlCrYN coatings increased fromlower than 0.2 to a high value of 0.6–0.7 after the formation of anadhesive tribofilm. Olofsson and Jacobson [20] found the coeffi-cient of friction was under 0.4 during the sliding contact betweentwo alumina smooth surfaces where no tribofilm was formed. Itincreased to above 0.7 after surface treatment by grit blastingwhich contributed to the formation of the tribofilm by buildingup the agglomerated wear particles. This was explained by thelow hardness of the tribofilm compared to the alumina counter-part. As the role of tribofilm is disputed and poorly understood, itis interesting to study the function of tribofilms on Ni–Co coatingswith varied triboloigical properties under the same tribologicalconditions.

In the present work, Ni–Co coatings having a controlled cobaltcontent were electrodeposited. In order to clarify the role of thetribofilms and debris, chemical and structural characterisations ofthe worn surfaces and debris were conducted by surface analysistechniques. The mechanisms of the tribofilm formation arediscussed.

Fig. 1. Variation of cobalt content as the function of the ratio of Co2þ/

(Co2þþNi2þ) in the electrolyte. The composition was determined by EDS.

2. Experimental details

Nanocrystalline Ni–Co coatings were electrodeposited fromthe all-sulphate acidic bath containing varying concentrations ofcobalt sulphate (0–200 g dm�3). The electrolytes contained nickelsulphate (200 g dm�3), sodium chloride (20 g dm�3), boric acid(30 g dm�3) and sodium dodecyl sulphate (0.2 g dm�3).Saccharin (2 g dm�3) was added to increase the compressivestress in order to compensate the tensile stress introduced bycobalt. The current density was 4 A dm�2 and the bath tempera-ture was maintained at 45 1C by a Grant LTD6G water bath. Thesolution was continuously stirred by a PTFE-coated magneticstirrer bar (6 mm diameter�30 mm length) at 200 rpm. Theanode for electrodeposition was a high purity Ni sheet (40 mm�10 mm�1 mm). The cathode substrate was mild steel AISI 1020(40 mm�10 mm�2 mm), which was held parallel to the anodewith an interelectrode gap of 25 mm. The substrate surface wasground with 320, 800 and 1200 grit SiC paper and fully cleanedwith soap solution followed by rinsing in distilled water toremove contamination. Subsequently, it was immersed into anaqueous 10% hydrochloric acid bath for 20 s to be activated, thenrinsed by deionised water. The electroplating process produced acoating of 5075 mm thickness. As a comparison, hard Cr coatingswere electrodeposited on mild steel substrates from a conven-tional chromic acid bath (CrO3 160 g dm�3, H2SO4 2.5 g dm�3)containing 2 g dm�3 CrCl3 with a direct current of 100 A dm�2

at 45 1C.The surface of the as-deposited Co–Ni coating was investigated

by an Agilent 5500 atomic force microscope (AFM). The wavinesscomponents were subtracted from the initial images to obtainroughness components using the Gaussian filter ð0:08 mm�0:08 mmÞ. A line profile ð5 mmÞ with height measurements wasused to determine the surface roughness. Five measurementswere performed on each coating. The composition of the alloyswas studied by an Oxford Instruments INCA 300 energy disper-sive X-ray spectrometer (EDS) equipped in a Jeol JSM 6500 SEMoperating at 15 kV. The hardness of deposits was measured by aVicker’s microhardness indenter under an applied load of 100 gfor 15 s. An average of five measurements was carried out on eachcoating.

The friction and wear behaviour were evaluated using areciprocating TE-77 tribometer (Phoenix, UK) under the samedry sliding conditions with relative humidity of 40–50% at roomtemperature (25 1C) in air. An AISI-52100 stainless steel ball(diameter 6 mm) with surface roughness (Ra) of 0:48 mm wasused as the counter body with a hardness of 700 HV. Thecalibrated load was 14 N corresponding to the initial Hertziancontact pressure of 2.5 GPa with the sliding frequency of 1 Hz.The sliding stroke was 2.69 mm, while the sliding speed was5.38 mm/s. The total sliding time is 25 min. The friction force wasmeasured by a piezo electric transducer and recorded automati-cally during the tests. The coefficient of friction was calculated bydividing the friction force by the normal load. Three tests wereperformed on each sample and the average dynamic friction forthe whole tests were obtained. The line profile of wear tracks wasmeasured by an Alicona InfiniteFocus Real3D surface profilometerafter wear testing, which illustrated the shape and depth of weartracks for calculating the wear volume. The average crosssectional area was determined by the profiles at five locationsperpendicular to the wear track. The wear volume was calculatedby multiplying the worn cross sectional area by the sliding stroke(i.e. the wear track length). Subsequently, the average wear rateswere determined in order to evaluate the anti-wear properties.Additionally, morphology and composition of the worn surface,debris and the counter body were studied by SEM and EDS. Thecollected debris powders were dispersed in absolute alcohol byultrasonic agitation. A droplet of each suspension was dried on acopper grid with carbon film for subsequent characterisation by aJeol JEM 3010 transmission electron microscope (TEM) operatedat 300 kV equipped with Oxford Instruments INCA 100 EDS.

