Science

Long-term stability of hydrogenated DLC coatings: Effects of aging on the structural, chemical and mechanical properties

Description
Long-term stability of hydrogenated DLC coatings: Effects of aging on the structural, chemical and mechanical properties
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Long-termstabilityofhydrogenatedDLCcoatings:Effectsofagingonthestructural, chemical and mechanical properties M. Cloutier a,b , C. Harnagea c , P. Hale c , O. Seddiki c , F. Rosei c,d , D. Mantovani a,b, ⁎ a Laboratory for Biomaterials and Bioengineering, CRC-I, Dept Min-Met-Materials Engineering, Laval University, Québec, QC, Canada b Laboratory for Biomaterials and Bioengineering, CRC-I, CHU de Québec Research Center>, Québec, QC, Canada c Centre Énergie, Matériaux et Télécommunications, Institut National de la Recherche Scienti  fi que, Université du Québec, Varennes, QC J3X 1S2, Canada d Center for Self-Assembled Chemical Structures, McGill University, H3A 2K6 Montreal, QC, Canada a b s t r a c ta r t i c l e i n f o  Article history: Received 8 January 2014Received in revised form 14 June 2014Accepted 7 July 2014Available online 14 July 2014 Keywords: Diamond-like carbona-C:HPlasma enhanced CVDAgingLong-term stabilitySurface characterizationMicrostructureMechanical propertiesCoatings Long-term stability is an essential condition for the commercial use of protective coatings, yet often remainsoverlookedintheliterature.Herewereporttheeffectsoflong-termenvironmentalagingonthepropertiesofhy-drogenated diamond-like carbon (DLC)  fi lms. A range of DLC coatings produced by plasma-enhanced chemicalvapor deposition were  fi rst thoroughly characterized and then stored for three years before the second set of analysis.Ramanspectroscopyshowedthatthe fi lmsexhibitedexcellentstructuralstabilityduringaging,observ-ing no sign of sp 3 to sp 2 conversion. Similarly, the hardness and smoothness of the DLC coatings remained un-changed, despite the observed relaxation of their intrinsic stress with time. However, X-ray photoelectronspectroscopy analyses provided evidence of aging-induced surface oxidation, which was con fi rmed by reducedhydrophobicity (water contact angle dropped to 65°). Overall, these  fi ndings suggest that DLC possesses a suit-able long-term stability when exposed to environmental conditions.© 2014 Elsevier B.V. All rights reserved. 1. Introduction The high stability of amorphous diamond-like carbon (DLC)  fi lmshas been actively promoted in recent years and has encouraged its useinvariousapplicationsasadurableprotectivecoating,rangingfrombio-active biomedical implants [1 – 4] to magnetic data storage disks [5,6]. This interest stems from the unique set of properties of DLC, which in-cludes high hardness and elastic modulus, excellent wear resistance,lowfrictioncoef  fi cient,highchemicalinertnessandatomicsmoothness[6 – 9]. Like diamond, several of their bene fi cial properties are conferredbythe strong σ  bonds of sp 3 hybridizedcarbon atoms. However, unlikediamond, they are achieved in an isotropic, amorphous thin  fi lm con-taining both sp 2 and sp 3 bonded carbon in nanometer-sized clustersand, in some cases, hydrogen [6]. This particular microstructure arises from the bombardment of energetic species duringgrowth via a physi-cal process termed subplantation [10 – 12]. Penetration of carbon orhydrocarbon ions creates local quenched-in increases in density,which causes the local bonding to convert to sp 3 , but also generateshighcompressivestressesinthe fi lm.Thelatterphenomenoncanresultin delamination, especially on non-optimized interfaces, and generallylimits the thickness of deposited  fi lms, as stresses can act as a drivingforce for debonding from the substrate [13].Although theimportance of these intrinsic stresses