Tribological properties of tribofilms formed from ZDDP in DLC/DLC and DLC/steel contacts

m reen , Gordon D. Lamb , Hugh A. Spikes ndon ings gine for ont C gi der airs. on (DL arts be ies. ditives es we r to fo now if ive wi re are m DLC coatings [1,12,13] while some have reported it did not form or from different suppliers and rubbed in a ZDDP solution to under- mens 2100 Rq. All Hardness and elastic modulus, coating thickness, chemical composi- Contents lists available at ScienceDirect els Tribology Int Tribology International 44 (2011) 165–174 chemical composition are listed in Tables 1 and 2, respectively.E-mail address: [email protected] (B. Vengudusamy). result from different and varying DLCs being studied. In this investigation, a wide range of DLC coatings have been obtained tion and hydrogen content of each of DLC coatings was measured using FischerscopeHM2000, FIBmilling (FEI FIB200-SIMS), SEM/EDX and ERD, respectively. Six different types of DLC were studied. Their properties and 0301-679X/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.triboint.2010.10.023 n Corresponding author. Tel.: +44 20 7594 7063; fax: +44 20 7594 7023. that no tribochemical reaction (or no evidences) occurred between a DLC and an additive [4,5,7,10]. Some of these discrepancies may the coatings were deposited at a temperature of less than 200 1C, so that no loss of the original hardness of the substrate should occur. Many researchers have evaluated the tribological performance of DLC coatings with antiwear and extreme pressure additives, including ZDDP [1–13]. The results are quite contradictory, as some authors have reported that a ZDDP-derived tribofilm did form on 2. Materials and experimental details In all cases, DLC coatings were deposited on the test speci generally used in the minitraction machine (MTM), i.e. AISI 5 steel balls anddiscs, of hardness 760 HVand roughness 10 nm, DLC coating and these may behave quite differently. Lubricant manufacturers must be aware of these differences and be able to produce lubricant formulations that are effective with many types of DLC coating. This paper describes a study of the performance of the antiwear additive zinc dialkyldithiophosphate (ZDDP) with a wide range of different DLC types. and notwith 30 at%H), this paperwill focus only on the similarities and differences in the tribological properties of the main types of DLC coatings. 1. Introduction In recent years diamond-like carb to appear attractive for automotive p friction and wear resistance propert Howevermost of the lubricant ad to control friction and wear in engin with ferrous-based surfaces. In orde these new surfaces it is essential to k and widely used additives are effect It is important to realise that the C) coatings have begun cause of their excellent that are currently used re designed to interact rmulate lubricants for the currently available th DLCs. any different types of stand and compare their film-forming and wear resistance proper- ties. For the bulk of the test work, DLC/DLC tribopairs have been used so as to focus on the interactionof the additive ZDDPwithDLC. However some tests have also been carried out using the more practically relevant DLC/steel tribopair. In this case it can be difficult to know whether any tribofilm observed on DLC formed on the DLC surface or whether it was transferred from the steel counterpart during rubbing [7]. Although some studies [4,13] reported that hydrogen has some film formation influence (i.e. tribofilm formation observed only with DLC having 14–16 at% H Tribological properties of tribofilms for in DLC/DLC and DLC/steel contacts Balasubramaniam Vengudusamy a,n, Jonathan H. G a Tribology Group, Department of Mechanical Engineering, Imperial College London, Lo b Castrol Ltd., Pangbourne, Reading RG8 7QR, UK a r t i c l e i n f o Article history: Received 16 August 2010 Received in revised form 19 October 2010 Accepted 20 October 2010 Available online 29 October 2010 Keywords: DLC coatings ZDDP Tribofilm AFM a b s t r a c t Diamond-like carbon coat and wear resistance of en additives such as ZDDP can they do in steel on steel c coatings. It is seen that ta- wear prevention. A ZDDP- W-DLC in DLC/DLC tribop journal homepage: www. SW7 2AZ, UK (DLCs) are considered to hold great promise for improvement in friction parts. It is hence interesting to know whether conventional engine oil m tribofilms and