- Email: [email protected]
Scratch resistance and tribological properties of DLC coatings under dry and lubrication conditions
Tribology International 56 (2012) 129–140
Contents lists available at SciVerse ScienceDirect
Tribology International journal homepage: www.elsevier.com/locate/triboint
Scratch resistance and tribological properties of DLC coatings under dry and lubrication conditions Neha Sharma, N. Kumar n, S. Dash, C.R. Das, R.V. Subba Rao, A.K. Tyagi, Baldev Raj Materials Science Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603102, TN, India
a r t i c l e i n f o
abstract
Article history: Received 18 July 2011 Received in revised form 23 April 2012 Accepted 27 June 2012 Available online 4 July 2012
Scratch resistance and tribological properties of hydrogenated and hydrogen free diamond like carbon (DLC) coatings were investigated in ambient atmosphere under unlubricated and lubricated conditions using layers of oleic (C18H34O2) and linoleic acid (C18H32O2). Scratch resistance property improved in hydrogenated DLC while using linoelic and oleic acids compared to unlubricated condition. Coefficient of friction (CoF) was found to decrease under ambient dry conditions, while under lubrication condition, it was found to increase with normal load. At low load of 1 N, hydrogenated and hydrogen-free coatings show a super low CoF of 0.001 and high wear resistance with oleic acid. & 2012 Elsevier Ltd. All rights reserved.
Keywords: DLC coatings Fatty acid Scratch resistance Friction coefficient
1. Introduction The diamond like carbon coatings have been explored extensively in recent years due to their low friction, high wear resistance properties and adhesive protection [1]. However, their low surface energy limits them to have any physical and chemical interaction with conventional lubricants and additives. Thus the influence of lubricants, lubricant additives and lubrication process on the friction and wear properties of these coatings needs to be investigated. When sliding occurs under unlubricated atmospheric conditions, two solid surfaces in contact produce surface layer of chemisorbed or physisorbed molecules, or a capillary condensed liquid bridge, between them. Each of these effects can drastically modify their response to adhesion and friction. Adhesion usually decreases, but in the case of capillary condensation, the additional Laplace pressure, or attractive capillary force between the surfaces, may further enhance the adhesion [2]. However, in lubricated sliding contact conditions, the molecules at the interface relax and/or rearrange to form a new equilibrium configuration that is distinctly different from the ones that prevail when the surface is kept in isolation. These rearrangements may involve simple positional and orientational changes of the surface molecules. In a complex condition, new molecular groups that are previously buried below the surfaces can re-appear and intermix with the interface. This behavior commonly occurs with surfaces whose molecules have both polar and nonpolar groups [3]. All these
n
Corresponding author. Tel./fax: þ 914427480081. E-mail address: [email protected] (N. Kumar).
0301-679X/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.triboint.2012.06.020
effects act to enhance the adhesion between the contacting surfaces. Shear stress is also an important factor which can convert the amorphous structure of liquids into crystalline ones when the liquid is confined at high pressure. Applying the critical stress liquid can lead to the formation of a monolayer (10 atomic layers). In this case, the viscosity of the liquid is very high which can cause to freeze the sliding hence strong adhesion acts [2,4]. Crystalline structure of a liquid can also have freezing behavior due to rapid rise in viscosity. Adhesion and friction become low if the viscosity is normal and the behavior of liquid emulates that of a crystalline medium. It can easily shear and provide slippage induced movement. In amorphous state, when the stress is below a critical value needed for transition, the liquid molecule sticks and hinders shearing which raises adhesion and friction. Adhesion and friction are related to slip induced movement and viscosity enhancement. These parameters depend on the structure and bonding fraction of materials participating in a sliding contact. DLC coatings consist of a fractional combination of sp2 and sp3 bonded carbon atoms with a considerable amount of hydrogen. If hydrogen content is about 40 at%, it is called as a hydrogenated DLC and if it is less than 1 at%, it is known as a hydrogen free DLC [5]. Depending upon sp2/sp3 ratio and content of hydrogen, different kinds of DLCs with different properties can be synthesized [6,7]. DLC coatings are called inert coatings because of their low surface energy [8,9]. Therefore, they do not react with various lubricants, oil additives as these coatings do not attract the polar groups from these oils and additives. There are conventional lubrication mechanisms for steels and other metals [10–13]. Most of the mechanical systems work under lubricated conditions. As DLC coatings exist in a wide variety possessing different properties, there
130
N. Sharma et al. / Tribology International 56 (2012) 129–140
is a huge demand for these coatings in various mechanical systems like, bearings [14], gears [15], piston rings and pins [16,17], directinjection fuel systems [18], cutting and forming tools [19,20]. In computer industry, these coatings have application as head–disk interface [21–24]. These coatings also find applications in orthopedic applications in medical industry [25,26]. To reduce the severity of the contact conditions, some of the above performance driven systems are to be operated under lubricated conditions [27,28]. Thus it is necessary to investigate mechanism of lubrication of these coatings. In this regard, some of the important studies have been carried out to address physical and chemical interaction of lubricants, oils and additives with DLC coatings in sliding contact conditions [29–36]. In the present study, the lubricating mechanisms and their influence on the scratch resistance and tribological properties of DLC coatings are investigated. Two types of diamond like carbon coatings, a hydrogen free amorphous carbon (a-C) and a hydrogenated amorphous carbon (a-C:H), were investigated under different loading conditions in the presence of unlubricated and lubricating layer of fatty acids.
2. Experimental In the present study, we designate commercial name of Graphit (hydrogen free) and Dymon (hydrogenated) coatings as G1 and D1, respectively. The G1 coating was deposited by closed field unbalanced magnetron sputter ion plating from the carbon target in a closed field arrangement [37]. Coating D1 was deposited by plasma enhanced chemical vapor deposition using a hydrocarbon gas precursor. Hydrogen content entrained in solid matrix during the deposition of D1 coating in CH4/Ar plasma amounts to 2%. The bias voltage applied to the substrate was pulsed direct current (PDC) and the plasma was further enhanced by the use of additional radio frequency (RF) electrodes [38,39]. Both these coatings are single layered deposited on a steel substrate. Adhesive strength of the coating was tested by using a commercial scratch tester (CSM Instruments, Switzerland) fitted with a Rockwell spherical diamond indenter (tip radius of 200 mm). Scratch-tests were performed using progressive loads from 1 to 40 N for a transverse scratch length of 3 mm in dry unlubricated and lubricated (linoleic and oleic acids) conditions. The scratch tester is equipped with an acoustic emission monitoring sensor which detects acoustic energy in the range of 20–80 kHz. To detect the failure of the coating, the scratches were examined using Scanning Electron Microscope (SEM) and correlated with the acoustic emission peaks. Tribological tests were performed using a ball on disk microtribometer (CSM Instruments, Switzerland) in linear reciprocating mode. A spherical steel ball (100Cr6 SS) of diameter 6 mm with a surface roughness 0.06 mm was used as a sliding body to measure the CoF. Tests were carried out at 1, 4 and 8 N loads with constant sliding speed of 2 cm/s and a stroke length of 3 mm in ambient dry and lubricated conditions with a relative humidity of 58%. For this purpose oleic and linoleic acids were used as lubricating media for the contacting surfaces. Wear dimension was measured using a Dektak 6 M stylus profiler (Veeco, USA). Friction coefficient was found to be reproducible in couple of tests carried out under similar conditions. Calculated wear rate was also found to be more or less similar. A nano-hardness tester (CSM, Switzerland) was used for extracting surface hardness of D1 and G1 samples at a constant load of 2 mN with a Berkovich diamond indenter. Load–displacement curves were generated and resulting curves were analyzed using the Oliver and Pharr method to obtain nano-hardness of the samples. Five indentations were performed on each sample. Raman measurements, using a 5 mW laser power and 514 nm wavelength, were performed to record the local chemical structure present at the coating surface as well as in the wear tracks. Raman measurements in the wear tracks were performed while
stopping the test after each sliding distance. These locations are indicated as d1, d2 and d3. X-ray photoelectron spectroscopy (XPS) was used to study the chemical state of constituents comprising this coating. Prior to acquisition of XPS spectra the specimen was sputtered to remove surface contaminants. An Ar þ ion beam sputtering of specimen surface was carried out for a duration of 3 min. Energy dispersive X-ray (EDX) was carried out to obtain the elemental analysis of the coating surface and the scratched region.
3. Results and discussion 3.1. XPS of coatings surface The sp3 and sp2 fractions of DLC films are deduced from XPS fitting for C (1s) core level peak. These peaks comprise of contribution from C–C (sp3) at a binding energy of 285.3 eV, graphite C–C (sp2) at 284.8 eV and C–O at 286.3 eV. Broad FWHM of C (1s) peak points to presence of carbon with several distinct local bonding environments. The C (1s) XPS in Fig. 1(D1) depicts these contributions. Since the area under each peak is directly related to the concentration of the corresponding phase, the sp3 content is estimated by taking the ratio of diamond peak area
Fig. 1. XPS of coating surface (D1) hydrogenated and (G1) hydrogen free.
N. Sharma et al. / Tribology International 56 (2012) 129–140
131
Table 1 Structural and mechanical properties of DLC coatings. No.
(ID/IG)
(sp3/sp2)
Dt (lm)
H (GPa)
E (GPa)
Ra (nm)
D1 G1
0.89 0.94
82/18 72/28
1.5 1.5
38 34
535 478
64 68
(Dt—coating thickness, H—hardness, E—elastic modulus, Ra—surface roughness of coatings).
over the sum of diamond and graphite peak areas. Each spectral component is a convolution of a Gaussian and a Lorentzian. The contribution of the background is approximated by the Shirley method. Coating composition dominated by sp2 fraction is shown in Fig. 1(G1). In this figure, sp2–C bonding intensity is prominent at a binding energy of 284.8 eV. Other bonding contributions such as C–O (286.3 eV) and HO–CQO (290. 1 eV) are ascribed to oxygenated carbon and carboxylic acid groups, respectively [38]. Structural and microstructural properties with surface roughness and coating thickness are presented in Table 1. These properties will be discussed in the following sections.
Fig. 2. Evolution of scratch induced acoustic emission and penetration depth under dry, linoleic and oleic acid conditions on D1 specimen.