in the formationof sp 3 bonds is still controversial (McKenzie et al. [14,15] vs Robertsonet al. [6,16]), they remain a de fi ning characteristic of DLC coatings,which brings forth questions about their long-term stability. The ther-mal stability of DLC coatings has been extensively studied [17 – 20], yetvery little work has been reported on the effects of long-term environ-mental aging (i.e. stability in environments that are not deliberatelyharsh), either on stress-related degradation or on the stability of theaforementionedproperties.However,thisisofparticularinterestforap-plications such as protective and barrier coatings where DLC  fi lms areexpected to maintain their protective behavior for several years[21 – 23]. Furthermore, it has been extensively reported that plasma-deposited coatings and other thin  fi lms materials are particularlyprone to progressive surface alterations induced by environmentalaging [24 – 27]. The study of the long-term stability and the occurrence Diamond & Related Materials 48 (2014) 65 – 72 ⁎  Correspondingauthorat:LaboratoryforBiomaterialsandBioengineering,CRC-I,DeptMin-Met-Materials Eng, University Hospital Research Center, Laval University, PLT1745,Québec, QC G1A 0V6, Canada. Tel.: +1 418 656 2131x6270. E-mail addresses:  maxime.cloutier.2@ulaval.ca (M. Cloutier), harnagea@emt.inrs.ca(C. Harnagea), rosei@emt.inrs.ca (F. Rosei), diego.mantovani@gmn.ulaval.ca (D. Mantovani). URL's:URL:E-mail addresses:E-mail address:  http://www.lbb.ulaval.ca (M. Cloutier),http://www.lbb.ulaval.ca (D. Mantovani).http://dx.doi.org/10.1016/j.diamond.2014.07.0020925-9635/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Diamond & Related Materials  journal homepage: www.elsevier.com/locate/diamond  of time-dependent degradation in DLC coatings is therefore necessarybefore they can be used in practical applications. The importance of in-vestigatingthelong-termstabilityofhardcoatingsatroomtemperatureforperiodsofuptothreeyearshasalreadybeenhighlightedinanexcel-lent review by Veprek et al. [28].The aim of our work was to study the effects of long-term aging onthe stability of the structure and properties of hydrogenated DLC  fi lms,to assess their use as  robust   protective coatings. The few studies pub-lishedonthissubjectuntilnowusuallytrackedtheevolutionofasingleproperty and studied one type of   fi lm only. On the other hand, this in-vestigationpresentsacomprehensiveanalysisofthechemical,mechan-ical,morphologicalandmicrostructuralpropertiesof  fi lmswitharangeof different microstructures. The properties of DLC coatings were mea-sured after deposition, and again after three years of environmentalaginginambientairatroomtemperature.Thechoiceofthistimeperiodwas based on several considerations, including theexpected lifetime of DLC coatings (from several months to several years depending on theapplication) and the reported timespan of time-dependent surfaceand structural changes in plasma deposited carbon coatings followingdeposition [28 – 31].DLC coatings were produced at different deposition powers byplasma-enhanced chemical vapor deposition (PECVD), to obtain  fi lmswith different microstructures. While the term DLC encompasses arange of materials with different characteristics, from hydrogen-freetetrahedral amorphous carbon to their hydrogenated counterparts, itwillbeusedinthistextinamorespeci fi csensetorefertohydrogenatedamorphous carbon (a-C:H)  fi lms. The present investigation includesthorough characterization of the specimens by Raman Spectroscopy,X-Ray Photoelectron Spectroscopy (XPS), water contact angle, AtomicForceMicroscopy(AFM),opticalmicroscopy,ScanningElectronMicros-copy (SEM), pro fi lometry and nanoindentation. 