reduce friction and wear in DLC contacts as effectively as acts. This paper compares the behaviour with ZDDP of six different DLC ves lower boundary friction than the other types while a-C:H gives better ived tribofilm forms on all DLCs but a pad-like structure is seen only on & 2010 Elsevier Ltd. All rights reserved. ed from ZDDP b b a evier.com/locate/triboint ernational ess 35 80 13 35 14 13 45 B. Vengudusamy et al. / Tribology International 44 (2011) 165–174166 Table 1 DLC coatings investigated. Materials No. of layers Adhesion layer Rq (nm) Hardn W-DLC 4 Cr 16 13187 WC-DLC 3 Cr 30 12507 Si-DLC 1 Ti 32 13157 ta-C 1 – 40 67937 a-C:H 1 – 13 24607 a-C 1 Cr 17 24657 AISI 52100 steel – – 10 7607 Table 2 Chemical composition (atomic %) of as-deposited DLC coatings. As-deposited C O Si Ti W-DLC 76 8.4 – – WC-DLC 90 – – – Si-DLC 91 4 4.5 0.5 ta-C 97 0.9 – – a-C:H 99 – 1.0 – Coatings were deposited by different techniques based on the type of coatings.W-DLC andWC-DLC coatingswere deposited by hybrid PVD/PECVD, Si-DLC by CVD, ta-C by arc PVD, a-C:H by PECVD and a-C by PVD technique. AFM topography images of all the DLCs before testing were processed (using the software developed by Horcas et al. [14]) and are shown in Fig. 1. In this figure the same topography and lateral force scales have been used for all the AFM images. Tests were conducted on DLC/DLC and DLC/steel tribopairs. The base fluid used in this study was an API Group III oil having density of 0.78 g/cm3, viscosity of 3.32 cP and effective pressure viscosity coefficient of 16.3 GPa�1 at the test temperature of 100 1C, respectively. The ZDDP additive had a primary alkyl structure andwas used at a concentration that gave 0.08 wt% P in the base oil solution. Friction tests were carried out using a ball-on-disc MTM, where a 19 mm diameter DLC-coated or steel ball was loaded and rubbed against a DLC-coated disc immersed in lubricant solution under rolling–sliding lubrication conditions (applied load¼31 N corre- sponding to a maximum Hertz contact pressure of 0.95 GPa based on substrate, entrainment speed¼0.1 m/s, slide–roll ratio¼0.5, temperature¼100 1C, test duration¼2 h). The calculated EHD minimumfilm thicknesswas 10 nm, corresponding to a theoretical lambda ratio ranging from 0.3 to 0.8 for the various DLCs; i.e. the operating lubrication regime was mixed-boundary lubrication. Periodically during a test, motion was halted and friction was then Fig. 1. AFM (a) topography and (b) lateral force images of DLC coatings before testin 0–120 mV). (HV) Elastic modulus (GPa) Coating thickness (mm) Hydrogen % 16573 1–5 14 14077 3–5 16 0 9073 3 20 0 473725 3 1 5 17577 3 20 5 19873 3 1 21075 – – Cr Fe Co Ni W – 0.6 2.0 – 13 – – – 2.0 8 – – – – – 0.1 2.0 – – – – – – – – measured over a range of entrainment speeds at slide–roll ratio¼0.5 (slide–roll ratio (SRR) is defined as the ratio of sliding speed to mean rolling speed). In some tests, optical interference images were also taken from the rubbed track on the ball using the spacer layer imaging method (SLIM) [15] before slow speed rubbing was resumed. After friction tests, the discs were rinsed in cyclohexane before the wear tracks were examined by use of a series of techniques or devices including the AFM (D3000with a NanoScope IIIa controller from Veeco Instruments, in contact mode), SEM/EDX (Hitachi S3400 VPSEM and Inca EDX system, 15 kV accelerating voltage, 60 s spectra collection time), ToF-SIMS IV instrument (ION-TOF GmbH, Mu¨nster, Germany), Raman spectroscopy (Renishaw 1000 confocal system), an optical white light interferometer (WYKO NT 9100) and an optical microscope. Wear tests were conducted under pure sliding conditions using the MTM that was operating in the disc reciprocating mode with the ball stationary (applied load¼31 N corresponding to a contact pressure of 0.95 GPa based on substrate, frequency¼10 Hz, stroke length¼4 mm, temperature¼100 1C and test duration¼4 h). After wear tests, the tested samples were rinsed in cyclohexane and the tribofilms (if any) were removed by EDTA as described in [16], beforemeasuring thewear volumesof the testedball anddisc using an optical white light interferometer (WYKO NT 9100). In a recent paper on DLC wear [17] the composite wear rate was studied, i.e. that obtainedby summing thewear volumes of the ball and the disc g (scan size¼2 mm�2 mm; vertical scale: topography¼0–60 nm, lateral force¼ and then calculating a system wear value. However, most other wear studies [8,13] calculated wear of individual components. In this paper, therefore, the composite wear rate and thewear rates of individual components, i.e. ball and disc,were calculated from their respective wear volumes using the following equations: kb ¼ Vb 2DxnW kd ¼ Vd 2DxnW kc ¼ VbþVd 2DxnW where Vb and Vd are the wear volumes of the ball and the disc, respectively; kb, kd and kc are the ball, the disc and the composite coefficients of wear, respectively,W the normal load, Dx the stroke length and n the number of cycles. 