Dynamic testing using oscillatory shear and dynamic compression.

Traditional dynamic testing of cured rubber samples has been performed in cyclic compression using dynamic mechanical analyzers and Goodrich Flexometers. More recently, dynamic mechanical rheological testers measure these properties in conjunction with viscosity and curing tests. This article focuses on a comparison between the three methods, and how using newer types of testing can add efficiency and provide equivalent information. Compounds with varying properties will be tested in order to evaluate a wide range of viscoelastic behavior.

The necessity of testing after-cure dynamic properties on test samples has grown in the last twenty years. Dynamic mechanical analyzers (DMAs) (ref. 1) which can work in various modes, and specialized dynamic mechanical rheological testers (DMRTs), which are generally used for processability and cure testing, have seen an increased usage for the testing of after-cured properties (refs. 2 and 3). The continuing need to create more innovative products, while reducing waste and production costs, has created the necessity for laboratory sophistication. At the same time, it is important to perform rapid tests on materials before assembling a final product, in order to attend to individual material properties and requirements. A one-instrument solution has long been desirable, yet the complexity of filled elastomers requires a range of test procedures and equipment.

Rubber compounds possess various responses to dynamic forces. Tires may have more than ten different compounds ranging from a higher modulus tread to a lower modulus inner liner. The dynamic properties of these compounds are crucial to the tire's performance. While incoming specifications can be relied on to understand a polymer's properties, a filled compound's modulus and loss factor will vary substantially from the unfilled polymer. A filler's particle size, loading and morphology greatly influence dynamic properties, including hysteresis (ref. 4). The range of viscoelastic properties in a tire presents a challenge not only to tire development, but to useful quality control procedures.

Tensile, modulus, elongation and resilience tests, including the Lupke Resilometer, Goodyear Healey Rebound and the Yerzley Oscilligraph, were initially used to help quantify a cured compound's properties (ref. 5). Static tests such as compression set and stress relaxation were also employed. The fatigue mode of tires, however, could not be fully understood by these tests alone. To better understand the hysteresis of a compound, the Goodrich Flexometer, which dynamically cycles a specimen in compression, was developed. Heat buildup, as an indicator of hysteresis, and blow out were measured with the flexometer. Later, more sophisticated tests could be performed using thermal analysis equipment. This gave the ability to test over a wide range of temperatures, frequencies and strains, depending on the type of thermal analysis. This related to tire properties such as rolling resistance, fatigue and tire decay (ref. 6).

Thermal analysis, and in particular thermal mechanical analysis, subjects a specimen to a sinusoidal strain and measures the resultant stress. Therefore, in compression, the complex dynamic modulus E *, the elastic modulus E', the loss modulus E" and tangent delta (E"/E') are reported. By utilizing these data and looking at the elastic modulus and loss factor (tangent delta), an indication of elasticity and hysteresis can be studied (ref. 7).

With this need for a dynamic test to measure elastic and loss modulus, Goodrich Flexometers have been instrumented to determine these properties, in addition to measuring heat buildup and blow out. While other methods discussed here provide some flexibility advantages, the Goodrich Flexometer has the practical advantage of being relatively simple to operate and cost-effective.

DMAs use the mechanics of applying a known strain to a specimen and measuring the complex, elastic and loss modulus. DMAs have a wide range of functionality, and are used in all industries. The DMA can work in compression, shear, tension, torsion or bending. A typical temperature range from -100[degrees]C to 400[degrees]C makes it possible to create temperature profiles which represent aging, glass transition and performance. A frequency range of less than 1 Hz to greater than 1,000 Hz, and a strain range of up to 300%, makes it possible to analyze samples in a very detailed fashion, replicating the intensive service conditions for a tire. An example of a use for the DMA is testing at sub-ambient temperatures to relate results to snow and ice traction. It is also used to test degradation over extended periods of time. Tests are done for days, or even weeks, to look at fatigue and crack growth, as well as other types of degradation. An important function for the tire industry is the DMA's ability to run a cyclic test with an overlay of an additional pulse frequency. This test can closely mimic tire stresses.