2. Materials & methods  2.1. Sample preparation and storage The a-C:H  fi lms were deposited on 100-oriented single crystal Sili-con substrates via a commercial RF (13.56 MHz) capacitively-coupledPECVD system (Plasmalab system 100, Oxford Instruments, Abingdon,UK).Substrateswerecutintopiecesof1×1cm 2 and1×10cm 2 ,loadedin the deposition chamber, which was then pumped down to a basepressure below 1 × 10 − 4 Pa. Before deposition, the Si substrates wereetched with an Ar plasma(50sccmAr, 40 Pa,150 W) for 20 min.Depo-sitionwasperformedusingapuremethaneplasma(50sccmCH 4 ,30Pa)at room temperature. Films between 100 and 250 nm were depositedby controlling and adjusting the deposition time, while preventing ex-cessiveheatingofthesubstratebyusingawatercoolingsystem.Differ-ent RF powers (50 W, 100 W, 150 W, 200 W, 250 W and 300 W) wereused to vary the ion energy during the process and obtain a variety of microstructures. Specimens were  fi rst characterized post-depositionand then stored in tissue-culture polystyrene dishes at room tempera-ture (T = 23 ± 1 °C), in contact with air and exposed to ambient hu-midity. They were kept in storage for three years before the second setof analyses.  2.2. Characterization The structural properties of the DLC  fi lms were analyzed by Ramanspectroscopy (Renishaw inVia spectrometer, Renishaw, Wootton-under-Edge,UK)usinganArlaserat488nmoperatedat20mW.Spectrawere collected with a 5× objective using a 1800 g/mm grating. Eachsamplewasscannedatthreedifferentpositionstomonitorthehomoge-neityofthecoating.Fortheanalysis,bothcharacteristicpeaks(GandD)were  fi tted with Gaussian functions, following the procedureestablished by Casiraghi et al. [32]. The photoluminescence (PL) back- ground was taken between 1050 cm − 1 and 1800 cm − 1 .ThesurfacecompositionwasinvestigatedusinganX-rayPhotoelec-tron Spectrometer (XPS  –  PHI5600-ci spectrometer, Physical EletronicsUSA, Chanhassen, MN, USA). Survey and high resolution spectra wereacquired at a detection angle of 45° using the K α  line of a standard alu-minum and magnesium X-ray source, respectively, operated at 300 W.The curve fi ttingsfor thehighresolutionC(1 s) peaks weredeterminedby means of the least-squares method using Gauss – Lorentz functionswith a Shirley background subtraction. Contact angles of the sampleswere measured by the sessile drop method with the VCA optima XE(AST Products, Billerica, MA, USA) using distilled water.Surface imaging was performed using an EnviroScope AFM (Veeco,Woodbury, NY, USA) operating in tapping mode and equipped with aT300R silicon tip (VISTAProbes, Nanoscience Instruments Inc., Phoenix,AZ, USA). The roughness was measured over 1 × 1  μ  m 2 areas. Sampleswere also studied by optical microscopy, with a BX41M microscope(Olympus, Center Valley, PA, USA) and scanning electron microscopy(SEM), using a JEOL 7500-F scanning electron microscope (JEOL Ltd.,Tokyo,Japan).Forthesemeasurements,priortoimaging,theagedspec-imenswerethoroughlycleanedwithdeionizedwaterandsubsequentlydriedwithnitrogentominimizethepresenceofdustandotherparticleson the surface.The fi lmandsubstratecurvaturesweremeasuredwithamicrostyluspro fi lometer(Dektak150,Veeco)andtheinternalstressesofthedepos-itedcoatingswerecalculatedusingStoney'sequation[33].Thethickness was considered constant for aged samples. Hardness measurementsweremadeusinganAFM(Dimension3100,Veeco)equippedwithacor-ner cube diamond indenting tip (PDNISP model, Bruker, Billerica, MA,USA) working in nanoindendation mode [34]. The maximum indenta- tion load used was 450  μ  N. Indented zones were imaged using thesame setup, so as to directly measure the projected contact area.The friction coef  fi cient was estimated at the nanoscale by scanningthe AFM tip in contact mode, at a fast scan direction perpendicular tothe cantilever axis (scan angle 90°), such that the friction force inducesthetorsionaldeformationsofthecantilever[35,36].Thespeedofthetip was set at 2  μ  m/s, and the normal force was increased up to 4  μ  N. Weused commercial doped diamond-coated silicon tips (DDESP-FM-10model, Veeco) with a tip radius of approximately 250 nm. These tipswere chosen because they are presumed to preserve their propertiesoverlongperiodsofsustainedscanningincontactmode.Thecalibrationof the cantilever was performed by measuring its resonance frequency,dimensions, and using the known relationships for the calculation of normal and torsional spring constants as described in [35]. 