Acta Polytechnica doi:10.14311/APP.2020.27.0062 Acta Polytechnica 27(0):62–66, 2020 © Czech Technical University in Prague, 2020 available online at http://ojs.cvut.cz/ojs/index.php/app HARDNESS OF HYBRID PVD-PECVD W-C:H COATINGS VS. SUBSTRATE TYPE František Lofaj∗, Margita Kabátová Slovak Academy of Sciences, Institute of Materials Research, Watsonova 47, 040 01 Košice, Slovakia ∗ corresponding author: flofaj@saske.sk Abstract. The effects of substrate on the measurement of hardness (and indentation modulus) - penetration depth profiles were investigated on hybrid PVD-PECVD W-C:H coatings made with the additions of C2H2 or CH4 using High Power Impulse Magnetron Sputtering (HiPIMS) method. The substrates involved 100Cr6 bearing steel, Al2024 aluminum alloy and (111) Si wafer. Although no clear influence of the substrate type on hardness and indentation modulus of the coatings at different acetylene and hydrogen additions was observed, when both measured properties were combined in HITEIT ratio, significant shift of HIT/EIT ratio dependencies obtained on Al alloy vs. steel and Si wafer substrates occurred. Thus, the effect of substrate type on the measurementlevaluation was confirmed. Keywords: Hardness, indentation modulus, substrate effect, W-C:H coating. 1. Introduction Tribological properties of hard coatings, especially their wear resistance, coefficient of friction (COF) de- pend on large number of parameters including loading and environmental conditions but also material system used for the coating, its structure, surface roughness and such properties as hardness, elastic modulus, and others. It was reported that the wear properties of hard coatings more strongly correlate with H/E ratio than just with the hardness itself [1]. Moreover, it was also found out that the H/E and/or H3/E2 ratios are related to the level of plasticity and can be used to estimate changes in the toughness of the coatings [2–4]. This is of significant interest because tough- ness is among the main parameters, besides hardness, Young’s modulus, COF, wear, scratch, oxidation and corrosion resistance of the coatings for engineering ap- plications. The measurement of mechanical properties, involving hardness, elastic modulus and toughness in the coatings with the thickness in the range of few micrometers is a challenge even by nanoindentation because of the inevitable influence of the substrate on the measurement of the coating properties. To reduce this influence, the indentations depths used for the determination of coating hardness should be substantially smaller than the thickness of the coating. Physical explanation is that the stress field under the indenter, corresponding to the zone with the stresses above the yield stress (zone of plasticity), has to be fully confined in the coating. So called “10 % (of the coating thickness) rule” is often applied to de- termine such maximum depth and true hardness of the coating, respectively [5–10]. Another approach to eliminate substrate influence and to determine true coatings properties is to use hardness (and elastic modulus) – indentation depth profiles obtained from continuous multi-cycle (CMC) or continuous stiffness measurements (CSM) [11]. The coating properties correspond to the maximum or plateau on the corre- sponding depth profile at the depths smaller than 10 % of the coating thickness. However, the liability of 10 % rule is limited by the accuracy of the measurement at coating thicknesses below 1 µm related to tip surface area calibration function, by indenter tip radius [12], surface roughness, pile-up or sink-in behavior [13], residual stresses as well as by the ratio between the yield stresses and the elastic moduli of the coating and the substrate, respectively [9, 14]. The measurements of indentation modulus are much more sensitive to substrate properties than hardness because the elastic stress field under the indenter would be much larger than the zone of plasticity and therefore, 10 % rule may not be sufficient. Thus, the problem of proper measurement of hardness simultaneously with the elas- tic modulus in thin hard coatings persists and it is even emphasized by the applicability of the H/E ra- tio for toughness and wear resistance estimations. In our earlier works, the influence of indenter tip radius on the hardness – depth profiles has been analyzed using a combination of