DMRTs also use the sinusoidal analysis, as discussed above, but operate in torsional shear rather than compression. The DMRT holds a small specimen in a heated, sealed cavity under pressure. Unlike the Goodrich Flexometer, a DMRT measures elastomeric properties from plastic flow, curing and finally after-cured dynamic properties. The test begins with an uncured specimen, making product sectioning impossible. The clear advantage of this type of testing is the ability to look at a wide variety of compound properties, during various chemical states, in one relatively fast test (refs. 8 and 9).

A further evolution of the DMRT is the "event driven test" (ref. 10). An event driven test monitors state of cure in real time, which allows test conditions to be changed at optimal times in order to achieve test sensitivity and constant state of cure. An example of an event driven test, as it applies to the testing of after-cured properties, is the use of a slope threshold to end the cure portion of a complete test. Traditionally, the end of the curing section is determined by time alone. The time to completion is generally taken from a trial and error process, and normally a long test time is set in order to guarantee complete cure. The variability of rubber compounds makes this method subjective, as a consistent state of cure is not used to end the cure section and start an after-cure test. With the event driven test, the slope threshold guarantees that there is not under- or over-cure (ref. 11).

The following experiment will explore the use of three instruments to obtain dynamic property values of selected tire compounds. A further look at the utilization of rapid tests and their correlation to more time-consuming, but scientifically accepted, tests will be evaluated. The article will look at the possibilities of providing relevant tangent delta values in a reliable and easy-to-interpret method.

Experimental

Three tire compounds, shown in table 1, were chosen to provide a range of viscoelastic properties. The general polymer type and percentage used is also shown in table 1. Prior to testing, the preparation and handling of the compounds included: The compound was milled to 0.2 cm and allowed to cool. Where cured specimens were applicable, they were press cured at 160[degrees]C for 20 minutes, quenched in water for 30 minutes and were allowed to rest 24 hours before testing.

The typical operational ranges of the instruments used are given in table 2. The following tests were performed on each compound:

* Dynamic mechanical analysis (DMA)--An MTS Model 831 with TestStar II software was used. Button specimens 17.8 mm in diameter and 25.4 mm height were press cured. The specimen was loaded between the instrument platens and the dynamic test performed in compression at 70[degrees]C, 10% strain, [+ or -] 2% DSA and 10 Hz. The test time was 10 seconds. The median of three measurements was reported. Results are given in tables 3 and 4.

* Goodrich Flexometer heat buildup (HBU)--Button specimens 17.8 mm in diameter and 25.4 mm height were press cured. The specimens were pre-conditioned in the Flexometer oven for one hour at 50[degrees]C. A 25-minute test was then performed on the sample at 50[degrees]C. The test was done at 30 Hz with a static load of 25 kg. The temperature was measured continually from the bottom platform of the flexometer. At the completion of the test, the temperature was recorded. Results are given in table 3.

* Goodrich Flexometer DMT--Button specimens 12.5 mm in diameter and 25 mm thick were press cured. The specimens were pre-conditioned in the Flexometer oven for one hour at 70[degrees]C. A test was run in cyclic compression at 70[degrees]C, 10% strain, [+ or -] 2% DSA and 10 Hz. The median of three measurements was reported. Results are given in tables 3 and 4.

* DMRT--The rheoTECH MDPT was used for these tests. An event driven cure test was run on each sample, with the test results shown in figure 1. The cure tests were run at 160[degrees]C, 2% strain and 1.67 Hz. When the slope of the cure curve achieved a slope threshold of 0.005, the test temperature was automatically changed to 100[degrees]C. When the instrument had stabilized at this temperature, the post-cure test was run at 2% strain and 10 Hz. Results are given in tables 3 and 4.

[FIGURE 1 OMITTED]

* Rebound--The rebound was measured with a Zwick Rebound Tester, Model B5109. Button specimens 29 mm in diameter and 12.5 mm in height were press cured at 160[degrees]C for 20 minutes. Tests were run in accordance with DIN 53512 (ref. 12).

Modulus values were recorded for all dynamic tests, and the results given in table 4. An analysis of the correlation between tests is shown in table 5.

Discussion and conclusion

All of the tests that were run for the reported experiment were done on equipment that is standard to a rubber research laboratory. An effort was made to evaluate relatively simple-to-perform tests, to see if a correlation existed between the methods. As the results in table 5 point out, all of the tests show very strong correlations. The validity of each of these tests to work for a wide range of tangent delta properties was confirmed.

