ICMCTF2011 Session B4-3: Properties and Characterization of Hard Coatings and Surfaces

The addition of a third element in transition metal nitride coatings is widely used to improve their properties: high hardness is obtained by addition of silicon or boron and these elements also increase the high temperature oxidation resistance. Since machining technique evolution, i.e. high speed machining, imposes coatings protection to be more and more efficient, TM-Si-N and TM-B-N films are used to coat cutting tools operating in severe conditions. However, the literature is mainly focused on the characterization of such films at room temperature after an annealing step at high temperature. Although this procedure is the most convenient, it brings information about the films properties that may differ from those required for high temperature applications. This presentation is focused on the in situ structural characterization of TM-Si-N and TM-B-N coatings during air oxidation using XRD, Raman and FTIR methods.

Nb-Si-N and Fe-Si-N films were deposited on stainless steel and silicon substrates by reactive magnetron co-sputtering of elemental targets, while Cr-B-N films were synthesised by reactive magnetron sputtering of composite Cr/B targets with various compositions.

The characterization at room temperature of oxidized Nb-Si-N clearly highlights the formation of a niobium oxide top layer on the remaining Nb-Si-N films. On the other hand, in situ XRD analyses shows that the NbN diffraction peak intensity decreases while no diffraction peak of niobium oxide are observed, indicating that the low temperature NbN oxidation induces the formation of amorphous niobium oxide. The crystallization of the oxide top layer occurs at 525°C. A further increase of the annealing temperature induces the grain growth of the Nb₂O₅ phase but the coalescence step occurs at higher temperatures for higher values of silicon content. Concerning Fe-Si-N films, in situ XRD analyzes clearly highlight a progressive denitridation of the FeN grains and this phase is fully reduced in Fe₂N at 425°C. Then, hematite diffraction peaks are noticed at 450°C. This decomposition-oxidation mechanism is confirmed by in situ Raman analyses.

The pure CrB₂ and three Cr–B–N films with various nitrogen contents were deposited by a bipolar asymmetric pulsed DC reactive magnetron sputtering system. The structures of thin films were characterized by X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), respectively. The surface and cross sectional morphologies of thin films were examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The surface roughness of thin films was explored by atomic force microscopy (AFM). A nanoindenter, a scratch tester and pin-on-disk wear tests were used to evaluate the hardness, adhesion and tribological properties of thin films, respectively.

The phases of Cr-B-N coating transferred from CrB₂ into CrN and BN as the nitrogen content increased. It was found that a high hardness value was observed on the pure CrB₂ coating. The hardness and elastic modulus of Cr-B-N thin films decreased with increasing nitrogen content. The influences of nitrogen concentration on the mechanical and tribological properties of thin films were discussed.

Using aberration-corrected high resolution transmission electron microscopy (HRTEM), electron energy loss spectroscopy (EELS), and ab-initio density functional theory (DFT) the interface microstructures of VN (CrN)/MgO (001) are closely examined. By HRTEM, under the conditions of negative spherical aberration we show an atomic resolution structure of epitaxially grown VN (CrN) film on MgO with a clearly resolved oxygen and nitrogen sublattice across the interface. As revealed by DFT calculation, the (002) interplanar spacing oscillates in the first several VN layers across the interface. Quantitative EELS analysis across the interface was carried out. Interfacial chemistry determined by EELS analysis shows the preponderance of O and V atom at the interface of VN/MgO, and V-L_2,3 and O-K edges at the VN/MgO interface obtained by ELNES and the spatial difference analysis show slightly discrepancy from the bulk, a small detectable core-level shift as compared to the bulk. The fine structures of Cr-L_2,3 and O-K edges across the CrN/MgO interface were also revealed.

Titanium aluminum nitride (TiAlN) films are one of the most common hard ceramic coatings in use today. Often out of lack of adequate measuring techniques, the characteristics of hard ceramic films are investigated at room temperature or after heat treatment. Although this approach is sometimes useful, it is sensible to test at temperatures, where dislocation activity is more significant. Using high temperature nanoindentation the typical hardness (H) and Young modulus (E) measurements were extended to calculate the fundamental deformation parameters, such as activation volumes (V*) and total activation energies (ΔG_tot).

Near-to epitaxial TiAlN films were grown on MgO single crystals (110) using a reactive magnetron sputtering system. The following chemical compositions were investigated: cubic(c)-Ti_0.44Al_0.Δ56N, c-Ti_0.68Al_0.32N and wurtzite(w)-Ti_0.36Al_0.64N. Additionally c-Ti_0.44Al_0.56N was annealed at 600C for 24 hours. High temperature nanoindentation experiments were carried out in a temperature range from 298 to 673K at 3 loading rates of 0.5, 1 and 10 mN/s.

