Influence of probe geometry in micro-scale impact testing of nano-multilayered TiAlCrN/NbN coatings deposited on WC-Co

https://doi.org/10.1016/j.ijrmhm.2020.105441Get rights and content

Highlights

  • Applied load and probe sharpness control deformation in micro-impact.

  • Damage can be tailored to affect coating, substrate or interface.

  • Ratio of coating thickness to probe radius influences deformation.

  • High t/R and lower load micro impacts generate damage within the coating.

  • Damage at lower load occurs by a three step process.

Abstract

Hard nano-multilayered TiAlCrN/NbN coatings on cemented carbide have shown promise in dry high speed machining applications involving repetitive contact, such as end milling of hardened H13 steel. In this study the fracture resistance of TiAlCrN/NbN coatings under repetitive dynamic high strain rate loading has been evaluated by the micro-scale impact test method. Although the fatigue mechanisms can vary with the ratio of coating thickness t to the indenter radius R, macro-scale tests of thin coatings using probe radii in the mm range are necessarily at low t/R. Micro-impact tests at higher t/R have been performed with a range of diamond indenter geometries (R = 8, 20, 100 μm) to investigate the role of varying t/R (0.03–0.375) on the deformation behaviour. With the largest radius probe there was no clear failure for the coatings or substrate under the test conditions. With the 8 and 20 μm radius probes the behaviour of the coatings was strongly load-dependent and they were more susceptible to impact-induced damage than the carbide substrate. As the load increased there was a change from coating to substrate dominated deformation behaviour as the stress field extended further into the substrate. At lower load the dominant fracture behaviour was coating fracture through ring cracking, radial cracking and chipping. At higher load chipping became less prevalent and break-up of the carbide substrate more extensive.

Introduction

Multilayer coatings have shown enhanced performance in tribological tests and machining applications [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]]. Interlinked factors responsible for the improvement include (i) higher mechanical properties, especially H3/E2 (ii) through-thickness graded properties (iii) crack resistance - multiple interfaces providing the ability to deflect cracks laterally rather than through-thickness (iv) rotation of columnar grains distorting bilayer period (v) interfaces providing enhanced thermal stability (vi) adaptive mechanisms from tribo-film formation. Despite generally improved performance, their behaviour at higher load can be compromised as materials with high H3/E2 necessarily have less available options for reducing stress by plastic deformation. Impact resistance is important for many of these applications, but it is commonly assessed through macro-scale cyclic impact/fatigue tests [[16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]], with relatively blunt (typically R = 1–3 mm) probes or nano-scale repetitive impact testing with very sharp (typically R ~ 50 nm) cube corner diamond indenters [14,15,[28], [29], [30], [31]]. Strong correlation between fracture resistance in the nano-impact test and cutting tool life has been reported [14,15,[28], [29], [30], [31]]. In an impact test, the severity of the test and positions of peak impact-induced stresses relative to the coating-substrate interface can be controlled by varying the applied load, accelerating distance and the probe geometry. The position of peak stresses relative to the coating-substrate interface is completely different in the nano- and macro-scale tests. FIB cut cross-sections through nano-impact test craters on multilayer TiAlSiN coatings on hardened steel revealed extensive chipping but there was no delamination [13].

To fill the gap between the nano- and macro-scale, a micro-impact test has been developed which uses impact loads in the micro- range (~0.5–5 N) together with spheroconical diamond probes with end radii of ~20 μm [[32], [33], [34], [35], [36]]. The maximum energy that can be supplied per micro-impact is ×100 greater than in nano-impact. By increasing the energy delivered per impact, blunter indenter geometries can be used to produce damage within short experimental timescales. Switching from sharp to blunter spherical indenters provides an intrinsic suitability for examining gradual damage processes [16]. However, as probe radii increase, the results become progressively less sensitive to coating properties [37] and become more strongly influenced by the substrate hardness and toughness, as has also been reported in erosion testing under severe conditions [38].

