Diamond has been considered an amazing coating material for a wide range of applications due to its many extraordinary properties including super hardness, high wear-corrosion resistance, low coefficient of friction, high thermal conductivity and excellent biocompatibility . High-quality diamond films have been coated on various substrate materials for enhanced physical, mechanical, chemical and biomedical properties [2–4]. Depositing diamond films on low-cost high-speed steel substrates and cemented carbide (WC-Co) tools will offer enhanced wear/corrosion/erosion resistance for applications in more aggressive conditions . However, it has been technologically difficult due to the weak interfacial adhesion of diamond films to the substrate materials, primarily caused by the preferential formation of a loose graphite intermediate layer on these ferrous substrates, and a big mismatch in the coefficients of thermal expansion between the diamond coatings and the substrate materials [6, 7]. As a direct consequence, a long incubation period is required to form continuous diamond films which, however, easily delaminate from the substrates during cooling process after deposition.
To solve these problems, various techniques have been developed, such as by applying additional interlayers or diffusion layers, or applying a comprehensive surface modification method on the transition metal substrates [8–14]. We have recently investigated the effects of a series of alloying elements on the quality of diamond films directly deposited on bulk steel substrates [15, 16]. The results showed that Al plays a unique role in promoting the formation of continuous and adherent diamond films. The great potential of Al as a high performance interlayer material for diamond adhesion enhancement on steels also comes from its low material cost and mature coating technologies. In this study, we investigated the feasibility of using Al as an interlayer on a typical iron-nickel-cobalt Kovar alloy substrate. This alloy has a much lower coefficient of thermal expansion in comparison with steels and it has been widely used as low-cost electronic packaging material in high performance ICs. However, this alloy has a relatively low thermal conductivity and can hardly meet the increasing demand on power dissipation. Therefore, combining diamond coatings with Kovar alloy substrate is expected to offer a possible solution to the thermal management challenge. The Al interlayer is designed as a barrier function to prevent graphite formation and inward diffusion of carbon, while the substrate material selected is expected to decrease the thermal stress developed in the coating/substrate system and consequently, enhance interface adhesion of diamond coatings produced on it. The general diamond deposition behavior on such a substrate has been described before , and this study is to further clarify the fundamental mechanism Al performs by determining its chemical state at the interfacial locations.
The substrate material used for diamond deposition was commercial Kovar alloy (Fe29Ni17Co, in mass %). The substrates were machined into specimens of 10 mm × 10 mm × 1 mm in size and mechanically polished with 600# SiC sandpaper, then ultrasonically cleaned in acetone bath and dried for Al film deposition. The aluminum film about 80 nm in thickness was coated on the Kovar substrate by an ion beam sputtering. Following this process, a 20-second surface scratching treatment was performed in a diamond powder suspension (2 g diamond of 0.5 μm in size, mixed with 20 ml of ethanol) to enhance diamond nucleation density. Deposition of diamond was carried out in a 2.45 GHz microwave plasma-assisted CVD system (Plasmionique) using H2 and 1 vol.% CH4 at a total flow rate of 100 sccm, and a working pressure of 30 Torr. The substrate temperature produced at a power of 800 W was about 690 (±50)°C as measured by a thermocouple mounted underneath the stainless steel substrate holder and calibrated by optical pyrometer. The surfacial and cross-sectional morphologies were observed by scanning electron microscopy (SEM). The composition, structures of the surface deposits and interfaces, the changes of the substrate microstructure were identified by micro-Raman (Renishaw 2000, Ar laser wavelength 514 nm), X-ray diffraction (Co Kα radiation), energy-dispersive X-ray spectroscopy (EDS) and X-ray photoelectron spectroscopy.
