ICMCTF2012 Session B4-1: Properties and Characterization of Hard Coatings and Surfaces

Seed layers are often used to control nucleation, phase formation and texture of top layers. Within this work we studied the growth of sputtered rutile TiO₂ layers on α-Al₂O₃ single crystals with (00.1) orientation, with the goal to illuminate the epitaxial relations between both materials and to determine stable interface structures by ab-initio modeling. Reactive magnetron sputtering from Ti targets in an Ar+N₂ atmosphere has been used; the substrate temperature was set to 770°C and a moderate bias voltage of -50 V has been applied. Films showing the rutile TiO₂ structure with a thickness of 450 nm have been grown. Transmission electron microscopy and pole figure measurements indicated a sharp TiO₂ preferred orientation, with TiO₂ (100) oriented parallel to the interface. Three different variants of TiO₂ (100) on (00.1) α-Al₂O₃ representing TiO₂ domains with three different but crystallographically equivalent orientations have been found. Their epitaxial relationship corresponds to [010]TiO₂//[10.0]Al₂O₃ and [001]TiO₂//[21.0]Al₂O₃. Ab-initio calculations indicate that this interface can be assumed to be oxygen-rich, with a stacking sequence of O – Al – Al – O – O – Ti, which is characterized by a high work of separation and Young’s modulus, compared to an oxygen-poor interface of type O – Al – Al – Ti – O – O.

Oxide ceramics are well known to be hard materials with high chemical stability and long durability. A drawback of these materials is, however, their brittleness after passing threshold energy during deformation. A general way, how to improve toughness of oxides, is an addition of ductile metal. In such a case, it is, however, necessary to determine thermal stability of such material system.

Recently, Al-Cu-O coatings prepared in our laboratories by co-sputtering of aluminum and copper in argon‑oxygen gas mixture have showed an enhanced hardness and resistance to cracking [1]. The present paper brings data on maximum thermal stability and transformation processes occurring in these coatings during their heating up to 1300°C. The coatings were deposited with various copper contents (up to 10 at.%) by dual pulsed dc magnetron sputtering on silicon wafers and aluminum foils. Powdered coatings removed from the aluminum substrates were investigated by means of differential scanning calorimetry and the structure of all the coatings was characterized by X‑ray diffraction. It was found that all the as-deposited coatings are nanocrystalline one-phase materials with a cubic lattice structure independently of the Cu content. The lattice parameter varies from γ-Al₂O₃ to CuAl₂O₄ with increasing Cu content. At lower Cu contents (1.5 and 5 at.%) the as-deposited cubic structure decomposes to α-Al₂O₃ and CuAl₂O₄ at about 950°C during exothermic reaction. The highest thermal stability of the as-deposited cubic structure is achieved for the Al-Cu-O coating with the Cu content about 10 at.% and reaches 1100°C. An endothermic reaction detected above 1100°C will be discussed as well.

[1] J.Blazek, J.Musil, P.Stupka, R.Čerstvý, J.Houška: Properties of nanocrystalline Al-Cu-O films reactively sputtered by dc pulse dual magnetron, Appl. Surf. Sci. (2011), in print.

Here we present experimental studies of various multicomponent nanostructured coatings with emphasis on the high-temperature tribological applications. The concept of the "comb" like nanocomposite structure, which shows a very high thermal stability in the temperature range of 25-1300^oC with hardness well above 37 GPa is presented. The influence of Al and Cr on the high-temperature oxidation resistance and tribological performance, fatigue failure under dynamic impact tests both in air and NaCl solution, as well as electrochemical behavior of TiCN and TiSiCN coatings is described. An approach to decrease the friction coefficient without influence on the hardness and wear resistance by the incorporation of solid lubricant phase into hard coatings is demonstrated. The results of cutting tool tests are also reported.

