Exposure of Test Pieces

Test pieces were exposed in air exchange ovens complying with ISO 188 [7] for a series of times and temperatures. The temperatures used for a given material were selected hearing in mind the known heat resistance of the material and the limitation that the longest exposure (lowest temperature) would he 6 months. The temperatures are shown on the individual graphs of property change with time. 

For all properties except compression set, 5 batches of test pieces were prepared for each compound which allowed a maximum of five temperatures to be used. The plan was to aim for a minimum of 4 temperatures per material. Each batch consisted of 8 sets of 3 dumbbells, a piece of sheet for hardness measurements and 8 strips (10 mm wide and 150 mm long) for DMTA tests. 

In 1964, the boxes were moved from Cairns to Innisfail which is 60 miles along the coast and which has a similar climate. 

Natural Ageing of Rubber - Changes in Physical Properties Over 40 Years 

The storage at Shawbury represents a temperate climate and because the location was a laboratory the conditions were relatively constant. The temperature range was 18-25°C which covers the two standard laboratory temperatures in use over the 40 years, 20 2°C and 23 2°C. Humidity was not controlled but was generally 50 10% with a total range of about 35-80%. 

The two Australian sites were in Queensland at what was known as the Joint Tropical Science Unit (JTSU). At both sites, the boxes were stored in well ventilated buildings protected from direct sunlight and rainfall. 

Cloncurry is about 250 miles from the nearest coast and is representative of hot, dry conditions. Over the years 1963 to 1998 the average daily minimum, mean and maximum temperatures were 18.8°C, 25.6°C and 32.4°C, respectively. In 8 months of the year, temperatures of over 40°C were recorded. The average daily minimum, mean and maximum relative humidities were 29%, 46% and 63%, respectively. It dipped below 10% in every month. Microbiological activity is reported to be insignificant, and in summer months, the site is subject to windblown dust. 

---

A general procedure for determining the effects of immersion in liquid chemicals is given in ISO 175. This covers exposure in the unstrained condition. The test pieces are immersed in the liquid held in (usually) a glass container. The volume of liquid should be at least 8 ml per square centimetre of test piece surface. The test pieces should be completely immersed and with all surfaces freely exposed, for example, by hanging on hooks. 

Whatever the other considerations may indicate, it is a simple fact of life that in virtually all trials the selection of properties will be affected by the cost and convenience of the experimental requirements. There are enormous differences between different properties in the cost of test piece preparation, testing time, number and size of test pieces, and apparatus requirements. In accelerated exposures the availability of exposure space is very frequently the limiting factor. 

In ageing trials the most important considerations are the numbers of test pieces, exposure times and levels of the degradation agent, together with ensuring that the test pieces used are representative. Ideally, the number of experimental points should be maximised but in practice cost considerations lead to restrictions. 

Initially, 5 of the 8 sets were exposed at each selected temperature and the first set tested after 3 days. Further exposure times were then selected, adding sets of test pieces if necessary, with the aim of obtaining useful results at a minimum of 5 exposure times with the longest time yielding approximately a 50% change in properties. Test pieces were conditioned for a minimum of 16 hours at 23 °C prior to test. 

Velocity effects can be achieved either by having the test-piece move through a presumably stationary liquid or by having a moving liquid come into contact with a stationary test-piece. Occasionally tests may involve both types of exposure. Details of test procedures are given in NACE TM 0270-70 Method of Conducting Controlled Velocity Laboratory Corrosion Tests. 

Since ozone attack on rubber is essentially a surface phenomenon, the test methods involve exposure of the rubber samples under static and/or dynamic strain, in a closed chamber at a constant temperature, to an atmosphere containing a given concentration of ozone. Cured test pieces are examined periodically for cracking. 

The position and angle of the test pieces to the sun (for both radiation and temperature) is important. A wide range of exposure methods and angles are available at most test stations. The more conventional and widely used are ... 

The construction of exposure racks will influence the effects of the environmental factors, such as the temperature of the test pieces. 