3.2. Scratch resistance test High adhesive strength of coating is one of the basic requirements to realize several technological applications. Scratch test is a technique widely used to measure the adhesion of a coating with the substrate. If adherence is insufficient, premature failures can arise from the detachment of the coating by interfacial fractures [37]. Scratch test reveals that the adhesion of D1 is stronger than G1 specimen in both dry and lubricated conditions as shown in Figs. 2 and 3, respectively. In dry condition, however, there is an early onset of spallation event in D1 (Fig. 2). In G1 (Fig. 3) specimen, there is a complete adhesive failure (known as the critical failure) as indicated from the periodic high amplitude acoustic emission (AE) peaks. Unlike D1, G1 does not show any signs of partial adhesive failure. Only microscopic cracks are formed in the beginning of the scratch test. Under oleic acid lubricated conditions, we observe an increase in the adhesive strength for D1 along with a steep rise in the penetration depth. In contrast, G1 specimen shows a complete adhesive failure with increasing normal load. Spallation event set in early in case of lubrication with linoleic acid. The recorded penetration depth is always higher in G1 compared to D1. The results are presented in Table 2. In G1 specimen, the penetration depth rapidly approaches a maximum value at a scratch load of 4 N. However, this increase is rather slow at higher loads. Even during the initial scratching stages, the penetration depth is observed to be higher than the coating thickness (1.5 mm) as shown in Fig. 3. It clearly establishes that the coating is harder than the substrate which plastically deforms and bends into the substrate along the applied normal scratch load. In D1 specimen, the penetration depth gradually increases with the scratch load (Fig. 2). The trend followed by penetration depth and adhesive failure loads are different in D1 and G1 specimens as seen from the micrographs presented in Figs. 4 and 5, respectively. Failure mode of D1 is completely different as compared to G1 specimen. In D1, the surface undergoes continuous periodic spallation with ductile type adhesive failure. The amount of chromium in the DLC coating provides a relaxation of the residual stress by enhancing ductility to resist the crack induced surface and interfacial fracture. EDX is carried out to study the elemental concentration on the surface and deformed region of the D1 and G1 coatings. In this regard, Cr concentration is found to be high and low Fe on the D1 coating surface (Fig. 6(a)) compared to the deformed region (Fig. 6(b)) where the surface of the
Fig. 3. Evolution of scratch induced acoustic emission and penetration depth under dry, linoleic and oleic acid conditions on G1 specimen.
Table 2 Adhesive failure and penetration depth during scratch test. Case no.
Adhesive failure (N)
Critical failure (N)
Penetration depth (mm)
D1—Dry D1—linoleic D1—oleic G1—dry G1—linoleic G1—oleic
– – – 18 9 7
30 36 39 28 22 19
0.055 0.06 0.08 0.1 0.22 0.15
substrate is exposed due to spallation and delamination. Similarly, C reaches the surface with Ar, Cr and Fe related peaks are observed on the surface of G1 coating as shown in (Fig. 7(a)). However, elemental concentration of C is found to be low in the spalled and delaminated regions of the coating (Fig. 7(b)). This is an indication of sharp and brittle type adhesive failure. In this case, even debris of the scratch induced surface may not be generated due to the high stress accumulated in the interface which causes a sharp interfacial brittle failure. A strong Ar peak observed in G1 coating ingress into matrix during ArþH2 þCH4 plasma media exposure. Cohesive spallation along the scratch track with conformal type of buckling and brittle crack generation is observed in G1 specimen. Due to progressive stress, these buckling cracks eventually bring about large area
132
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 4. Optical micrograph of failure mode during scratch test of D1 specimen under (a) linoleic (b) oleic acid and (c) dry conditions.
Fig. 5. Optical micrograph of failure mode during scratch test of G1 specimen under (a) linoleic (b) oleic acid and (c) dry conditions.
interfacial spallation of the coatings. The cracking pattern is opposite to the scratch direction (SD) as indicated in Fig. 5. This indicates that cracks originate from the action of tensile stress. The rupture or the
flaking of coatings leads to discrete acoustic emission peaks with high amplitude. The process is dynamically followed on line where the onset of AE bursts can be reliably linked to failure. In the
N. Sharma et al. / Tribology International 56 (2012) 129–140
beginning of scratching, when the normal load reaches the critical load required for crack initiation, semicircular cracks are formed in the rear end of the contact zone located on the sliding axis. This crack propagates in a quasi-semicircular manner toward the back of the indenter, following the direction of the principal tensile stress in the surface. As the indenter continues to advance with progressive load, it inevitably crosses previously generated cracks, at which point further accumulation of stress occurs. When the indenter advances forward, it is possible to generate the nucleation of cracks and the induced spallation and flaking of the coating. Under all these three conditions, the modes of failure are similar. However, coating failed early under linoleic acid. In dry condition it failed at 28 N load with a larger area spallation. In case of fatty acids, the surface reactivity is enhanced due to the presence of free carbon atoms which interact among each other providing larger adhesion with diamond tipped scratch indenter.
Fig. 6. EDX analysis of D1 coatings (a) surface and (b) deformed region.
133
3.3. Tribological properties 3.3.1. Tribology of DLC in dry contact conditions Fig. 8 shows the variation of CoF with the sliding distance with respect to loading in D1, G1 and steel substrate tested under dry condition at 1, 4 and 8 N loads. Figs. 8(D1) and (G1) show that CoF decreases with sliding distance and then saturates at each loading. Initial high value of CoF is attributed to the roughness at the surface of the specimen. CoF of steel falls in the same region with different loading conditions, Fig. 8(steel). It is known that D1 coating comprises a metastable hydrogenated amorphous carbon (a-C:H) containing a mixture of sp2 and sp3 bonds [5]. In general, the reduction in the steady state value of CoF of a-C:H is attributed to the formation of a lubricious graphite-like or amorphized transfer layer formed during contact sliding at the interface of coating and the counterbody [40,41]. Generally, this tribolayer is generated when hydrogen is released from the coating surface during sliding as a consequence of friction induced localized heating and melting of contact asperities. This causes sp3 bonds to become unstable [41]. As the normal load increases, the contact temperature between the ball and the coating increases. Hence higher is the applied normal load, rapid is the formation of the graphite-like lubricious tribolayer. Typical Raman spectroscopic measurements were conducted on the coating surface and in different wear track locations. Such spectra obtained at different sliding distances traversed on D1 are shown in Fig. 9. It shows broad features of D band compared to G band. The G and D bands’ peak positions and the intensities are found to increase with sliding distance while the ratio ID/IG decreases. One possible reason for a slight decrease in the intensity ratio with sliding distance (for D1 in Fig. 9) could be as following. As the contact area increases with sliding distance, the operating contact pressure decreases (from d1 to d3). This results in greater graphitization/amorphization initially when contact pressure is high. Reduction in contact pressure obviously suppresses this step towards later stage. This conforms formation of more ordered and relatively defect free structure of sp2 bonded carbon network on the sliding surface. In these curves a broad feature at 1180 cm 1
Fig. 7. EDX analysis of D1 coatings (a) surface and (b) deformed region.
134
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 8. Coefficient of friction of D1, G1 and steel specimens at (a) 1 N (b) 4 N and (c) 8 N under ambient and unlubricated conditions.
Fig. 9. Raman spectroscopy of D1 specimen at (CS) virgin surface of coating before the test and at various sliding distances (d1) 10 m (d2) 58 m and (d3) 120 m.
Table 3 FWHM of D and G band positions with ID/IG ratio of D1 coating. No.
D band FWHM (cm 1)
G band FWHM (cm 1)
ID/IG
(CS) (d1) (d2) (d3)
324 327 332 338
123 124 127 126
0.89 0.97 0.96 0.94
CS represents the virgin surface of the coating before the test.
is designated as transpolyacetylene (TPA) peak [1]. The results are presented in Table 3. Broad peaks recorded at 866 cm 1 correspond to the formation of amorphous chromium carbide (CrC). This peak is found to slightly increase as the sliding distance increases. In case of G1 specimen (Fig. 8(G1)), reduction in CoF during the test is due to the formation of a lubricious transfer layer on the counter face. During the test, the re-crystallization process takes place at the top surface of the coating and allows
the formation of a lubricious graphite-like film between coating and the counterpart. The higher is the normal load, the greater is the recrystallization at the top surface of the coating and lower is the steady-state value of CoF [37,41,42]. As the ball on disk test is carried out as unidirectional rubbing test, the applied normal load also encourages the re-orientation of the re-crystallized top layers of the graphite coating so that the basal planes become parallel to the sample surface. However, higher normal loads result in further re-orientation and much lower CoF. To understand the structural changes occurring in the coatings, Raman spectroscopic measurements were performed at different locations of the wear track at different sliding distances. The spectra of G1 DLC coatings (Fig. 10) have G and D peaks. On the coating surface, the intensity of these bands (G and D) is less pronounced compared to the spectra obtained from the wear track at different sliding distances. With the increase in the sliding distance, the intensity of both G and D band increases. The D band position is shifted to higher wave number (1363 cm 1) in Fig. 10(g1). This band has a
N. Sharma et al. / Tribology International 56 (2012) 129–140
135
Fig. 10. Raman spectroscopy of G1 specimen at (CS) virgin surface of coating before the test and at various sliding distances (g1) 8.5 m (g2) 44 m and (g3) 120 m.
Table 4 FWHM of D and G band positions with ID/IG ratio of G1 coating. No.
D band FWHM (cm 1)
G band FWHM (cm 1)
ID/IG
(CS) (g1) (g2) (g3)
278 282 298 304
259 120 116 121
0.94 1.25 1.30 1.32
CS represents the virgin surface of the coating before the test.
lower wave number 1337 cm 1 as shown in Fig. 10(CS). (CS) represents the virgin surface of the coating before commencing the test. The position of D band frequency decreases to 1359 cm 1 and 1356 cm 1 with sliding distances as shown in Fig. 10(g2) and (g3). This is presented in Table 4. The G band position also decreases with the sliding distance for G1 coatings. The D band FWHM increases while G band FWHM decreases and the ID/IG ratio increases with the sliding distance of G1 coatings as shown in Fig. 10. However, these minor differences in ID/IG ratio can arise from a change in the chemical behavior of the sliding surfaces. The change in the peak position from low and high wave numbers occurs due to strain–stress state of sliding surface. Broad peaks at 397 cm 1 and 693 cm 1 (Fig. 10) correspond to the formation of amorphous a-Fe2O3. We have provided additional results of Raman spectra for curve (a) of D1 specimen and curve (b) of G1 specimen in Fig. 6 to prove that there is a correlation between chemical structure of coatings and friction coefficient. These spectra are obtained on the similar location marked by line in Fig. 6 and these locations are mentioned as d1, d2, d3 in Fig. 6(D1) and g1, g2, g3 in Fig. 6(G1). In both these cases, Id/Ig is shown to increase in D1 and G1 specimens with the sliding distance as shown in Figs. 11 and 12, respectively. These results match well with Figs. 9 and 10. TPA peak does not fit well in Fig. 11 due to large broadening. During the initial passes, low value of CoF is derived from reduced contact area between the interfaces of both D1 and G1 coatings as shown in Fig. 8. Coating surface roughness is normally high and due to sliding, the rough surface asperities smoothen which in turn increases the contact area that is directly linked to an increase in CoF (Fig. 8). This value is maximum, when the surface is highly smooth yielding large sliding contact area. After initiation of sliding, this value gradually decreases which may be linked to the formation of graphitized/amorphized tribolayer [10,12]. This is evident from the Raman spectra measured with the sliding distance which clearly shows an increase in the intensity of D and G bands and the corresponding Id/Ig ratio are shown in Figs. 7–11. The formation of tribolayer is easy if the surface asperities are highly developed which in turn melt during the sliding [26,38]. At a low load, surface asperities resist for
Fig. 11. Raman spectroscopy of curve (a) of D1 coating from Fig. 8 at three different locations of (d1) 10 m (d2) 58 m and (d3) 120 m.