the experimental and finite element modelling approaches [12]. However, the in- fluence of the Young’s modulus of the substrate on the nanoindentation measurements was not investi- gated. Therefore, the aim of the present work is to investigate the influence of the substrate properties on the measurements of hardness, and indentation modulus in thin hard coatings used for engineering applications. The measurements were performed on HiPIMS (High Power Impulse Magnetron Sputter- ing) W-C and W-C:H coatings deposited using hybrid PVD-PECD (Physical Vapor Deposition – Plasma Enhanced Chemical Vapor Deposition) process with different acetylene and hydrogen additions on three different substrates. 62 http://dx.doi.org/10.14311/APP.2020.27.0062 http://ojs.cvut.cz/ojs/index.php/app vol. 27/2020 Coatings vs. substrate type hardness Figure 1. Load – indentation depth curves from the nanoindentation tests on the HiPIMS W-C:H coatings deposited with 2 sccm C2H2 on Si wafer, 100Cr6 bearing steel and Al 2024 alloy substrates – a) The depth profiles of hardness – b) and indentation modulus – c). 2. Experimental material and methods The studied W-C:H coatings were simultaneously de- posited on three different substrates: polished tem- pered bearing steel 100Cr6 discs (diameter – 25mm, thickness – 3mm ), polished Al2024 aluminum alloy discs with similar size and on approximately 1cm2 po- lished (111) Si wafer fragments with ∼ 200nm Cr bond layer. The deposition was carried out at a minimum working pressure of 0.5 Pa in a magnetron deposition system (Cryofox Discovery 500, Polyteknik, Denmark) using High Power Impulse Magnetron Sputtering (Hip- IMS) source at 350 W average power achieved at 2.62% duty cycle related to the frequency of 150Hz and impulse length of 175 µs. A set of 14 coatings with variable additions of C2H2 precursor (0, 2, 4, 6, 8 sccm) and (0, 5, 10, 20 sccm) into Ar atmosphere was prepared for the study. The thicknesses of the coatings were in the range from 0.7 µm up to almost 3 µm depending on the amount of precursor gas addi- tions [15–18]. The nanoindentation tests were carried out on a nanoindenter (model G200, Agilent, USA) using a diamond Berkovich tip in CSM mode with the preset maximum penetration depth of 1000 nm, strain rate of 0.05 s−1, frequency of 45 Hz and amplitude of 2 nm on a set of 16 indents. The hardness and indentation modulus depth profiles were obtained by averaging the results of at least 10 measurements. The Young’s modulus and Poisson’s ratio of the diamond tip used for calculations were Etip = 1141 GPa and νtip = 0.07 , respectively. The indenter tip was calibrated on a fused silica reference sample. The Poisson’s ratios of the coatings of v = 0.25 was used for the evaluation of the corresponding properties. 3. Results and discussion 3.1. Substrate effects on hardness and indentation modulus in HiPIMS W-C:H coatings Fig. 1a - 1c illustrate the differences between load - indentation depth curves, hardness – indentation depth and indentation modulus – depth profiles in the studied W-C:H coating deposited with the addition of 2 sccm C2H2, respectively. The load – displacement curve in the coating on Al alloy differs significantly from those on Si and steel substrates, which were prac- tically identical, at the displacements above 400 nm 63 František Lofaj, Margita Kabátová Acta Polytechnica Figure 2. The comparison of the hardness, HIT − a) and indentation modulus, EIT − - b) in HiPIMS W-C:H coatings deposited at different acetylene (and hydrogen) additions on three types of substrates. The data-points corresponding to 5, 10 and 20 sccm H2 additions were intentionally shifted by 0.2, 0.4 and 0.4 sccm respectively, for better visibility. (Fig. 1a). However, this difference is not important for the determination of the hardness and indentation modulus, because they have to be determined from the indentation depths within the range of 100 - 300 nm due to the earlier mentioned 10 % rule. Fig. 1b shows the depth profile of hardness: on Si and steel sub- strates, very wide plateau at their maxima of around 20 GPa extending up to almost 500 nm depth can be seen. In contrary, the influence of substrate is much stronger in the case of measurement on Al substrate: the onset of hardness decrease occurred already at 250 nm. Such difference is a natural consequence of lower Young’s modulus of Al alloy in