H values of the all cubic samples were stable in the measured temperature range, with a slight decrease at 673K. H for the wurtzite sample on the other hand decreased continuously from 19.8±0.9 GPa at 298K to 16.9±1.4 GPa at 673K. E values for all samples remained constant throughout the temperature range: 344 GPa±34 for Ti_0.68Al_0.32N, 336 GPa±23 for as-deposited- Ti_0.44Al_0.56N, 356 GPa±21 for annealed- Ti_0.44Al_0.56N and 219 GPa±11 for w-Ti_0.36Al_0.64N. H of the films was found to be strain rate sensitive at all temperatures, leading to a percentual increase of up to 20% for the cubic films.

Fundamental deformation parameters were furthermore calculated. V* values ranged from 0.05-0.2 (burgers vectors⁾³ (b³) and 0.4-1.2 b³ for cubic and wurtzite sample respectively. ΔG_tot for cubic samples were very close and ranged from 5.1-5.8 .10^-3 µb³. ΔG_tot values rose linearly with the temperate, whereby the mechanical work negligibly influenced the calculated values.

Summing up for the tested temperature range the crystallographic structure had the most notable influence on all parameters. The rate-controlling mechanism was determined to be lattice-resistance for the cubic films, as V* values were smaller than 1b³ and ΔG_tot values of very low order. For wurtzite films a dislocation-dislocation interaction mechanism is assumed.

This investigation proved that high temperature nanoindentation can be used, to extract fundamental deformation parameters and to conclude, which rate-controlling deformation mechanism is taking place.

Ti-Si-N is a newly developed superhard coating material. The common issue for the hard coating is the delamination due to high residual stress, especially as the coating thickness is larger than 1 μm. However, for the industrial application to deposit a superhard coating with a thickness larger than 1 μm is usually required. In addition, since the delamination is mostly caused by high residual stress, the measurement of residual stress on the Ti-Si-N coating is also important for understanding the adhesion issue. In this study, Ti-Si-N films were deposited on P-type (100) Si wafers using unbalanced magnetron sputtering (UBMS) at different deposition durations. The purposes of this study were to investigate the mechanical properties of Ti-Si-N films with different thickness, especially the hardness and residual stress, and to prepare Ti-Si-N films with high hardness and large thickness suitable for industrial applications. To study the intrinsic properties of the coating, no interlayer or buffer layer was inserted between the Ti-Si-N coating and Si substrate. The thickness of the thin films increased with increasing deposition time ranging from 281 to 2044 nm. The structure of the nanocomposite coatings characterized by X-ray diffraction (XRD) showed that the crystalline phase was TiN with (200) or (111) preferred orientation depending on thickness. The results of X-ray photoelectron spectroscopy (XPS) indicated the existence of Si₃N₄ bonding in the nanocomposites. Therefore, the Ti-Si-N films were TiN/SiN_x nanocomposite. Nanocomposite specimen with a good combination of hardness and thickness was obtained, where the hardness was 37 GPa with a thickness of 2 μm. Optical laser curvature method and XRD cos²αsin²ψ method were used to measure the average residual stress and stress of the TiN phase in the nanocomposites, respectively. The results indicated that the amorphous SiN_x in the TiN/SiN_x nanocomposite could significantly relieve the average residual stress ranging from 19 to 68%. The degree of stress relief increased with increasing film thickness, which may be the reason that the thickness of nanocomposite can reach 2 μm. The critical stress of film delamination for each specimen can be estimated from fracture mechanics, and the stress decreased as film thickness increased.

This study aims at understanding the structures, morphologies, and mechanical properties of TaN-Cu nanocomposite thin films after multiple aging and annealing. These films were deposited by reactive co-sputtering on Si and tool steel substrates. The films were then annealed using RTA (Rapid Thermal Annealing) at 400°C to induce the nucleation and growth of Cu particles in TaN matrix and on film surface. Cu nano-particles emerged on the surface of TaN-Cu thin films were then removed after the samples were tested for their structural and mechancial behaviors. The samples were then re-annealed (rejuvenated), and re-tested for their properties. This process was repeated for four times to investigate the annealing effect. The results reveal that the Cu particles would re-appear on film surface after each annealing cycle. The particle size appears to be smaller with the increase of the number of annealing cycles. The hardness of the samples would decrease gradually with the repeated annealing. This appears to be due to the increased porosity in the films. The wear rate and friction coefficient of these repeatedly annealed samples depends mainly on the formation of lubricious Cu films, not on the hardness values.