Designed To Beat The Heat

Carl A. Cura
Manager, Application Development
Damon Brown
Laboratory Technician
LNP Engineering Plastics
Exton, Pa.

Edited by Jean Hoffman

All design engineers must address temperature and chemical resistance during the development process. Temperature and chemical exposure must not be taken lightly — even when choosing high-performance materials such as polyetheretherketone (PEEK), polyphenylene sulfide (PPS), polyethersulfone (PES), and even 4/6 and 6/6 nylons. High temperature and even seemingly benign chemicals can many times cause or contribute to failure of these and other thermoplastic materials. This is especially true for components under stress.

Although plastics do not corrode like metals, they are prone to chemical attack. More importantly, they can experience environmental stress crazing and cracking (ESC). Every polymer has at least one solvent that will cause chemical reactions and molecular degradation even in the absence of physical stress. But it is relatively easy to design around this type of solvent degradation. The bigger design problem is ESC. A relatively benign chemical acting together with stress can often reduce the effective strength of the material far below what would be expected.

It is interesting to study some of the latest compounded thermoplastics. Some of the most widely used consist of a resin compounded with either 30% glass or carbon fiber.

Many of these are general-purpose materials that can serve in a variety of applications. There are, however, specialty resins that may outperform some of these materials in one particular area, but often serve only a single market. Examples include fluoropolymers used in special fluid-handling applications or connectors made from liquid-crystal polymers.

A HANDLE ON TEMPERATURE
Glass-transition temperature (T_g) is an important property of any plastic. During heating, the glasstransition temperature is the point at which an amorphous polymer or the amorphous phase of a semicrystalline polymer displays moltenlike behavior. At this temperature, the random molecules that are locked in place after molding begin to move, rotate, and slide around. Hard regions become soft and flexible. One practical example of T_g can be seen when ironing polyester clothing. The hot iron softens amorphous regions in the polyester and forms a crease. The crease becomes permanent when the cloth cools off and drops below the T_g.

Mechanical properties stay fairly constant below Tg, but gradually change when molecules within the material start to move at T_g. For semicrystalline materials containing both crystalline and amorphous phases or regions, movement of the polymer chains is phase dependent. Ordered crystalline regions remain intact above T_g, but will quickly "unravel" as the temperature approaches the plastic's melting point. Conversely, at T_g, the amorphous regions soften more quickly, becoming pliable and leatherlike. In general, above T_g, materials lose strength and modulus, while gaining elongation.

Two high-temperature amorphous materials, PES and PEI, have the highest T_gs. In contrast, nylons have the lowest. This is particularly true after treatment with plasticizers such as moisture to boost flexibility. Nylons, however, have small amorphous regions and large crystalline areas. This gives them better properties above the T_g. It is important for designers to choose resins with T_gs above the maximum operating temperature if mechanical properties must remain constant.

Another useful parameter is the relative thermal index (RTI), historically called the continual use temperature. A material's RTI value is determined by tests outlined by a rating defined by UL 746B. The specification parameters, however, are subject to some variability which may skew test results. In addition, test specimens are heat aged under no load and then properties such as tensile strength are tested at room temperature. For most practical applications, materials experience heat and load at the same time.

Furthermore, the rating doesn't indicate how strong or stiff the material will be at the RTI temperature. But, the RTI is still a valuable piece of information because it indicates how heat will degrade base resins. Of the widely used materials, PEEK sees only a small amount of deterioration. And PPS, PES, polyetherimide (PEI), and polyphthalamide (PPA) resins along with nylon 4/6 and SPS have similar performance and are classified as high-temperature resins.

Other important temperature effects include heat deflection temperatures under load (HDTUL) and stressstrain properties. HDTUL represents the maximum temperatures compounds can withstand under no load conditions and for less than 5 min exposures. It is only useful for short-duration manufacturing steps like overmolding or high-temperature painting and curing. When the part hits the real world, the HDTUL is meaningless. Reinforcement often boosts HDTUL because it is a bending test that doesn't involve heat aging. Most materials at the HDTUL will behave like elastomers. And the material will likely fail quickly if actually used at these temperatures.