comparison with steel and Si wafer. However, the maximum of the hardness profile obtained on Al alloy overlapped with those on the other two substrates and essentially no difference of the substrate on hardness can be seen. However, the situation was different in the case of indentation modulus profiles. A plateau at around 225 GPa extending up to 500 nm can be seen on steel substrates but the measurement on Si wafer resulted in a maximum plateau at around 220 GPa only up to ∼ 300 nm. The effect of Al alloy substrate was much more pronounced: the maximum was around 175 GPa and it extended only to 150 nm indenta- tion depth. The values of HIT and EIT determined from the corresponding depth profiles according to the above mentioned way are summarized in Fig. 2a and 2b, respectively. They reveal number of effects and differences. They involve the effects of substrate type but also of acetylene and hydrogen additions during hybrid PVD-PECVD. The PVD process without additional precursor gases resulted in (possibly over-stoichiometric [18]) WC coatings exhibiting much lower values of hardness (22 GPa) and modulus (< 220 GPa) in the coatings de- posited on Al alloy substrates than on Si and steel sub- strates (26 – 29 GPa and 280 – 340 GPa, respectively). It is known from earlier works on DC magnetron sput- tered and HiPIMS W-C:H coatings [17, 18] that the structure of the PVD coatings without additions of acetylene may vary from nano-columnar to nanocrys- talline, which resulted in the variations of hardness from 34 GPa to 28 GPa, respectively. The additions of acetylene caused gradual increase of the content of hydrogenated free carbon phase and a transition from nanocomposite to amorphous structure accompanied by a decrease of hardness [16–18]. The substrate itself may also play significant role in the formation of the coating structure and its morphology [15]. Without detail structure observations, it is difficult to decide, which of the mentioned effects could be responsible for the changes of hardness and indentation modulus in the studied W-C coatings. As expected, the additions of acetylene resulted in gradual decrease of these prop- erties. There were certain variations among the values obtained on various substrates, but they were within the scatter bars of the measurements and the differ- ences decrease, especially in the case of elastic moduli (Fig. 2b). Apparently, the amorphization of the struc- ture at high acetylene additions seems to result in the unification of the properties and has stronger effect than the possible influence of the substrate. Thus, the differences in coating properties deposited on different substrates seem to be statistically significant only in the cases without or with very small acetylene addi- tions. However, they may interfere with the effects of coating structure and therefore, cannot be fully attributed to the substrate effect. Figs. 2 also show the influence of hydrogen additions; it should be noted that the data-points corresponding to 5, 10 and 20 sccm H2 additions were intentionally shifted by 0.2, 0.4 and 0.4 sccm from the position of the corresponding acetylene addition to show the tendency of the changes without overlap. It can be seen that the hydrogen additions have different effects on HIT and EIT in PVD W-C coatings and in hybrid PVD-PECVD W-C:H coatings made with acetylene additions. In the first case, hydrogen additions re- sulted in the increase of hardness and elastic modulus 64 vol. 27/2020 Coatings vs. substrate type hardness whereas they were decreased in the coatings made with 4 sccm and 6 sccm acetylene additions regardless of the substrate. Moreover, the corresponding values principally followed the dependencies described for the coatings without hydrogen additions and agree with the above conclusions concerning exclusivity of W-C coatings compared to W-C:H coatings. The above results indicate only a small influence of substrate type on hardness and indentation modu- lus. However, they are measured simultaneously and it is reasonable to check the simultaneous effect of both properties on different substrates. Thus, the results from Fig. 2a and 2b were replotted as a func- tion of HIT/EIT ratio on acetylene (and hydrogen) additions in Fig. 3. The obtained plots show that the acetylene and hydrogen additions usually result in gradual HIT/EIT increase and that the