3. Results and discussion  3.1. Structural properties The Raman spectra of as-deposited samples (Fig. 1a) showed thecharacteristic fi ngerprints ofhydrogenatedDLC fi lms:theGpeak(orig-inating in the stretching vibration of pairs of sp 2 sites), the D peak(breathing mode of sp 2 sites within six-fold rings) and an increasingPL background [6,37,38]. Fitting and analysis of both peaks providedvaluable information on the evolution of the bonding structure andthe occurrence of structural changes with deposition and aging condi-tions. The parameters of interest are the position and full width at half maximum oftheGpeak,Pos(G) andFWHM(G) respectively, theinten-sityratioofthepeaksI(D)/I(G)andthehydrogencontentdeducedfromtheintensityofthePLbackground.Theseparametersareplottedagainstdeposition power in Fig. 1, for as-deposited and aged specimens. Ac-cording to the three stage model introduced by Ferrari and Robertson[37], the Raman spectrum is primarily in fl uenced by the clustering of thesp 2 phase.However,Pos(G)andI(D)/I(G)canofferanindirecteval-uation of the sp 3 bondingin the fi lm,since thepresence of hydrogen ina-C:H fi lmslinkstheamountandcon fi gurationofthesp 2 phasewiththeoverall sp 3 content (C \ C & C \ H sp 3 ) [6,32,37]. Similarly, for hydroge-nated DLC  fi lms, the FWHM(G) is in fl uenced by structural disorder, as 66  M. Cloutier et al. / Diamond & Related Materials 48 (2014) 65 – 72  itismainlyaprobeofbondangleandlengthdistortion.BothpropertiesultimatelydependontheamountofC \ Csp 3 bondsandFWHM(G)canhencebeusedto derivethestructuralandmechanicalpropertiesof the fi lms [32]. Lastly, the hydrogen content can be empirically determinedbyusingtheratiobetweentheslopemofthePLbackgroundandthein-tensity of the G peak, m/I(G) [32].Foras-deposited fi lms,Pos(G)andI(D)/I(G)exhibitaroughlylinearincrease with deposition power (Fig. 1b and c), indicating, as expected,an evolution of the structural order of the fi lms with ion energy. It alsosuggeststhat fi lms depositedatlower power/bias energyhave a highersp 3 content since they have lower Pos(G) and I(D)/I(G) values [37]. Ontheotherhand,theFWHM(G) of as-deposited fi lmsexhibits a differenttrend, with a local maximum at 100 W surrounded by decreasingFWHM(G) values for other deposition powers (Fig. 1d). This differenceresults from the fact that both Pos(G) and I(D)/I(G) probe the overallsp 3 content (C \ C + C \ H sp 3 ), while FWHM(G) is only in fl uenced bytheC \ C sp 3 fraction.Hence, even if they do have a lower C \ C sp 3 con-tent, the higher sp 3 fraction for 50 W  fi lms calculated with Pos(G) andI(D)/I(G)couldbeindeedduetothehigherhydrogencontentmeasuredin such  fi lms (see Fig. 1e). The behavior observed in Fig. 1d with the Fig. 1.  