Physical properties and their meaning.

Rubber/fluid compatibility and swelling

Rubber/fluid compatibility is an important consideration during manufacture of rubber products. A fluid or plasticizer will bloom to a rubber surface if the solubility limit of the fluid is exceeded. Bloom can cause problems such as poor building tack. sticky surfaces and poor adhesion.

A useful indicator of rubber/fluid compatibility and swelling behavior is the solubility parameter [Delta]. Close values of [Delta] for a fluid and a rubber indicate compatibility. Values of [Delta] for several fluids and rubbers are shown in table 3 (ref. 52).

Table 3- values for several rubbers and fluids

Rubber   [Delta of rubber]        Fluid          [Delta of fluid]

EPDM          7.5(*)          Paraffinic oil                  7.5
SBR           8.54            Aromatic oil                    8.9
NBR           9.25            Ester or highly      8.9 or greater
                                aromatic oil

(*) estimated

It should be emphasized that o is only an indicator of compatibility and swelling. Other factors such as hydrogen bonding also affect compatibility.

Swelling behavior of a rubber compound is important after it is vulcanized and subsequently exposed to fluids. Swelling is especially important in the case of rubber seals like o-rings, where a large volume of fluid often contacts a small o-ring. To seal effectively. o-rings are designed to swell very slightly in such fluids.

In the case of an automotive engine mount, the volume ratio of fluid to rubber is generally extremely low. A few drops of engine oil may fall on a mount. This small amount is unimportant, even though the oil may be quite compatible with the rubber (often NR) in the mount.

Another factor with mounts is the penetration rate. This rate is affected by factors such as temperature and oil viscosity. For example, an SAE 40 motor oil will penetrate one mm (0.04 in.) into an NR vulcanizate in four weeks. It may take 100 years to penetrate 40 mm (1.6 in.) (ref. 53).

Among methods to assess swelling, ASTM D 471 is probably the most widely used. It determines both swelling and deterioration properties of rubber caused by exposure to liquids.

Electrical properties

The use of rubber in electrical applications places demanding requirements on rubber properties. Often, these applications require a combination of electrical properties and other requirements that are difficult to achieve. Other requirements often include specific mechanical properties, along with ozone and aging resistance. Four important electrical properties are: resistivity (its reciprocal, conductivity), dielectric strength, dielectric constant and power factor (ref. 35).

Resistance and conductivity measurements are made by ASTM D 257 using direct current. When surface conductivity is negligible compared to conductivity through a specimen, volume resistivity is measured by ASTM D 991. Volume resistivity is the ratio of the electric potential gradient to the current density when the gradient is parallel to the current in a material. This electrical property varies widely. For example, the volume resistivity of certain NBR compounds varies by more than nine decades (from about 10 to [10.sup.10] ohm-cm) (ref. 54).

Dielectric strength is determined by ASTM D 149. It is defined as the ratio, dielectric breakdown voltage/thickness of insulating material.

Dielectric constant and power factor are measured by ASTM D 150. Dielectric constant is the ratio, [C.sub.x]/[C.sub.v], and [C.sub.x] and [C.sub.v] are measured by a given set of electrodes, where:

* [C.sub.x] is determined with an experimental material as the dielectric;

* [C.sub.v] is determined using a vacuum as the dielectric but air may be used for most purposes.

Power factor is the ratio, power dissipated in a material/the product of effective sinusoidal voltage and current.

An extensive list of references for electrical properties is appended to ASTM D 150.

Nondestructive testing (NDT)

NDT methods detect and measure discontinuities or anomalies in materials, components or assemblies. They do this without impairing the serviceability of the object tested (ref. 55). While NDT methods are directed presently toward specific rubber products, more general use of these methods is expected in the future. NDT techniques are currently being evaluated for, or applied to, products like tires and engine mounts. For tires, infrared, holographic, microwave and ultrasonic techniques are used (ref. 56). For example, monitoring of the surface of running tires identifies hot spots in a tire.

Tire surface displacement is measured by holography. A newer technique, shearography, is claimed superior to holography as it more easily identifies defects (ref. 57). With both of these techniques, tires are stressed by placing them in a vacuum chamber so that belt edge separations can be identified.

Microwave is used to penetrate tires and detect changes in physical properties (ref. 56). By ultrasonic testing, ultrasonic waves travel a short distance through water where they then enter a submerged tire. Reflected ultrasonic waves are used to detect dimensional irregularities in a tire.

An ultrasonic test also detects voids at the interface in bonded, rubber/metal, engine mounts (ref. 58). During test, a stress (tension) of at least 70 kPa (10 psi) is applied to the bonded part. An ultrasonic probe is then moved over the exposed metal surface, opposite the bonded surface, to detect voids. This technique can be visualized by referring to figure 21, where a probe could move over the flat exposed steel surfaces of the stressed adhesion specimen.

[Figure 21 ILLUSTRATION OMITTED]

This ultrasonic test detects only voids in engine mounts. It does not measure bond strength. Destructive tests are run on random samples of mounts to determine bond strength.

Computers and rubber properties

Computers are involved presently with many aspects of rubber properties and computer use in these activities is growing rapidly. Computers are used in such varied applications as experimental design, property optimization, control of test and process variables, and automatic data logging.

Experimental design replaces traditional experimental methods in which one factor, e.g. sulfur level, is varied one at a time. With experimental design, levels for each factor are distributed according to a design table and the desired data generated (ref. 13). Properly used with a table, equations are developed for the data. These equations are then used to determine response values between data points where experiments were not run. This approach is very efficient and reduces the amount of experimentation.