In comparison to macro-scale tests, there are potential benefits of assessing coating fatigue resistance with nano or micro-impact tests. Short-duration experiments with automatic scheduling of test matrices on single samples enable rapid screening to evaluate the performance of novel coating compositions. The tests have the flexibility to alter loading level and severity of impact loading so that peak stresses can be positioned in the coating or at interface rather than in the substrate when results are less sensitive to coating properties. The tests are depth sensing, enabling accurate recording of cycles to failure and providing information on the fatigue failure mechanism. In contrast, macro-impact tests are limited by not being depth-sensing, so the precise time at which the coating failed is not clearly defined and it is only possible to say that the coating has survived or failed after the test is ended. The failure criterion can be arbitrary with average measures such as failure area at a given number of cycles being used or the coating wear depth at the end of the test. However, as the substrate deformation may not be fully elastic to deconvolute the coating wear from substrate plasticity requires FIB cross-sectioning. Due to the large size of the test probe, the response is averaged out over a larger area of the coating surface, making it insensitive to situations where the fatigue behaviour varies across a sample. Differences between the new micro-impact test, nano-impact and conventional macro-scale impact testing are summarised in Table 1.

The micro-impact test has been used to study the impact resistance of monolayer AlTiN and TiAlCrN coatings on cemented carbide [32], mono- and multilayer TiAlSiN coatings on cemented carbide [33], graded DLC coatings on hardened steel [34] and uncoated cemented carbides [35,36]. A strong sensitivity of the damage tolerance on the applied load was reported in all these studies, which used diamond indenters of end radius ~ 20 μm as the impact probe. With these indenters it was possible to produce severe damage within 5–10 min tests on all the samples tested.

In this current study, TiAlCrN/NbN coatings have been used as a model system to develop our understanding of the micro-impact technique, to study the evolution of damage in repetitive small-scale mechanical contact with probes of different sharpness. TiAlCrN/NbN is a promising coating for high speed machining of hardened steels, with longer tool life than other nano-laminated coatings [[1], [2], [3]]. In this application the coating system displays adaptive behaviour forming beneficial tribo-oxides. These complex (AlOx/CrOx/NbOx) tribo-films work in synergy by protecting the surface (like alumina tribo-films), lubricating the cutting zone (like chromia tribo-films) and dissipating energy (like NbOx films) [1]. To understand whether enhanced crack resistance from the multi-layer structure could also be important, results in the micro-impact are compared to tests on monolayer coatings (TiAlCrN) with similar mechanical properties. The results were also compared with tests on the uncoated cemented carbide substrate under the same conditions to understand the influence of substrate fatigue on the response of coated system.

Section snippets

Samples

Ti0.25Al0.65Cr0.10N/NbN nanomultilayer coatings were supplied by Kobe Steel Ltd. They were deposited using a hybrid coating system with combined plasma-enhanced cathodic arc source and unbalanced magnetron sputtering unit, using a powder metallurgical Ti25Al65Cr10 alloy as target for the arc cathode and Nb as the sputtering target. Further details are given in refs [1–3]. Three compositions were tested, which varied in the power supplied to the sputter source (0.5, 1 and 2 kW) which influenced

Results

The nanoindentation data summarised in Table 2 show that the mechanical properties of the three nano-multilayer TiAlCrN/NbN coatings were very similar, all being harder but less stiff than the WC-Co substrate. The H/E ratio of the coatings was virtually the same although H3/E2 values were marginally higher for 0.5 and 1 W than 2 W.

SEM images of impact craters produced with the 20 μm probe are shown in Fig. 1. All three TiAlCrN/NbN coatings showed the presence of some unreacted metal

Discussion

The mechanical properties of the hard multilayer coatings were broadly consistent with previous reports on similar coatings [1]. In monolayer hard coatings, alloying TiAlN with significant NbN has been reported to reduce elastic modulus whilst keeping hardness approximately constant [39], Although similar behaviour might be expected for TiAlCrN, the small fraction of Nb of these TiAlCrN/NbN coatings did not result in clear differences between them. Although there were some slight differences

Conclusions

Micro-impact tests with a range of diamond indenter sharpness (R = 8, 20, 100 μm) and applied load have shown how the dimensionless parameter t/R influences the deformation behaviour. With the 100 μm probe there was no clear failure for the coatings or substrate under the test conditions. With the 8 and 20 μm radius diamond probes the behaviour of the TiAlCrN/NbN nano-multilayer coatings was strongly load-dependent and they were more susceptible to impact-induced damage than the carbide

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

Funding to develop the novel micro-impact test technique through the Innovate UK Project No: 132369 – “Nano-to Micro-Impact Testing: An in-situ test for UK SEAC sector” is gratefully acknowledged. JLE thanks the financial support from the Basque Government Industry Department (ELKARTEK Intool2). C. Kimpton and X.W. Liu (both Cranfield University) are both acknowledged for their technical assistance with the SEM imaging and M. Rueda-Ruiz (IMDEA, Spain) is thanked for useful discussions on impact

References (43)

Cited by (0)

View full text