ICMCTF1999 Session F1/E4: Mechanical Properties and Adhesion

This paper describes methods combining finite element analysis and experimental indentation testing to determine coating modulus and hardness independent of substrate effects.

To determine coating modulus, the contact modulus of the coated surface is first determined from the elastic portion of an experimental load/displacement curve. The modulus of the coating alone is then determined from a function derived by a parametric fimite element analysis of indentation behaviour.

The hardness of the coating alone is determined by comparing an experimentally determined mean contact pressure for the coated surface with parametric finite element results for the same contact geometry and known hardness ratios.

An accurate determination of mechanical properties of film substrate compounds and thin layers itself becomes increasingly important due to its economic relevance. On the other hand it is desirable to come to a calculated film design regarding the mechanical properties to limit the empirically based search for a suitable film. For both cases it is advantageous to use spherical indenters because they allow a description with analytical models and furthermore the splitting of the penetration process into a completely elastic an an inelastic part.

The results of a new analytical model for the calculation of elastic stresses and deformations during normal and tangential loading of a sphere will be compared with experimental results. It will be shown that the calculated load-depth data for different film substrate combinations agree with the results of indentation measurements.

Further, the change of value and position of the stress maximum and the measurable stiffness in dependence on the elastic properties and the thickness of the film are given. This is used for the evaluation of the critical load during scratch tests.

Finally, an outlook will be given for the calculation and evaluation of elastoplastic deformations due to spherical indentation in the case of homogeneous materials. An impoved analysis method is proposed that uses a pressure distribution after plastic deformation which is more realistic than that of the Hertzian theory.

TiO₂ is an insulator well known for its optical properties. High hardness and corrosion resistance are in favour of mutiple industrial applications.

In the present work we study PVD multi-layers with alternating Ti and O compositions obtained at low deposition pressures (<400K) by magnetron sputtering in a reactive plasma of varying, low pressure (0.2 0.3 Pa) Ar O carrier gas compositions.

Structural properties are assessed by standard metallurgical techniques like SEM, TEM and XRD. They are studied in relation with the surface mechanical properties assessed by depth sensing micro and nanoindentation, standard and microscratch testing as well as multipass friction fatigue in the online ,triboscopic operation mode.

The main issue in the present study is the relation between multilayer coating architecture and brittle mechanisms in tribologic applications

New generations of microelectronics and microsystem devices call for the utilization of a variety of new materials and the combination of materials with a wide spread of their mechanical and thermal characteristics. Reliability and functionality of microsystems, i.e. of small scale integrated electronic, mechanical and optical components, largely depend on their mechanical and thermal constitution. Thermo-mechanical aspects of component and system reliability become more and more important with growing miniaturization as the local physical parameters and field quantities show an increase in sensitivity due to inhomogeneities in local stresses, strains and temperature fields. Since there is usually a lack of information about the local material parameters, a pure field simulation cannot as a rule solve the problem. The state of the art of microsystem design more and more requires direct "coupling" between simulation tools (including e.g. FE modelling) and advanced physical experiments.

The author has combined various laser techniques, scanning microscopy, thermography and acoustomicroscopy with advanced numerical field simulation. The input data from the experiment are directly led into the simulation codes to get more realistic results for reliability estimation and life time prediction. The investigations have been directed towards electronic packaging of microsystems, crack and fracture behavior of multilayer systems ("fracture electronics") within the interconnected regions surrounding the chips and to other problems, e.g. occurring due to thermal "misfit" between different materials within the critical regions of microsystems. The author also deals with the microDAC method, a new powerful tool for microdeformation analysis within a scanning electron microscope applying the grey level image correlation method to obtain the strain fields and related local fracture quantities of microcomponents.

Measurement of the small scale mechanical properties of surface engineered materials at elevated temperatures opens up significant new possibilities for the materials scientist. For example, wear resistant coatings and surfaces generally experience temperatures significantly above room temperature during normal service, and it is well known that both the hardness and modulus of such materials can be strongly dependent on temperature. It is clearly desirable to optimise the mechanical properties at the service temperature rather than to extrapolate the results of room temperature measurements. In addition, many microelectronic thin film structures are produced and processed at temperatures of several hundred degrees. A complete understanding of the interactions between the materials used can be obtained only by making measurements across the full temperature range rather than only at room temperature.

Insruments for routinely measuring the mechanical properties of materials on a very small scale have hitherto been limited to room temperature operation. One factor responsible for this is that in most nano-scale instruments the displacement transducer is placed above the specimen, thus leading to unacceptably high thermal drift. In the present work, a novel transducer arrangement has been used which obviates this. The configuration avoids any significant radiant heating, conductive heating, as well as convection currents. Specimen temperatures of 500 °C can be achieved with a thermal drift rate below 0.004 nm/s.

Nanohardness and nanoscratch results have been obtained on a variety of materials, including metals, thin films and ceramics. In many cases, these show a substantial temperature dependence of the mechanical properties. For the particular case of silicon, the well-known crystalline-to-amorphous transition behaviour has now been characterised from room temperature to 500°C.

As the substitution for the conventional aluminum metalized film, the transparent gas barrier films consisting of a 10~100nm thick SiOx film deposited on a 12µmm-thick polyethylene terephthalate (PET) substrate are widely noticed in the food packaging industry. In the manufacturing process, such as the lamination or the heat sealing, films are often loaded under high temperature conditions.

Therefore, in order to investigate the temperature influence on the fracture strength, the multiple cracking progress in SiOx films during tensile tests in the furnace were in-situ observed. It was shown that at the temperature of 82°C, the crack onset applied strain was higher than that at 33°C. In addition, at 82°C, increasing rate of cracks with the applied strain became slower than that at 33°C.

Next, to take into account the influence of the mechanical properties of the substrate to the cracking, tensile tests of bare PET films were performed at different temperatures. The remarkable reduction of the Young's modulus and the yield stress were seen around 70°C. The residual strains in films which are also significant factors, were also estimated from the curvature of SiOx/PET specimens at different temperatures. With the consideration of these factors, the cracking progress in SiOx films with different temperatures were reasonably predicted by the modified shear lag analysis.