Results and discussion
After exposure in the CH4–H2 gas mixture, a thick black deposit has formed on the surface of the as-polished Kovar substrate. Figure 1 shows typical surface and cross-sectional SEM images. The top deposit containing both fine particles and filaments (Figure 1(a)) is loosely packed and can be easily scratched away from the substrate. The cross-sectional observation (Figure 1(b)) shows a deep internal attack zone. Figure 2 shows the Raman spectrum and XRD patterns of the Kovar substrate after exposure in CH4–H2 mixture. The Raman spectrum identifies that the surface deposit is primarily graphitic carbon while diamond is absent. The XRD analysis shows that a significant amount of graphite and cementite has been produced. The formation of such a composite is closely associated with a metal dusting mechanism because of the strong catalytic properties of the Fe, Co and Ni components for a preferential formation of graphitic carbon, rather than a diamond structure . The metal dusting procedure in carburizing atmospheres has been well described before by Grabke . It includes an initial carbon decomposition from the carbonaceous gases and then adsorbs on the metal surface. The oversaturated carbon combines with the metal Fe and a carburization occurs. The formed cementite is metastable and a following decomposition occurs: Fe3C = 3Fe + C. Therefore, graphite soot and iron in forms of precipitated particles are produced on the alloy surface. Carbon can also diffuse into the metal matrix and cause an internal carburization. Accordingly, the diamond nucleation and growth needs a much longer incubation period as the carbon source available has been occupied.
Applying an intermediate Al layer on the Kovar substrate has significantly enhanced diamond nucleation and growth, as shown in Figure 3. Diamond nucleation starts at an early deposition stage and becomes continuous quickly with prolonged deposition. The Raman spectrum (Figure 4) measured from the continuous diamond coating indicates a strong diamond characteristic peak centered at 1338 cm−1. The upshift of the peak position from the natural diamond line at 1332 cm−1 is associated with compressed stress inside the diamond film . This peak shift can be regarded as a signal of enhanced adhesion of diamond coating to the substrate. Once the diamond film delaminates from the substrate, the diamond peak position will come back to its standard position at 1332 cm−1 as a result of the stress release. These data demonstrate that the Al interlayer has effectively inhibited the formation of graphite phase and substrate damage due to carbon diffusion, while the nucleation, growth and interfacial adhesion of diamond is enhanced.
The positive role Al plays for enhanced diamond nucleation and growth on steel substrate has been attributed to its ability of suppressing the catalytic effect of the transition metals. For instance, it was assumed that the partially filled 3d shell of the transition metal iron (or nickel, cobalt) has strong catalytic activity for sp2 bonding and therefore promotes deposition of sp2 bonded carbon phases. With Al addition into the steel substrate, electron transfer between transition metals and Al occurs and the 3d orbital of iron becomes fully filled. However, our recent investigation reveals that for FeCrAl alloy substrate, on which an adherent diamond coating can directly deposit even without any interlayer or surface treatment, a very thin Al-rich amorphous oxide sublayer has formed between the diamond film and the underlying substrate interface . Similarly, an Al-rich interfacial oxide layer has been detected on the Kovar substrate surface after applying an Al interlayer. Figure 5 shows an EDX analysis measured from the diamond coating–substrate interface. Obviously, this intermediate layer is very rich in both oxygen and Al. An interfacial analysis by XPS further confirms the formation of an alumina layer on the substrate surface, as seen in Figure 6. The main peaks of O1s and Al2p XPS spectra are located close to 532.1 eV and 74.5 eV, respectively, corresponding to the Al–O bond. This can be explained by the fact that aluminum has a very high reactivity with oxygen. It reacts with the residual oxygen in the reaction chamber and forms an Al2O3 layer, acting as a barrier to prevent carbon diffusion and protect the substrate from the atmosphere. This function is just like the protection it provides for high temperature alloys exposed in carburizing environments . Consequently, graphite formation is restricted and fast nucleation of diamond is enhanced along with improves interface adhesion.