3. Results In order to understand the role of ZDDP in friction and wear, DLCs were tested in base oil and ZDDP solution. Friction versus mean speed (Stribeck) curves were obtained for DLC/DLC and DLC/steel tribopairs. 3.1. DLC/DLC tribopairs The friction coefficients obtained initially and after 2 h of rubbing of DLCs in base oil are shown in Fig. 2(a, b). Initially, most coatings give a low friction coefficient of 0.02 at high entrainment speed, characteristic of full film EHL lubrication. The initial friction curves for WC-DLC and Si-DLC show high friction at intermediate speeds, which suggests that a full EHL separating film is only formedat high speeds, above 3 m/s. For Si-DLC this probably results from the initial surface roughness of this coating, but no clear correlation was noticed with respect to the lambda ratio. It may be that the initial structure of this coating, with high nodules separated by deep valleys, may inhibit EHL film formation. The ta-C was also initially rough, but its low boundary friction meant that it did not produce high friction at intermediate speeds seen in Si-DLC. It should be noted from Fig. 1 thatmost of the coatings have nodular structures and these structures were either smoothed or covered by very thin oxygen-rich layers after 2 h of rubbing. The possibility of the formation of an oxygen-rich film on DLCs at room and elevated temperatures has been reported previously by Kalin et al. [11,12]. Friction after 2 h of rubbing is slightly reduced for all coatings and the reduction is especially significant for Si-DLC, probably because of surface smoothing (Rq reduced from 32 to 12 nm). It is also interesting to note that a-C gave very low a) in B. Vengudusamy et al. / Tribology International 44 (2011) 165–174 167 Fig. 2. Stribeck curves for DLC/DLC tribopairs ( Fig. 3. Stribeck curves for DLC/DLC tribopairs (a) initia itially and (b) after 2 h of rubbing in base oil. lly and (b) after 2 h of rubbing in ZDDP solution. boundary friction initially, probably because of its high sp2 nature but this low frictionwas not retained during subsequent rubbing in base oil. Despite the varying friction behaviour of DLCs, all their friction coefficients were lower than the very high value of m¼0.22 measured for steel/steel in base oil at low entrainment speed. With the ZDDP solution, an immediate reduction in frictionwas observed for most DLCs [compare Figs. 2(a) and 3(a)]. This clearly shows that ZDDPs interact with the DLC surfaces. DLCs also retain this reduced friction behaviour during the prolonged rubbing of 2 h. The boundary frictions of all the DLCs were significantly lower than steel/steel. Normally, and as shown in Fig. 3(b), steel/steel lubricated by ZDDP gives increased friction in the intermediate speed region after rubbing, which has been ascribed to its thick, pad-like structure. Such behaviourwas not observedwith theDLCs. One similarity between base oil and ZDDP was that ta-C gave low boundary friction in both cases, followed by a-C/a-C:H, Si-DLC and W-DLCs. This indicates the dependence of friction on the type of DLC. It is interesting to note thatwhen lubricated by ZDDP solution, the Si-DLC continues to give quite high friction at high entrainment speeds, even after 2 h of rubbing. This suggests that the running in of this DLC observed with base oil is retarded by the ZDDP. The wear coefficients of the ball and the disc tested in base oil and ZDDP solution are presented in Table 3. For most DLCs, disc wear rate is one order of magnitude higher than ball wear rate and therefore the composite wear is dominated by the disc wear. Composite wear coefficients are shown in Fig. 4. In each base oil and ZDDP solution, Si-DLC, a-C and a-C:H coatings give negligible (i.e. no measurable) wear whereas other DLCs show similar wear coefficients in the order of 10�17 m3/N m as can be seen from Table 3 and Fig. 4. It should be noted that there were no significant changes in friction during these tests (one example of friction response shown in Fig. 5), suggesting that the contacts remain in the mixed lubrication regime throughout a rubbing test. The wear tracks of two coatings, W-DLC and a-C:H after 4 h of rubbing in ZDDP solution are compared in Fig. 6. It is evident from Fig. 6 that a-C:H coating shows no measurable wear on the rubbed surface. 