3. Results

3.1. Composition and surface roughness

Fig. 1 shows that the cobalt content in the coatings graduallyincreased with the increase of the Co2þ/(Co2þ

þNi2þ) ratio in theelectrolytes. The cobalt content of the coatings was alwaysgreater than expected from the bath composition reference line.The anomalous deposition has been explained by the formationand adsorption of metal hydroxide ions on the cathode surface[21]. As the adsorption ability of Ni(OH)þ is lower than Co(OH)þ ,the formation of a Co(OH)þ-enriched film on the substrate

Fig. 2. AFM images of as-deposited Ni–Co alloys: (a) pure Ni, (b) 58 at.% Co–Ni from the electrolyte with 40 g dm�3 CoSO4, (c) 75 at.% Co–Ni from the electrolyte with

100 g dm�3 CoSO4 and (d) 83 at.% Co–Ni from the electrolyte with 200 g dm�3 CoSO4.

Fig. 3. Variation of microhardness with cobalt content.

C. Ma et al. / Wear 306 (2013) 296–303298

surface inhibits the adsorption of Ni(OH)þ and consequently thereduction of cobalt is promoted.

The AFM images of Co–Ni coatings deposited from electrolyteshaving different concentrations of cobalt sulphate are shown inFig. 2. The coatings deposited from the bath with 0, 40 and100 g dm�3 CoSO4 (Fig. 2a–c) are characterised by their smoothsurface with low value of roughness ðRa ¼ 1:522 nmÞ. The depos-its from the bath containing 150 g dm�3 CoSO4 exhibits the samedense and smooth surface morphology. In contrast, 83 at.% Co–Nialloy produced from the bath with 200 g dm�3 CoSO4 has thelens-shaped structure with relatively high surface roughness(Ra¼30 nm) as shown in Fig. 2d. It was further noticed that thehigh Co2þ concentration led to the formation of a lens-shapedmorphology, which can be inhibited by adding more saccharin asa stress reliever [22].

3.2. Microhardness

Fig. 3 presents the microhardness as the function of cobaltcontent in the coatings. The microhardness of pure Ni coating is480 HV. It remains in the range of 470–500 HV as the cobaltcontent increases up to 82 at.%. However, it drops sharply to270 HV for the 83 at.% Co–Ni coating deposited from the electro-lyte with 200 g dm�3 CoSO4. The low microhardness can beattributed to the existance of pores introduced by the looselens-shaped microstructure [22] as shown in Fig. 2d.

3.3. Friction and wear behaviour

As shown in Fig. 4, the coefficient of friction varies in the rangeof 0.45–0.5 as the cobalt content in the coatings increases from

0at.% to 58 at.%. It is reduced to approximately 0.25 when thecobalt content exceeds 70 at.%. The variation of the wear rate asthe function of cobalt content is shown in Fig. 5. The wear rate ofpure nickel coating with higher hardness is relatively low. Itincreases to approximately 2.5�10�4 mm3 N�1 m�1 with 5 at.%cobalt and remains constant until it reaches the highest value asthe cobalt content is 58 at.%. With further increase in the cobaltcontent, the wear rate decreases to 2.2�10�5 mm3 N�1 m�1,which is comparable to that of hard chromium coatings under thesame conditions. Fig. 6 shows the wear tracks of the 83 at.% Co–Nicoating are much narrower and shallower compared with those ofthe pure nickel coating and Ni-58 at.% alloy. The worn surface of

C. Ma et al. / Wear 306 (2013) 296–303 299

the pure nickel coating is shown in Fig. 7a. As listed in Table 1, thehighlighted area A in the wear track contains a high amount ofiron (20 at.%), which is transferred from the pin because of thesevere adhesive wear. Fig. 8a shows the EDS spectrum of the wornsurface compared to that of the as-deposited pure nickel coating.The low-energy peaks O-Ka and the peaks representing Fe wereonly found in the spectrum of the worn surface. It is worthmentioning that the analysis depth of EDS with the acceleratingvoltage of 15 kV is approximately 1 mm. Although the EDS results

Fig. 4. Evolution of coefficient of friction as a function of cobalt content.

