The sensitivity of frictional characterization of elastomers and elastomeric composites.

There is a countless variety of elastomers and elastomeric composites working in friction-related field applications. They differ in composition, construction, processing, design and other variables. Physical and mechanical properties of products would change dramatically depending on each of these variables. To make the problem more complicated, the actual application conditions are hardly predictable as well. As a result, the exact descriptions of material properties and product applications are hard to achieve. However, at the same time, there is still a need for robust and sensitive express test methods to describe the elastomer behavior in a last, effective and cheap way suitable for material benchmarking, selection and optimization.

It is a well known and well recognized fact that friction is one of the major factors of service behavior for any product when material contact is involved. Friction phenomena can be quantitatively characterized by numerous laboratory tests, including standardized and custom designed methods. Poor sensitivity of tests to application parameters dictates the necessity of designing a wide variety of friction-related methodologies. Traditionally, there are groups of friction-related tests designed for each product application. This study does not target a detailed review of all products and corresponding test methods, however just to touch on the subject, it would be interesting to list a few.

One of the most friction-involved products is a tire, where one of the main performance characteristics is tread-life. Therefore, it is essential for tire development to be able to accurately describe the condition of the worn surface of a tire, as well as the frictional properties of the materials involved. There are multiple ASTM procedures that cover wear and friction measurements for tires. One, ASTM F 408 (ref. 1), determines wet traction, braking coefficient, sliding and other friction related parameters. This test method is stated to be "suitable for research and development purposes and not suitable for regulatory statuses and specification acceptance because the values obtained may not necessarily agree to correlate or rank traction performance level with those obtained under different conditions (surface, environmental, etc.)." This is a good example of a test's low sensitivity to input parameters.

Another example of a product working in friction mode is conveyor belts, where wear and friction are measured differently. The overview of the wear-resistant elastomeric materials for mining industries (ref. 2) stresses the importance of understanding the physical properties of elastomers by proper testing. One of the numerous traditional tests is ASTM D 2228 (ref. 3), known as pico abrasion, widely used in the conveyor belting industry. A rubber sample is rotated under loaded tungsten carbide knives at controlled speed, time and force, and the amount of wear is reported. However, it is noted in the pico test that "even though the test may be used to estimate the relative abrasion resistance of different rubber compounds, no correlation between this test and service performance is given or implied, due in part to widely varying nature of surface conditions." It has been proven for the conveyor belt industry that known standard and custom abrasion tests often do not show a correlation with belt field performance because of low sensitivity to test parameters (ref. 4).

The last example of high wear resistant products is soles and heels of footwear, with a wide variety of standard tests developed for that application. Numerous footwear abrasion test machines were suggested through the years (ref. 5). However, most tests have sensitivity limitations, such as environmental, surface and sample-geometry conditions, and do not correlate with other tests (ref. 5).

Based on the above short review of existing methods to quantify friction, it is obvious that there is no one single method sensitive enough to satisfy all friction-related applications.

The objective of this phenomenological work is to present a method of material characterization that is based on some sort of friction contact that mimics the frictional performance of elastomers and elastomeric composites. This test must be sensitive to major field application parameters, like load, velocity or contacting media shape, as well as environmental conditions. The test should be applicable for purposes of material benchmarking, rating, selection and cost optimization.

Concept of characterization

This article suggests a phenomenological approach to friction characterization. It states that a product is changing somehow through its field life. Processes that occur within the product are complicated and cannot be analytically described. However, some field application (or input) parameters, such as load, temperature, media shape, etc., are known. As a result of the input parameters' influence onto the product, final field property parameters (or output), such as product performance, strength, wear rate, cost, etc., are formed.

To apply the phenomenological approach to friction characterization with high sensitivity, a new dynamic test is introduced. The test involves a combination of loading and torsion properties. An experimental approach is proposed in this study to mimic a specific mode of contact behavior (ref. 5).

Contacting objects are presented in the test as simple axisymmetric bodies (indentors). All selected indentors are general-purpose abrasive points (ref. 6) with different shapes of stones, including spherical, conical, flat, etc. Examples of indenter geometry are shown in figure 1.

[FIGURE 1 OMITTED]

An indentor is compressed into a fixed specimen, defined as a flat piece of a specific elastomer with known thickness. The compression may be controlled by either force or displacement. A torque load is applied to the indenter and is recorded as a function of angular displacement and compression. Resistance to the torque is assumed to be due to friction at the contact surface between the indenter and elastomeric specimen.