At elevated temperatures, the metastable Ti_1-xAl_xN solid solution decomposes in cubic AlN and cubic TiN. Besides the chemical driving force, the residual stresses can be assumed to control whether the decomposition starts with more volume consuming TiN or smaller cubic AlN domains. Thus, within this work the effect of residual stresses on the decomposition of sputtered Ti_1-xAl_xN coatings was investigated. Using different bias voltages, a series of Ti_1-xAl_xN coatings (x=0.63) with stresses ranging from +670 to -550 MPa was synthesized on silicon (100) substrates. Vacuum annealing treatments and subsequent X-ray diffraction measurements showed that those coatings having compressive stresses start with the precipitation of AlN at temperatures of about 950 °C, while tensile stresses lead to the formation of TiN domains at slightly lower temperatures. These findings have been corroborated by high-temperature X-ray diffraction and differential scanning calorimetry investigations.

The cubic phase Ti_1-xAl_xN has been used to coat cutting tools since the late 1980’s. Extensive research has shown that the excellent tool performance is closely related to age hardening at elevated temperature where c-Ti_1-xAl_xN decomposes to domains of c-TiN and c-AlN. However, the number of in-situ studies is limited and especially isothermal annealing experiments of the transformation are lacking in the literature. Here, we use a combination of in-situ x-ray scattering experiments during isothermal annealing and phase-field simulations to study the evolution and coarsening of the domains. Further we investigate how the composition and annealing temperature affect the decomposition.

Coatings of two compositions, Ti_0.34Al_0.66N and Ti_0.50Al_0.50N, were grown by reactive arc evaporation using Ti₃₄-Al₆₆ and Ti₅₀-Al₅₀ compound cathodes in a N₂ atmosphere. In-situ high energy small-angle x-ray scattering (SAXS) were used to follow the decomposition at two isothermals, 850 and 900 °C, which were reached with a heating rate of >130 °C/min. Simultaneous wide angle scattering (WAXS) measurements allowed for determining of the strains in the coatings during the annealing. The x-ray scattering experiments were performed at the synchrotron x-ray at the Advanced Photon Source, Illinois, USA, beamline 1-ID.

The maximum entropy method was used to fit and extract a domain size from the SAXS patterns. At both 850 and 900 °C the growth rate of the domain size is slightly faster for the coating with higher Al content, which is consistent with its predicted higher driving force for decomposition. The results further show a significantly faster growth rate of the domain at 900 °C compared to 850 °C for both compositions, which is also predicted by phase field simulations. A peak shift in the SAXS pattern over time suggests that the growth of the domains is due to coarsening i.e. the spinodal decomposition is in its latter stage. STEM imaging and EDX mapping of the post annealed coatings reveal that the Ti_1-xAl_xN has decomposed and coarsened to domains rich of Al and Ti with a size coinciding with the size extracted from the SAXS fitting. With this work we show the difference in microstructure evolution at different compositions and temperatures. The results are discussed in terms of enthalpy of mixing and metal diffusivities in this alloy system.

Diamond scratch styli are defined by the Rockwell geometry specification in ISO6508 part 2. However, it is well known that the main uncertainty responsible for poor scratch test reproducibility is variability in Rockwell diamond stylus geometry (EC (S,M&T) contract No.MAT1-CT 940045 ‘FASTE’ & EC contract No.SMT4-CT98-2238 ‘REMAST’) and a certified reference material (BCR-692) exists to identify poor stylus performance. Part of the problem is an inadequate definition of Rockwell stylus geometry and measurement in ISO6508. Improvements are being actively considered, both at ISO, and by the defining metrology committee for hardness CCM-WGH. The main problem appears to be variability in radius over the spherical cap of the stylus. A VAMAS intercomparison (VAMAS TWA22 project 6) has been set up to develop improved measurement specifications to define better this key parameter and the results are presented here.