Fig. 12. Raman spectroscopy of curve (b) of G1 coating from Fig. 8 at three different locations (g1) 10 m (g2) 58 m and (g3) 120 m.
longer sliding distance and the formation of tribolayer is not efficient. Similar phenomenon is observed in G1 sample (Fig. 8). But at low loads, CoF is lower as compared to that obtained at
136
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 13. Coefficient of friction at (A) 1 N (B) 4 N and (C) 8 N on the (a) steel (b) D1 and (c) G1 specimens under oleic acid.
higher loads. This is attributed to the fact that the contact area is low due to rough asperities present on the surface which does not smoothen out at low load. However, in case of steel, CoF is much higher than that of D1 and G1 with no significant effect of loading. This is shown in Fig. 8(steel).
3.3.2. Tribology of DLC in lubricated contact conditions Conventional lubrication mechanisms involving steel surfaces are based on physical and chemical adsorptions. The anti-friction additives and fatty acids having polar groups play a key role in interactions with the metal surfaces. The most efficient tribological performance is expected from vegetable sunflower base oil. This oil consists of a considerable amount of fatty acids with unsaturated bonds [28,32,42,43]. When the test is performed with oleic and linoleic acids as lubricants, CoF gets modified as shown in Figs. 13 and 14, respectively. Summary of coefficient of friction under different sliding conditions with load is presented in Fig. 15. Both the acids are used in cis-form. The oleic acid has a single double bond and linoleic acid has two double bonds. As we know, that the presence of double bond freezes the conformation of a molecule. The oleic acid has a kink and linoleic acid has a pronounced bend in its conformal arrangement. It is well known that more the number of double bonds that a chain has in cisform, less is the flexibility of the molecule, thus limiting the ability of fatty acids to be densely packed. As the load increases, the number of active sites, the rate of diffusion of the lubricant molecules and their interaction with the surface increases. The chemically active centers present in the lubricant interact with these sites. The chemical structure is the dominant factor that determines the magnitude of polar attraction of the molecules towards active sites present on the surface of interaction [29,33,44,45].
In our case, it can be easily seen that in both of these cases, CoF of D1 and G1 increases by increasing the load and this in turn increases in linoleic acid (Fig. 14) much more than oleic acid (Fig. 13). In case of oleic acid, it is found that oleic acid forms a loose-packed monolayer on the surface of DLC coating with low shear-strength. This loose-packed monolayer of oleic acid, oriented parallel to the surface, facilitates easy sliding between the coating surface and the counterbody. As the sliding distance during the test increases, this monolayer is removed and CoF approaches to a value of that of pure DLC coating. This is the reason why CoF at each particular load decreases with the increasing sliding distance. However, another investigation of the friction and wear pertaining to steel surfaces in the presence of oleic, linoleic, and linolenic acid in n-hexadecane showed that the performance of the three systems are in fact similar. This arises possibly due to the severity of test conditions that occur with high load and low sliding velocity [22]. In a different study, the CoF was found to decrease with increasing degree of unsaturation [34,46]. It is known that better oxidation stability provides poor tribological properties. Oxidation stability of an organic molecule depends on the number of reactive groups present and the extent of saturation assisted with fatty acids. More double (or triple) bonds suggest easier oxidation. On the other hand, oils with unsaturated molecules and polar groups (–COOH) have more sites for reactions and/or adsorption with the metal surfaces. This can lead to superior boundary lubrication effects [31,36,43]. When the surfaces are separated by a lubricant film, a low friction is achieved if the film is a solid, crystalline one but it shows incommensurate registry with the substrate surface. Such a situation appears in the so-called ‘‘hard’’ lubricant system, where the intermolecular interaction strength within a lubricant is stronger than the lubricant–substrate interaction. In that case, the shape of the lubricant molecules is not relevant because
N. Sharma et al. / Tribology International 56 (2012) 129–140
137
Fig. 14. Coefficient of friction at (A) 1 N (B) 4 N and (C) 8 N on the (a) steel (b) D1 and (c) G1 specimens under linoleic acid.
Fig. 15. Summary of coefficient of friction under different test conditions.
friction is determined uniquely by the substrate–lubricant interface architecture [30,47,48].
3.3.3. Wear rate of DLC in dry unlubricated and lubricated conditions Wear rate is found to be high on the steel surface in dry conditions. This value is significantly less under oleic and linoleic lubricated condition (Fig. 16). It is the lowest at 1 N load under oleic acid followed by low CoF. In dry condition, both D1 and G1 specimens show high wear rates. This value is less while using
linoleic acid on both the specimens D1 and G1. However, wear rate is low on G1 compared to D1 specimen. A minimum wear rate of 1.2 10 11 mm/Nm is measured at the maximum normal load of 8 N applied to G1 specimen in the presence of linoleic acid. D1 shows high wear rate under dry sliding conditions due to high adhesion which is caused by the capillary force induced adhesion. This kind of strong adhesion is possible when large amounts of hydrogen atoms and molecules are present in the coating matrix. The D1 coating contains internal hydrogen which interacts with external microdroplets of water present in humid tribo-atmosphere. This interaction is strong in nature and causes a high wear rate of hydrogenated D1 coatings compared to G1 (Fig. 16). However, in lubricated conditions, internal hydrogen has negligible effect on the surface due to the microscopic layer of fatty acids in the sliding contact conditions. This fatty acid is constituted by the long chained hydrocarbons. Lubricated layers are sufficient to passivate the covalent bonds and interact with internal hydrogen present in D1 coatings which does not react directly with the counter body. This does not cause large differences in wear rates of D1 and G1 coatings in the lubricated conditions.
3.3.4. Optical microscopy of wear track It is evident from optical microscopy that wear tracks inscribed in dry condition are significantly wider compared to those recorded in lubrication condition. This is true in cases of D1, G1 and steel specimens as shown in Fig. 17. In dry condition, track of steel specimen is severely deformed as shown in (c). In lubricated condition, wear track width as shown in images (d) and (e) is less compared to (a) and (b), respectively. Using oleic acid, the wear track width is less on D1 as compared to G1 specimen. It is well known that less wear depth correspond to
138
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 16. Wear rate of specimen vs applied normal load on (a) steel (b) D1 and (G1) under dry,oleic and linoleic conditions.
Fig. 17. Typical optical image of wear tracks formed at 4 N normal loads on (a) D1—dry (b) G1—dry (c) steel—dry (d) G1—oleic (e) D1—oleic (f) steel—oleic (g) G1—linoleic (h) D1—linoleic and (i) steel—linoleic conditions.
lower CoF. Similarly, wear width is shallow on D1 specimen in the presence of linoleic acid lubricant, which results is ultralow value of CoF. In lubricated conditions, residual surface roughness of the coating is visible in the wear tracks which clearly indicate the formation of protective lubricant layer on the contacting interface. In this condition, CoF mostly depends on the lubricant properties. In some cases, it is shown that CoF gradually decreases with the sliding distance. It may be related to the smoothness of the surface and the formation of crystalline lubricant tribolayer.
4. Scratch and friction model Scratch test reveals that the interfacial adhesion in D1 coating is stronger than G1 coating in both dry and lubricated conditions as shown in Figs. 2 and 3, respectively. It is indicated that diamond like sp3 fraction is higher in D1 which results in high hardness and elastic modulus compared to G1 coating. High fraction of sp3 bonding and Cr content causes ductile deformation to dominate the scratch test while brittle type fracture is observed in G1 coating due to the large amount of sp2 content.
N. Sharma et al. / Tribology International 56 (2012) 129–140
The sp2 fraction possesses shearable basal planes. These basal planes slide and exhibit brittle type fracture. Low friction coefficient of D1 coatings at higher loads of 4 and 8 N (Fig. 8(D1)) and after laps of longer sliding distance is dominated by two factors: (i) the formation of graphitized/amorphized tribolayer and (ii) the passivation of covalent bonds due to the chemisorption of hydrogen atoms/molecules [5,7]. At low load of 1 N, graphitization/amorphization and chemisorption of hydrogen atoms/molecules do not occur sufficiently. Similarly, G1 coating shows super low friction coefficient after longer sliding distance at higher load of 8 N (Fig. 8(G1)). At high loads, self-alignment of graphitized/ amorphized tribolayer and effective passivation of dangling bonds by hydrogen molecules present in ambient atmosphere are possible factors that reduce the friction coefficient to such an extent [5,38–40]. As the sliding distance progresses, the friction coefficient of D1 and G1 coatings is found to decrease. This is related with the formation of stabilized graphitized/amorphized tribolayers [38–40]. However, increase in friction coefficient of G1 and D1 coatings under lubricated condition using oleic and linoleic acids (Figs. 13 and 14 and summarized in Fig. 15) occurs with an increase in load. This remarkable behavior of friction is related to the formation of surface tribolayer. In lubricated conditions, the formation of graphitized/amorphized tribolayer does not act as a determining factor. Formation of these layers is unstable and restricted by lubricants present at the sliding interface while the friction mechanism is mainly driven by internal energy (viscosity) of the fluid related to the interfacial stress and lubricant confinement [25,26,30,32]. At lower load, the layer of the liquid is thicker compared to what occurs in higher load. Thickness dependent lubricant layer follows the hydrodynamic, elastohydrodynamic and boundary lubrication processes. At low load, in hydrodynamic and/or elastohydrodynamic conditions, the contact deformation is low which is enhanced at high loads when boundary lubrication dominates. In this case, friction model depends on the viscosity of the lubricant as well as on the underlying deformation mechanism. At high loads, the viscosity of the lubricant confined at interfaces is high and the deformation of the contacting surface is also severe. These are the dominating factors which contribute in the increase of the friction coefficient with the increase in load.
