Rapid plasticity testing of silicones.
The simplicity of compression (parallel) plate plastimeters led to their early adoption and continuing popularity. The compression plate plastimeter consists of two parallel plates used to compress a sample with a constant applied force; the thickness reduction is measured after a set time. Ira Williams (ref. 2) first reported this type of plastimeter in 1924, and the Williams plastimeter is still widely used today. However, the Williams measurement took between three minutes and ten minutes, excluding sample preparation. For effectiveness, results had to be known quickly, and there was seen to be a need lot a rapid plasticity test. Experimental work was carried out in the 1940s to determine the optimum test conditions (for rubber), resulting in the manufacture of the first Wallace rapid plastimeter in 1951. This instrument provided a test that was quicker and less sensitive to operator error. The test pre-compresses the sample for 15 seconds to 1 mm (at a temperature of 100[degrees]C for natural rubber) and then further compresses the sample for 15 seconds under a load of 100 N. The thickness reduction gives a plasticity value. Although the instrument design has changed over the years, the principle has remained the same. In both instruments, the plasticity number is a measure of the final sample thickness, reflecting the degree of flow within the sample. In all industries, repeatability and the sensitivity of the instrument to slight compound variations are important. While these two plastimeters have been studied previously, much of the work relates to the natural rubber industry (ref. 3). Traditionally, the rapid plastimeter has been used for testing unvulcanized rubber, particularly raw natural rubber. This article studies the repeatability and sensitivity of these instruments with respect to the silicone rubber industry.
Standard Williams (ASTM D926-98 [ref. 4]) and Wallace rapid plasticity tests (ISO 2007 [ref. 5]) were carried out on a range of silicone rubber samples. Both the instruments used were designed and manufactured by H.W. Wallace.
For the Williams plastimeter (Wallace P2), two 5 [cm.sup.2] pieces of 12 [micro]m mylar film were used to set the zero position. A 2.0 [cm.sup.3] (approximately 16 mm diameter and 10 mm thick) sample of silicone rubber was cut from a sheet using a rotary cutter. The sample was placed between the two pieces of film and located between the plastimeter platens. The load was applied to the sample for three minutes. The final result was read from a dial gauge and converted to a WP (Williams plasticity) value. One unit represents a change in thickness of 0.01 ram. For the rapid plastimeter (Wallace P12E), 0.4 [cm.sup.3] samples were prepared using the standard constant volume sample cutter (mylar film was used to maintain consistency). The rapid plastimeter was operated semi-automatically with unheated platens. The sample was pre-compressed to 1 mm for 15 seconds, and then the load was applied for a further 15 seconds. The result (in rapid plasticity units) was digitally displayed. One unit represents a change in thickness of 0.01 mm (ref. 6).
Several tests were carried out on a range of typical silicone rubber samples to determine result repeatability, sensitivity and standard deviation. The sample mix was altered slightly (by 1%) to determine the sensitivity of both instruments to small compound variations. Standard Williams and rapid plastimeter samples were taken from both the original and the new mix and tested on the instruments.
A range of silicone rubber samples was tested on both types of instrument. The standard deviations were normalized to full range. While the samples tested on the Williams plastimeter were found to have slightly lower standard deviations than those given by the rapid plastimeter (table 1), the spread of results was found to be acceptable on both instruments.
The sample range was re-tested (with each sample having a slightly different mix as described above). Result averages were calculated and the percentage change in plasticity value before and alter the mix change was tabulated (table 2). It was found that, for three out of four samples measured with the rapid plastimeter, the percentage change was greater. Only the lower plasticity number sample showed a larger percentage change when measured with the Williams plastimeter.
The altered compound samples were also tested for repeatability, and the same trend for a small spread of results was noted on both instruments.
Other factors influence accuracy and repeatability, such as specimen preparation and instrument operation. The rapid plastimeter uses standard samples, produced in a few seconds with a constant volume cutter. A range of silicone rubber samples was prepared with the rapid plastimeter sample cutter and then weighed. The sample weights (nominally 0.5 g) were found to be consistent to within +0.005 g (i.e., [+ or -] 1%). In contrast, the Williams plastimeter uses samples prepared with a rotary cutter. In practice, a constant diameter is maintained, but the sample height varies. In order to obtain the correct sample volume (conforming to the standard), the sample's weight is adjusted by adding or removing small amounts of compound. Therefore, the accuracy depends on the operator and the balance used. The Williams sample preparation procedure can easily take two to three minutes.
The operator may influence the results when using both types of instrument. Several different samples were tested on both instruments by two different operators, and standard deviations were calculated from their results. In the majority of cases, the percentage difference between the two operators' standard deviations was greater for the Williams than for the same sample measured on the rapid plastimeter, indicating better repeatability between operators when using the rapid plastimeter.