3.2. DLC/steel tribopairs 3.2.1. Influence of steel counterpart (comparison of DLC/DLC and DLC/steel) As can be seen in Fig. 7(a), with a steel counterpart, most DLCs tested in base oil initially show higher friction than the corresponding DLC/DLC, the one exception being Si-DLC. However for most DLCs the friction obtained after 2 h of rubbing was comparable with that of DLC/DLC, as can be seen by comparing Figs. 2(b) and 7(b). AswithDLC/DLC, ta-C gave the lowest boundary friction whereas all other types gave similar friction behaviour. As shown in Fig. 8, all DLC/steel combinations lubricated with ZDDP solution give higher boundary and mixed friction than the corresponding DLC/DLC, initially and after 2 h of rubbing. For most DLCs the boundary friction values approach the friction coefficient of steel/steel, with the friction coefficients of a-C:H and Si-DLC even being slightly higher than those for steel/steel. This indicates that Table 3 Coefficient of wear for balls and discs in DLC/DLC tribopair with BO and ZDDP solution (NMW—no measurable wear). Material (DLC/DLC) Ball wear rate (DLC) �17 3 Disc wear rate (DLC) �17 3 B. Vengudusamy et al. / Tribology International 44 (2011) 165–174168 [10 m /N m] [10 m /N m] BO ZDDP BO ZDDP W-DLC/W-DLC 0.2 0.2 4 2 WC-DLC/WC-DLC 0.3 0.6 4 5 Si-DLC/Si-DLC NMW NMW 0.2 0.3 ta-C/ta-C 1 0.5 2 1 a-C:H/a-C:H NMW NMW NMW NMW a-C/a-C NMW NMW 0.1 NMW Steel/steel 3 NMW 1 0.1 Fig. 4. Composite wear coefficients for (a) DLC/DLC and (b) D Fig. 5. Friction coefficients from pure sliding wear tests for DLC/DLC contacts in ZDDP solution. LC/steel tribopairs tested in base oil and ZDDP solution. B. Vengudusamy et al. / Tribology International 44 (2011) 165–174 169 even though DLCs are known to exhibit low boundary friction properties,whenone of the contact surfaces is steel this low friction is lost or degraded during reaction of ZDDP with steel, as can be seen by comparing Figs. 3 and 8. Friction in the mixed lubrication regime at intermediate speeds is lower than for steel/steel but higher than for DLC/DLC. The one exception is ta-C which maintains relatively low friction with ZDDP even when rubbed against steel. Comparison of Tables 3 and 4 and Fig. 4 shows that the wear properties of DLCs were changed when the counterface was steel. This is evident from the decrease in wear coefficients of the DLC discs when they were rubbed against a steel ball in base oil. This is not surprising as the steel ball surface, being softer than the DLC disc, should impart less wear on the DLC disc. One could have expected more wear on the steel balls since they were rubbed Fig. 6. Wear tracks on balls and discs of (a) W-DLC and (b) a-C:H coatings after 4 h of scale¼�5 to 1 mm). against relatively harder DLC discs. However, the steel balls show wear coefficients similar to those of DLC balls except when rubbed against ta-C coatings, where the steel balls show more wear, probably because of the large difference in hardness. In contrast, the wear coefficients of the DLC discs increased when rubbed against steel in ZDDP solution and showed a particularly large increase with Si-DLC. It appears that the combination of ZDDP and steel counter-surface degrades the wear properties of Si-DLC. 3.2.2. Influence of ZDDP in the presence of steel counterpart (comparison of DLC/steel in base oil and ZDDP solution) By comparing Figs. 7 and 8, it can be observed that only ta-C is able to retain lowboundary friction propertieswhen ZDDP solution lubricates DLC/steel tribopairs. For all other DLCs, ZDDP increases rubbing in base oil and ZDDP solution (disc sizes in mm; ball sizes in mm; vertical B. Vengudusamy et al. / Tribology International 44 (2011) 165–174170 boundary and mixed friction, initially and after 2 h of rubbing. It is possible that ZDDP molecules reduce the effectiveness of the low-friction surface of sp2-dominated DLCs. All DLCs except ta-C show a friction trend after 