difference between substrates was strongly magnified. Although the differences between HIT/EIT in the coatings on 100Cr6 steel and Si substrates were negligible in the studied additions range, the corresponding values on Al alloy substrates followed the same tendency but they were by 0.02 – 0.025 higher. Since both hardness and indentation modulus decrease with the acetylene addition increase, the increase of HIT/EIT has to result from faster decrease of indentation moduli. Another consequence of the shift of the dependencies between steel, Si and Al alloy substrates was that all HIT/EIT values obtained on Al alloy were above the plastic- ity limit of 0.1 while the identical coatings deposited on Si and steel substrates at 0 and 2 sccm acetylene (and hydrogen) additions would be below that limit. Obviously, such systematic differences in HIT/EIT on the same coatings do not have real physical reason. They have to be related to the influence of the sub- strate type on the measurement method and/or data evaluation rather than on true hardness and inden- tation modulus changes. It is a clear indication that the substrate type definitely affects the nanoindenta- tion measurements. The question is what is the main parameter controlling the influence of the substrate. Young’s moduli of steel and Si substrates are rather similar (210 GPa vs. 193 GPa) while that of Al alloy is considerably lower (∼ 73 GPa) and this difference is reflected in the difference obtained Fig. 3. However, the yield strength and hardness of substrate materials should play a role because they control plasticity of the substrate 4. Conclusions The investigations of the influence of the substrate type on the measurements of hardness and indentation modulus in HiPIMS W-C and W-C:H coatings with different acetylene and hydrogen additions on Si wafer, 100Cr6 steel and Al 2024 aluminum alloy substrates revealed: • Significant influence of substrate type on HIT and EIT in PVD W-C coatings deposited without the 0 2 4 6 8 0.075 0.100 0.125 0.150 HiPIMS W-C:Hcoatings 5,10,20 H 2 5,10,20 H 2 H IT /E IT ra tio xC 2 H 2 (+ yH 2 ) flow, sccm Alsubstrate Sisubstrate Fe substrate 5,10,20 H2 Al 2024 substrate Sisubstrate 100Cr6 substrate Figure 3. The dependence of HIT /EIT ratio in HiP- IMS W-C:H coatings on the substrate type at different acetylene (and hydrogen) additions. Note that the data-points corresponding to 5, 10 and 20 sccm H2 additions to the corresponding acetylene flows were intentionally shifted to emphasize the role of hydrogen additions. additions of acetylene. However, these differences may overlap with the influence of substrates on the coating structure. • An enhanced scatter of the measured properties in hybrid PVD-PECVD W-C:H coatings with the additions of acetylene on the studied substrates. The differences between the substrates were within the scatter of measurement. • When the effects of substrate type on hardness and indentation modulus were combined in HIT/EIT ratio, significant shift between the dependencies ob- tained on Al alloy vs. steel and Si wafer substrates was obtained. Thus, the effect of substrate type on the measurement/evaluation was confirmed. • The effect of substrate properties on nanoindenta- tion measurements cannot be related only to the difference in Young’s moduli of the substrates and further investigations are required to find the way to eliminate this influence. The above conclusions imply that the reliable eval- uation of the wear resistance and fracture toughness of thin coatings based on HIT/EIT ratio requires the elimination of the influence of the substrate type on HIT/EIT ratio. Acknowledgements The support provided by the projects APVV-17-0320, APVV 17-0059, APVV-17-0049 and VEGA 2/0017/19 is acknowledged. The equipment used in the work was acquired from the projects “Research Centre of Advanced Materials and Technologies for Recent and Future Appli- cations” PROMATECH, ITMS: 26220220186. 