Predicting plastic part life

Paul C. Haschke
Bodycote Broutman Inc.
Chicago, Ill.

Edited by Jean M. Hoffman

DMA testing helps determine the effect of different curing conditions on thermoset polyester products. DMA demonstrates the degree of cure based on the stiffness and glass transition temperature of the samples. Optimum cure is important to insure that thermoset materials retain the required stiffness needed at elevated temperatures.

A variable sinusoidal stress is applied to a sample and the resultant sinusoidal strain is measured.

The relationships between complex modulus, storage modulus, and loss modulus are often shown as a right triangle. The hypotenuse is complex modulus. The tangent of the phase angle from a DMA test equals the ratio loss modulus/storage modulus.

A DMA comparison of three polymers — epoxy, polyethylene terephthalate (PET), and polyethersulfone (PES) — shows all have similar deflection temperatures under load (DTUL). However, the mechanical response below and above the DTUL of each is different. This is important when the product will experience short-term temperature excursions. A short-term excursion above the DTUL for PES could result in a product failure because of the sharp drop in storage modulus above 218°C.

Dynamic mechanical analysis (DMA) measures mechanical response of materials subjected to periodic stress at operating temperature. Tests can be configured for pure tension or compression, single or dual cantilever, and three-point bend modes.

DMA works for a wide range of materials from metals, ceramic, and composites to coatings and adhesives. But it's particularly useful for polymers because of their combined viscous and elastic response. DMA can be set to vary time, temperature, frequency, and deformation amplitude, all of which may adversely affect polymer mechanical properties. This helps designers predict how well a plastic part will perform in the field.

Test procedure
DMA applies a sinusoidal stress to the sample and measures the resultant strain. In purely elastic materials, the phase difference between the stress and strain sine waves is 0°. Here, the two waves are in phase and the input energy is completely returned as kinetic energy. Purely viscous materials, on the other hand, phase shift 90°, which effectively cancels the imposed stress. Applied energy dissipates as heat. Most polymers combine viscous and elastic response (viscoelastic), so the measured phase angle falls between 0 and 90°.

The phase angle and the stress and strain amplitudes measure a material's elastic and viscous response or storage and loss modulus. It also reveals damping efficiency (tan ), creep and stress relaxation properties, and transition temperatures.

DMA quantifies other parameters including complex and dynamic viscosity, storage and loss compliance, and complex loss and storage modulus. Viscosity is a measure of a material's resistance to flow. Complex viscosity accounts for energy loss and storage. Dynamic viscosity is basically the same as complex viscosity but instead measures sample viscosity during mechanical oscillation as a function of temperature, frequency, time, or both.

Complex compliance or yield is the ratio of strain to stress. Complex modulus is its reciprocal. Complex compliance includes both storage compliance and loss compliance. Storage compliance deals with elastic yield (energy storage) and loss compliance, viscous yield or energy loss.

These parameters help quantify a thermal operating range, impact resistance, stiffness as a function of load, and flow under dynamic load. Other metrics identified by DMA include sound absorption, the rate and degree of cure, and short or long-term effects of molecular weight, cross-links, and chain entanglements.

Materials are often evaluated for long-term, high-temperature performance under short-term tests such as deflection temperature under load (DTUL). However, materials can exhibit similar deflection temperatures yet show vastly differing mechanical properties above and below the deflection temperature. DMA better evaluates material mechanical response over time and a wide temperature range.

It should be emphasized that DMA can't predict all material properties for any polymer. However, DMA can check the effects of temperature, load, deformation, and frequency.

Commercial DMA instruments typically have somewhat limited frequency ranges and time scales. However, one technique known as time temperature superpositioning (TTS) helps boost the amount of experimental data. TTS assumes higher temperatures accelerate molecular relaxation. In other words, elevated temperatures reduce the time over which these processes occur, shifting the data.

Measurements at higher temperatures can shift along a time or temperature axis. The technique uses a method of reduced variables to shift frequency scans at various temperatures and compares them to a reference curve at one temperature. The resulting master curves predict the property of interest at a specific temperature over a broad time scale or wide frequency range.

