Jan-Åke Schweitz

Fracture testing of silicon microelements in situ in a scanning electron microscope

Fracture testing of silicon cantilever beams (thicknesses 10–20 μm) was performed in situ in a scanning electron microscope by means of an equipment specially designed for this purpose. Beams of various sizes and orientations (〈011〉 and 〈001〉) were manufactured in Si (100) wafers by two different micromachining procedures. The beams were tested by simple bending to fracture, and a number of fundamental fracture parameters were determined from an analytical model of elastic fracture. To verify its validity, the model was utilized to evaluate an experimental E modulus, which was found to agree well with previous results. Fracture limits, fracture strains, and initiating flaw sizes were determined. The maximum fracture limit was very high; about 10 GPa. The strengths of different beams scattered from this value down to practically zero strength, with an average close to 4 GPa. The corresponding fracture strains and initiating flaw sizes were 6% and 3 nm, respectively (maximum strength), and 2% and 17 nm (ave...

Micromechanical fracture strength of silicon

In order to test the statistical influence of some process and micromachining parameters on the fracture strength of silicon microelements, arrays of identical microsized cantilever beams were bulk micromachined in single‐crystalline silicon wafers. The beams were exposed to various surface treatments (diamond polishing with different grades, oxidization, stripping of oxide) in different combinations. The influence on fracture strength was investigated by bending the beams to fracture in a micromanipulator mounted in situ in a scanning electron microscope while registering force‐versus‐deflection curves. Average fracture strengths, standard deviations, Weibull moduli, crack‐initiating flaw sizes, and in some cases elastic moduli were evaluated. Diamond polishing was found to decrease the fracture strength drastically, but polishing followed by oxidization not only restored the original strength, but actually increased it, due to crack healing. Polishing, oxidization, and subsequent stripping of oxide resu...

In situ tensile strength measurement and Weibull analysis of thick film and thin film micromachined polysilicon structures

Abstract A method is introduced in which tensile tests can be performed in situ on micromachined structures. The testing equipment consists of a testing unit mounted on a micromanipulator in a scanning electron microscope. The fracture loads of micromachined beam structures made from thick and thin film polysilicon were measured, and the fracture strengths were then calculated via measurements of the fracture surface areas. Characterization of the film materials was also performed with transmission electron microscopy, atomic force microscopy and scanning electron microscopy to locate the critical defects in the materials. The statistical scatter of the fracture strength values was evaluated using Weibull statistics, which yielded the mean fracture strength and the Weibull modulus—a measure of the amount of scattering. In this study, for the first time, Weibull theory was also applied to a real micromechanical structure, i.e. standard test results were transformed into expected strength limits of a more complex structure.

Influence of surface coatings on elasticity, residual stresses, and fracture properties of silicon microelements

Microscale silicon cantilever beams of (100)〈100〉 and (100)〈110〉 orientations were magnetron sputtered with submicron layers of Al, Ti, or TiN, or thermally coated with SiO2. Theoretical expressions for the elastic deflection induced by residual stresses were derived, and utilized to deduce such stresses from observed deflections. A theory for the elastic stress distribution in coated beams exposed to external bending moments was utilized to deduce maximum stress levels at fracture in the coatings and in the substrates. The fracture tests were performed in situ in a scanning electron microscope by means of specially designed equipment. For uncoated beams, the average fracture stress was 6 GPa (maximum 13 GPa) for 〈100〉 beams, and 4 GPa (maximum 6 GPa) for 〈110〉 beams. Most coatings proved to have a strength‐reducing effect, particularly the brittle, thin coatings of TiN, but also the Ti coatings (which displayed brittle fracture behavior). Ductile, thin coatings of Al were either neutral, or induced a sma...

Hardness and fracture toughness of semiconducting materials studied by indentation and erosion techniques

Abstract In recent years, the growing field of semiconductor micromechanics has created an increasing demand for strength data on semiconductors and for adequate tests and evaluations of their mechanical properties. In a recently published paper, the present authors have demonstrated that the solid particle erosion rate can be taken as a simple and highly reproducible statistical measure of the susceptibility of silicon and GaAs to contact damage in the micron range. In the present work the scope is broadened to include several new crystal orientations (and one new doping level), as well as three other materials: germanium, InP and InAs, for which the hardness and fracture toughness K Ic values are determined by means of the indentation technique. K Ic values are also derived from erosion data by means of a recently reported brittle fracture model, based on non-lateral spalling in single-crystal semiconductors. These values are compared with results obtained by the indentation technique and conventional test methods reported in the literature. Fracture surface energies are deduced from the experimental K Ic results. The materials tested are ranked with respect to elastic properties, microhardnesses, fracture toughnesses, and sensitivities to contact damages in general. The influence of crystallographic orientation on room temperature microfracture properties is clearly established, but the corresponding influence on microhardness is found to be rather limited. The influence of doping on the room temperature mechanical properties is noticeable but small.