When considering a type of test to run for after-cured dynamic properties, it is then important to consider other relevant factors. Advantages exist for all of the above methods.

The rebound test correlates well with the tangent delta values obtained by the other three methods. This test requires the same sample preparation as the DMA and Flexometer. The advantage of the test is that it is simple, fast and cost-effective. The disadvantage is that the test conditions are limited, and dynamic modulus information is not available.

A DMA has a wide range of abilities to analyze material characteristics in tension, compression, shear, torsion and bending. Only with a DMA can properties which require extremely large temperature and frequency ranges be properly measured. The DMA tests, however, are often time consuming, requiring 16-96 hours conditioning and test preparation.

The Goodrich Flexometer provides a versatile, yet relatively simple set of tests, which can be used to examine heat buildup, blow out and the dynamic properties, as described in this experiment. The Goodrich Flexometer also has the sample preparation complication of the DMA, and can only operate in compression. The Flexometer and the DMA have the ability to work on a sectioned portion of a tire or a press-cured sample.

The dynamic modulus and tangent delta as measured by both the DMA and Goodrich Flexometer are equivalent.

The benefit of the DMRT test is the ability to evaluate flow characteristics, curing data and after-cure dynamic properties in one rapid test. The value of G * is, as expected, less than that of E *, due to testing in shear rather than compression (ref. 13). However, the values of tangent delta from the DMRT were similar to those obtained on the DMA and flexometer. The correlations are shown in table 5.

The DMRT test is inherently a fast way to measure the curing and after-cure properties. This specific test was further optimized by use of the cure slope threshold. This method allowed each cure portion of the test to be performed to exactly the same state of cure as determined by the cure slope threshold. When this threshold was met, the test conditions were changed to allow the post-cure testing. This is all done in one test without operator intervention and without the need to estimate the expected cure time of each specimen. This can be seen in figure 1. For both the DMA and Flexometer tests, 30 minute rheometer tests were performed to determine the cure time for the test specimens. In the case of the DMRT, the specimens were cured to a time determined by the slope threshold and were, hence, optimized for each specimen.

[FIGURE 1 OMITTED]

Summary

Dynamic properties of rubber can be as simple as considering the rebound characteristics of a compound or as detailed as measuring the viscoelastic properties of elastic modulus, loss modulus and tangent delta. Traditional DMA test results of E *, E', E" and tangent delta correlated well with the results obtained from a fully instrumented Goodrich Flexometer. In addition to the values of E *, E' and E", the Flexometer can also provide heat buildup and blow out information. The use of the DMRT test provided additional details regarding the curing of a specimen and, in the same test, information about the dynamic properties. The tangent delta values from the DMRT correlated well with the DMA and Flexometer; whereas the modulus values were less, as expected, since the DMRT measured in shear rather than compression. Finally, the DMRT test was optimized by measuring the curing characteristics until a cure slope threshold was met. Upon meeting this threshold, the test conditions were automatically changed to evaluate the after-cure properties.

The choice of testing instrument for dynamic properties should be made with consideration to the required range of test and/or application conditions, the type of data needed and the time constraints of the test.

References

(1.) ASTM D 5992-96, "Dynamic testing of vulcanized rubber and rubber-like materials using vibratory methods," ASTM International, West Conshohocken, PA, Vol. 9.01 (2003).

(2.) ASTM D6204-01, "Standard test method for measurement of unvulcanized rheological properties using rotorless shear rheometers, "' ASTM international, West Conshohocken, PA, Vol. 9.01 (2003).

(3.) ASTM D5289-95 (2001), "Standard test method for rubber property--vulcanization using rotorless cure meters, " ASTM International, West Conshohocken, PA, Vol. 9.01 (2003).

(4.) M.J. Wang, "Effect of polymer-filled and filler-filler interactions on dynamic properties of filled vulcanizates, "Rubber Chemistry and Technology, vol. 71, p. 522, (1998).

(5.) R.F. Ohm, "The Vanderbilt Rubber Handbook," R.T. Vanderbilt Publishers, (Norwalk, CT 1990), p. 531.