Fig. 5. Evolution of wear rate as a function of cobalt content.

Fig. 6. Surface profile of the wear tracks of pure ni

of the worn surface may contain the signal from the coating beneaththe tribofilm, the presence of oxygen and iron confirms theexistence of the tribofilm because none of them was detected inas-deposited coatings. Most of the fine debris particles are found atthe end of the wear track, and some of them are attached on thewear scar (Fig. 7a). The debris contains 40 at.% O, 50 at.% Ni and10 at.% Fe as listed in Table 2. The worn surface of the 58 at.% Co–Ni(Fig. 7b) exhibits three different areas: the smooth area B containing7 at.% Fe, the transition zone with superficial cracks on it, and thearea covered by the debris, which contains 2 at.% Fe. There is moredebris on the worn surface. The reason why the 58 at.% Co–Nicoating has a higher wear rate than the Ni-rich coatings can beattributed to its higher internal stress. As more cobalt atoms areaccommodated in the fcc lattice of nickel, a larger mismatch of thelattices (0.4%) generates higher internal stress, which causes cracksand more debris on the worn surface shown in Fig. 7b. As the cobaltcontent increases to 75 at.%, no iron is detected either on the smootharea C as shown in Fig. 7c or in much reduced amounts of debris.The EDS spectra shown in Fig. 8b confirms that no tribofilmcontaining oxygen and iron was formed. The worn surface hassimilar composition as the as-deposited coating. The 83 at.% Co–Nicoating shows the least debris on the worn track (Fig. 7d), which iscorresponding to its lowest wear rate among present samples. Theasperities of the as-deposited coating shown in Fig. 7d was uni-formly removed to form the smooth wear scar. Iron was not foundeither in area D or in the debris. Furthermore, there is no evidence ofoxidation during the sliding tests in areas C and D. The pin wearscars were also examined by SEM and EDS. The size of the pin wearscar is in accordance with the wear rate. The pin sliding on the83 at.% Co–Ni coating has the smallest wear scar. The pin surfacesare not totally covered by the transferred debris.

The structure of the debris accumulated at the end of the wearscars was further investigated by TEM. Fig. 9 shows the corre-sponding selected area electron diffraction (SAED) patterns. Thecontinuous rings indicate the presence of the nanocrystals.As shown in Fig. 9a a new set of diffraction rings appears comparedto the diffraction pattern of nanocrystalline nickel coatings [1].Based on the calculations of atomic plane spacings, it is found thatthe debris of the pure nickel coating contains a mixture of twophases: nickel face centred cubic (fcc) metallic phase and NiO fccoxidised phase. The structure of iron or iron oxide cannot bedetected. The patterns are indexed in Fig. 10, and all diffractionsare consistent with the diffraction patterns in Fig. 9. While thecomposition of the debris varies as listed in Table 2, the four SAEDpatterns are similar to each other due to the similar latticeconstants of fcc NiO (a¼0.4177 nm) and fcc CoO (a¼0.4260 nm).

ckel, 58 at.% Co–Ni and 83 at.% Co–Ni coatings.

Fig. 7. Worn surface morphology of (a) pure nickel, (b) 58 at.% Co–Ni, (c) 75 at.% Co–Ni and (d) 83 at.% Co–Ni coatings.

Table 1Composition of the area highlighted in the four white rectangles on the worn

surface shown in Fig. 7.

Content (at.%) Worn surface

Area A Area B Area C Area D

O 63 48 0 0

Ni 17 18 23 16

Co 0 27 77 84

Fe 20 7 0 0

Fig. 8. Comparison of EDS spectra of the worn surface and the as-deposited

coating (a) pure nickel and (b) 75 at.% Co–Ni.

C. Ma et al. / Wear 306 (2013) 296–303300

4. Discussion

4.1. Influence of tribofilm formation

In the present work, the results show that there is an oxidationproduct containing Fe on the worn surface of the pure Ni and the58 at.% Co–Ni coatings showing a higher coefficient of friction andwear rate. However, for the 75 at.% Co–Ni and 83 at.% Co–Nicoatings shown in Fig. 7c and d, there is no evidence of oxidationor material transfer from the pin. Without formation of thetribofim, the latter surprisingly exhibit remarkable friction reduc-tion and improved wear resistance.