2 h of rubbing similar to that of steel/ steel. The reason may be that a ZDDP-derived tribofilm forms on the steel counterpart and, as a result, the contact exhibits friction behaviour similar to that of a ZDDP tribofilm-coated steel/steel. Because of this, thewear resistance ofDLC/steel tribopairsmight be expected to be better than that of DLC/DLC. However this is not noted. Instead, an increase in wear coefficient was noted for doped DLCs (W and Si) when rubbed in ZDDP solution. Their composite Fig. 7. Stribeck curves for DLC/steel tribopairs (a) i Fig. 8. Stribeck curves for DLC/steel tribopairs (a) initi Table 4 Coefficient of wear for balls and discs in DLC/steel tribopair with BO and ZDDP solution (NMW—no measurable wear). Material (DLC/steel) Ball wear rate (steel) [10�17 m3/N m] Disc wear rate (DLC) [10�17 m3/N m] BO ZDDP BO ZDDP W-DLC/steel 0.2 0.2 3 5 WC-DLC/steel 0.2 0.2 3 6 Si-DLC/steel NMW 0.2 0.2 3 ta-C/steel 2 0.9 0.8 NMW a-C:H/steel NMW NMW NMW NMW a-C/steel NMW NMW NMW NMW Steel/steel 3 NMW 1 0.1 nitially and (b) after 2 h of rubbing in base oil. wear coefficients were in the range 2�10�17–6�10�17 m3/N m as can be seen from Fig. 4(b), comparable to that of steel/steel in base oil. 4. Discussion In this study, six different types of DLCs have been investigated and it is clear that ta-Cs show consistently lower boundary friction than the other types. DLC/DLC combinations exhibit slightly high boundary friction in base oil than in the ZDDP solution. Addition of ZDDP reduces the boundary friction almost immedi- ately, indicating some surface adsorption on all DLCs. DLCs in DLC/steel combinations show slightly higher friction in the ZDDP solution than in base oil, probably reflecting a negative impact on the friction of the ZDDP tribofilm, on the steel surface, for the ZDDP used in this study. In DLC/DLC andDLC/steel tribopairs tested in the ZDDP solution, 2 h of rubbing has a negligible further influence on friction reduction. However it was found from surface analysis that all DLCs show tribofilm formation after 2 h of rubbing. This was evidenced by a patchy,white layer in SEMmicrographs as shown in Figs. 9 and 10. A similar patchy layer was observed by Equey et al. [1]. EDX measurements were taken from this patchy layer and showed the presence of P, S and Zn elements, as summarised in Tables 5 and 6. Since EDX is not a surface sensitive technique ally and (b) after 2 h of rubbing in ZDDP solution. n DL P so Ti – – 0 – – n DL B. Vengudusamy et al. / Tribology International 44 (2011) 165–174 171 Fig. 10. SEM micrographs of DLC surfaces i Table 5 EDX measurements (atomic %) on DLC surfaces of DLC/DLC tribopairs tested in ZDD ZDDP C O Si P S W-DLC 64 14 – 0.7 0.4 WC-DLC 82 7 – 0.6 0.7 Si-DLC 82 11 3.9 0.6 0.6 ta-C 90 4 – 0.2 0.4 a-C:H 95 3 1.0 0.1 0.1 Fig. 9. SEM micrographs of DLC surfaces i ToF-SIMS was used to map the surface elements and it was found that the surfacemainly comprised P-, S- and Zn-containing species, including thiophosphates, as can be seen in Fig. 11. Brighter regions in ToF-SIMS images indicate an abundant presence of triboele- ments and darker regions indicate their absence. Thus, the ability of DLCs to form the ZDDP-derived tribofilm is clearly evident from the measurements of SEM, EDX and ToF-SIMS. This is in agreement with some reports in the literature [1–3,9,11–13], but contrary to others [4,5,7,10]. It is also interesting to know whether any of the pad-like tribofilm structures that are normally observed on steel/steel rubbed in ZDDP-containing oils form on DLC surfaces. ZDDP was found to form pad-like structures only on the W-DLC surface, but even these were very tiny (o30 nm diameter) compared to those formed on steel/steel (1–5 mm). This is evident from Fig. 12(a) which shows tiny pads (surrounded by dotted lines for clarity) of very small ZDDP-derived tribofilmonW-DLC. Darker regions in the AFM lateral force maps indicate low lateral force and brighter regions indicate high lateral force, so it can be seen that the tiny pads exhibit low lateral force [Fig. 12(b)]. It appears that ZDDP pads are able to formonly onDLCs containing reactivemetallic elements like tungsten and only on DLC/DLC contacts (compare Figs. 12 and 13). For other types of DLC, the ZDDP-derived tribofilms appear as patches (for ta-C and Si-DLC) or rolled debris (for a-C:H) along the track, as can be seen from Figs. 9 and 12. The thickness