The test program has been described (ref. 5) and is performed in this study by means of a torsional MTS dynamic machine (ref. 7). An elastomer slab with constant thickness is placed under a rotating indentor. Thickness of elastomeric composites corresponds to actual thickness of products. The indentor is pressed into the elastomeric sample in the direction perpendicular to the specimen surface down to a selected level of load, given in terms of axial force, F, or axial displacement, [DELTA]. After reaching the selected load or displacement and holding it for 10 seconds, the indenter is rotated in load control mode at a selected rate of angular speed, [omega]. The load (torque T), displacement (angle [theta]) and time are recorded each 0.5 seconds. The total rotation angle is set for 200[degrees].

The typical outcome of the test is schematically described in figure 2. There is a peak of maximal torque, which may be quantified by [T.sub.max] and [[theta].sub.max]. This peak represents a classic static frictional resistance corresponding with the effort required to move one surface over another from a static position as it is described in literature (for example, ref. 8). Values [T.sub.a], show the torsional resistance to rotational movement and may be used to quantify the classic dynamic friction, or the resistance to move one surface over another at a given velocity.

There were two groups of materials considered in this study, including four general-purpose production elastomers (1, 2, 3, 4) and two elastomeric composites (A, B) with fabric reinforcement. Elastomers are based on blends of natural rubber (NR) with synthetic rubbers at various contents, with carbon black (CB) fillers. Tested composites had one and four reinforcement layers of polyester and/or nylon fabric of different density (figure 3), with cover rubber compounds selected from regular general purpose rubbers based on synthetic and NR blends, with CB fillers.

[FIGURE 3 OMITTED]

Sensitivity of frictional characterization

Effects of indentor shape, axial load, test temperature and torsional speed were investigated for selected materials, with results presented in figures 4-8.

[FIGURES 4-8 OMITTED]

The effect of indentor shape represents the impact of differently shaped objects contacting a product, for example similar to the difference of impact of a sharp stone versus a round stone compressed into a shoe sole or dropped onto a conveyor belt.

Indentors of cone, sphere and flat sphere shapes were loaded at axial force into a representative rubber material (NR with sulfur cure system and carbon black fillers) and turned 200[degrees] against it at room temperature (RT). Resulting torsional resistance (N*mm) is shown in figure 4. The shape of produced curves is similar to the traditional trend (as figure 2). The peak of maximum torque corresponding with static friction is well represented in figure 4a.

There is a significant difference in torsional resistance created by the different shapes of the indentors. For example, at an axial force of 300 N, the round cone indentor causes material to resist torque at 1,000 N*mm, while the sphere shape indentor produces about 750 N*mm (figure 4a).

The next aspect in the testing is to investigate the effect of axial loads applied to the same representative rubber material. This test would duplicate, for example, the effect of stones of various weights dropped on a selected material from equal heights. As expected, the increase in axial load resulted in a significant gain of torsional resistance for all shapes of indentors. Torsional resistance more than doubled with a load increase from 300 N to 700 N (figure 4). It is interesting to note that the rating of material resistance to selected shapes of indentors changes with load increase. For example, the torsional resistance of the material to the spherical indentor was slightly lower than its resistance to the flat sphere shape indentor at a lower axial load, but the rating switched at higher loads. This result indicates the importance of proper selection of laboratory test parameters in order to appropriately mimic each field application.

The influence of temperature on material frictional properties was studied as well, in order to model the changing environmental temperature in field applications. For example, the torsional resistance of the same material to a flat shaped indentor was tested at room temperature, 60[degrees]C and 80[degrees]C and at 300 N, 500 N and 700 N axial load (figure 5). The torsional resistance of the material decreased with increasing temperature, definitely as a result of rubber softening. The same trend is observed for cone shaped indentors (figures 6 and 7).

The effect of speed was previously studied using one elastomeric material (ref. 5), where a very consistent dependency was noted for torsion values recorded at 300 N axial force: The torsional resistance values increased with increasing speed. In the present work. the effect of changing the speed from 1[degrees] per second to 2[degrees] per second was studied at a higher axial force of 700 N for multiple materials, including four rubber compounds (1,2,3 and 4) and two elastomeric composites (A and B), (figure 8). The torsional resistance ranking of the rubber compounds did not change with the speed increase, and the higher speed did correlate with higher torque results for all tested materials.

Material benchmarking.

The relative rating of any two compounds (benchmarking) can depend on the conditions of the test, as was stated back in the 1950s by A. Schallamach (ref. 9). Benchmark tests are needed in order to rate and compare materials and products. In this study, the four representative rubber compounds and two composites were tested at room temperature by indenting a sphere shaped stone at different speeds, and their torsional responses were rated (figures 8a or 8b). It is important to note the significant differences between the results for different materials due to the high test sensitivity to material properties. These results illustrate the importance of benchmarking at specifically identified test parameters in order to avoid misleading results in material ranking.