Single-wavelength Raman spectroscopy ( λ  = 488 nm) analysis of DLC samples. a) Typical Raman spectra showing both deconvoluted peaks and the  fi tted linear background,b) Pos(G), c) I(D)/I(G) ratio, d) FWHM(G) and e) hydrogen content of as-deposited (squares) and aged (triangles) DLC  fi lms as a function of deposition power.67 M. Cloutier et al. / Diamond & Related Materials 48 (2014) 65 – 72  presence of an optimum ion energy value for maximizing the C \ C sp 3 content was also observed by several other authors for a-C and a-C:Hcoatings[39 – 43].Theobservedincreaseinhydrogencontentatlowde-positionpowerisconsistentwiththeliterature [38,44] andcouldbeat-tributed to incomplete methane dissociation at lower power, whichmay cause an increased incorporation of H-bonded carbon atoms intothe  fi lm. Conversely, it was also suggested that this behavior could beassociated with the increased release of H in the collision cascades in-duced by higher energy atoms [45].After aging, Pos(G), I(D)/I(G) and FWHM(G) all show the sametrends as their as-deposited counterparts, with highly similar values(Fig. 1b, c, d). This suggests that there was nosigni fi cant phase change,conversionofsp 3 intosp 2 carbonorothermodi fi cationinthestructuralbonding of hydrogenated DLC  fi lms during aging, regardless of thedeposition power. An increase, between 18 and 27% of the hydrogenconcentrationwasnotedforallsamples(Fig.1e).Filmsofvariousthick-nesses were subsequently analyzed by Raman spectroscopy to assesstheoriginofthe%Hincrease.However,nocorrelationwasobservedbe-tween thickness and increase in hydrogen content, suggesting that the%H increase was due to surface adsorbed water rather than bulk diffu-sion of hydrogen or other bulk structural modi fi cations. This wouldalsoimplythatthehydrogenated fi lmsdidnotundergoagraphitizationprocess at room temperature, despite the long exposure time [46].  3.2. Surface chemistry ThesurfaceofDLC fi lmswasinvestigatedbyXPS,soastocharacter-ize the as-deposited surface chemistry and the changes induced byaging. XPS survey analyses of as-deposited and aged specimens re-vealed the presence of both carbon (284.6 eV) and oxygen (532.5 eV)(Fig. 2). Plasma-deposited carbonaceous materials are known topromptly oxidize upon air exposure, through the formation of surfaceoxygen functional groups [47,48].The in fl uence of depositionpower andagingonthesurface chemis-try of the  fi lm was further investigated with high resolution XPS. Anexample of the C1s carbon region is depicted in Fig. 3a, showing thecharacteristic asymmetric tailing due to the contribution of surfaceoxygen groups. The C1s high resolution spectrum of carbon wasdecomposed into four components, assigned to C \ C and C \ H (BE =284.6 eV), C \ O (BE = 286.4 eV), C _ O (BE = 287.8 eV) and O \ C _ O(BE = 289.1 eV) [49 – 51]. Fig. 3b compares the proportion of oxygen components of the C(1s) peak before and after aging at different depo-sition powers. For as-deposited  fi lms, results show an increasing pro-portion of O groups with deposition power. A similar behavior can beobserved on aged  fi lms. Although it is not possible with the currentdata to pinpoint the exact mechanism involved in this phenomenon,this increase could be explained by the higher density of danglingbonds created at higher deposition power, caused by ions impactingthe surface with more energy, thus enhancing the surface's reactivityto the highly reactive oxygen molecules [52]. Fig. 3b also shows a clear increase in the proportion of all O groups in the C1s region due toaging. The trend was con fi rmed with similar increases of the oxygencontent measured from the survey spectra (Fig. 2). This behavior isnot due to physisorbed molecules, since increases in pumping timedid not induceanychange ontheresultingspectra.Whilechemisorbedmolecules may partially explain the oxygen increase due to aging, theyareunlikelytoaccountforthewholevariation(between70%and140%).For amorphous carbon, surface oxidation has rather been associatedwithUV-inducedbreakingofC \ CandC \ Hbondsandsubsequentfor-mation of C \ O bonds [9] and hydrolysis of carbon – carbon bonds [53].Besides altering the surface composition of DLC, aging also affectedthe wetting behavior of the  fi lms. The measured increase in oxygenwas associated with an increase in surface