of the ZDDP tribofilms could be assessed using AFM line profiling from a topography image that spanned the edge of the rubbed contact. Table 6 EDX measurements (atomic %) on DLC surfaces of DLC/steel tribopairs tested in ZDDP s ZDDP C O Si P S T W-DLC 72 12 – 0.2 0.1 – WC-DLC 80 9 – 0.7 0.5 – Si-DLC 86 9 3.4 0.3 0.2 0 ta-C 58 24 – 0.9 2.0 – a-C:H 94 4 – 0.2 0.2 – C/steel tribopairs tested in ZDDP solution. lution. Cr Fe Co Ni Zn W – 0.6 3.1 – 1.2 16 – – – 1.3 1.4 7 .4 – 0.2 – – 1.3 – 0.1 4.7 – – 0.6 – – 0.2 – – 0.3 – C/DLC tribopairs tested in ZDDP solution. This is shown for W-DLC in Fig. 14. In general the film thicknesses measured were in the range of 10–30 nm on the DLC discs and 50–150 nm on the steel balls, depending on the type of DLC. Raman spectroscopy was also carried out before and after rubbing to estimate the degree of graphitization from the ratio ofD andG intensities [18]. No graphitisationwasnoted in this study when DLCs were rubbed in base oil. However when rubbed against steel in the ZDDP solution WC-DLC showed some graphitization. When conditions are favourable for the surface to be graphitised, the tribofilm formation appears to be disrupted and one example of this kind can be observed with WC-DLC/steel tested in the ZDDP solution, as shown in Fig. 15. Fig. 15 shows a series of SLIM interference images taken from a steel ball rubbed against W-DLC and WC-DLC during a friction test. It can be seen that for the steel ball rubbed against WC-DLC, a tribofilm forms quite rapidly on the ball but this then reduces in thickness (from 90 to 50 nm) with further rubbing. Raman spectrawere taken fromthe steel ball at the end of 2 h of rubbing and these revealed some carbon layer transfer from theWC-DLC coating onto the steel ball. A slight increase in the intensity ratio could be noted for the transfer layer (Fig. 16), probably because theWC-DLC had graphitised, and the graphitised carbon layer transferred to the steel ball during subsequent rubbing. Although such disrupting behaviour is noted, the presence of triboelements was seen (Table 6). In order to evaluate and compare the friction properties of tribofilms formed on DLC surfaces using AFM the same load was maintained (by maintaining the same set point) between the AFM olution. i Cr Fe Co Ni Zn W – – 2.6 – 0.5 13 0.3 0.4 – 1.3 1.2 6 .1 – 0.2 – – 0.6 – 0.2 13 – – 1.5 – – 0.1 – – 0.5 – B. Vengudusamy et al. / Tribology International 44 (2011) 165–174172 tip and the sample in all AFM imaging tests. As can be seen in Figs. 12 and 13, in DLC/DLC and DLC/steel, a-C:H and a-C show tribofilms of debris-like and patchy nature, respectively, whereas, W-DLC show tiny pads with DLC/DLC but not with DLC/steel. Fig. 12. AFM (a) topography and (b) lateral force images of DLC surfaces in DLC/DLC trib 20 mm�20 mm for steel/steel; vertical scale for DLCs: topography¼0–60 nm, lateral fo Fig. 13. AFM (a) topography and (b) lateral force images of DLC surfaces in DLC/steel tri 20 mm�20 mm for steel/steel; vertical scale for DLCs: topography¼0–60 nm, lateral fo Fig. 11. ToF-SIMS chemical mapping of DLC surfaces The obtained trace and retrace lateral forces were subtracted to nullify the topographical effect and then a histogramwas obtained of the resultant lateral force. The averaged histogram over the load gave friction coefficients (AFM friction results not shown) thatwere opairs after 2 h of rubbing in ZDDP solution (scan size¼2 mm�2 mm for DLCs and rce¼0–120 mV). bopairs after 2 h of rubbing in ZDDP solution (scan size¼2 mm�2 mm for DLCs and rce¼0–120 mV). in DLC/DLC tribopairs tested in ZDDP solution. ball tested against W-DLC and WC-DLC in ZDDP solution. in D B. Vengudusamy et al. / Tribology International 44 (2011) 165–174 173 quite comparable with the friction coefficients obtained by MTM. This shows that AFM is a potential and very useful tool for evaluation of the boundary friction properties of tribofilms. Fig. 15. MTM SLIM interference images taken on steel Fig. 14. Film thickness measurement for W-DLC coating In terms of wear, in the DLC/DLC tribopair, W-DLC in base oil shows higher wear thanmost of the other coatings and thismay be due to surface graphitization, as previously reported by Kalin et al. [12]. However, the addition of ZDDP markedly improves the wear resistance of the W-DLC coating, probably due to the presence of the tiny, ZDDP-derived anti-wear pads on the W-DLC surface. By