Role of energy component

All previously discussed tests were performed using a force-controlled axial compression mode, representing a situation where the weight of the penetrating object causes a contact stress: for example, a conveyor belt carrying a rock. In some field applications, the deformation is essential, such as a seal under fixed compression. That means that the lab test must be programmed in displacement control mode. And, finally, some applications are orchestrated by the energy factor, for example a shock absorber or a mount performance. For these cases, there is an attempt to combine load- and deformation-control modes into one energy component-controlled test, and then compare the rating of torsional resistance obtained at load versus the rating from an energy component-controlled test.

In the first part of this experiment, the four rubber compounds and two composites were tested at three axial loads at room temperature as presented in figure 9. The sphere shaped indentor was loaded into a sample, turned 200[degrees] at 1[degrees] per second velocity and completely released. This loading was repeated three times at axial force 300 N, 500 N and 700 N for a total 600[degrees] rotation (200[degrees] repeated three times). As in previous tests, there was a significant difference in torque for the representative materials. The general torque rating shows material #4 to be highest, followed by #1, #3 and #2. The torque values of the two composites were quite close. It is interesting to note the clear appearance of a classic static friction maximum on the charts obtained at 300 N load, versus the monotonically growing torque at higher loads.

[FIGURE 9 OMITTED]

The second part of the experiment was to determine the energy component for each material by running load-deflection curves using the sphere shaped indentor at 10 N/sec. compression rate at room temperature (figure 10). The area under the curve was calculated for each material and assigned a name of energy component E (N*mm). Loads corresponding with three levels of energy components were determined for each material (figure 11).

[FIGURE 10-11 OMITTED]

Then the first part of the experiment was repeated, but the applied loads were not the same for all materials. Instead, the axial force was individually determined for each material corresponding with three levels of energy component: 700 N*mm, 1,000 N*mm and 1,500 N*mm, as described in figure 11. The obtained torque values are presented in figure 12. The ranking of the materials' torsional resistance was completely changed, with compound #3 and composite B now espousing the highest torque as a result of changing the control mode of the test from load (figure 9) to energy component (figure 12). However. note that some materials seem to maintain their rating: for example, compound #2 and composite A are still on the low side of the torsional resistance level.

[FIGURE 12 OMITTED]

This is a valuable illustration of how important it is to properly select the testing control conditions in accordance with a specific field application of the part.

Conclusions

A method of material characterization was proposed and developed for elastomers and elastomer-based composites to emphasize their unique frictional performance.

Based on the method, a family of material parameters was selected to quantify frictional performance under complex dynamic loading and environmental conditions.

High sensitivity of proposed parameters was observed with respect to loading conditions (pressure, velocity), environmental conditions (temperature), contact pattern (shape of indentor) and material composition (different elastomers and composites).

Sensitivity of the proposed material parameters may be considered as an important feature for possible phenomenological benchmarking of elastomers for specific products and applications.

The benchmarking may be considered as an accelerated express method for elastomeric selection and optimization based on available limited information of field performance.

References

(1.) ASTM F408-94. Standard Test Method for Tires for Wet Traction in Straight-Ahead Braking, Using a Towed Trailer, Annual Book of ASTM Standards, vol. 09.02.

(2.) I.R. Sare, J.I. Mardel and A.J. Hill, "Wear resistant metallic and elastomeric materials in the mining and mineral processing industries--an overview," Wear, vol. 250, 1, 2001.

(3.) ASTM D2228-88. Standard Test Method for Rubber Property--Abrasion Resistance (Pico Abrader), Annual Book of ASTM Standards, vol. 09.01.

(4.) M. Scherbakov, etc., "Extending conveyor belt life by improving abrasion resistance," proceedings of Society of Mining Engineers meeting, February 25-27, 2002.

(5.) M. Scherbakov and M.R. Gurvich. "An approach of frictional characterization for elastomers and elastomeric composites," J. Elastomers and Plastics, vol. 35, number 4, October 2003, 335.

(6.) Mounted Abrasive Points, McMaster-Carr Supply Company, (2000). Catalog 102, Cleveland, OH, p. 2,377.

(7.) Model 810, MTS Systems. (2001). www.mts.com, Eden Prairie, MN.

(8.) ASTM D1894-01. Standard Test Method for Static and Kinetic Coefficients of Friction of Plastic Film and Sheeting, Annual Book of ASTM Standards, vol. 08.01.

(9.) A. Schallamach. Friction and Abrasion of Rubber Wear, vol. 1, 1957/1958, 384.