5. Conclusions Scratch resistance and tribological properties of hydrogenated and hydrogen free diamond like carbon coatings were investigated under dry unlubricated and lubricated conditions in ambient atmosphere using layers of oleic and linoleic acids. During scratch test, ductile type failure was observed in hydrogenated coatings while brittle failure of crack induced larger area spallation was observed in hydrogen free coatings. Scratch resistance was improved in hydrogenated coating using linoelic and oleic acids compared to unlubricated dry condition. Under similar conditions, scratch resistance property was found to be low in the hydrogen free coatings. Under ambient dry conditions, CoF was found to decrease, while under lubrication condition, it was found to increase with normal load for both hydrogenated and hydrogen free DLC coatings. The CoF was found to be low with oleic acid as compared to linoleic acid. At low load of 1 N, hydrogenated and hydrogen-free coatings show a super low CoF of 0.001 with oleic acid. Even, steel surface shows a very low CoF of 0.005 in such conditions. Optical micrograph shows wider wear track under dry conditions while this was shallow under linoelic acid compared to dry and lubricated oleic acid condition. Severe deformation of the wear track was observed on steel surface while sliding under ambient dry conditions.
139
Acknowledgments The authors would like to thank the reviewer for providing their useful comments, suggestions and guidelines during the course of revision to improve the technical quality of the present paper. The authors would also like to thank Teer Coatings Ltd., UK, to provide the specimens for tribological studies. The authors are grateful to Dr. C.S. Sundar, Director, Materials Science Group, and Shri C.S. Chetal, Director, Indira Gandhi Centre for Atomic Research, for encouragement and support.
References [1] Robertson J. Deposition and properties of diamond like carbon. Materials Research Society Symposium Proceedings 1999;555:12. [2] Braun OM, Manini N, Tosatti E. Role of lubricant molecular shape in microscopic friction. Physical Review B 2008;78:195402–12. [3] Etsion I, Halperin G, Becker E. The effect of various surface treatments on piston pin scuffing resistance. Wear 2006;261:785–91. [4] Castle RC, Bovington CH. The behaviour of friction modifiers under boundary and mixed EHD conditions. Lubrication Science 2003;15:253–63. [5] Erdemir A. The role of hydrogen in tribological properties of diamond-like carbon films. Surface and Coatings Technology 2001;146–147:292–7. [6] Neuville S, Matthews A. A perspective on the optimization of hard carbon and related coatings for engineering applications. Thin Solid Films 2007;515:6619–53. [7] Sanchez-Lopez JC, Erdemir A, Donnet C, Rojas TC. Friction-induced structural transformations of diamondlike carbon coatings under various atmospheres. Surface and Coatings Technology 2003;163–164:444–50. [8] Grischke M, Hieke A, Morgenweck F, Dimigen H. Variation of the wettability of DLC-coatings by network modification using silicon and oxygen. Diamond and Related Materials 1998;7:454–8. [9] Mortier RM, Orszulik ST. Chemistry and technology of lubricants. 2nd ed. Glasgow: Blackie Academic Professional; 1993. [10] Dorinson A, Ludema KC. Mechanics and chemistry in lubrication. Tribology series. Amsterdam: Elsevier; 1985 [p. 9]. [11] Pawlak Z. Tribochemistry of lubricating oils. Tribology and interface engineering series. Amsterdam: Elsevier; 2003 [p. 45]. [12] Franklin SE, Baranowska J. Conditions affecting the sliding tribological performance of selected coatings for high vacuum bearing applications. Wear 2007;263:1300–5. [13] Jiang JC, Meng WJ, Evans AG, Cooper CV. Structure and mechanics of W-DLC coated spur gears. Surface and Coatings Technology 2003;176:50–6. [14] Tung SC, Gao H. Tribological characteristics and surface interaction between piston ring coatings and a blend of energy-conserving oils and ethanol fuels. Wear 2003;255:1276–85. [15] Etsion I, Halperin G, Becker E. The effect of various surface treatments on piston pin scuffing resistance. Wear 2006;261:785–91. [16] Hershberger J, Ozturk O, Ajayi JB, Woodford JB, Erdemir A, Erck RA, Fenske GR. Evaluation of DLC coatings for spark-ignited direct-injected fuel systems. Surface and Coatings Technology 2004;179:237–44. [17] Dai M, Zhou K, Yuan Z, Ding Q, Fu Z. The cutting performance of diamond and DLC-coated cutting tools. Diamond and Related Materials 2000;9:1753–7. [18] Sato T, Besshi T, Tsutsui I, Morimoto T. Anti-galling property of a diamondlike carbon coated tool in aluminium sheet forming. Journal of Materials Processing Technology 2000;104:21–4. [19] Tan AH. Corrosion and tribological properties of ultra-thin DLC films with different nitrogen contents in magnetic recording media. Diamond and Related Materials 2007;16 [467–47]. [20] Zhao X, Bhushan B. Comparison studies on degradation mechanisms of perfluoropolyether lubricants and model lubricants. Tribology International 2000;9:187–97. [21] Gatzen HH, Beck M. Tribological investigations on micromachined silicon sliders. Tribology International 2003;36:279–83. [22] Numata T, Nanao H, Mori S, Miyake S. Chemical analysis of wear tracks on magnetic disks by TOF-SIMS. Tribology International 2003;36:305–9. [23] Sheeja D, Tay BK, Nung LN. Feasibility of diamondlike carbon coatings for orthopaedic applications. Diamond and Related Materials 2004;13:184–90. [24] Shi B, Ajayi OO, Fenske G, Erdemir A, Liang H. Tribological performance of some alternative bearing materials for artificial joints. Wear 2003;255: 1015–21. [25] Erdemir A. Review of engineered tribological interfaces for improved boundary lubrication. Tribology International 2005;38:249–56. [26] Hsu SM, Gates RS. Boundary lubricating films: formation and lubrication mechanism. Tribology International 2005;38:305–12. [27] Karin P, Rickard G. Tribological performance of a DLC coating in combination with water-based lubricants. Tribology International 2003;36:851–5. [28] Takeshi Y, Junji A, Takuya T, Naoko T, Mamoru T, Atsushi M, Toshihide O, Kazuo H. Sliding velocity dependency of the friction coefficient of Si-containing diamond-like carbon film under oil lubricated condition. Tribology International 2011;44:1296–303.
140
N. Sharma et al. / Tribology International 56 (2012) 129–140
[29] Vercammen K, Van Acker K, Vanhulsel A, Barriga J, Arnsek A, Kalin M, Meneve J. Tribological behaviour of DLC coatings in combination with biodegradable lubricants. Tribology International 2004;37:983–9. [30] Haque T, Morina A, Neville A, Kapadia R, Arrowsmith S. Study of the ZDDP antiwear tribofilm formed on the DLC coating using AFM and XPS techniques. Journal of ASTM International 2007;4:1–7. [31] Kalin M, Roman E, Vizintin J. The effect of temperature on the tribological mechanisms and reactivity of hydrogenated, amorphous diamond-like carbon coatings under oil-lubricated conditions. Thin Solid Films 2007;515:3644–52. [32] Kalin M, Vizintin J, Barriga J, Vercammen K, van Acker K, Arnsek A. The effect of doping elements and oil additives on the tribological performance of boundary-lubricated DLC/DLC contacts. Tribology Letters 2004;17:679–88. [33] Vengudusamy B, Green JH, Lamb GD, Spikes HA. Tribological properties of tribofilms formed from ZDDP in DLC/DLC and DLC/steel contacts. Tribology International 2011;44:165–74. [34] Equey S, Roos S, Mueller U, Hauert R, Spencer ND, Crockett R. Reactions of zinc-free anti-wear additives in DLC/DLC and steel/steel contacts. Tribology International 2008;41:1090–6. [35] Podgornik B, Vizintin J. Tribological reactions between oil additives and DLC coatings for automotive applications. Surface and Coatings Technology 2005;200:1982–9. [36] Podgornik B, Jacobson S, Hogmark S. Influence of EP additive concentration on the tribological behaviour of DLC-coated steel surfaces. Surface and Coatings Technology 2005;191:357–66. [37] Field SK, Jarratt M, Teer DG. Tribological properties of graphite-like and diamond-like carbon coatings. Tribology International 2004;37:949–56.
[38] Liu Y, Erdemir A, Meletis EI. A study of the wear mechanism of diamond-like carbon films. Surface and Coatings Technology 1996;82:48–56. [39] Hongxuan L, Tao X, Chengbing W, Jianmin C, Huidi Z, Huiwen L. Tribochemical effects on the friction and wear behaviors of a-C:H and a-C films in different environment. Tribology International 2007;40:132–8. [40] Liu Y, Erdemir A, Meletis EI. An investigation of the relationship between graphitization and frictional behavior of DLC coatings. Surface and Coatings Technology 1996;86–87:64. [41] Erdemir A, Bindal C, Pagan J, Wilbur P. Characterization of transfer layers on steel surfaces sliding against diamond-like hydrocarbon films in dry nitrogen. Surface and Coatings Technology 1995;76–77:59. [42] Zaidi H, Djamai A, Chin KJ, Mathia T. Characterization of DLC coating adherence by scratch testing. Tribology International 2006;39:124–8. [43] Stachowiak GW, Batchelor AW. Engineering tribology. Tribology series 24. Amsterdam: Elsevier; 1993. [44] Weller DE, Perez JM. A study of the effect of chemical structure on friction and wear. Part 1. Synthetic ester base fluids. Lubrication Engineering 2000;56:39–44. [45] Waleska C, Weller DE, Kraipat C, Perez Joseph M. The effect of chemical structure of basefluids on antiwear effectiveness of additives. Tribology International 2005;38:321–6. [46] Castle RC, Bovington CH. The behaviour of friction modifiers under boundary and mixed EHD conditions. Lubrication Science 2003;15:253–63. [47] Braun OM, Naumovets AG. Nanotribology: microscopic mechanisms of friction. Surface Science Reports 2006;79 [79–58]. [48] Braun OM, Manini N, Tosatti E. Role of lubricant molecular shape in microscopic friction. Physical Review B 2008;78:195402–12.