For the pure Ni and the 58 at.% Co–Ni coatings, the debrisgenerated by the asperity contact between the pin and thecoatings form a three-body contact. It leads to physical andchemical interactions and the adhesive tribofilms are developed(areas A and B in Fig. 7a and b). Other debris particles aregenerated, oxidised, detached and most of them are pushed tothe end of the wear scars, i.e. not over-rolled. Formation of thetribofilm depends on the adhesion between the coating and thecounterpart. With the strong adhesion, the tribofilm consisting ofthe oxides of both surfaces is favoured to be generated due to theextensive transfer from the counterpart and the frictional heat

[23]. Conversely, no tribofilm is found on the worn surface of the75 at.% Co–Ni and 83 at.% Co–Ni coatings which have hcp struc-ture with a (0001) preferred orientation [22]. The typical hcp

C. Ma et al. / Wear 306 (2013) 296–303 301

structure has smaller number of slip systems (one slip plane withthree slip directions) than that of fcc-structured metals (four slipplane each with three slip directions). The preferred slip plane incobalt-rich coatings is the (0001) plane with the greatest spacingbetween planes (c/a¼1.623), which leads to less cohesive forcesbetween (0001) planes. Therefore, much lower rate of shearingoccurs and the extent of adhesion between the 75 at.% Co–Ni and83 at.% Co–Ni coatings and the pins is significantly reduced. Inthis case, the tribofilm cannot be formed under the condition ofmuch weaker adhesion and lower friction heating. No irontransfers from the pin to the worn surface. The enhanced lubricityof the hcp structure of cobalt results in less debris generated and alower coefficient of friction compared with the that of the pure Niand 58 at.% Co–Ni coatings. The high friction coefficient of thelatter can be attributed to higher ductility of fcc structure [24]and the lack of a low-shear strength lubricious component in the

Table 2Composition of the debris collected from pure nickel, 58 at.% Co–Ni, 75 at.% Co–Ni

and 83 at.% Co–Ni coatings.

Content (at.%) Debris

Ni coating 58 at.% Co–Ni 75 at.% Co–Ni 83 at.% Co–Ni

O 40 39 36 49

Ni 50 28 16 4

Co 0 31 48 47

Fe 10 2 0 0

Fig. 9. SAED patterns of the wear debris collected from (a) pure nicke

tribofilm [19]. Inman and Datta [12] reported that the superiorwear resistance of Stellite 6 was attributed to the fewer slipplanes and shear-introduced alignment of the operational basaplane of hcp phases, which agrees with the present study. It wasalso found that at elevated sliding temperature the phase trans-formation from hcp to fcc led to enhanced material removal ofStellite 6 [25].

l, (b) 58 at.% Co–Ni, (c) 75 at.% Co–Ni and 83 at.% Co–Ni coatings.

Fig. 10. Schematic showing the indices of the diffraction patterns with the

solid and dashed lines, respectively, representing the oxidised phase and

metallic phase.

C. Ma et al. / Wear 306 (2013) 296–303302

4.2. The role of the debris on friction and wear properties

The EDS results and SAED patterns show that the pulveriseddebris is partially oxidised. As shown in Fig. 9, the debrisgenerated on the four coatings with different compositions hasa similar fcc phase structure, the performance of which as a solidlubricant is not as good as the hcp structure. Peterson et al. [26]found that the coefficient of friction of Ni-based alloys (Ni–Ta,Ni–W and Ni–Cu) sliding against Al2O3 at room temperature wasaround 0.6, much higher than that at 600 1C (0.3–0.4). Thereduction of friction at elevated temperatures was attributed tothe formation of the metal oxides NiO film. Although the debriscould also be oxidised at room temperature because of the highlocal temperatures in contact areas, the particles could not formthe efficient film to eliminate the metallic contact. In the presentresearch, most of the wear debris accumulated at the end of thewear track. The debris left on the wear track is found at the edgesof the smooth areas of the pure Ni coatings as shown in Fig. 7a.The detached debris of the 58 at.% Co–Ni coating is trapped in thevalley of the broken tribofilm as shown Fig. 7b. There is much lessdebris on the worn surface of 75 at.% Co–Ni and 83 at.% Co–Nicoatings (Fig. 7c and d). Hence, the loose debris is insufficient toseparate the two sliding surfaces. The role of wear debris on thetribological properties depends on the its own performance as asolid lubricant and the formation of a complete oxide film toavoid the metal-to-metal contact. Therefore, the debris has alimited effect on the contact of the rubbing surfaces in thepresent study.