The geometry of 30 Rockwell diamonds (nominal radius 200 μm) from various manufacturers have been measured, using different metrological methods, to determine tip radius as a function of distance from the tip. Radius values were obtained from fits to 2D profiles of each tip using different width fitting windows (40 μm, 80 μm, 120 μm and 160 μm) centred on the stylus central axis. Styli, even those compliant with the ISO 6508:2 tolerances of 200 ± 15 μm when averaged over a wide fit window, frequently were very significantly outside tolerance over the smaller windows; results ranging from 140 μm < R < 320 μm for a fit window of 40 μm. This is a serious problem when Scratch testing hard coatings. Scratch tests on an a-CH DLC showed that the contact diameter at the first critical load was 40 μm. In these contact conditions, the strong variation in geometry are directly correlated to poor reproducibility of the Scratch Test.

It is becoming clear that the demands placed upon a Rockwell stylus by scratch testing are more stringent than those required for Rockwell hardness testing in which only the “superficial” scales (A, N & T) have test depths within the spherical cap region. To evaluate the effect of the observed radius variability on the Rockwell test, Rockwell HRN tests were performed on a Rockwell N scale hardness reference block. A significant effect of tip geometry on the Rockwell HRN test result was found.

TiSiN coatings were deposited on high speed steel substrates in a PVD technique using a combination of DC and RF magnetron sputtering by varying DC power. The optimized sample was coated at 500 W, and controlled samples were deposited at 300 and 400 W. Surface morphology showing that rectangular shaped grains of the optimized sample were observed in stack. Trend of surface roughness was correlated with the observed surface morphology. Cross sectional coatings were compared with structural zone model. There was no evidence of columnar growth that associated with the impurity of Si content in the coating for above 3.0 at.%. Nevertheless, columnar growth was observed for lower Si content of 1.0 at.%. Approximated method was utilized to calculate crystallite size and micro strain via XRD line broadening. The approximated crystallite size was varied but could be associated with the DC power, thickness and surface morphology of the coatings. The XRD showed all samples were oriented at (111), (200) and (220). A plane of (111) was dominant in the optimized sample that was corresponded to low micro strain and compressive stress. Orientations at (200) and (220) were preferred in the other samples since these were the less dense planes when the Si content was above 4.5 at.%. There was evidence of nanocrystalline TiN and an amorphous phase of Si N indicating that all samples were successfully deposited with TiSiN coating.

Plasma deposited coatings of TiAlON and CrAlN are a promising approach to decrease wear and friction of tools within plastics processing such as extrusion and injection molding. Up to now the exact mechanisms of interactions between the coating and the polymer melt are not well understood yet.

An experimental approach is presented that analyzes the surface chemistry of TiAlON and CrAlN coatings as a function of chemical composition and deposition parameters by means of X-ray photoelectron spectroscopy (XPS) and infrared spectroscopy (FTIR). Surface energies of CrAlN and TiAlON coatings were measured by static and dynamic contact angle experiments. The coatings were synthesized via High Power Pulse Magnetron Sputtering (HPPMS) in case of TiAlON and direct current (DC), pulsed middle frequency (MF) sputtering or HPPMS in case of CrAlN, respectively.

It could be shown that the surface composition of CrAlN coating significantly differs from the bulk composition. Although the CrAlN phase is supposed to be inert under a wide range of environmental conditions the surface is oxidized to a CrAlON phase upon contact with air. These effects have to be taken into account considering adhesion phenomena and the interaction with the polymer melt.

Additionally, a contamination layer of low-weight (hydro-)carbon species being omnipresent in the environmental atmosphere is formed on the surface of the coatings, which was followed from contact angle experiments and XPS analysis. The observed surface energy values measured by dynamic and static contact angle experiments were in the range of 35±10 mJ/m² which is much smaller than the expected high surface energy values in the range of hundreds of mJ/m² for TiAlON derived from DFT calculations.

The properties of the contamination layer and the oxidized CrAlON surface were investigated in detail via angle resolved XPS and depth profiling experiments with respect to their influences on the surface chemistry of the subjacent coating. To remove the organic contamination layer for advanced surface analysis under UHV conditions a cleaning process via a He plasma was evolved. It could be shown by means of in-situ photomodulated infrared spectroscopy (PM-IRRAS) that this process of He plasma cleaning is applicable to remove organic contaminations from the film surface without changing its surface chemistry.