Some experimental designs permit optimizing a given rubber property while other properties are constrained within limits. Say for example tear strength is lower than a specification permits; other properties: tensile hardness, etc., are within specification limits. Assume that compositional changes are made according to a proper experimental design and the resulting data are entered into a computer. Tear strength could be optimized while constraining other properties within specification limits. Additional experiments will be necessary if tear strength is still outside specification limits.

Computers also control processing and testing. A dedicated microcomputer system controls a strip-winding process to build up tire treads (ref. 59). Computer control makes it economical to increase the number of test conditions imposed on bonded rubber parts during test (ref. 60).

Often computers are dedicated to specific test equipment where they record a variety of data. Advantages include:

* rapid data capture, e.g. from high-speed impact tests;

* significant savings in labor;

* orderly presentation of data;

These many advantages and capabilities of computers assure their widespread use in the future.

Property correlations - laboratory/end use

Ideally, rubber properties determined in the laboratory would correlate directly with properties determined for the same compositions in end-use or service applications. Unfortunately, lack of correlation is often the rule. This lack of correlation is no doubt the reason that ASTM methods, such as D 865 (accelerated aging in heat), are considered comparative only. These methods are quite useful. They provide preliminary data that can be used later to control properties of rubber compositions for end-use products.

There are a number of reasons for the lack of correlation of laboratory properties with service properties. Only several are considered here. Laboratory test specimens often consist only of rubber. End-use products are often complex composites of rubber and fabric, or rubber and metal. Common examples are tires and reinforced hose. The design of these products might affect product performance more than the rubber compounds used in them. For instance, considerably higher stresses might occur in a product in service than the stresses that occur in laboratory tested specimens.

Conditions experienced in service might not occur in the laboratory. For example, antiozonants are commonly added to rubber compounds to protect them from ozone attack. The protected compounds are tested at high ozone levels in the laboratory. The compounds are not exposed to acid rain in laboratory tests as they might be when exposed outdoors (ref. 61). Acid rain is a potentially important variable because it can leach antiozonants (generally these are alkaline). Hence, a factor can reduce ozone resistance that was not included in laboratory tests.

Another problem with determining rubber properties is hysteresis (ref. 60), which causes increased temperature in rubber parts when they flex. Some rubber properties change considerably with changes in temperature. Carried to an extreme, hysteresis can cause premature part failure. It might even cause failure by a mechanism different than the one that determines the life of a part in service. Hence the severity of strain cycles must be limited in some tests to avoid excessive hysteresis.

Properties of full-scale tires are commonly measured by indoor tests because test conditions can be carefully controlled (ref. 11). Of course, this is an advantage. A disadvantage is that a wide variety of conditions that occur in service are not experienced in the indoor test, such as different climactic conditions and variations in road surfaces. For these reasons outdoor testing is needed to finally prove products like tires.

Sometimes correlation is excellent between properties determined in the laboratory and in service. An example was cited in the section on creep. Settling (creep) of a building mounted on natural rubber bearings has exactly followed the predictions of short-term laboratory experiments (ref. 40).

As greater insight is gained into rubber behavior and properties, improved correlation is expected between behavior in the laboratory and end-use applications.

Acknowledgement

John Sommer gratefully acknowledges the assistance of GenCorp in the presentation of this chapter.

References

[11.] J.R. Beatty, "Physical testing," presented at the Tenth Annual Lecture Series (Akron Rubber Group), Akron, OH, February 5, 1973.

[13.] F.S. Conant, "Physical testing of vulcanizates," chapter 5 in Rubber Technology, M. Morton, Ed., Van Nostrand Reinhold Company, Second Edition, 1973, p. 114.

[35.] A.E. Juve, chapter 19, "Physical testing" in Introduction to Rubber Technology, M. Morton, ea., Reinhold Publishing Corp., New York, 1959, p. 462.

[40.] C.J. Derham, Paper F in "Rubber in engineering - 1973 Conference," Natural Rubber Producers Research Association, p. F/1.

[52.] M.L. Studebaker and J.R. Beatty, "The rubber compound and its composition, " chapter 9 in ref. 17, p. 367.

[53.] E. Southern, Rubber Developments, 18, 51 (1965).

[54.] W.A. Hartz, D.A. Meyer and J.G. Sommer (to The General Tire & Rubber Company), U.S. Patent 3,620,901 (November 16, 1971).

[55.] H. Berger, Rubber Chem. Technol. 54, 996 (1981).

[56.] F.J. Kovac "Tire manufacturing and engineering, " chapter 14 in ref. 17, p. 569.

[57.] Y.Y. Hung and RM. Grant, Rubber Chem. Technol. 54, 1042 (1981).

[58.] Based on a presentation by S. Peer, European Rubber Journal 156, 42 (1974).

[59.] R.E. Kazares and J.D. Becker, "Computer technology applied to process control," Abstract in Rubber Chem. Technol. 51, 369 (1978).

[60.] R.L. Clinard, Industrial Research and Development 24, 130 (1982).

[61.] M. J. Nix, Plastics, Paint and Rubber 19, 22 (1975).

General references

R.P. Brown, "Physical testing of rubbers," Applied Science Publishers Ltd., London, 1979,

L.R.G. Treloar, "The physics of rubber elasticity," Clarendon Press, Oxford, Third Edition, 1975.

John Sommer, an educator and consultant, is president of Elastech, Inc., a firm specializing in elastomer technology. His publications on elastomers include 30 technical papers and book chapters, in addition to 16 U.S. patents. He is a registered professional engineer in Ohio.