It is worth noting that no hcp structure has been found in thedebris of the 75 at.% Co–Ni and 83 at.% Co–Ni coatings as shownin Fig. 9c and d. As mentioned above, the Co-rich alloys mainlycontain hcp structure. The absence of hcp structure in debriscannot be explained by the hcp–fcc allotropic phase transforma-tion because the hcp phase is not reformed after cooling to roomtemperature. Instead, it is suggested that all the cobalt in debrishas been oxidised prior to the complete oxidation of nickel. It hasbeen reported that the oxide surface of Ni–Co deposit onlyconsists of CoO after annealing at 300 1C. The CoO layer formedby the outward diffusion of Co prevented the oxidation of nickel[27]. It is indicated that nickel has better oxidation resistancethan cobalt in Ni–Co alloys. In the present research, due to thelocal temperature rise introduced by the friction heat, cobalt indebris has been fully oxidised and the formed CoO can protectnickel against oxidation. Therefore, no hcp structure is found indebris and fcc nickel is the only metallic phase as shown in Fig. 9.The explanation is confirmed by the reduced intensity of the (200)ring of metallic phase as the nickel content decreased. Other ringsare closed to those of oxidised phase and difficult to identify theirintensity.

5. Conclusion

With an increase in the cobalt sulphate concentration in theplating bath, the cobalt content of the electrodeposited Ni–Cocoatings gradually increased. The low microhardness of the83 at.% Co–Ni coating is related to its loose lens-shaped micro-structure. By examining the role of the third bodies on thetribological behaviour, the following conclusions have beendrawn:

(1)

The coefficient of friction dropped from 0.45–0.5 to ca. 0.25 asthe cobalt content exceeded 70 at.%. At the same time, thewear rate gradually decreased. The pure nickel coating highhardness exhibited relatively low wear rate.

(2)

A tribofilm consisting of oxides of both the coating and thecounterpart was found on the worn surface of the Ni–Cocoatings having a cobalt content less than 60 at.%. Theformation of the tribofilm was not beneficial to the frictionreduction and the wear resistance.

(3)

Low coefficients of friction were obtained without the forma-tion of tribofilm on the Co-rich coatings (more than 70 at.%Co) and the stainless steel pin contact. The composition of thesmooth worn surface is the same as that of the as-depositedcoating. No pin material transfer from the pin to the coatingwas observed. The improved wear resistance is mainly due toa large portion of hcp structure of the Co-rich coatings, whichcan act as the solid lubricant.

(4)

The debris was nanocrystalline, containing a mixture of twophases: nickel fcc metallic phase and the nickel oxide/cobaltoxide fcc oxidised phase. No hcp structure has been found inthe debris due to the complete oxidation of cobalt. Most of thedebris was pushed to the end of the wear scars. At roomtemperature (25 1C), the loose oxides were insufficient toseparate the two sliding surfaces. The debris had a limitedeffect on the contact of rubbing surfaces.

References

[1] D. Jeong, F. Gonzalez, G. Palumbo, K. Aust, U. Erb, The effect of grain size onthe wear properties of electrodeposited nanocrystalline nickel coatings,Scripta Materialia 44 (3) (2001) 493–500.

[2] S. McFadden, R. Mishra, R. Valiev, A. Zhilyaev, A. Mukherjee, Low-temperature superplasticity in nanostructured nickel and metal alloys,Nature 398 (6729) (1999) 684–686.

[3] L. Wang, Y. Gao, Q. Xue, H. Liu, T. Xu, Microstructure and tribologicalproperties of electrodeposited Ni–Co alloy deposits, Applied Surface Science242 (3) (2005) 326–332.

[4] R. Mishra, R. Balasubramaniam, Effect of nanocrystalline grain size on theelectrochemical and corrosion behavior of nickel, Corrosion Science 46 (12)(2004) 3019–3029.

[5] G. Qiao, T. Jing, N. Wang, Y. Gao, X. Zhao, J. Zhou, W. Wang, High-speed jetelectrodeposition and microstructure of nanocrystalline Ni–Co alloys,Electrochimica Acta 51 (1) (2005) 85–92.

[6] B. Tury, M. Lakatos-Varsanyi, S. Roy, Ni–Co alloys plated by pulse currents,Surface and Coatings Technology 200 (24) (2006) 6713–6717.

[7] N. Myung, K. Nobe, Electrodeposited iron group thin-film alloys: structure-property relationships, Journal of the Electrochemical Society 148 (2001)C136.