Possibly the most important material property to consider is the actual stress-strain properties at the part's operating temperature. Tensile strength and tensile modulus are frequently given as is a stress-strain curve. Amorphous materials, that include PES and PEI have a relatively constant moduli up to 350°F. But at 400°F PES and PEI are close enough to their T_g values that they are no longer structurally useful. On the other hand, many materials start to deteriorate by 200°F because of lower glass-transition temperatures. And by 300°F, even PEEK shows the effects of its glass transition temperature. Many semicrystalline materials have tensile strength beyond 400°F, though syndiotactic polystyrene (SPS) and polyketone (PK) are exceptions.

PES, PEI, and PEEK tend to have good mechanical properties up to 350°F, with minor exceptions such as the stiffness of SPS and PPS at room temperature. PPA and nylon 4/6 have good properties as well. At 500°F nylon 4/6 outperforms many other high-temperature plastics and is strong and stiff.

Stress-strain curves at a given temperature are often critical for planning specific designs. They show fracture elongation and the material modulus beyond the initial approximate straight line portion of the curve. For example, 4/6 has good elongation, 8%, at 300°F. This indicates 4/6 has good impact resistance. Similarly, curves show that PEI and PES are stiff at stresses above 5,000 psi.

The nonlinear nature of thermoplastic materials at elevated temperatures is important. The best way of testing such a behavior involves first heat-aging the material and then completing a stress-strain test at the same aging temperature. But the plastic industry has yet to define such a test. Fortunately, however, if a design engineer uses high-temperature stressstrain curves heat-aging has only a small impact. And heat-aging degradation can be handled by the addition of a material safety factor.

EFFECT OF CHEMICALS
Surprisingly, the effect of chemicals on these materials also involves T_g. Researchers at GE's corporate research facility, Schenectady, N.Y., in the 1970s discovered that a resin's susceptibility to ESC from a given chemical relates to swelling and a lowering of T_g. The degree of swelling that the chemical causes can be correlated to its solubility parameters and those of the polymer. That is, the closer the two solubility parameters are, the easier it is for the chemical to enter the polymer and make it swell. Regions of a part that are under high stress or strain tend to aggravate ESC because they absorb more of the chemical. Such effects can reduce T_g in a localized region of the part, often down to a temperature not far above ambient. This also reduces the critical strain needed to cause crazing because the amorphous regions in the material get softer, more flexible, and weaker at the T_g. Critical strains for many base resins can be as low as 0.1 to 0.2%

.

Therefore, ESC directly relates to how easily a chemical enters the polymer. Logically, one might expect that a chemical would be able to enter more freely if polymer molecules are stretched apart. And resins that form more-ordered crystalline regions create a tighter barrier to chemicals. Likewise, elevated operating temperatures dramatically reduce chemical resistance because T_g need not drop as much to reach the flexible state. It turns out that a polymer's degree of crystallinity is the most critical variable for chemical resistance. Temperature is also important, as is time of exposure and concentration. In addition, operating temperatures above the standard T_g make it easier for chemicals to attack because the polymer's coefficient-of-expansion is two to three times greater.

Other polymer variables affect chemical resistance. They included the polymer backbone, the degree of cross linking between the long polymer chains, and the stability of the bonds in the chain itself. For this reason, some amorphous materials will outperform crystalline resins in certain environments.

There are only a few laboratory tests that measure ESC. The most widely used involves introducing the chemical to specimens under constant strain or stress. Bending the specimen in different ways applies various strains. The stress and strain below which no crazes or cracks develop is taken as the material "strength" in the given environment. The trade-off with this test is the time to wait for cracking and the fact that the constant strain subjects the polymer to stress relaxation.

Abbreviated ESC tests are the norm for companies with short-development cycles. One such test employs the stress/strain test as described, but varies the temperature to make the situation aggressive enough to cause cracking within hours and days. The approach is to establish an Arrhenious equation from the data with the natural logarithm of time plotted against stress. This makes it possible to extrapolate time-to-crack at lower stresses and temperature. The technique offers a means of evaluating ESC without actual prototype testing for polymers such as unreinforced amorphous materials that craze relatively easily.