When Good Parts Go Bad
At times, a single technique can make great strides in determining cause of failure, but in many cases, multiple methods will be needed to make a confident diagnosis and find the underlying problem or combination of issues causing the failure.
Kevin Battjes, M.S.
Associate Scientist, Impact Analytical
Many OEMs and contract manufacturers have the facilities to test basic physical properties to assure the quality and performance of their products. This works well for the vast majority of day-to-day business needs, but most companies have been faced at some point with the unexpected failure of a proven design, often in the absence of any conscious change in materials or production.
The consequences can be disastrous, depending largely on how early the problem is discovered and how efficiently it can be analyzed and corrected. The results can range from failures identified in routine quality control, which allow the manufacturer to halt further production until the problem can be investigated; to failures in the field, which can be far more grave. Depending on the application, the potential safety hazards, product liability, recalls, and damaged relationships can all loom large.
This article discusses the many analytical technologies available for determining the cause of failures in particular as they relate to failure of plastic parts. Technologies covered include optical microscopy, transmission electron microscopy, scanning electron microscopy, atomic force microscopy, energy dispersive X-ray spectroscopy, thermal analysis, dynamic mechanical analysis, and chromatography.
The initial step in analyzing a failure is to get a complete history on the part and material, including the service environment, load, physical composition, and molding conditions. If the failure involves a fracture, the initiation area is typically examined with a low-power stereo microscope, as technicians look for patterns in the break and/or the wear surface that give clues to the mechanism and ultimate cause of failure.
Optical microscopy: Optical microscopy provides the opportunity to observe the part and the fractures at low enlargement factors. The stereomicroscope is especially useful as an initial review to gain an overall sense of the failure, and it offers great flexibility in specimen size, orientation, and lighting choices. The compound light microscope is also valuable for examining crystalline materials in polarized light.
Transmission electron microscopy (TEM): Often, the imaging process advances to TEM to view detailed morphological images that define the failure mechanisms. Imaging the fracture point and analyzing the elemental composition can help detect foreign particles, non-homogeneous mixing, cold welds, cyclic loading, poor adhesion to reinforcing fibers, and void formation.
Scanning electron microscopy
Scanning electron microscopy (SEM) is an excellent tool for examining surfaces at high magnifications.
It produces a pseudo three-dimensional image without many of the aberrations that are troublesome in optical microscopy. Magnification typically ranges from as low as 20X to as high as 50,000X or more. (Because of their low signal yield, good insulating properties and chemical composition, most polymer materials have a practical magnification limit of 10,000X to 20,000X.)
In many cases, the fracture surfaces of a failed part indicate the failure mechanism by characteristic features along the fracture. Microscopy can discover many clues to the failure, from identifying the point of origin, to the propagation direction, to the mechanical properties.
This photo is an SEM micrograph of a fractured glass-reinforced nylon part. The image allows analysts to study details of the glass-matrix interface and the fracture surface. Magnification can be as high as 50,000X or more, although most polymer materials have a practical limit of 10,000X to 20,000X.
Specimen preparation for SEM is relatively simple and quick, generally consisting of excising the sample area of interest and applying a vacuum deposited conductive layer to avoid charging in the electron beam. In blends of incompatible polymers or block copolymers, phase separation has a significant impact on the end use properties of the material, so it's important to know the morphology of the components. Additive amounts and processing conditions also affect the phase domain type and size.
To generate SEM images of the separated phases, specimens are frozen and fractured in a reproducible manner. Preparation of the sample usually involves carbon coating by evaporation and/or sputter coating with a heavy atom. (Gold or gold/palladium are most common.) The component phases are often readily identified, and micrographs are recorded for measurement and reference.
Polymer parts and adhesives
Frequently at issue in failure are plastic components or polymeric adhesives, which continue to replace traditional metal parts and fasteners at an increasing rate, as manufacturers seek to reduce weight, lower production costs and/or improve the corrosion resistance of their products. Compounding the problem for many engineers is a general lack of experience with plastics as opposed to metals, where modes of failure and their underlying causes are drastically different.
The motivators that drive the choice of polymeric materials can be either economic or performance related. Weight reduction is often a key objective, particularly in automotive applications. Polymeric materials offer features such as sound absorption, aesthetics and cost savings, as well as the versatility to conform and/or mold into complex shapes. In some parts (such as valve covers, fans, fuel and emissions components, vacuum control systems and fluid reservoirs), the flexibility and/or corrosion resistance of a polymer can deliver superior performance over a metal counterpart. Within the realm of polymeric materials, the selection of specific grades of plastics, adhesives, and elastomers can also have a significant impact on profitability. Even the difference of a few cents per pound from one resin or supplier to another can be enough of an impetus to make a change, at least on paper. But when a part fails, the anticipated savings can quickly evaporate from the ensuing testing, analysis, and corrections.
