ICMCTF2016 Session H2-2: Advanced In-situ Mechanical Testing of Films and Coatings

Residual stress plays an important role in the application of thin films and coatings concerning their mechanical performance under loading. Using focused ion beam (FIB) milling and digital image correlation (DIC), residual stress analysis has become possible for a wide variety of coating systems regardless of whether they are crystalline or amorphous. The residual stress is relaxed by FIB milling of certain geometries like pillar, double or single slit etc. and the resulting displacement field is tracked by high resolution scanning electron microscopy and DIC. Finally, the residual stress state of the coating can be calculated based on the determined deformation either by analytical solutions or finite element analysis (FEA).

However, the exact correlation between the local displacement gradients determined by DIC and the internal stress relief of the coating for different milling geometries isn’t fully clear yet. In this work, as a part of the European collaborative research project iSTRESS, diamond coatings have been used as reference material for residual stress analysis by FIB-DIC as well as Raman spectroscopy. Using a new in-situ µ-Raman spectrometer installed in a FIB, the Raman peak shift during FIB milling of different geometries like pillar, double and single slit was analyzed to directly observe the stress relief inside the diamond coating. Moreover, the obtained displacement fields and stress distributions were compared to FEA.

A good correlation between the experimentally observed displacement, by DIC, and stress, by µ-Raman, fields and the corresponding FEA was found. The combination of FIB-DIC and in situ µ-Raman allows deeper and direct insights in the relaxation of residual stress of thin films and coatings. The results might be useful for further improvement of the relaxation geometries used for FIB-DIC on thin films and coatings.

MEMS-based quantitative in-situ TEM nanomechanical testing techniques have emerged over the past decade as promising techniques to characterize and investigate the mechanical deformation of nanostructures. We previously developed a MEMS device which makes use of two identical capacitive sensors on either side of a nanospecimen providing independent measurements of force and displacement in the sample, and therefore dedicating TEM imaging only for high magnification observations. The sensing is based on differential measurement of the two capacitive sensors (with and without a specimen), which requires the assumption of linear elastic behavior of the specimen to calculate stress and strain. In this work, a sensing scheme for independent measurements of the two capacitive sensors has been implemented, thus eliminating this requirement and enabling force-controlled tests. The implementation and calibration of this new sensing scheme is shown in this work. The scheme is validated by performing monotonic and stress relaxation tests on 100-nm-thick Au films. Additionally, the capability of the technique to do advanced tests such as multiple stress relaxation and stress dip tests has been shown.

The point deflection method has recently emerged as a possible alternative to current micromechanical techniques for measuring the mechanical properties of thin films. A point deflection experiment consists in deflecting a clamped membrane in its center with a nanoindenter tip. The widespread availability of the required equipment makes the method very promising for future applications. These outlooks were further enhanced by the recent extension of the evaluation theory to rectangular membranes, which – unlike circular ones – are easily fabricated by standard lithographic techniques.

In this work, the recent theoretical advances were critically reviewed and an improved experimental method based on the measurement of the contact stiffness was developed. The new method was applied to the measurement of the residual stress of 100-nm thick SiNx and TiO₂ membranes. The accuracy of the point deflection experiments was assessed by comparing them to measurements performed with the reference bulge test technique on the very same samples. It is shown that the new experimental method dramatically improves the reproducibility of the measurements, and suggestions are made to improve the current evaluation scheme.

Ophthalmic lenses are made of plastic polymeric substrates usually coated with functional treatments composed of 5 to 15 layers, ranging from micrometers to nanometers. The first treatment consists of a primer, conferring impact resistance properties to the lens. A hardcoat with nanoparticles, is then deposited on top of this primer, bringing anti-scratch properties to the system. Both primer and hardcoat are within the micrometer scale and are deposited by wet chemical methods. Nanometric anti-reflective stacks are then evaporated onto the hardcoat by vapor deposition technology, to enhance wearers’ comfort.

Each of these interfaces may lead to delamination due to poor adhesion, and therefore affect the vision and comfort of wearers. The interface between the anti-reflective stack and the hardcoat is particularly sensitive because of chemical and mechanical contrast of its materials.

To better understand the mechanisms that lead to loss of adhesion between the SiO₂ anti-reflective layer deposited on the anti-scratch hardcoat, the authors first focused on characterizing mechanical properties of materials composing the layers and the interface. Nano-indentation and AFM experiments not only allowed to obtain topographical information, but also to access mechanical and adhesion properties. The operating range of a nano-indenter and an AFM being overlapped, an interesting study to confront elastic moduli obtained by both techniques is being carried on. In parallel, the best suited method to quantify adhesion will be chosen among different techniques, including 3-point bending, centrifugation technology, micro-compression and micro-tensile experiments.

