Testing elastomers at low temperatures

As the operating temperature reduces, the polymer chains in elastomers become closer together, and their ability to move becomes restricted. In practical terms, when they cool beyond a certain point they become ‘leathery’, remaining elastic, though with a poorer response. This continues to worsen as they approach the glass transition temperature (Tg), which is the point where they cease to retain any elastomeric properties, and become ‘glass like’. At a specific temperature materials will reach their brittle point, where seals subjected to shock loadings can potentially fracture.

In order to provide transparency when comparing the performance of different compounds we have used two common laboratory test methodologies: Temperature retraction and Gehman torsional modulus. Whilst these tests do not provide absolute figures for performance in application, they do offer a basis for the accurate comparison of low temperature capability between different materials. From a sealing viewpoint, there are a number of key values that need to be considered. These include:

• Glass transition temperature (Tg). We compare values using differential scanning calorimetry (DSC). Other techniques are available.
• Temperature retraction (TR), where a sample is stretched and frozen in an alcohol bath, then slowly warmed and the temperature at which it recovers by 10% noted. This is referred to as the TR10 temperature
• Gehman torsional modulus, where a sample is cooled in an alcohol bath and twisted using a calibrated wire from which the torsional modulus can be calculated. The temperature at which the torsional modulus reaches 70 MPa is known as the ‘T70’ temperature and is often considered the limit of its technically useful flexibility.

These tests offer a degree of accuracy and are less susceptible to the flexibility of interpretation of other low temperature performance assessments.TR10 and T70 values can both be used to predict the minimum operating temperature of an elastomer in terms of responsiveness. Although they measure different properties (stretching and twisting), they generally give values that are quite close together, and just above the Tg.

The TR10 and Gehman T70 values give a good guide to a general minimum material operating temperature.

So, in addition to using standard laboratory tests described above, James Walker also assess functional low temperature sealing capability using a variety of techniques. The most common utilises a standard O-ring type seal in a test fixture.

Low temperature sealing is a complicated subject and is dependent on many variables. These results only relate to the product in the format and conditions tested, and offer no accurate guidance to performance in application where the media, pressure, product design, housing design and housing surface finish will all have a bearing on performance.

It is our belief that, when selecting and specifying materials, relying solely on published low temperature claims utilising undisclosed test methodologies is unwise and potentially dangerous. The only true method to determine whether the low temperature sealing capability of a material will be suitable, is to undertake product configured tests under conditions as close as possible to those found in application.