Contents lists available at SciVerse ScienceDirect
Tribology International journal homepage: www.elsevier.com/locate/triboint
Scratch resistance and tribological properties of DLC coatings under dry and lubrication conditions Neha Sharma, N. Kumar n, S. Dash, C.R. Das, R.V. Subba Rao, A.K. Tyagi, Baldev Raj Materials Science Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603102, TN, India
a r t i c l e i n f o
abstract
Article history: Received 18 July 2011 Received in revised form 23 April 2012 Accepted 27 June 2012 Available online 4 July 2012
Scratch resistance and tribological properties of hydrogenated and hydrogen free diamond like carbon (DLC) coatings were investigated in ambient atmosphere under unlubricated and lubricated conditions using layers of oleic (C18H34O2) and linoleic acid (C18H32O2). Scratch resistance property improved in hydrogenated DLC while using linoelic and oleic acids compared to unlubricated condition. Coefficient of friction (CoF) was found to decrease under ambient dry conditions, while under lubrication condition, it was found to increase with normal load. At low load of 1 N, hydrogenated and hydrogen-free coatings show a super low CoF of 0.001 and high wear resistance with oleic acid. & 2012 Elsevier Ltd. All rights reserved.
Keywords: DLC coatings Fatty acid Scratch resistance Friction coefficient
1. Introduction The diamond like carbon coatings have been explored extensively in recent years due to their low friction, high wear resistance properties and adhesive protection [1]. However, their low surface energy limits them to have any physical and chemical interaction with conventional lubricants and additives. Thus the influence of lubricants, lubricant additives and lubrication process on the friction and wear properties of these coatings needs to be investigated. When sliding occurs under unlubricated atmospheric conditions, two solid surfaces in contact produce surface layer of chemisorbed or physisorbed molecules, or a capillary condensed liquid bridge, between them. Each of these effects can drastically modify their response to adhesion and friction. Adhesion usually decreases, but in the case of capillary condensation, the additional Laplace pressure, or attractive capillary force between the surfaces, may further enhance the adhesion [2]. However, in lubricated sliding contact conditions, the molecules at the interface relax and/or rearrange to form a new equilibrium configuration that is distinctly different from the ones that prevail when the surface is kept in isolation. These rearrangements may involve simple positional and orientational changes of the surface molecules. In a complex condition, new molecular groups that are previously buried below the surfaces can re-appear and intermix with the interface. This behavior commonly occurs with surfaces whose molecules have both polar and nonpolar groups [3]. All these
n
Corresponding author. Tel./fax: þ 914427480081. E-mail address: [email protected] (N. Kumar).
0301-679X/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.triboint.2012.06.020
effects act to enhance the adhesion between the contacting surfaces. Shear stress is also an important factor which can convert the amorphous structure of liquids into crystalline ones when the liquid is confined at high pressure. Applying the critical stress liquid can lead to the formation of a monolayer (10 atomic layers). In this case, the viscosity of the liquid is very high which can cause to freeze the sliding hence strong adhesion acts [2,4]. Crystalline structure of a liquid can also have freezing behavior due to rapid rise in viscosity. Adhesion and friction become low if the viscosity is normal and the behavior of liquid emulates that of a crystalline medium. It can easily shear and provide slippage induced movement. In amorphous state, when the stress is below a critical value needed for transition, the liquid molecule sticks and hinders shearing which raises adhesion and friction. Adhesion and friction are related to slip induced movement and viscosity enhancement. These parameters depend on the structure and bonding fraction of materials participating in a sliding contact. DLC coatings consist of a fractional combination of sp2 and sp3 bonded carbon atoms with a considerable amount of hydrogen. If hydrogen content is about 40 at%, it is called as a hydrogenated DLC and if it is less than 1 at%, it is known as a hydrogen free DLC [5]. Depending upon sp2/sp3 ratio and content of hydrogen, different kinds of DLCs with different properties can be synthesized [6,7]. DLC coatings are called inert coatings because of their low surface energy [8,9]. Therefore, they do not react with various lubricants, oil additives as these coatings do not attract the polar groups from these oils and additives. There are conventional lubrication mechanisms for steels and other metals [10–13]. Most of the mechanical systems work under lubricated conditions. As DLC coatings exist in a wide variety possessing different properties, there
130
N. Sharma et al. / Tribology International 56 (2012) 129–140
is a huge demand for these coatings in various mechanical systems like, bearings [14], gears [15], piston rings and pins [16,17], directinjection fuel systems [18], cutting and forming tools [19,20]. In computer industry, these coatings have application as head–disk interface [21–24]. These coatings also find applications in orthopedic applications in medical industry [25,26]. To reduce the severity of the contact conditions, some of the above performance driven systems are to be operated under lubricated conditions [27,28]. Thus it is necessary to investigate mechanism of lubrication of these coatings. In this regard, some of the important studies have been carried out to address physical and chemical interaction of lubricants, oils and additives with DLC coatings in sliding contact conditions [29–36]. In the present study, the lubricating mechanisms and their influence on the scratch resistance and tribological properties of DLC coatings are investigated. Two types of diamond like carbon coatings, a hydrogen free amorphous carbon (a-C) and a hydrogenated amorphous carbon (a-C:H), were investigated under different loading conditions in the presence of unlubricated and lubricating layer of fatty acids.
2. Experimental In the present study, we designate commercial name of Graphit (hydrogen free) and Dymon (hydrogenated) coatings as G1 and D1, respectively. The G1 coating was deposited by closed field unbalanced magnetron sputter ion plating from the carbon target in a closed field arrangement [37]. Coating D1 was deposited by plasma enhanced chemical vapor deposition using a hydrocarbon gas precursor. Hydrogen content entrained in solid matrix during the deposition of D1 coating in CH4/Ar plasma amounts to 2%. The bias voltage applied to the substrate was pulsed direct current (PDC) and the plasma was further enhanced by the use of additional radio frequency (RF) electrodes [38,39]. Both these coatings are single layered deposited on a steel substrate. Adhesive strength of the coating was tested by using a commercial scratch tester (CSM Instruments, Switzerland) fitted with a Rockwell spherical diamond indenter (tip radius of 200 mm). Scratch-tests were performed using progressive loads from 1 to 40 N for a transverse scratch length of 3 mm in dry unlubricated and lubricated (linoleic and oleic acids) conditions. The scratch tester is equipped with an acoustic emission monitoring sensor which detects acoustic energy in the range of 20–80 kHz. To detect the failure of the coating, the scratches were examined using Scanning Electron Microscope (SEM) and correlated with the acoustic emission peaks. Tribological tests were performed using a ball on disk microtribometer (CSM Instruments, Switzerland) in linear reciprocating mode. A spherical steel ball (100Cr6 SS) of diameter 6 mm with a surface roughness 0.06 mm was used as a sliding body to measure the CoF. Tests were carried out at 1, 4 and 8 N loads with constant sliding speed of 2 cm/s and a stroke length of 3 mm in ambient dry and lubricated conditions with a relative humidity of 58%. For this purpose oleic and linoleic acids were used as lubricating media for the contacting surfaces. Wear dimension was measured using a Dektak 6 M stylus profiler (Veeco, USA). Friction coefficient was found to be reproducible in couple of tests carried out under similar conditions. Calculated wear rate was also found to be more or less similar. A nano-hardness tester (CSM, Switzerland) was used for extracting surface hardness of D1 and G1 samples at a constant load of 2 mN with a Berkovich diamond indenter. Load–displacement curves were generated and resulting curves were analyzed using the Oliver and Pharr method to obtain nano-hardness of the samples. Five indentations were performed on each sample. Raman measurements, using a 5 mW laser power and 514 nm wavelength, were performed to record the local chemical structure present at the coating surface as well as in the wear tracks. Raman measurements in the wear tracks were performed while
stopping the test after each sliding distance. These locations are indicated as d1, d2 and d3. X-ray photoelectron spectroscopy (XPS) was used to study the chemical state of constituents comprising this coating. Prior to acquisition of XPS spectra the specimen was sputtered to remove surface contaminants. An Ar þ ion beam sputtering of specimen surface was carried out for a duration of 3 min. Energy dispersive X-ray (EDX) was carried out to obtain the elemental analysis of the coating surface and the scratched region.
3. Results and discussion 3.1. XPS of coatings surface The sp3 and sp2 fractions of DLC films are deduced from XPS fitting for C (1s) core level peak. These peaks comprise of contribution from C–C (sp3) at a binding energy of 285.3 eV, graphite C–C (sp2) at 284.8 eV and C–O at 286.3 eV. Broad FWHM of C (1s) peak points to presence of carbon with several distinct local bonding environments. The C (1s) XPS in Fig. 1(D1) depicts these contributions. Since the area under each peak is directly related to the concentration of the corresponding phase, the sp3 content is estimated by taking the ratio of diamond peak area
Fig. 1. XPS of coating surface (D1) hydrogenated and (G1) hydrogen free.
N. Sharma et al. / Tribology International 56 (2012) 129–140
131
Table 1 Structural and mechanical properties of DLC coatings. No.
(ID/IG)
(sp3/sp2)
Dt (lm)
H (GPa)
E (GPa)
Ra (nm)
D1 G1
0.89 0.94
82/18 72/28
1.5 1.5
38 34
535 478
64 68
(Dt—coating thickness, H—hardness, E—elastic modulus, Ra—surface roughness of coatings).
over the sum of diamond and graphite peak areas. Each spectral component is a convolution of a Gaussian and a Lorentzian. The contribution of the background is approximated by the Shirley method. Coating composition dominated by sp2 fraction is shown in Fig. 1(G1). In this figure, sp2–C bonding intensity is prominent at a binding energy of 284.8 eV. Other bonding contributions such as C–O (286.3 eV) and HO–CQO (290. 1 eV) are ascribed to oxygenated carbon and carboxylic acid groups, respectively [38]. Structural and microstructural properties with surface roughness and coating thickness are presented in Table 1. These properties will be discussed in the following sections.
Fig. 2. Evolution of scratch induced acoustic emission and penetration depth under dry, linoleic and oleic acid conditions on D1 specimen.