Assembling three-dimensional microstructures using gold-silicon eutectic bonding

Abstract An assembly method for three-dimensional microelements is presented. The assembly is done in situ with a micromanipulator in an SEM using Au-Si eutectic bonding. Microblocks bonded to larger silicon substrates are used for evaluation of the mechanical strength and a microarch is presented to demonstrate the possibilities of the technique. The microelements are fabricated by bulk micromachining, and sputter deposited with chromium and gold. Etched (111) faces have been successfully bonded. TEM investigation of samples from vacuum furnace experiments show large gold grains with smaller chromium silicide grains in the bonded region. The silicon in the eutectic liquid precipitates epitaxially on both silicon faces. Mechanical bending tests on the microblocks give sufficiently high fracture stresses for the intended applications in microrobotic systems. Average fracture stresses of 65 MPa are measured for one set of parameters. Problems encountered are misalignments of the microelements during processing and void formation in the bonds. It is believed this is connected to the experimental equipment and set-up. The microarch, which consists of three assembled microblocks, reaches a tensile stress of 16 MPa, encouraging further development. In conclusion, strong microbonds are achieved using a solidified gold-silicon eutectic melt as an adhesive, and it is demonstrated for the first time that three-dimensional microassembly by means of eutectic bonding of micromachined elements can be performed by manipulation and processing on a locally heated specimen table.

A transmission electron microscopy study of hillocks in thin aluminum films

Hillocks, small outgrowths on a film surface, form when compressional stresses in an aluminum film are relaxed at elevated temperature (≥90 °C), for instance during the phase of rising temperature in an annealing cycle. This paper reports a study of hillock formation in Al films of thicknesses in the interval 0.25–2.2 μm and which have been deposited by electron beam evaporation. Hillock sizes, shapes, number and formation temperatures were determined, the latter on a heating stage in situ in a scanning electron microscope. The internal structure of the hillocks was studied by cross‐sectional transmission electron microscopy technique. These studies provided strong support for the idea that hillocks are formed by migration of material along grain boundaries, presumably at triple junctions, up to the surface where it is deposited in a growing hillock. Initially, the hillocks are separated from the original film surface by a grain boundary‐like interface, but prolonged annealing will cause underlaying grain...

High-sensitivity surface micromachined structures for internal stress and stress gradient evaluation

The internal stress and stress gradient of thick () and thin () polysilicon films were evaluated with surface micromachined test structures. The structure that measured internal stress consisted of actuator beams rotating an indicator through an angle corresponding to the stress. The indicator deflection was measured in an SEM. Finite element analysis (FEA) was used both to optimize the design and to calibrate the structure. A folded beam design was used to minimize the total area the structure occupied so that it could be incorporated in the wafer layout of other surface micromachined details, and used for online process diagnostics. The indicator was provided with a Vernier scale to facilitate quick evaluation in an optimal microscope. The stress gradient was measured from the deflection of long () cantilever beams. The deflection was measured in an optical microscope and the output was calibrated with FEA calculations.

Evaluation of mechanical materials properties by means of surface micromachined structures

For all micromechanical devices, mechanical properties such as elasticity constants, internal stresses, fracture limits and, for ductile materials, yield limits and strain-hardening behaviour, are ...

Microfabrication of three‐dimensional boron structures by laser chemical processing

A method for microfabrication of three‐dimensional structures in free space is presented. Laser‐assisted chemical vapor deposition is used to grow a material at a point where a laser beam locally heats the substrate. By moving the substrate relative to the laser beam with a micropositioning system, three‐dimensional shapes can be created. Helical shapes are generated utilizing three linear translational axes as well as an additional rotational axis. Tilting the substrate to align the growth direction with the laser beam direction facilitates improved process control. The smallest structures that can be grown with this technique are about 1 μm. Amorphous boron fibers and crystalline boron springs have been manufactured as two examples of micromechanical elements. The amorphous boron fibers show excellent mechanical properties: A modulus of elasticity of 420–450 GPa, a fracture strain of 2.7%–3.7%, and a fracture stress of 12–17 GPa. The crystalline boron springs produced so far display only moderate mechan...