Overall, rapid plastimeter repeatability was acceptable, but lower than expected. Previous work (ref. 7) indicated that, when testing natural rubber, the rapid plastimeter generally exhibited better repeatability than shown here. One factor highlighted in the previous work showed that the type of paper or film used affected the end result. Since mylar film was used on both instruments throughout the experimental work, some rapid plastimeter tests were repeated using Rizla paper (recommended in the natural rubber standard, ISO 2007 [ref. 51).
It was found that paper improved the standard deviation of the rapid plastimeter results. Paper use gave results with standard deviations that were comparable, or better than those obtained using mylar film. Different operators also experienced better repeatability when using paper in place of film (table 3). It was also noted that lower plasticity samples appeared to be more affected by the use of paper or film, tending to give similar standard deviations.
For the range of silicone rubber samples tested, it was found that the results obtained from the Williams plastimeter had a lower normalized standard deviation than those obtained from the rapid plastimeter (when using mylar). However, the spread was found to be acceptable on both instruments. When Rizla paper was used instead of mylar film on the rapid plastimeter, the standard deviations were found to be better than, or comparable to, those obtained from the Williams plastimeter.
While the rapid plastimeter is semi-automatic in operation, the Williams plastimeter result is more operator dependent, both in sample preparation and during measurement. Therefore, more care must be taken when using the Williams plastimeter.
One additional area of concern with the Williams plastimeter is the method of sample preparation. While the rapid plastimeter uses accurately prepared samples from special cutter pliers, the Williams uses samples of a constant diameter, with the correct volume attained by adding or subtracting small amounts of the sample.
It was found that the rapid plastimeter was generally more sensitive to sample mix variations than the Williams plastimeter. This is important when trying to discriminate between different samples. Additionally, if different operators obtain results, confusion can arise when small compound variations are observed on the Williams (these could be due to either the operator or the sample mix). Previous work (ref. 7) showed that when using a rapid plastimeter to test natural rubber samples, little variation between operator is seen.
While these tests were carried out at room temperature, testing may also take place at elevated temperatures. The rapid plastimeter uses electrically heated platens, allowing the sample temperature to be easily controlled. In contrast, the Williams plastimeter requires the entire instrument to be placed in an oven (sample temperature control is not straightforward or simple).
In addition, the sample used for the rapid plastimeter is only about 20% of the volume required for a Williams plastimeter. This is also an advantage when measuring samples at a higher temperature, since temperature equilibrium may be reached more quickly.
The measurements show that the spread of results between the two plastimeters is comparable for the range of silicone rubber tested. Several advantages in using the rapid plastimeter have been presented.
From the work carried out, it has been possible to see the effects of a change in sample mix and operator on the two plastimeters. The results indicate that for the range of samples tested, the rapid plastimeter is more sensitive to small changes in compound mix and less operator dependent than the Williams plastimeter.
Most importantly, testing can be carried out in a fraction of the time on the rapid plastimeter, a 90% time saving is realizable, once sample preparation time is included.
Table 1--result repeatability Sample Williams std. dev. Rapid plastimeter std. dev. 1 0.005 0.008 2 0.017 0.021 3 0.007 0.008 4 0.004 0.015 5 0.002 0.005 6 0.003 0.004 Table 2--sensitivity to compound change Sample % change between sapmle mix Williams value Williams Rapid (original) plastimeter plastimeter 1 684 43 54 2 454 35 60 3 317 5 7 4 254 21 11 Table 3--result repeatability between operators using mylar film and Rizla paper Sample Rapid plastimeter s.d. Williams Operator 1 Operator 2 plastimeter std. dev. Mylar Rizla Mylar Rizla 1 0.017 0.033 0.006 0.031 0.011 2 0.004 0.011 0.010 0.005 0.017 3 0.002 0.002 0.005 0.003 0.002 4 0.003 0.010 0.004 0.010 0.005
(1.) RABRM Bulletin, Special Issue on Plasticity, Vol. 8, No. 4 (1954).
(2.) I. Williams, Industrial and Engineering Chemistry, Vol. 16, No. 4, pp. 362-364 (1924).
(3.) Physical Testing of Rubber, R.P. Brown, pp. 85-98, second edition (1986).
(4.) ASTM D926-98. Standard Test Method for Rubber Property--Plasticity and Recovery (Parallel Plate Method).
(5.) ISO 2007: 1991, Physical Testing of Rubber, Part A59, Methods using plastimeters, Section 59.1, Determination of the rapid plasticity number.
(6.) B.M. McGarry, "Rapid plasticity and plasticity retention index testing," National Rubber Conference, India (1994).
(7.) R. Morgans, S. Lackovic, B. McGarry, G. Dinnage and B. Pearce, "Importance of experimental parameters on rapid plasticity testing for PRI," International Rubber Conference, India (1998).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Process Machinery|
|Date:||Nov 1, 2003|
|Previous Article:||Polyurethane consumption dips slightly.|
|Next Article:||Third generation metallocene EPDMs.|