3.2. Scratch resistance test High adhesive strength of coating is one of the basic requirements to realize several technological applications. Scratch test is a technique widely used to measure the adhesion of a coating with the substrate. If adherence is insufficient, premature failures can arise from the detachment of the coating by interfacial fractures [37]. Scratch test reveals that the adhesion of D1 is stronger than G1 specimen in both dry and lubricated conditions as shown in Figs. 2 and 3, respectively. In dry condition, however, there is an early onset of spallation event in D1 (Fig. 2). In G1 (Fig. 3) specimen, there is a complete adhesive failure (known as the critical failure) as indicated from the periodic high amplitude acoustic emission (AE) peaks. Unlike D1, G1 does not show any signs of partial adhesive failure. Only microscopic cracks are formed in the beginning of the scratch test. Under oleic acid lubricated conditions, we observe an increase in the adhesive strength for D1 along with a steep rise in the penetration depth. In contrast, G1 specimen shows a complete adhesive failure with increasing normal load. Spallation event set in early in case of lubrication with linoleic acid. The recorded penetration depth is always higher in G1 compared to D1. The results are presented in Table 2. In G1 specimen, the penetration depth rapidly approaches a maximum value at a scratch load of 4 N. However, this increase is rather slow at higher loads. Even during the initial scratching stages, the penetration depth is observed to be higher than the coating thickness (1.5 mm) as shown in Fig. 3. It clearly establishes that the coating is harder than the substrate which plastically deforms and bends into the substrate along the applied normal scratch load. In D1 specimen, the penetration depth gradually increases with the scratch load (Fig. 2). The trend followed by penetration depth and adhesive failure loads are different in D1 and G1 specimens as seen from the micrographs presented in Figs. 4 and 5, respectively. Failure mode of D1 is completely different as compared to G1 specimen. In D1, the surface undergoes continuous periodic spallation with ductile type adhesive failure. The amount of chromium in the DLC coating provides a relaxation of the residual stress by enhancing ductility to resist the crack induced surface and interfacial fracture. EDX is carried out to study the elemental concentration on the surface and deformed region of the D1 and G1 coatings. In this regard, Cr concentration is found to be high and low Fe on the D1 coating surface (Fig. 6(a)) compared to the deformed region (Fig. 6(b)) where the surface of the
Fig. 3. Evolution of scratch induced acoustic emission and penetration depth under dry, linoleic and oleic acid conditions on G1 specimen.
Table 2 Adhesive failure and penetration depth during scratch test. Case no.
Adhesive failure (N)
Critical failure (N)
Penetration depth (mm)
D1—Dry D1—linoleic D1—oleic G1—dry G1—linoleic G1—oleic
– – – 18 9 7
30 36 39 28 22 19
0.055 0.06 0.08 0.1 0.22 0.15
substrate is exposed due to spallation and delamination. Similarly, C reaches the surface with Ar, Cr and Fe related peaks are observed on the surface of G1 coating as shown in (Fig. 7(a)). However, elemental concentration of C is found to be low in the spalled and delaminated regions of the coating (Fig. 7(b)). This is an indication of sharp and brittle type adhesive failure. In this case, even debris of the scratch induced surface may not be generated due to the high stress accumulated in the interface which causes a sharp interfacial brittle failure. A strong Ar peak observed in G1 coating ingress into matrix during ArþH2 þCH4 plasma media exposure. Cohesive spallation along the scratch track with conformal type of buckling and brittle crack generation is observed in G1 specimen. Due to progressive stress, these buckling cracks eventually bring about large area
132
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 4. Optical micrograph of failure mode during scratch test of D1 specimen under (a) linoleic (b) oleic acid and (c) dry conditions.
Fig. 5. Optical micrograph of failure mode during scratch test of G1 specimen under (a) linoleic (b) oleic acid and (c) dry conditions.
interfacial spallation of the coatings. The cracking pattern is opposite to the scratch direction (SD) as indicated in Fig. 5. This indicates that cracks originate from the action of tensile stress. The rupture or the
flaking of coatings leads to discrete acoustic emission peaks with high amplitude. The process is dynamically followed on line where the onset of AE bursts can be reliably linked to failure. In the
N. Sharma et al. / Tribology International 56 (2012) 129–140
beginning of scratching, when the normal load reaches the critical load required for crack initiation, semicircular cracks are formed in the rear end of the contact zone located on the sliding axis. This crack propagates in a quasi-semicircular manner toward the back of the indenter, following the direction of the principal tensile stress in the surface. As the indenter continues to advance with progressive load, it inevitably crosses previously generated cracks, at which point further accumulation of stress occurs. When the indenter advances forward, it is possible to generate the nucleation of cracks and the induced spallation and flaking of the coating. Under all these three conditions, the modes of failure are similar. However, coating failed early under linoleic acid. In dry condition it failed at 28 N load with a larger area spallation. In case of fatty acids, the surface reactivity is enhanced due to the presence of free carbon atoms which interact among each other providing larger adhesion with diamond tipped scratch indenter.
Fig. 6. EDX analysis of D1 coatings (a) surface and (b) deformed region.
133
3.3. Tribological properties 3.3.1. Tribology of DLC in dry contact conditions Fig. 8 shows the variation of CoF with the sliding distance with respect to loading in D1, G1 and steel substrate tested under dry condition at 1, 4 and 8 N loads. Figs. 8(D1) and (G1) show that CoF decreases with sliding distance and then saturates at each loading. Initial high value of CoF is attributed to the roughness at the surface of the specimen. CoF of steel falls in the same region with different loading conditions, Fig. 8(steel). It is known that D1 coating comprises a metastable hydrogenated amorphous carbon (a-C:H) containing a mixture of sp2 and sp3 bonds [5]. In general, the reduction in the steady state value of CoF of a-C:H is attributed to the formation of a lubricious graphite-like or amorphized transfer layer formed during contact sliding at the interface of coating and the counterbody [40,41]. Generally, this tribolayer is generated when hydrogen is released from the coating surface during sliding as a consequence of friction induced localized heating and melting of contact asperities. This causes sp3 bonds to become unstable [41]. As the normal load increases, the contact temperature between the ball and the coating increases. Hence higher is the applied normal load, rapid is the formation of the graphite-like lubricious tribolayer. Typical Raman spectroscopic measurements were conducted on the coating surface and in different wear track locations. Such spectra obtained at different sliding distances traversed on D1 are shown in Fig. 9. It shows broad features of D band compared to G band. The G and D bands’ peak positions and the intensities are found to increase with sliding distance while the ratio ID/IG decreases. One possible reason for a slight decrease in the intensity ratio with sliding distance (for D1 in Fig. 9) could be as following. As the contact area increases with sliding distance, the operating contact pressure decreases (from d1 to d3). This results in greater graphitization/amorphization initially when contact pressure is high. Reduction in contact pressure obviously suppresses this step towards later stage. This conforms formation of more ordered and relatively defect free structure of sp2 bonded carbon network on the sliding surface. In these curves a broad feature at 1180 cm 1
Fig. 7. EDX analysis of D1 coatings (a) surface and (b) deformed region.
134
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 8. Coefficient of friction of D1, G1 and steel specimens at (a) 1 N (b) 4 N and (c) 8 N under ambient and unlubricated conditions.
Fig. 9. Raman spectroscopy of D1 specimen at (CS) virgin surface of coating before the test and at various sliding distances (d1) 10 m (d2) 58 m and (d3) 120 m.
Table 3 FWHM of D and G band positions with ID/IG ratio of D1 coating. No.
D band FWHM (cm 1)
G band FWHM (cm 1)
ID/IG
(CS) (d1) (d2) (d3)
324 327 332 338
123 124 127 126
0.89 0.97 0.96 0.94
CS represents the virgin surface of the coating before the test.
is designated as transpolyacetylene (TPA) peak [1]. The results are presented in Table 3. Broad peaks recorded at 866 cm 1 correspond to the formation of amorphous chromium carbide (CrC). This peak is found to slightly increase as the sliding distance increases. In case of G1 specimen (Fig. 8(G1)), reduction in CoF during the test is due to the formation of a lubricious transfer layer on the counter face. During the test, the re-crystallization process takes place at the top surface of the coating and allows
the formation of a lubricious graphite-like film between coating and the counterpart. The higher is the normal load, the greater is the recrystallization at the top surface of the coating and lower is the steady-state value of CoF [37,41,42]. As the ball on disk test is carried out as unidirectional rubbing test, the applied normal load also encourages the re-orientation of the re-crystallized top layers of the graphite coating so that the basal planes become parallel to the sample surface. However, higher normal loads result in further re-orientation and much lower CoF. To understand the structural changes occurring in the coatings, Raman spectroscopic measurements were performed at different locations of the wear track at different sliding distances. The spectra of G1 DLC coatings (Fig. 10) have G and D peaks. On the coating surface, the intensity of these bands (G and D) is less pronounced compared to the spectra obtained from the wear track at different sliding distances. With the increase in the sliding distance, the intensity of both G and D band increases. The D band position is shifted to higher wave number (1363 cm 1) in Fig. 10(g1). This band has a
N. Sharma et al. / Tribology International 56 (2012) 129–140
135
Fig. 10. Raman spectroscopy of G1 specimen at (CS) virgin surface of coating before the test and at various sliding distances (g1) 8.5 m (g2) 44 m and (g3) 120 m.
Table 4 FWHM of D and G band positions with ID/IG ratio of G1 coating. No.
D band FWHM (cm 1)
G band FWHM (cm 1)
ID/IG
(CS) (g1) (g2) (g3)
278 282 298 304
259 120 116 121
0.94 1.25 1.30 1.32
CS represents the virgin surface of the coating before the test.
lower wave number 1337 cm 1 as shown in Fig. 10(CS). (CS) represents the virgin surface of the coating before commencing the test. The position of D band frequency decreases to 1359 cm 1 and 1356 cm 1 with sliding distances as shown in Fig. 10(g2) and (g3). This is presented in Table 4. The G band position also decreases with the sliding distance for G1 coatings. The D band FWHM increases while G band FWHM decreases and the ID/IG ratio increases with the sliding distance of G1 coatings as shown in Fig. 10. However, these minor differences in ID/IG ratio can arise from a change in the chemical behavior of the sliding surfaces. The change in the peak position from low and high wave numbers occurs due to strain–stress state of sliding surface. Broad peaks at 397 cm 1 and 693 cm 1 (Fig. 10) correspond to the formation of amorphous a-Fe2O3. We have provided additional results of Raman spectra for curve (a) of D1 specimen and curve (b) of G1 specimen in Fig. 6 to prove that there is a correlation between chemical structure of coatings and friction coefficient. These spectra are obtained on the similar location marked by line in Fig. 6 and these locations are mentioned as d1, d2, d3 in Fig. 6(D1) and g1, g2, g3 in Fig. 6(G1). In both these cases, Id/Ig is shown to increase in D1 and G1 specimens with the sliding distance as shown in Figs. 11 and 12, respectively. These results match well with Figs. 9 and 10. TPA peak does not fit well in Fig. 11 due to large broadening. During the initial passes, low value of CoF is derived from reduced contact area between the interfaces of both D1 and G1 coatings as shown in Fig. 8. Coating surface roughness is normally high and due to sliding, the rough surface asperities smoothen which in turn increases the contact area that is directly linked to an increase in CoF (Fig. 8). This value is maximum, when the surface is highly smooth yielding large sliding contact area. After initiation of sliding, this value gradually decreases which may be linked to the formation of graphitized/amorphized tribolayer [10,12]. This is evident from the Raman spectra measured with the sliding distance which clearly shows an increase in the intensity of D and G bands and the corresponding Id/Ig ratio are shown in Figs. 7–11. The formation of tribolayer is easy if the surface asperities are highly developed which in turn melt during the sliding [26,38]. At a low load, surface asperities resist for
Fig. 11. Raman spectroscopy of curve (a) of D1 coating from Fig. 8 at three different locations of (d1) 10 m (d2) 58 m and (d3) 120 m.