(6.) D.J. Burlett, "Thermal techniques to study complex elastomer/filler systems," Journal of Thermal Analysis and Calorimetry, vol. 75 (2004), p. 533.

(7.) A.I. Medalia and G Kraus, "Reinforcement of elastomers by particulate fillers. "Science and Technology of Rubber, J.E. Mark, B. Erman and F.R. Eirich, ed., (Academic Press, 1994), p. 406, 407.

(8.) J.B. Putman and M.C. Putman, "A simplified approach to QC testing. "Rubber World, vol. 229, January 2004.

(9.) J.S. Dick, "Dynamic aging of vulcanizates using the rubber process analyzer," ACS Rubber Division paper #96, (Oct. 2000).

(10.) J.B. Putman and M.C. Putman, "Event driven DMRT analysis, "A CS Rubber Division paper #53, (May 2005).

(11.) J.B. Putman and M.C. Putman, "Optimization of DMRT testing for cured dynamic properties," ACS Rubber Division paper #17, (Oct. 2004).

(12.) DIN Standard DIN53512 (1983) "Determination of the rebound resilience of rubber using the Schob pendulum," Berlin, Germany.

(13.) Richard W Hertzberg, "Deformation and fracture mechanics of engineering materials," John Wiley and Sons, New York, NY, Fourth Edition.

Matthew C. Putman, Tech Pro, and Gary Bradnick, Kumho

Table 1--description of compounds

A   Low tan [delta]      Under tread   85% NR/15% BR
B   Medium tan [delta]   Rim flange    50% NR/50% BR
C   High tan [delta]     Race tread    100% SBR

Table 2--typical range of instrument operating conditions

             Temperature
               range,      Frequency    Strain
             [degrees]C    range, Hz     range

DMRT            60-230      0.1-45      0.1-125%
Flexometer      23-110        5-40      0-8.25 mm
DMA           -100-300      .01-400      0-30 mm

                     Mode of strain
             Shear    Compression     Tension

DMRT          Yes          No           No
Flexometer    No          Yes           No
DMA           Yes         Yes           Yes

Table 3--results--tan [delta], heat buildup and rebound

                  tan [delta]

       DMRT       Flexometer        DMA
    tan [delta]   tan [delta]   tan [delta]

A      0.079         0.067         0.065
B      0.179         0.168         0.170
C      0.361         0.463         0.476

           Flexometer

    Heat buildup,   Comp.
      [degrees]C     set    Rebound, %

A         22         1.3       66.8
B         55         3.1       44.0
C        134        32.0        9.2

Table 4--results--dynamic modulus

            DMRT

    G *,    G',     G",
    kPa     kPa     kPa

A   1.631   1.626   0.129
B   4.211   4.145   0.743
C   3.570   3.358   1.213

    Flexometer

    E *,    E',     E",
    kPa     kPa     kPa

A   4.439   4.429   0.295
B   8.605   8.486   1.424
C   7.821   7.094   3.294

            DMA

    E *,    E',     E",
    kPa     kPa     kPa

A   4.715   4.705   0.306
B   9.683   9.548   1.612
C   7.712   6.969   3.303

Table 5--correlation of results

                                                       Correlation
Comparison of:                                        coefficient, R

DMRT tan [delta] to Flexometer tan [delta]                 0.994
DMRT tan [delta] to DMA tan [delta]                        0.994
DMRT tan [delta] to Flexometer heat buildup                0.998
DMRT tan [delta] to Flexometer comp. set                   0.954
DMRT tan [delta] to rebound                               -0.999
Flexometer tan [delta] to DMA tan [delta]                  1.000
Flexometer tan [delta] to Flexometer heat buildup          0.999
Flexometer tan [delta] to Flexometer comp. set             0.981
Flexometer tan [delta] to rebound                         -0.988
DMA tan [delta] to Flexometer heat buildup                 0.999
DMA tan [delta] to Flexometer comp. set                    0.981
DMA tan [delta] to rebound                                -0.988
DMRT G * to Flexometer E *                                 0.998
DMRT G * to DMA E *                                        0.987
Flexometer E * to DMA E *                                  0.974
DMRT G" to Flexometer heat buildup                         0.952
Flexometer E" to Flexometer heat buildup                   0.996
DMA E" to Flexometer heat buildup                          0.987