We have also observed on repeated samples that in case the Al interlayer is not continuous or not thick enough, such as by partial damage during diamond suspension treatment, detachment of diamond films and attack of the substrate will occur. Figure 7a shows a surface image of diamond coting formed on the Al-interlayered kovar substrate on which local spallation of diamond coating is observed. On some areas where diamond coating still remains on the substrate, a small amount of graphite has grown out of the diamond phase, as shown in Figure 7(b). With prolonged deposition time, the volume fraction of diamond decreases and it is gradually incorporated into the rapidly formed graphite soot. Figure 7(c) shows the corresponding XRD patterns of such a Kovar substrate on which graphite and diamond coexist, while the substrate has been simultaneously carburized due to inward diffusion of carbon. It should be mentioned that the substrate peak in the patterns overlaps with the diamond peak (111), while the existence of diamond has been identified from the SEM observation and Raman spectrum analysis. The failure of diamond film deposited on Kovar substrate should be closely associated with the deficiency inside the Al interlayer. A discontinuity of the Al interlayer, likely being partially removed during diamond suspension scratching will lead to insufficient protection against carbon diffusion and graphitization induced by the substrate elements.
Direct deposition of adherent diamond films on low thermal expansion FeCoNi Kovar alloy substrate has been proved unsuccessful due to the rapid formation of graphite intermediate layer and substrate carburization. Applying an Al thin film interlayer has significantly suppressed the formation of non-diamond phase and enhanced nucleation and adhesion of diamond. The results demonstrate that Al can be a promising cost-effective interlayer material for diamond coating on ferrous substrates. The major role Al plays is by forming an alumina interfacial barrier layer.
This research has been sponsored by the Canada Research Chair Program and by the Natural Sciences and Engineering Research Council of Canada.
May PW. Phil T R Soc A 2000;358:473–95. Google Scholar
Askari SJ, Chen GC, Lu FX. Mater Res Bull 2008;43:1086–92. Google Scholar
Wei QP, Yu ZM, Ashfold MNR, Ye J, Ma L. Appl Surf Sci 2010;256:4357–64.Google Scholar
Zhang CZ, Li YS, Tang Y, Yang L, Zhang L, Sun Y, et al. Thin Solid Films 2013;527:59–64. Google Scholar
Neto VF, Vaz R, Ali N, Oliveira MS, Grácio J. Diamond Relat Mater 2008;17:1424–8. Google Scholar
Buijnsters JG, Shankar P, Ter Meulen JJ. Surf Coat Technol 2007;201:8955–60. Google Scholar
Silva FJG, Fernandes AJS, Costa FM, Baptista AP, Pereira E. Diamond Relat Mater 2004;13:828–33. Google Scholar
Buijnsters JG, Shankar P, van Enckevort WJP, Schermer JJ, ter Meulen JJ. Phys Stat Sol (a) 2003;195:383–95.Google Scholar
Endler I, Leonhardt A, Scheibe H-J, Born R. Diamond Relat Mater 1996;5:299–303.Google Scholar
Bareiβ C, Perle M, Rosiwal SM, Singer RF. Diamond Relat Mater 2006;15:754–60.Google Scholar
Xu F, Xu JH, Yuen MF, Zheng L, Lu WZ, Zuo DW. Diamond Relat Mater 2013;34:70–5.Google Scholar
Polini R, Mattei G, Valle R, Casadei F. Thin Solid Films 2006;494:116–22.Google Scholar
Glozman O, Hoffman A. Diamond Relat Mater 1997;6:796–801. Google Scholar
Li YS, Hirose A. Surf Coat Technol 2007;202:280–7. Google Scholar
Li YS, Hirose A. Chem Phys Lett 2006;433:150–3. Google Scholar
Li YS, Yang LZ, Ma HT, Feng RF, Yang Q, Hirose A. Mater Res Innovations 2014;18:S4-979–S4–982. Google Scholar
Grabke HJ. Corrosion 1995;51:711–20. Google Scholar
About the article
Published Online: 2015-01-14
Published in Print: 2015-12-01