Growth defects in WC:H layers for tribological applications
Graphical abstract
Introduction
DLC layers are frequently used for mechanical protection of various industrial components. Since the very beginning of the DLC layer research dating back to the 1970's it has always been very tempting to use these layers for various engineering applications because of their high hardness and low coefficient of friction. On top of that, these layers also exhibit other interesting properties (optical, electrical, biological etc.) leading to even wider range of applications [1].
Along with the first application attempts the problem of adhesion of the layers when deposited on real metal parts emerged and was effectively solved by the application of an additional metal adhesive layer [2,3].
DLC layers actually represent a larger variety of structures based on amorphous carbon, with or without hydrogen and possibly with additional metallic or non-metallic components [4]. The metal-carbon layers (specifically combinations with tungsten) were further studied both by the deposition system manufacturers directly (e.g. Leybold AG, [5]) or by the research institutes [6] due to their advantageous combination of properties inherited from the DLC layers (high hardness, low coefficient of friction) and additionally increased temperature resistance due to the metallic doping. Another practical benefit of the WC:H layers is their relatively high deposition rate [6] (due to the combination of PVD and PACVD regimes – see Section 2), high target utilization, good process stability even for the scaled-up systems for large substrates and relatively modest requirements for the substrate heating before and during the deposition (below 200 °C), all of which makes the layers very attractive for the cost-effective industrial mass production processes. Due to the advantageous characteristics, the R&D departments of automotive producers readily started testing the WC:H layers on real automotive components - as for example published in Ref. [7] by Mercedes Benz AG.
The WC:H layer itself can be then specifically optimized for different applications, e.g. by varying the W/C ratio, to achieve targeted values of hardness, wear rate or coefficient of friction. For example, a rather wide range of friction coefficient's values ranging from 0.1 to 0.8 as a function of the C content (30 - 90 %) in the WC:H layers was reported by Voevodin et al. already in 1999 [8].
With the further development of analytical techniques (especially microscopic methods like AFM and FIB/SEM) it was subsequently shown that the structure of the DLC layers could contain various structural defects - see for example the pioneering works [9,10]. The studies still continue and the published works, e.g. by Vetter et al. [11] with an overview of defects in the DLC-like layers or by Panjan et al. [12] with even more detailed overview of defects in the layers prepared by PVD as well as with the suggestions of certain procedures for the defects elimination, continuously expand the general knowledge about the growth and structure of various modifications of the DLC layers.
In this paper we solely focus on a selected type of a DLC-based multilayer (further on denoted as WC:H) due to the fact that it is based on hydrogenated amorphous carbon (a-C:H) with targeted metallic doping, namely tungsten. This layer is currently used in certain automotive applications as a protective and wear resistant layer. Its mechanical and tribological properties were already reported earlier [13]. For the sake of its further optimization, i.e. to achieve a layer with a very smooth surface, we studied the micro-structure of this layer with the focus on the presence of structural defects. We visualized the defects and their cross-sections by advanced analytical techniques that helped us identify possible root causes of their nucleation. Subsequently, we attempted to control their growth, resulting in variations of their size, shape and density by means of practically useable adjustments of selected deposition parameters.
Section snippets
Material and methods
The WC:H layers in this study were prepared in a hybrid regime combining magnetron sputtering (PVD) and Plasma Assisted CVD (PACVD) in a commercial deposition system Flexicoat 1200 by Hauzer Techno Coating equipped with 5 rectangular planar magnetrons (target size 1000× 170 mm) with WC and Cr targets at the power of 9 kW in the Ar + C2H2 gas mixture. The process pressure varied for individual sublayers in the range of 0.1–0.3 Pa at a fixed Ar gas flow of 225 sccm and variable C2H2 gas flow from
Observations of layer structure and defects
A desired functional feature of the WC:H multilayer for our application is an ideally smooth surface. However, this was not the case of the layer shown in Fig. 1a, where the unwanted structural defects are displayed as dark circles with a density of up to 10 units (typically around 3–4) per selected 10 μm × 10 μm area. The density of the defects is comparable for Si and steel substrates. For the corresponding SEM image with a higher magnification (Fig. 1b) the sample was tilted by 52° in order
Discussion
During the study of the structure of WC:H multilayers, specifically their conically shaped defects, we employed the powerful combination of analytical techniques FIB/SEM/EDX in order to identify where and how these defects originate. The first observations of FIB craters brought us to the conclusion that smaller defects (leftmost defect in Fig. 3a) might originate within the bulk of the WC:H multilayer. However, due to the conical shape of defects combined with tilted sample geometry, the FIB
Conclusions
In this study we explored the appearance, internal structure and composition of the conical structural defects (hillocks) in a certain WC:H tribological multilayer. We found that the defects slightly differ from the surrounding layer not only in the structure (as shown in SEM and cross-sectional FIB/SEM images), but also in the elemental composition, i.e., defects contain more C and also evenly more H, which is preferably bound to C instead of W. We also found many similarities with structural
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors would like to thank Ing. Jan Maňák for the FIB/SEM measurements.
This work was funded by the HVM Plasma, spol. s r.o. and by the TA CR Project TH03020004. We acknowledge the CzechNanoLab Research Infrastructure supported by MEYS CR (LM2018110).
This work was also supported by Operational Programme Research, Development and Education financed by EU Structural and Investment Funds and the MEYS CR Project No. CZ.02.1.01/0.0/0.0/16_026/0008382 (CARAT).
References (33)
60 years of DLC coatings: historical highlights and tech. review of various DLC types, and their evolution for industrial applications
Surf. Coating. Technol.
(2014)- et al.
Improvement of carbon-based coatings for use in the cold forming of non-ferrous metals
Surf. Coating. Technol.
(1994) - et al.
History of diamond-like carbon films – from first experiments to worldwide applications
Surf. Coating. Technol.
(2014) - et al.
Surf. Coating. Technol.
(1991) - et al.
Preparation of W-C: H coatings by reactive magnetron sputtering
Surf. Coating. Technol.
(1993) - et al.
Metal-carbon layers for industrial applications in the automotive industry
Surf. Coating. Technol.
(1993) - et al.
Tribological performance and tribochemistry of nanocrystalline WC/amorphous diamond-like carbon composites
Thin Solid Films
(1999) - et al.
Cross-sectional scanning force microscopy analysis of arc-discharge-deposited diamond-like carbon films
Ceram. Int.
(1996) - et al.
A model for particle growth in arc deposited armophous carbon films
Diam. Relat. Mater.
(1998) - et al.
Growth effects in carbon coatings deposited by magnetron sputtering
Surf. Coating. Technol.
(2003)