energy, as revealed by con-tactanglemeasurements(Fig.4).Weobservedadropofthewatercon-tact angle from 81 ± 3°for as-deposited samples to65± 3° after agingandweassociatedthiswiththeslow,albeitsigni fi cant,formationofhy-drophilicfunctionalgroupssuchascarbonyl(O _ C),carboxyl(O – C _ O)and ether (O \ C) and the consequent loss of hydrophobic hydrocarbon(CH 3  and CH 2 ) groups at the surface of DLC  fi lms.Thischangeinsurfacechemistrycouldimpactthelong-termperfor-manceofDLCcoatingsinseveral fi elds,butespeciallyinbiomedicalandtribological applications. For instance, for biomedical coatings, the ob-serveddecreaseinchemicalinertnessmayresultinadditional,undesir-able interactions with surrounding cells and bodily  fl uids.  3.3. Surface morphology Both as-deposited and aged  fi lms revealed the typical nanoscalesmooth topography of DLC coatings (see Fig. 5) [7]. The measured R  RMS for as-deposited  fi lms was 0.11 ± 0.04 and 0.20 ± 0.03, for 100 – 300 Wand 50 W deposition power, respectively. This rougher topography ob-tained at lower deposition power is due to the inability of incomingions to penetrate the surface, leading to the formation of orderedsp 2 -richsurfaceclusters whichincreasestheroughness,asdescribedbyPengetal.[54]Thesamepatternwasobservedafteraging,withR  RMS valuesof0.10±0.02nmfor fi lmsdepositedbetween100Wand300Wandof0.17±0.02nmfor fi lmsdepositedat50W.Thistopologicalsta-bility suggests the absence of any stress/strain-induced rougheningmechanisms during the aging process.OpticalmicroscopyandSEMwerealsousedtoassessthepresenceof defectsatthesurface.DLCdepositedoninadequatesurfacescanunder-go stress-relieving processes that lead to buckling, cracking or evencomplete delamination [13,21,55]. In our case, the analyses revealedtheabsenceof suchfeatures;nostress-related surface defects were ob-servedatthemacro-,micro-andnano-scalelevelsforallsamplesbeforeand after aging. The aged  fi lms were very similar to as-deposited coat-ings, indicating that the coating adhesion and cohesion were suf  fi cientto resist the high compressive stresses of DLC for extended periods of time.  3.4. Mechanical properties Fig. 6 shows the measured stress of as-deposited and aged speci-mens as a function of deposition power. For as-deposited  fi lms, thestress curve is typical of DLC  fi lms. [11,56]. The initial increase in com-pressive stress with increasing deposition power is the result of the in-tensi fi cationofimplantationasthebombardingcarbonionshavehigherenergy. However, the stress reaches a maximum at 100 W RF power,whichisexplainedbythecreationoflocalthermalspikesuponimplan-tation with high energy ions. These local increases in temperature Fig. 2.  Representative XPS survey spectra of DLC  fi lms (deposition power 200 W) beforeand after aging. Only carbon (C1s) and oxygen (O1s) were detected on both samples.68  M. Cloutier et al. / Diamond & Related Materials 48 (2014) 65 – 72  promote a thermally activated relaxation of the bonding into sp 2 [12],resulting in the observed decrease in stress.Althoughbothas-depositedandagedsamplesexhibitahighlevelof stress in the GPa range, we observed a decrease of the compressivestress levels for all aged samples, with a more pronounced reductionfor specimens deposited at lower power. A similar time-induced relax-ation of residual stresses was observed by Bull and Hainsworth [29] inDLCcoatingsdepositedbyion-beam-assisteddeposition(IBAD).Theor-iginofthisrelaxationis,however,problematictopinpoint.Itmaynotbeattributable to a signi fi cant sp 3 to sp 2 conversion [57], since such atransformationwasnotdetectedintheRamananalysis(Fig.1b – d).An-other possibility would be a time and strain-induced roughening, thatwould allow a partial relaxation through the formation of grooves andpits at the surface [58]. Other common