65 František Lofaj, Margita Kabátová Acta Polytechnica References [1] A. Leyland, A. Matthews. On the significance of the h/e ratio in wear control: a nanocomposite coating approach to optimised tribological behaviour. Wear 246(1-2):1– 11, 2000. doi:10.1016/s0043-1648(00)00488-9. [2] J. Musil. Hard and superhard nanocomposite coatings. Surface and Coatings Technology 125(1-3):322–330, 2000. doi:10.1016/s0257-8972(99)00586-1. [3] J. Musil. Hard nanocomposite coatings: Thermal stability, oxidation resistance and toughness. Surface and Coatings Technology 207:50–65, 2012. doi:10.1016/j.surfcoat.2012.05.073. [4] J. Musil, M. Jirout. Toughness of hard nanostructured ceramic thin films. Surface and Coatings Technology 201(9-11):5148–5152, 2007. doi:10.1016/j.surfcoat.2006.07.020. [5] B. Jönsson, S. Hogmark. Hardness measurements of thin films. Thin Solid Films 114(3):257–269, 1984. doi:10.1016/0040-6090(84)90123-8. [6] A. Korsunsky, M. McGurk, S. Bull, T. Page. On the hardness of coated systems. Surface and Coatings Technology 99(1-2):171–183, 1998. doi:10.1016/s0257-8972(97)00522-7. [7] R. Saha, W. D. Nix. Effects of the substrate on the determination of thin film mechanical properties by nanoindentation. Acta Materialia 50(1):23–38, 2002. doi:10.1016/s1359-6454(01)00328-7. [8] R. Saha, W. Nix. Soft films on hard substrates — nanoindentation of tungsten films on sapphire substrates. Materials Science and Engineering: A 319- 321:898–901, 2001. doi:10.1016/s0921-5093(01)01076-0. [9] J. Chen, S. Bull. On the factors affecting the critical indenter penetration for measurement of coating hardness. Vacuum 83(6):911–920, 2009. doi:10.1016/j.vacuum.2008.11.007. [10] X. Cai, H. Bangert. Hardness measurements of thin films-determining the critical ratio of depth to thickness using FEM. Thin Solid Films 264(1):59–71, 1995. doi:10.1016/0040-6090(95)06569-5. [11] X. Li, B. Bhushan. A review of nanoindentation continuous stiffness measurement technique and its applications. Materials Characterization 48(1):11–36, 2002. doi:10.1016/s1044-5803(02)00192-4. [12] T. Csanádi, D. Németh, F. Lofaj. Mechanical properties of hard w-c coating on steel substrate deduced from nanoindentation and finite element modeling. Experimental Mechanics 57(7):1057–1069, 2016. doi:10.1007/s11340-016-0190-x. [13] A. Bolshakov, G. M. Pharr. Influences of pileup on the measurement of mechanical properties by load and depth sensing indentation techniques. Journal of Materials Research 13(4):1049–1058, 1998. doi:10.1557/jmr.1998.0146. [14] S. Kataria, S. Goyal, S. Dash, et al. Evaluation of nano-mechanical properties of hard coatings on a soft substrate. Thin Solid Films 522:297–303, 2012. doi:10.1016/j.tsf.2012.09.001. [15] F. Lofaj, P. Hviščová, P. Zubko, et al. Mechanical and tribological properties of the high target utilization sputtering w-c coatings on different substrates. International Journal of Refractory Metals and Hard Materials 80:305–314, 2019. doi:10.1016/j.ijrmhm.2016.12.015. [16] F. Lofaj, M. Kabátová, M. Klich, et al. Tribological behavior of hydrogenated w-c/a-c:h coatings deposited by three different sputtering techniques. Cerâmica 65(suppl 1):58–69, 2019. doi:10.1590/0366-6913201965s12577. [17] F. Lofaj, M. Kabátová, L. Kvetková, et al. Hybrid PVD-PECVD w-c:h coatings prepared by different sputtering techniques: The comparison of deposition processes, composition and properties. Surface and Coatings Technology 375:839–853, 2019. doi:10.1016/j.surfcoat.2019.07.078. [18] F. Lofaj, M. Kabátová, J. Dobrovodský, G. Cempura. Hydrogenation and hybridization in hard w-c:h coatings prepared by hybrid PVD-PECVD method with methane and acetylene. International Journal of Refractory Metals and Hard Materials 88:105211, 2020. doi:10.1016/j.ijrmhm.2020.105211. 66 https://doi.org/10.1016/s0043-1648(00)00488-9 https://doi.org/10.1016/s0257-8972(99)00586-1 https://doi.org/10.1016/j.surfcoat.2012.05.073 https://doi.org/10.1016/j.surfcoat.2006.07.020 https://doi.org/10.1016/0040-6090(84)90123-8 https://doi.org/10.1016/s0257-8972(97)00522-7 https://doi.org/10.1016/s1359-6454(01)00328-7 https://doi.org/10.1016/s0921-5093(01)01076-0 https://doi.org/10.1016/j.vacuum.2008.11.007 https://doi.org/10.1016/0040-6090(95)06569-5 https://doi.org/10.1016/s1044-5803(02)00192-4 https://doi.org/10.1007/s11340-016-0190-x https://doi.org/10.1557/jmr.1998.0146 https://doi.org/10.1016/j.tsf.2012.09.001 https://doi.org/10.1016/j.ijrmhm.2016.12.015 https://doi.org/10.1590/0366-6913201965s12577 https://doi.org/10.1016/j.surfcoat.2019.07.078 https://doi.org/10.1016/j.ijrmhm.2020.105211 Acta Polytechnica 27(0):62–66, 2020 1 Introduction 2 Experimental material and methods 3 Results and discussion 3.1 Substrate effects on hardness and indentation modulus in HiPIMS W-C:H coatings 4 Conclusions Acknowledgements References