To simulate service conditions tests may be needed with exposure being on one side of the test piece only, which can achieved by using a simple jig in which the test piece forms one end of a cylindrical container. If immersion under pressure is needed a special jig would have to be developed. The standard does not include cases of partial immersion nor immersion under pressure. 

Natural weathering (see Section 5.2.5) can be accelerated directly by exposure in a climate more severe than that expected in service. There are established test sites for this purpose in Australia and in the hotter states of the USA. The severity of exposure can also be maximised by arranging for the plane of the test pieces to automatically follow the sun. Another approach is to use a Fresnel mirror concentrating device that increases the intensity of sunlight falling on the test piece. These procedures are standardised in ISO 877 [25]. 

The economical use of material, the almost non-destructive nature, cheapness and simplicity make hardness measurements very attractive for monitoring surface degradation. A single test piece can be used for successive ageing periods, improving reproducibility and reducing the space required for exposure. 

Further uncertainty arises from the environmental exposure. For accelerated tests, repeatability can be estimated from exposure of replicate test pieces and minimised by control of the exposure conditions. Particular points to consider are spatial variation in temperature as well as mean temperature and air flow in ovens. In accelerated weathering apparatus, spatial variation and variation with time of light sources can be very significant. 

The exposure can be either in air or a liquid chosen to simulate service conditions. Commonly, ring test pieces are chosen for liquid exposure, so simulating the geometry of practical seals and giving a relatively large surface area to volume ratio so that equilibrium swelling is reached reasonably quickly. It should be noted that the swelling effect of the liquid will affect the relaxation pattern measured and an increase in stress may be seen over a limited time period if there is a volume increase. 

At the end of the test exposure period, the test piece is allowed to cool under one of three alternative methods. In method A, the preferred procedure, the strain is released and the test piece allowed to recover for 30 min at ambient temperature in method B the recovery is at the test temperature for 30 min followed by 30 min at ambient temperature, whilst in method C the strained test piece is cooled for 30 min at ambient temperature, the strain then released and the measurement made after a further 30 min. The preferred recovery procedure corresponds to the traditional one for compression set and is a marked change from earlier versions of the standard where only method C was specified. Set is expressed as a percentage of the applied extension. 

The procedure for exposure of a test piece to a liquid on one side only is applicable to relatively thin sheet materials which are exposed this way in service. A suitable jig is used to contain the liquid and the change of weight measured. The result is expressed as change in mass per unit surface area. 

The normal procedure for measurement of change after exposure is to test immediately after removing the test piece from the liquid. However, ISO 1817 does allow the alternative procedure, which cannot be expected to give the same results, of drying the test pieces to constant mass at 40°C and at reduced pressure, recondition at standard laboratory temperature and then test. The question to dry or not to dry should be answered on the basis of the relevance to service and quite possibly both figures would be of interest. 

Because antiozonants and waxes, which to be effective must form a surface bloom, are used to enhance ozone resistance it is usual to condition test pieces in the strained state before exposure. The usual conditioning period is between 48 and 96h and the test pieces should be kept in the dark and in an ozone-free atmosphere. For this treatment to be effective, the test piece surface must not of course be touched in the course of subsequent handling. Where specifications wish to specifically exclude compounds which rely on an adequate wax film for protection, the conditioning period is dispensed with. Hill and Jowett47 in a criticism of ozone test methods strongly make the point that the conditioning process should be relevant to service conditions if a discriminating evaluation of waxes is to be made. 

Most specifications give a set strain and exposure period but it is preferable to examine test pieces at a series of times such that data can be obtained on the relationship between strain and time to appearance of cracks. ISO 1431 requires examination to be carried out with a lens of x7 magnification but, unfortunately, any examination of cracks is to some extent dependent on the eyesight of the operator. In practice, many workers say a crack is only a crack if they can see it with the naked eye. The alternative procedure of measuring relaxation in stress will be discussed later. An optical method of automatically detecting cracks has been described by Zeplichal51 but this is relatively complicated and has not been considered for standardisation. 

---