contrast, a-C:H and a-C showed no measurable wear in either base oil or ZDDP solution. In DLC/steel contacts, because of the high hardness of the DLC coatings relative to steel, the steel balls in base oil-lubricated DLC/ steel contacts were expected to showmore wear than steel balls in steel/steel contacts. However thiswas not the case. Thismaybedue to the fact that the steel balls acquired carbon transfer layers, hence making the contacts more like the DLC/DLC rather than DLC/steel. Alternatively it may simply be due to lesser adhesion between DLC and steel than steel and steel. However, in the presence of ZDDP, steel balls in steel/steel performed excellently with no measurable wear whereas steel balls that rubbed against DLC coatings showed some wear. This indicates that the anti-wear properties of ZDDP tribofilms formed, when steel rubs against a steel disc, are superior to those formed when steel rubs against a DLC. This is probably influenced by other dynamic events like graphitization or carbon transfer layer that could change the tribological situation. For example, the steel balls tested against DLCs in base oil and ZDDP showed very similar wear rates (indicating negligible effect of ZDDP on wear resistance of the ball), rather than the expected lowerwear rates in the ZDDP solution than in base oil. One possible reason for this could be that in DLC/steel a mixture of tribofilm and a carbon layer forms on the steel surface (evident in Fig. 15) and not just tribofilm as observed on steel/steel. It would appear that this mixed layer exhibits wear resistance properties that are similar to those of DLC/DLC rubbed in base oil. LC/DLC contacts after 2 h of rubbing in ZDDP solution. DLC discs in DLC/steel rubbed in the ZDDP solution show slightly higher wear coefficients than in base oil. This indicates that ZDDP hinders the inherent wear-resistant properties of DLCs in DLC/steel contacts. In DLC/DLC contacts, the DLC disc wear coefficients in the ZDDP solution were either same as or slightly higher than in those base oil except for theW-DLC coating. In wear tests on W-doped DLCs, the coatings were almost worn through with more than 50% of the coating was removed at the end of the 4 h test. It should be noted that even though the hardnesses of W-DLC, WC-DLC and Si-DLC coatings are similar, their wear properties are different, as can be seen in Fig. 17. This indicates that hardness is not the only parameter that decides the wear resistance of DLC coatings in lubricated contacts, and it appears thatwear properties can be influenced by the counterfacematerial, the presence or the absence of tribofilms and graphitisation. Overall, the results suggest that although there was no significant variation in friction amongW-DLCs, Si-DLC, a-C:H and a-C, their wear behaviour was very significantly different, andwas dependent on the Fig. 16. Raman spectra of (a) WC-DLC coating before testing and (b) transfer layer on steel ball tested in ZDDP solution, in the region marked as ‘A’ in Fig. 15. ings like a-C:H and a-C showed excellent wear resistance properties ffici B. Vengudusamy et al. / Tribology International 44 (2011) 165–174174 irrespective of whether ZDDP was present or not. Despite the varied wear behaviour, the results clearly demonstrate that DLC coatings are potential candidates in terms of friction andwear even in the absence of widely used anti-wear additives such as ZDDP. Indeed, for most of the DLCs studied, wear was higher with ZDDP than without, in DLC/DLC and DLC/steel contacts. 5. Conclusions The friction and wear resistance properties of W-DLC, WC-DLC Si-DLC, ta-C, a-C:H and a-C in ZDDP solution have been studied. A general friction trend based on the DLC type has been observed irrespective of the presence or absence of ZDDP, i.e. ta-Cs give lower boundary friction, followed by a-C:H/a-C, Si-DLC and W-DLCs. The formation of ZDDP-derived tribofilms on all DLC surfaces is clearly evident and has been demonstrated by means of SEM, EDX, ToF-SIMS andAFM.HoweverZDDPpads formonlyonW-DLC.Acleardependence ofwear resistance properties on the DLC type aswell as ZDDP has been observed.W-DLCandWC-DLCexperiencehigherwear compared to the other coatingswhereas a-C:H and a-C shownomeasurablewear in the presence or absence of ZDDP. Despite its excellent friction behaviour, ta-C shows higher wear than the other types of undoped DLC. Acknowledgements DLC type and the