Fig. 12. Raman spectroscopy of curve (b) of G1 coating from Fig. 8 at three different locations (g1) 10 m (g2) 58 m and (g3) 120 m.
longer sliding distance and the formation of tribolayer is not efficient. Similar phenomenon is observed in G1 sample (Fig. 8). But at low loads, CoF is lower as compared to that obtained at
136
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 13. Coefficient of friction at (A) 1 N (B) 4 N and (C) 8 N on the (a) steel (b) D1 and (c) G1 specimens under oleic acid.
higher loads. This is attributed to the fact that the contact area is low due to rough asperities present on the surface which does not smoothen out at low load. However, in case of steel, CoF is much higher than that of D1 and G1 with no significant effect of loading. This is shown in Fig. 8(steel).
3.3.2. Tribology of DLC in lubricated contact conditions Conventional lubrication mechanisms involving steel surfaces are based on physical and chemical adsorptions. The anti-friction additives and fatty acids having polar groups play a key role in interactions with the metal surfaces. The most efficient tribological performance is expected from vegetable sunflower base oil. This oil consists of a considerable amount of fatty acids with unsaturated bonds [28,32,42,43]. When the test is performed with oleic and linoleic acids as lubricants, CoF gets modified as shown in Figs. 13 and 14, respectively. Summary of coefficient of friction under different sliding conditions with load is presented in Fig. 15. Both the acids are used in cis-form. The oleic acid has a single double bond and linoleic acid has two double bonds. As we know, that the presence of double bond freezes the conformation of a molecule. The oleic acid has a kink and linoleic acid has a pronounced bend in its conformal arrangement. It is well known that more the number of double bonds that a chain has in cisform, less is the flexibility of the molecule, thus limiting the ability of fatty acids to be densely packed. As the load increases, the number of active sites, the rate of diffusion of the lubricant molecules and their interaction with the surface increases. The chemically active centers present in the lubricant interact with these sites. The chemical structure is the dominant factor that determines the magnitude of polar attraction of the molecules towards active sites present on the surface of interaction [29,33,44,45].
In our case, it can be easily seen that in both of these cases, CoF of D1 and G1 increases by increasing the load and this in turn increases in linoleic acid (Fig. 14) much more than oleic acid (Fig. 13). In case of oleic acid, it is found that oleic acid forms a loose-packed monolayer on the surface of DLC coating with low shear-strength. This loose-packed monolayer of oleic acid, oriented parallel to the surface, facilitates easy sliding between the coating surface and the counterbody. As the sliding distance during the test increases, this monolayer is removed and CoF approaches to a value of that of pure DLC coating. This is the reason why CoF at each particular load decreases with the increasing sliding distance. However, another investigation of the friction and wear pertaining to steel surfaces in the presence of oleic, linoleic, and linolenic acid in n-hexadecane showed that the performance of the three systems are in fact similar. This arises possibly due to the severity of test conditions that occur with high load and low sliding velocity [22]. In a different study, the CoF was found to decrease with increasing degree of unsaturation [34,46]. It is known that better oxidation stability provides poor tribological properties. Oxidation stability of an organic molecule depends on the number of reactive groups present and the extent of saturation assisted with fatty acids. More double (or triple) bonds suggest easier oxidation. On the other hand, oils with unsaturated molecules and polar groups (–COOH) have more sites for reactions and/or adsorption with the metal surfaces. This can lead to superior boundary lubrication effects [31,36,43]. When the surfaces are separated by a lubricant film, a low friction is achieved if the film is a solid, crystalline one but it shows incommensurate registry with the substrate surface. Such a situation appears in the so-called ‘‘hard’’ lubricant system, where the intermolecular interaction strength within a lubricant is stronger than the lubricant–substrate interaction. In that case, the shape of the lubricant molecules is not relevant because
N. Sharma et al. / Tribology International 56 (2012) 129–140
137
Fig. 14. Coefficient of friction at (A) 1 N (B) 4 N and (C) 8 N on the (a) steel (b) D1 and (c) G1 specimens under linoleic acid.
Fig. 15. Summary of coefficient of friction under different test conditions.
friction is determined uniquely by the substrate–lubricant interface architecture [30,47,48].
3.3.3. Wear rate of DLC in dry unlubricated and lubricated conditions Wear rate is found to be high on the steel surface in dry conditions. This value is significantly less under oleic and linoleic lubricated condition (Fig. 16). It is the lowest at 1 N load under oleic acid followed by low CoF. In dry condition, both D1 and G1 specimens show high wear rates. This value is less while using
linoleic acid on both the specimens D1 and G1. However, wear rate is low on G1 compared to D1 specimen. A minimum wear rate of 1.2 10 11 mm/Nm is measured at the maximum normal load of 8 N applied to G1 specimen in the presence of linoleic acid. D1 shows high wear rate under dry sliding conditions due to high adhesion which is caused by the capillary force induced adhesion. This kind of strong adhesion is possible when large amounts of hydrogen atoms and molecules are present in the coating matrix. The D1 coating contains internal hydrogen which interacts with external microdroplets of water present in humid tribo-atmosphere. This interaction is strong in nature and causes a high wear rate of hydrogenated D1 coatings compared to G1 (Fig. 16). However, in lubricated conditions, internal hydrogen has negligible effect on the surface due to the microscopic layer of fatty acids in the sliding contact conditions. This fatty acid is constituted by the long chained hydrocarbons. Lubricated layers are sufficient to passivate the covalent bonds and interact with internal hydrogen present in D1 coatings which does not react directly with the counter body. This does not cause large differences in wear rates of D1 and G1 coatings in the lubricated conditions.
3.3.4. Optical microscopy of wear track It is evident from optical microscopy that wear tracks inscribed in dry condition are significantly wider compared to those recorded in lubrication condition. This is true in cases of D1, G1 and steel specimens as shown in Fig. 17. In dry condition, track of steel specimen is severely deformed as shown in (c). In lubricated condition, wear track width as shown in images (d) and (e) is less compared to (a) and (b), respectively. Using oleic acid, the wear track width is less on D1 as compared to G1 specimen. It is well known that less wear depth correspond to
138
N. Sharma et al. / Tribology International 56 (2012) 129–140
Fig. 16. Wear rate of specimen vs applied normal load on (a) steel (b) D1 and (G1) under dry,oleic and linoleic conditions.
Fig. 17. Typical optical image of wear tracks formed at 4 N normal loads on (a) D1—dry (b) G1—dry (c) steel—dry (d) G1—oleic (e) D1—oleic (f) steel—oleic (g) G1—linoleic (h) D1—linoleic and (i) steel—linoleic conditions.
lower CoF. Similarly, wear width is shallow on D1 specimen in the presence of linoleic acid lubricant, which results is ultralow value of CoF. In lubricated conditions, residual surface roughness of the coating is visible in the wear tracks which clearly indicate the formation of protective lubricant layer on the contacting interface. In this condition, CoF mostly depends on the lubricant properties. In some cases, it is shown that CoF gradually decreases with the sliding distance. It may be related to the smoothness of the surface and the formation of crystalline lubricant tribolayer.
4. Scratch and friction model Scratch test reveals that the interfacial adhesion in D1 coating is stronger than G1 coating in both dry and lubricated conditions as shown in Figs. 2 and 3, respectively. It is indicated that diamond like sp3 fraction is higher in D1 which results in high hardness and elastic modulus compared to G1 coating. High fraction of sp3 bonding and Cr content causes ductile deformation to dominate the scratch test while brittle type fracture is observed in G1 coating due to the large amount of sp2 content.
N. Sharma et al. / Tribology International 56 (2012) 129–140
The sp2 fraction possesses shearable basal planes. These basal planes slide and exhibit brittle type fracture. Low friction coefficient of D1 coatings at higher loads of 4 and 8 N (Fig. 8(D1)) and after laps of longer sliding distance is dominated by two factors: (i) the formation of graphitized/amorphized tribolayer and (ii) the passivation of covalent bonds due to the chemisorption of hydrogen atoms/molecules [5,7]. At low load of 1 N, graphitization/amorphization and chemisorption of hydrogen atoms/molecules do not occur sufficiently. Similarly, G1 coating shows super low friction coefficient after longer sliding distance at higher load of 8 N (Fig. 8(G1)). At high loads, self-alignment of graphitized/ amorphized tribolayer and effective passivation of dangling bonds by hydrogen molecules present in ambient atmosphere are possible factors that reduce the friction coefficient to such an extent [5,38–40]. As the sliding distance progresses, the friction coefficient of D1 and G1 coatings is found to decrease. This is related with the formation of stabilized graphitized/amorphized tribolayers [38–40]. However, increase in friction coefficient of G1 and D1 coatings under lubricated condition using oleic and linoleic acids (Figs. 13 and 14 and summarized in Fig. 15) occurs with an increase in load. This remarkable behavior of friction is related to the formation of surface tribolayer. In lubricated conditions, the formation of graphitized/amorphized tribolayer does not act as a determining factor. Formation of these layers is unstable and restricted by lubricants present at the sliding interface while the friction mechanism is mainly driven by internal energy (viscosity) of the fluid related to the interfacial stress and lubricant confinement [25,26,30,32]. At lower load, the layer of the liquid is thicker compared to what occurs in higher load. Thickness dependent lubricant layer follows the hydrodynamic, elastohydrodynamic and boundary lubrication processes. At low load, in hydrodynamic and/or elastohydrodynamic conditions, the contact deformation is low which is enhanced at high loads when boundary lubrication dominates. In this case, friction model depends on the viscosity of the lubricant as well as on the underlying deformation mechanism. At high loads, the viscosity of the lubricant confined at interfaces is high and the deformation of the contacting surface is also severe. These are the dominating factors which contribute in the increase of the friction coefficient with the increase in load.