Used for analyzing fillers, additives and contaminants in polymer materials, Energy Dispersive X-Ray Spectroscopy evaluates the elemental composition of deposits and residues found in problematic materials. This method is especially effective in identifying unknown particulates and foreign bodies in failed parts, and helps identify potential sources.
Transmission electron microscopy
Transmission electron microscopy (TEM) provides high magnification imaging and electron diffraction capability for the ultra-structural analysis of polymer systems. Magnification typically ranges from 500X to 500,000X. Specimen formats may be whole mounts, thin films, replicas, or ultra-microtomed thin sections. Staining techniques enhance image contrast based on functional group constitution. TEM is the tool of choice when characterizing the phase morphology of blended or copolymerized materials. Polymeric materials such as high impact polystyrene, ABS, TPO, and filled plastics are also readily characterized by TEM methods.
Energy dispersive X-ray spectroscopy
Energy dispersive X-ray spectroscopy (EDS) is based on an electron beam that strikes the specimen, which then emits X-rays characteristic of the material. Thus, EDS enables identification of the elemental composition of the material in question. This technique is especially effective in determining the chemical composition of unknown particulates and foreign bodies in materials or parts that are causing failure issues.
EDS is appropriate for analyzing fillers, additives and contaminants in polymer materials. It also helps answer questions regarding the composition of deposits and residues found in problematic systems and materials.
Fourier transform infrared spectroscopy
Fourier transform infrared spectroscopy (FT-IR) is a popular tool for identifying contaminants and determining the presence of functional groups in chemical systems. Because IR absorption depends upon a dipole moment change during each molecular vibration and/or molecular rotation, it's possible to conduct functional group analysis, as well as molecular identification and characterization.
Some FT-IR systems include a polymers and additives digital library, which provides an excellent tool for identifying an unknown polymer. A small amount of the material can be excised from the molded article and dissolved in a solvent. A thin film of the polymer is cast and its infrared spectrum is measured. By searching the library, a match can often be quickly found for the spectrum, thereby identifying the material. An experienced analyst is invaluable in interpreting the spectra and their implications for the material in question.
Fourier Transform lnfrared Spectroscopy is a tool for identifying unknown contaminants by extracting and analyzing non-volatile additives. The peaks in this graphic correspond to functional groups and chemical bond types. Peak positions and shapes help analysts identify specific chemistries that might contribute to part failure.
Thermal analysis includes a number of techniques that characterize the thermal and physical properties of a polymeric material. Specific determinations include glass transition temperature, enthalpy of phase transition, kinetic constant, crystalline melting point, specific heat, crystallinity, degradation temperature, viscoelastic properties, percent weight loss, material softening point, and coefficient of thermal expansion. The thermal and crystalline properties of polymeric materials play a significant role in understanding part performance in the intended applications.
Thermogravimetric analysis (TGA) is a thermal analysis technique in which changes in the weight (mass) of a sample are measured as a function of temperature and/or time. TGA is frequently chosen to determine polymer degradation temperatures, residual solvent levels, and absorbed moisture content. The inorganic glass fiber content of polymer and composite samples can also be readily analyzed by TGA.
In operation, the polymer resin is degraded and burned off in heating these specimens to high temperatures in an air atmosphere. The noncombustible glass fiber is left behind as a residue, so the weight fraction of glass fiber in the composite can be determined by a TGA residue analysis routine.
This test is widely used for analysis of filled systems to evaluate the material against the manufacturer's or supplier's specifications. It can also serve to examine moisture content in resins or molded articles.
Gas chromatography (GC) is the most common analytical technique for determining the number and concentration of components in a volatile mixture, or the presence of volatile impurities. It can also aid in the positive identification of a polymer compound. Additives extracted from the polymer can be detected and identified by GC, as well.
In operation, a sample, usually a liquid or a solid dissolved in solvent, is injected into the gas chromatograph. The sample is flash-vaporized and brought into the column by the carrier gas. As the sample passes through the column, it is separated into its individual components, and as each one passes through the detector, it appears as a deflection on a chart. The area under this deflection is proportional to the concentration of the component in the original sample.
When combined with mass spectrometry (MS), the sample mixture is first separated by gas chromatography. Upon entering the mass spectrograph, the components in each peak are identified by their molecular ion and fragmentation pattern.
High-performance liquid chromatography (HPLC) is the method of choice for separating nonvolatile, thermally unstable and/or polar components. Solid or liquid samples are dissolved in an appropriate solvent and injected into a liquid chromatograph, where the components are separated by selective retention within the stationary phase. As the analytes flow through the detector, there is a deflection on the chart. As with gas chromatography, the area underneath the curve is proportional to the concentration of the analyte in solution.