stress relief mechanisms ob-served at the surface of amorphous carbon  fi lms with high intrinsicstressesarebucklingandcracking[13].However,asrevealedpreviouslyby AFM and optical microscopy, the surface morphology did not showany signature of such stress relief mechanisms .Bondlengthandangledistortions(i.e.deviationsfromtheirequilib-riumvalue)aretheprimarycontributionstothehighresidualcompres-sivestressofDLC fi lms[15,59,60].Accordingly,Ferrarietal.pointedoutthat the stress could be relieved by a tiny strain relaxation [16]. Due tothehighYoung'smodulusoftheDLCcoatings(125 – 200 GPa),thestrainrelaxation necessary to produce a large stress reduction would be min-imal and thus not necessarily perceptible with Raman analysis. As a re-sult, the observed  fi lm relaxation could be due to the release of bonds'length and orientation distortions created during the deposition pro-cess. A similarly slow relaxation process was also noted by Bouzeraret al. in plasma deposited hydrogenated DLC  fi lms [30]. In our case,the relaxation phenomenon appears more pronounced for specimensdeposited at lower power. We attribute this to the fact that these  fi lmshave not already undergone the thermal spike relaxation associatedwith implantation of high energy ions.ThehardnesswasmeasuredusinganAFMequippedwithadiamondindenter.TheresultsareshowninTable1.Whilethistechniqueisusefulfor investigatingmechanical properties of  fi lmsatthenanometer scale,nanoindendation is best suited for making relative hardness measure-mentsandtheresultsshouldnotbeusedtomakeabsolutecomparisonswith data from the literature. Both tested aging conditions showed thesame tendency, with a clear hardness maximum at 100 W, con fi rmingthe trend observed previously with the Raman and stress experimentsfor as-depositedsamples. Samples deposited at other powers exhibitedroughly similar hardness levels of ~11 GPa. After aging, there was nosigni fi cant decrease of the measured hardness as the observed differ-ences were within the experimental error. This con fi rms our previous fi nding that the  fi lms did not undergo a signi fi cant sp 3 conversion,since their C \ C sp 3 content, which governs the mechanical properties,appears to be unchanged after aging.  3.5. Tribological properties ToassesstheeffectofagingonthetribologicalpropertiesofDLC fi lms,nanoscale friction studies were performed using a diamond AFM tip incontact mode as the counterface(Fig. 7). The initial friction coef  fi cientwas found to be 0.24, which is consistent with other studies performedin normal atmosphere [61,62]. The measurements were repeated afteraging and we observed an increase of the friction coef  fi cient to 0.37.Theincreaseinthefrictioncoef  fi cientcanbeattributedtoseveralfac-tors.First,theobservedstressrelaxationthatweinterpretedasbeingduetothereleaseofbond'slengthmayalsoin fl uencethe fi lm'sYoungmod-ulus,makingitmorecompliant.Asaresult,therealcontactareabetweentipand fi lmwouldincrease,and,accordingtothetheoryofBowdenandTabor [63], lead to an increase of the (effective) friction coef  fi cient.A second effect on the friction coef  fi cient may come from the in-creaseinthesurfaceenergy.Itiswell-knownfromtheadhesivecontacttheory [63] that the surface energy between the two bodies in contact Fig. 3.  High resolution (HR) XPS analysis of the investigated  fi lms. a) HR XPS C(1s) spectrum of the DLC  fi lms with corresponding peak deconvolution (deposition power 150 W, post-aging) and b) Proportion of oxygen components of the C(1s) peak for the as-deposited and aged DLC  fi lms. Fig. 4.  Representative water contact angle measurements of a) as-deposited (81 ± 3°) and b) aged DLC coatings (65 ± 3°).69 M. Cloutier et al. / Diamond & Related Materials 48 (2014) 65 – 72
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x