5. Conclusions Scratch resistance and tribological properties of hydrogenated and hydrogen free diamond like carbon coatings were investigated under dry unlubricated and lubricated conditions in ambient atmosphere using layers of oleic and linoleic acids. During scratch test, ductile type failure was observed in hydrogenated coatings while brittle failure of crack induced larger area spallation was observed in hydrogen free coatings. Scratch resistance was improved in hydrogenated coating using linoelic and oleic acids compared to unlubricated dry condition. Under similar conditions, scratch resistance property was found to be low in the hydrogen free coatings. Under ambient dry conditions, CoF was found to decrease, while under lubrication condition, it was found to increase with normal load for both hydrogenated and hydrogen free DLC coatings. The CoF was found to be low with oleic acid as compared to linoleic acid. At low load of 1 N, hydrogenated and hydrogen-free coatings show a super low CoF of 0.001 with oleic acid. Even, steel surface shows a very low CoF of 0.005 in such conditions. Optical micrograph shows wider wear track under dry conditions while this was shallow under linoelic acid compared to dry and lubricated oleic acid condition. Severe deformation of the wear track was observed on steel surface while sliding under ambient dry conditions.
139
Acknowledgments The authors would like to thank the reviewer for providing their useful comments, suggestions and guidelines during the course of revision to improve the technical quality of the present paper. The authors would also like to thank Teer Coatings Ltd., UK, to provide the specimens for tribological studies. The authors are grateful to Dr. C.S. Sundar, Director, Materials Science Group, and Shri C.S. Chetal, Director, Indira Gandhi Centre for Atomic Research, for encouragement and support.
References [1] Robertson J. Deposition and properties of diamond like carbon. Materials Research Society Symposium Proceedings 1999;555:12. [2] Braun OM, Manini N, Tosatti E. Role of lubricant molecular shape in microscopic friction. Physical Review B 2008;78:195402–12. [3] Etsion I, Halperin G, Becker E. The effect of various surface treatments on piston pin scuffing resistance. Wear 2006;261:785–91. [4] Castle RC, Bovington CH. The behaviour of friction modifiers under boundary and mixed EHD conditions. Lubrication Science 2003;15:253–63. [5] Erdemir A. The role of hydrogen in tribological properties of diamond-like carbon films. Surface and Coatings Technology 2001;146–147:292–7. [6] Neuville S, Matthews A. A perspective on the optimization of hard carbon and related coatings for engineering applications. Thin Solid Films 2007;515:6619–53. [7] Sanchez-Lopez JC, Erdemir A, Donnet C, Rojas TC. Friction-induced structural transformations of diamondlike carbon coatings under various atmospheres. Surface and Coatings Technology 2003;163–164:444–50. [8] Grischke M, Hieke A, Morgenweck F, Dimigen H. Variation of the wettability of DLC-coatings by network modification using silicon and oxygen. Diamond and Related Materials 1998;7:454–8. [9] Mortier RM, Orszulik ST. Chemistry and technology of lubricants. 2nd ed. Glasgow: Blackie Academic Professional; 1993. [10] Dorinson A, Ludema KC. Mechanics and chemistry in lubrication. Tribology series. Amsterdam: Elsevier; 1985 [p. 9]. [11] Pawlak Z. Tribochemistry of lubricating oils. Tribology and interface engineering series. Amsterdam: Elsevier; 2003 [p. 45]. [12] Franklin SE, Baranowska J. Conditions affecting the sliding tribological performance of selected coatings for high vacuum bearing applications. Wear 2007;263:1300–5. [13] Jiang JC, Meng WJ, Evans AG, Cooper CV. Structure and mechanics of W-DLC coated spur gears. Surface and Coatings Technology 2003;176:50–6. [14] Tung SC, Gao H. Tribological characteristics and surface interaction between piston ring coatings and a blend of energy-conserving oils and ethanol fuels. Wear 2003;255:1276–85. [15] Etsion I, Halperin G, Becker E. The effect of various surface treatments on piston pin scuffing resistance. Wear 2006;261:785–91. [16] Hershberger J, Ozturk O, Ajayi JB, Woodford JB, Erdemir A, Erck RA, Fenske GR. Evaluation of DLC coatings for spark-ignited direct-injected fuel systems. Surface and Coatings Technology 2004;179:237–44. [17] Dai M, Zhou K, Yuan Z, Ding Q, Fu Z. The cutting performance of diamond and DLC-coated cutting tools. Diamond and Related Materials 2000;9:1753–7. [18] Sato T, Besshi T, Tsutsui I, Morimoto T. Anti-galling property of a diamondlike carbon coated tool in aluminium sheet forming. Journal of Materials Processing Technology 2000;104:21–4. [19] Tan AH. Corrosion and tribological properties of ultra-thin DLC films with different nitrogen contents in magnetic recording media. Diamond and Related Materials 2007;16 [467–47]. [20] Zhao X, Bhushan B. Comparison studies on degradation mechanisms of perfluoropolyether lubricants and model lubricants. Tribology International 2000;9:187–97. [21] Gatzen HH, Beck M. Tribological investigations on micromachined silicon sliders. Tribology International 2003;36:279–83. [22] Numata T, Nanao H, Mori S, Miyake S. Chemical analysis of wear tracks on magnetic disks by TOF-SIMS. Tribology International 2003;36:305–9. [23] Sheeja D, Tay BK, Nung LN. Feasibility of diamondlike carbon coatings for orthopaedic applications. Diamond and Related Materials 2004;13:184–90. [24] Shi B, Ajayi OO, Fenske G, Erdemir A, Liang H. Tribological performance of some alternative bearing materials for artificial joints. Wear 2003;255: 1015–21. [25] Erdemir A. Review of engineered tribological interfaces for improved boundary lubrication. Tribology International 2005;38:249–56. [26] Hsu SM, Gates RS. Boundary lubricating films: formation and lubrication mechanism. Tribology International 2005;38:305–12. [27] Karin P, Rickard G. Tribological performance of a DLC coating in combination with water-based lubricants. Tribology International 2003;36:851–5. [28] Takeshi Y, Junji A, Takuya T, Naoko T, Mamoru T, Atsushi M, Toshihide O, Kazuo H. Sliding velocity dependency of the friction coefficient of Si-containing diamond-like carbon film under oil lubricated condition. Tribology International 2011;44:1296–303.
140
N. Sharma et al. / Tribology International 56 (2012) 129–140
[29] Vercammen K, Van Acker K, Vanhulsel A, Barriga J, Arnsek A, Kalin M, Meneve J. Tribological behaviour of DLC coatings in combination with biodegradable lubricants. Tribology International 2004;37:983–9. [30] Haque T, Morina A, Neville A, Kapadia R, Arrowsmith S. Study of the ZDDP antiwear tribofilm formed on the DLC coating using AFM and XPS techniques. Journal of ASTM International 2007;4:1–7. [31] Kalin M, Roman E, Vizintin J. The effect of temperature on the tribological mechanisms and reactivity of hydrogenated, amorphous diamond-like carbon coatings under oil-lubricated conditions. Thin Solid Films 2007;515:3644–52. [32] Kalin M, Vizintin J, Barriga J, Vercammen K, van Acker K, Arnsek A. The effect of doping elements and oil additives on the tribological performance of boundary-lubricated DLC/DLC contacts. Tribology Letters 2004;17:679–88. [33] Vengudusamy B, Green JH, Lamb GD, Spikes HA. Tribological properties of tribofilms formed from ZDDP in DLC/DLC and DLC/steel contacts. Tribology International 2011;44:165–74. [34] Equey S, Roos S, Mueller U, Hauert R, Spencer ND, Crockett R. Reactions of zinc-free anti-wear additives in DLC/DLC and steel/steel contacts. Tribology International 2008;41:1090–6. [35] Podgornik B, Vizintin J. Tribological reactions between oil additives and DLC coatings for automotive applications. Surface and Coatings Technology 2005;200:1982–9. [36] Podgornik B, Jacobson S, Hogmark S. Influence of EP additive concentration on the tribological behaviour of DLC-coated steel surfaces. Surface and Coatings Technology 2005;191:357–66. [37] Field SK, Jarratt M, Teer DG. Tribological properties of graphite-like and diamond-like carbon coatings. Tribology International 2004;37:949–56.
[38] Liu Y, Erdemir A, Meletis EI. A study of the wear mechanism of diamond-like carbon films. Surface and Coatings Technology 1996;82:48–56. [39] Hongxuan L, Tao X, Chengbing W, Jianmin C, Huidi Z, Huiwen L. Tribochemical effects on the friction and wear behaviors of a-C:H and a-C films in different environment. Tribology International 2007;40:132–8. [40] Liu Y, Erdemir A, Meletis EI. An investigation of the relationship between graphitization and frictional behavior of DLC coatings. Surface and Coatings Technology 1996;86–87:64. [41] Erdemir A, Bindal C, Pagan J, Wilbur P. Characterization of transfer layers on steel surfaces sliding against diamond-like hydrocarbon films in dry nitrogen. Surface and Coatings Technology 1995;76–77:59. [42] Zaidi H, Djamai A, Chin KJ, Mathia T. Characterization of DLC coating adherence by scratch testing. Tribology International 2006;39:124–8. [43] Stachowiak GW, Batchelor AW. Engineering tribology. Tribology series 24. Amsterdam: Elsevier; 1993. [44] Weller DE, Perez JM. A study of the effect of chemical structure on friction and wear. Part 1. Synthetic ester base fluids. Lubrication Engineering 2000;56:39–44. [45] Waleska C, Weller DE, Kraipat C, Perez Joseph M. The effect of chemical structure of basefluids on antiwear effectiveness of additives. Tribology International 2005;38:321–6. [46] Castle RC, Bovington CH. The behaviour of friction modifiers under boundary and mixed EHD conditions. Lubrication Science 2003;15:253–63. [47] Braun OM, Naumovets AG. Nanotribology: microscopic mechanisms of friction. Surface Science Reports 2006;79 [79–58]. [48] Braun OM, Manini N, Tosatti E. Role of lubricant molecular shape in microscopic friction. Physical Review B 2008;78:195402–12.