Since its introduction by Uchic et al. [1], micro-compression testing of pillar samples has risen to be one of the primary techniques for interrogating the deformation behavior of small volumes. The technique offers site specific interrogation of microstructural features and a simple uniaxial stress state. This uniaxial stress state provides several advantages over other techniques such as nanoindentation, which imposes complex triaxial stresses. However, fabrication of micropillars is usually performed using focused ion beam (FIB) machining techniques. These have been shown to impose significant ion damage into small pillar and cause changes in deformation behavior in metallic samples compared to pillars produced without FIB techniques [2]. Even the differences in ion beam dosage caused by different milling techniques has been shown to introduce significant variations in deformation behavior [3]. Polycrystalline aluminum micropillars are expected to be especially susceptible to gallium ion damage [4] due to the well-known embrittlement of aluminum by gallium caused by the segregation and concentration of gallium at the grain boundaries [5]. Therefore, polycrystalline aluminum samples are an ideal case for comparing the relative damage introduced by focused ion beam machining using xenon ions as an alternative to gallium ions for machining of metallic materials. Micropillars were manufactured by FIB using both ion species to produce pillars of similar sizes. The grain boundary content of the pillars will be varied using by varying the intrinsic grain size of the polycrystalline aluminum and single crystalline samples. The strengths and rate sensitivity of the resulting deformation behavior will finally be investigated by strain rate jump microcompression testing of the pillars.

References

[1] Uchic MD, Dimiduk DM, Florando JN, Nix WD. Sample dimensions influence strength and crystal plasticity. Science. 2004;305:986-9.

[2] Bei H, Shim S, George EP, Miller MK, Herbert E, Pharr GM. Compressive strengths of molybdenum alloy micro-pillars prepared using a new technique. Scripta Materialia. 2007;57:397-400.

[3] Hütsch J, Lilleodden ET. The influence of focused-ion beam preparation technique on microcompression investigations: lathe vs. annular milling. Scripta Materialia. 2014;77:49–51.

[4] Kiener D, Motz C, Dehm G, Pippan R. Overview on established and novel FIB based miniaturized mechanical testing using in-situ SEM. International Journal of Materials Research. 2009;100:1074-87.

[5] Schmidt S, Sigle W, Gust W, Rühle M. Gallium segregation at grain boundaries in aluminium. Z Metallk. 2002;93:428-31.

In many applications, RF connectors are subjected to severe environmental vibrations. Vibrations induce micro displacements, leading to fretting wear damages. Former investigations [1] on reals connectors subjected to vibration confirms that fretting wear increase the DC electrical contact resistance but also RF microwave transmission and above all generate a significant additive phase noise. However connector vibration tests are not suitable to quantify the fretting damages of plated gold coating according the contact normal force and the sliding amplitude can be measured. To palliate such limitation a new micro fretting test system consisting in a fretting piezo actuator implemented in a commercial micro indentation system was developed.

Hence, sliding amplitude, normal force, but also tangential force and friction energy can be measured. The RF contact assembly consists in a micro strip line fixed to the piezo actuator and on a band mounted on the head of the micro indenter (Fig.1). The piezo actuator excites the micro strip line with a controlled displacement while the indenter impose a constant normal force on the band. During the test DC electrical contact resistance and phase noise are measured simultaneously.

When the gold layer is worn out by fretting gross slip sliding, oxide debris are formed from the non noble subsurface degradation. These oxide debris which are trapped between the interface decays the DC contact resistance but also the microwave signal transmission. The analysis confirms that the noise degradation is related to fretting wear rate of top gold layer. The thicker, the gold thickness, the longer the fretting phase noise endurance Nc (φN> φN_threshold).

Like for DC applications,[2] we shows that HF endurance N(φN> φNth) is not proportional to the gold coating thickness(e) but can be approximated by a power low function with an exponent p larger than 2: N.αe^p

References:

[1]: R. Enquebecq, S. Fouvry, E. Rubiola, M. Collet, L. Petit, J. Legrand, L. Boillot, “Effect of fretting wear damage in RF connectors subjected to vibration: DC contact resistance and phase-noise response”, accepted HOLM IEEE, 2015.

[2]: J. Laporte, O. Perrinet, S. Fouvry,”Prediction of the electrical contact resistance endurance of silver-plated coatings subject to fretting wear, using a friction energy density”, Wear, 330–331(2015) 170-181.