One of the more commonly employed sample examination techniques involves cross-sectioning to reveal internal structural and compositional information. This destructive analysis technique is usually used to elucidate defects or failure mechanisms in electronic devices.

In the majority of cases, the specimen of interest is placed in a small plastic cylindrical container, whose inside is lubricated with some type of mold release agent, typically Teflon or silicone based. A two part epoxy is mixed in the appropriate proportions and poured into the cylinder to encapsulate the sample. The epoxy filled cylinder is then placed in a small vacuum chamber for several minutes in order to remove any bubbles resulting from the mixing process, and to force the epoxy into any porous parts of the specimen. After removal from the vacuum system, the epoxy is then allowed to cure for twenty four hours at room temperature. This curing process can be accelerated to four hours by heating the material in an oven at 50°C.

Once the epoxy has cured, the encapsulated sample is removed from the plastic mold, and is ready to be cross-sectioned. If the area of interest lies deep inside the sample, the process begins by removing the overlying material with a small diamond saw. The saw blade is made of a copper alloy and has its outer perimeter imbedded with fine industrial diamond particles. The slowly rotating blade is water cooled during the cutting process to prevent frictional heating of the specimen.

The specimen is now ready for polishing. The initial coarse grinding is accomplished by placing the surface of interest in contact with a circular rotating wheel containing 120 grit sand paper. De-ionized water is dripped onto the wheel to lubricate, cool, and remove the abraded particles. Once the sample is uniformly abraded and flat, it is meticulously cleaned with de-ionized water to remove any remaining loose particles and blown dry with nitrogen gas. This process is repeated using a cloth on the wheel with successively finer grit particles, typically diamond or alumina particles suspended in water. The final polishing is accomplished with 0.05µm alumina (Figure 3-1). Over polishing is to be avoided to prevent rounded edges at the junction between materials of differing hardness.

The sample is now ready for examination. Typical tools used for this purpose are optical and scanning electron microscopes (SEM). Both feature digital cameras which allow images to be captured and stored. Optical microscopes are useful up to approximately 1200X magnification, where they are able to visualize details as small as 0.5µm.

For higher magnifications, and visualization of finer detail, scanning electron microscopes are employed. A SEM functions by rastering a finely focused electron beam across the sample, and collecting the secondary electrons generated by the sample due to impingement of the primary beam. This secondary electron signal is used to modulate the brightness of a television screen (CRT, cathode ray tube), and point by point creates a visual image of the surface of the specimen. This image is strictly topographic in nature, and magnifications of several hundred thousand times can be obtained, allowing visualization of features as small as several tens of angstroms.

Qualitative relative compositional information can be determined through the use of a backscattered electron detector. This annular detector is placed around the primary electron beam and senses the electrons that are ejected from the specimen at very large angles. The intensity of the backscattered signal depends on the atomic number of the material, so additional contrast related to composition can be obtained. Elements with larger nuclei generate more backscattered electrons, and thus appear brighter on the CRT.

Quantitative compositional information can be obtained using an energy dispersive spectrometer (EDS) to examine the x-rays generated by the primary electron beam bombardment of the sample. Since each element has distinct energy levels, an x-ray spectrum can be used to identify the elements present, and determine their concentration. Additionally, the intensity of individual peaks in the spectrum can be used to modulate the video signal, providing a two dimensional compositional map of the sample.

Examples of an optical micrograph (Figure 3-2), SEM image and EDS spectra (Figure 3-3), and EDS map (Figure 3-4) from a variety of specimens are shown.

In summary, the ability to cross-section, visualize, and analyze specimens is a crucial capability in order to develop a more complete understanding of the functioning and failure mechanisms of modern electronics. For more information about the failure analysis capabilities at the EMPF, please contact Ken Friedman at 610.362.1200, extension 279 or via email at kfriedman@aciusa.org.