[8] D. Golodnitsky, Y. Rosenberg, A. Ulus, The role of anion additives in the electro-deposition of nickel-cobalt alloys from sulfamate electrolyte, Electrochimica Acta 47(17) (2002) 2707–2714.

[9] W. Brainard, D. Buckley, Preliminary Friction and Wear Studies of Cobalt–Rhenium Solid Solution Alloy in Air and in Vacuum, NASA Report NASA-TN-D-6165, 1971.

[10] I. Inman, Compacted Oxide Layer Formation Under Conditions of LimitedDebris Retention at the Wear Interface During High Temperature SlidingWear of Superalloy, Ph.D. Thesis, University of Northumbria at Newcastle,2003.

[11] I. Inman, S. Rose, P. Datta, Studies of high temperature sliding wear ofmetallic dissimilar interfaces II: Incoloy MA956 versus Stellite 6, TribologyInternational 39 (11) (2006) 1361–1375.

[12] I. Inman, P. Datta, Studies of high temperature sliding wear of metallicdissimilar interfaces III: Incoloy MA956 versus Incoloy 800HT, TribologyInternational 43 (11) (2010) 2051–2071.

[13] D. Persson, E. Coronel, S. Jacobson, S. Hogmark, Surface analysis of lasercladded Stellite exposed to self-mated high load dry sliding, Wear 261 (1)(2006) 96–100.

[14] M. Godet, The third-body approach: a mechanical view of wear, Wear 100(1–3) (1984) 437–452.

[15] H. Kato, K. Komai, Tribofilm formation and mild wear by tribo-sintering ofnanometer-sized oxide particles on rubbing steel surfaces, Wear 262 (1)(2007) 36–41.

[16] S. Gupta, D. Filimonov, V. Zaitsev, T. Palanisamy, M. Barsoum, Ambient and550 C tribological behavior of select MAX phases against Ni-based super-alloys, Wear 264 (3–4) (2008) 270–278.

[17] Z. Zhou, W. Rainforth, Q. Luo, P. Hovsepian, J. Ojeda, M. Romero-Gonzalez,Wear and friction of TiAlN/VN coatings against Al2O3 in air at room andelevated temperatures, Acta Materialia 58 (8) (2010) 2912–2925.

[18] S. Biswas, Some mechanisms of tribofilm formation in metal/metal andceramic/metal sliding interactions, Wear 245 (1–2) (2000) 178–189.

C. Ma et al. / Wear 306 (2013) 296–303 303

[19] Q. Luo, Z. Zhou, W. Rainforth, M. Bolton, Effect of tribofilm formation on thedry sliding friction and wear properties of magnetron sputtered TiAlCrYNcoatings, Tribology Letters 34 (2) (2009) 113–124.

[20] J. Olofsson, S. Jacobson, The influence of grain size and surface treatment onthe tribofilm formation on alumina components, Wear 292–293 (2012)17–24.

[21] A. Bai, C. Hu, Composition controlling of Co–Ni and Fe–Co alloys using pulse-reverse electroplating through means of experimental strategies, Electro-chimica Acta 50 (6) (2005) 1335–1345.

[22] C. Ma, S.C. Wang, Q. Luo, C.T.J. Low, R.J.K. Wood, F.C. Walsh, The Hardnessof Porous Nanocrystalline Co–Ni Electrodeposits, in press.

[23] Z. Zhou, W. Rainforth, C. Tan, P. Zeng, J. Ojeda, M. Romero-Gonzalez,P. Hovsepian, The role of the tribofilm and roll-like debris in the wear ofnanoscale nitride PVD coatings, Wear 263 (7–12) (2007) 1328–1334.

[24] I. Hutchings, Tribology: Friction and Wear of Engineering Materials, vol. 40,Butterworth-Heinemann Ltd, 1992.

[25] I. Inman, P. Datta, H. Du, J. Burnell-Gray, S. Pierzgalski, Q. Luo, Studies of hightemperature sliding wear of metallic dissimilar interfaces, Tribology Inter-national 38 (9) (2005) 812–823.

[26] M. Peterson, S. Calabrese, S. Li, X. Jiang, Friction of alloys at high temperature,Journal of Materials Science and Technology 10 (1994) 313–320.

[27] L. Wang, J. Zhang, Z. Zeng, Y. Lin, L. Hu, Q. Xue, Fabrication of a nanocrystallineNi–Co/CoO functionally graded layer with excellent electrochemical corrosionand tribological performance, Nanotechnology 17 (18) (2006) 4614.