additive used [8]. As a result, it was difficult to establish a general rule on their behaviour. However, selected coat- Fig. 17. Comparison of initial hardnesses and composite wear coe The authors wish to thank Castrol Limited, UK for funding this project. The authors are extremely grateful to Dr. David Scurr and Dr. Xinyong Chen of The University of Nottingham, UK for use of their ToF-SIMS and AFM facility and for their assistance. Also, the authors would like to thank Dr. Chris Jeynes of University of Surrey for carrying out the hydrogen content measurements using ERD. References [1] Equey Sebastien, Roos Sigfried, Mueller Ulrich, Hauert Roland, Spencer Nicholas D, Crockett Rowena. Tribofilm formation from ZnDTP on diamond- like carbon. Wear 2008;264:316–21. [2] Equey Sebastien, Roos Sigfried, Mueller Ulrich, Hauert Roland, Spencer Nicholas D, Crockett Rowena. Reactions of zinc-free anti-wear additives in DLC/DLC and steel/steel contacts. Tribology International 2008;41:1090–6. [3] Topolovec-Miklozic Ksenija, Lockwood Frances, Spikes HA. Behaviour of boundary lubricating additives on DLC coatings (2007). Wear 2008;265: 1893–901. [4] Haque T, Neville A, et al. Non-ferrous coating/lubricant interactions in tribological contacts: assessment of tribofilms. Tribology International 2007;40:1603–12. [5] Podgornik B, Jacobson S, Hogmark S. DLC coating of boundary lubricated components—advantages of coating one of the contact surfaces rather than both or none. Tribology International 2003;36:843–9. [6] Podgornik B, Vizintin J. Tribological reactions between oil additives & DLC coatings for automotive applications. Surface & Coatings Technology 2005;200:1982–9. [7] Podgornik B, Jacobson S, Hogmark S. Influence of EP additive concentration on the tribological behaviour of DLC-coated steel surfaces. Surface & Coatings Technology 2005;191:357–66. [8] KalinM, Vizintin J. Differences in the tribologicalmechanismswhenusing non- doped, metal-doped (Ti, WC), and non-metal doped (Si) diamond-like carbon against steel under boundary lubrication, with and without oil additives. Thin Solid Films 2006;515:2734–47. [9] De Barros Bouchet MI, Martin JM, Le-Mogne T, Vacher B. Boundary lubrication mechanisms of carbon coatings by MoDTC and ZDDP additives. Tribology International 2005;38(3):257–64. [10] KalinM, Vizintin J, Barriga J, Vercammen K, van Acker K, Arnsek A. The effect of doping elements and oil additives on the tribological performance of bound- ary-lubricated DLC/DLC contacts. Tribology Letters 2004;17:679–688. [11] 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. [12] Kalin M, Roman E, Ozbolt L, Vizintin J. Metal-doped (Ti, WC) diamond-like carbon coatings: reactions with extreme pressure oil additives under tribo- logical and static conditions. Thin Solid Films 2010;518:4336–44. [13] 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(7):1–7 (online). [14] Horcas I, Fernandez R, Gomez-Rodriguez JM, Colchero J, Gomez-Herrero J, ents for (a) DLC/DLC and (b) DLC/steel contacts in ZDDP solution. Baro AM. WSxM: a software for scanning probe microscopy and a tool for nanotechnology. Review of Scientific Instruments 2007;78(013705): 1–8 (online). [15] Spikes HA, Cann PM. The development and application of the spacer layer imaging method for measuring lubricant film thickness. Proceedings of the Institution of Mechanical Engineers 2001;215(Part J):261–77. [16] Juliane Benedet Jonathan H, Green Gordon D, Lamb Hugh A. Spikes, spurious mild wear measurement using white light interference microscopy in the presence of antiwear films. Tribology Transactions 2009;52:841–6. [17] Waesche Rolf, Hartelt Manfred, Weihnacht Volker. Influence of counterbody material on wear of ta-C coatings under fretting conditions at elevated temperatures. Wear 2009;267:2208–15. [18] Sanchez-Lopez JC, Erdemir A, Donnet C, Rojas TC. Friction-induced structural transformations of diamond-like carbon coatings under various atmospheres. Surface & Coatings Technology 2003;163–164:444–50. Tribological properties of tribofilms formed from ZDDP in DLC/DLC and DLC/steel contacts Introduction Materials and experimental details Results DLC/DLC tribopairs DLC/steel tribopairs Influence of steel counterpart (comparison of DLC/DLC and DLC/steel) Influence of ZDDP in the presence of steel counterpart (comparison of DLC/steel in base oil and ZDDP solution) Discussion Conclusions Acknowledgements References