Typically, a material is analyzed for a mechanical response at a series of frequencies at selected temperatures. Static tests such as creep and stress relaxation are performed for a specified duration over a range of isothermal conditions. By shifting data with respect to time or frequency, viscoelastic changes from operations at higher temperatures can be made to appear to take longer or happen at lower frequency.

DMA predictions from TTS can be confirmed by checking the mechanical response of samples with known service lives to the predicted response from TSS data. However, TTS doesn't work for all materials. Reportedly, polyethylene and polystyrene can't be shifted, while others such as natural and synthetic rubbers can. Generalized method development in different modes is made easier by ASTM and ISO standards covering DMA testing.

Complementing DMA are data from other tests such as differential thermal analysis (DSC), thermogravimetric analysis (TGA), and rheology. These techniques examine thermal transitions, compositional differences based on weight loss and shear rate, and viscosity or flow properties. The solution to a particular design problem often comes after performing several different thermal analysis techniques.

For example, an otherwise excellent design may be spoiled by poor quality control. Manufacturers may carefully control part dimensions and tolerances yet ignore material moisture content, % regrind, and other variables. Such manufacturing variances may cause a part to fail prematurely in the field.

Potential product failures are categorized by severity. For instance, in nuclear power plants, certain locations can't tolerate any plastic part failures during certain types of accidents. In such cases, a significant amount of testing is required to understand the failure mechanism. Then models that accelerate aging demonstrate a large safety factor under the conditions of interest. In other cases, a failure may be bothersome, but tolerable. Here, models and testing are still important, but a smaller safety factor suffices.

In either case, it is essential to periodically check the accuracy of a model by testing naturally aged parts that have been in service. In this way, the replacement interval can be adjusted to reflect factors affecting real service life that can't be accurately modeled or predicted.

Designing with DMA

Tan is the ratio of loss modulus to storage modulus and indicates damping efficiency. Two candidate rubber materials serve as an illustration. They were tested for damping efficiency at 200 Hz over a wide temperature range. The first sample (A) shows a larger relative magnitude of tan than (B) and, therefore, has better damping properties at this frequency.

A polypropylene sample was tested for creep at five temperatures from 10 to 70°C. A mastercurve is generated from the data and indicates creep compliance at 25°C over nine decades of time.

A DMA thermogram show the storage modulus for new and old rubber conveyor belts. The storage modulus values at –25, 0 and 25°C indicate the old belt has stiffened over time.

The thermogram shows the loss modulus of polycarbonate parts with good and bad impact resistance. Comparison of the curves shows that the good sample has a lower glass-transition temperature and an additional slight damping peak from 40 to 120°C.

DMA thermograms may indicate changes to material cross-link density, crystallinity, and molecular weight. Here are a few design examples that help illustrate the benefits of DMA testing:

• Rubber is used to damp vibration in many applications. Using DMA, the tangent of the phase angle, indicates damping efficiency. The ratio loss modulus/storage modulus, tan , is not affected by sample geometry. A high ratio is desired over a wide temperature range for the frequencies of operation. The ratio, loss modulus/storage modulus, shows how effectively the rubber loses energy to molecular rearrangements and internal friction.

• TTS simulates long-term creep in thermoplastic parts. First, short-term creep experiments are performed at increasing isothermal temperatures. Then data is translated to a master curve which shows creep behavior over many decades of time.

• Oxidation can also change material properties. For example, rubber conveyor belts, made from styrene butadiene rubber, oxidize over time and stiffen. DMA tracks embrittlement over time and helps establish proper replacement intervals.

• Key in many thermoplastic applications is resistance to cracking from impact. Impact resistance can be qualified by measuring the loss modulus and glass transition of a material. Loss modulus measures damping or energy dissipation. Good impact resistance is related to the magnitude of the loss modulus curve, a low-glass-transition temperature, and the presence of additional damping peaks below the glass-transition temperature.