Stiffness measurement for evaluating heat resistance of rubber compounds.

Stiffness measurement for evaluating heat resistance of rubber compounds

Increasing requirements in the automotive and manufacturing industries have led to the need for more effective methods of determining the aging behavior of rubber compounds in different media and at different temperatures.

According to international standards, heat resistance testing in rubber compounds is normally based on a comparison of certain properties before and after aging. All standards and various publications suggest testing:

* tensile strength, ultimate elongation;

* modulus;

* hardness;

* tear strength; or different derivatives of these properties, such as:

* product of tensile and elongation;

* energy of disrupture (area under stress-strain curve);

* retention index of different properties.

The results of these tests are used to calculate heat resistance, which is a function of the properties before and after the aging test. Sometimes the absolute value of rubber properties is used, for instance elongation after a 1000 hour aging test.

Unfortunately, any laboratory testing of this type cannot be compared with the expected field life of a rubber part, which increases significantly year after year. For instance, in the automotive industry, the life of a rubber part currently has to be predicted for a period of ten (10) years or 9 x [10.sup.4] hours. In our studies, another approach was taken. We proposed to measure the kinetic of the rubber properties change, and to quantify the kinetic rate.

In order to do this, the first question to be answered was: What rubber properties are most responsible for long term life? We considered a number of popular test methods to determine this:

Figure 1 shows the tensile change for NBR compound with semi EV cure in SAE-30W oil at different temperatures from 70[degrees] to 177[degrees]C (158[degrees] to 350[degrees]F). At 70[degrees]C, the tensile strength did not change significantly during aging. When the temperature increased from 70[degrees] to 125[degrees]C, the magnitude of tensile diminished. In the range from 135[degrees] to 177[degrees]C, the behavior of tensile changed. As the temperature increased, so did the tensile. At 177[degrees]C, the tensile grew with the increase in the aging time

This is a typical picture of rubber aging, taking into account the fact that different rubbers will have different temperature limits.

This leads us to ask: What is better for rubber heat resistance, tensile increase or decrease? We were not able to find an answer to this question, either in the literature or in our own study. Therefore, we concluded that tensile strength change could not be used as an index for heat resistance evaluation. This conclusion relates to all equations which include tensile measurement.

The next test we considered related to hardness change. Figure 2 illustrates the hardness IRHD change for NBR at 125[degrees]C (257[degrees]F) in different oils. In some oils, hardness diminished after an initial period of time, then increased. In other oils, hardness increased after immersion without the initial diminution. The same question asked for tensile is applicable for hardness: What is better - when hardness increases or decreases? We were also unable to answer this question, and concluded that hardness change should not be used for heat resistance evaluation in media where rubber can swell.

Based on a comparison of how different rubber properties change in oil, the decision was made that ultimate elongation would be the best characteristic to test for loss of rubber elasticity.

In a previous publication (ref. 1), we described tests of rubber resistance or oil aggressiveness using the kinetics of elongation change. The following recaps the base results given in the previous publication. Figure 3 shows the elongation change of medium acrylonitrile NBR with semi-EV cure in SAE-30W oil at different temperatures. To quantify the kinetic rates of elongation loss due to aging, the following equation was used to describe the process: (1) E[degree]/E = exp ([bt.sup. 1/2]) Where:

E[degree] - unaged ultimate elongation, %

E - ultimate elongation after aging, %

b - regression index

t - aging time in hours

Using equation (1), straight lines are obtained when E[degree]/E vs. [t.sup.1/2] is plotted. The slope of "b" lines characterizes the kinetics of aging (figure 4). The HRI (heat resistance index) was quantified by the equation: (2) HRI = (1/b) x [10.sup.-1] Thus, the higher the value of HRI, the more heat resistant the compound.

Stiffness change measurement

The above described method of testing heat resistance is based on measurements of the ultimate elongation change, namely, the extension of the sample at ultimate deformations or deformations at the moment of sample disrupture. Values for these deformations can reach high levels, but under practical operating conditions deformations do not exceed 5 to 15%. Differences in rubber properties at "large" and "small" deformations are shown by many authors (refs. 2-4). The type of deformation critical for the change in rubber properties depends on the type of polymer, and especially on the type and amount of filler. At "small" deformations, change of the rubber structure occurs due to distortion of the junctions between the polymer and filler. At "large" deformations, disruption of the crosslinks occurs between the molecules and the polymer chains.

Figure 5 shows the stress-strain curves for two types of HNBR rubber compound. Type A compounds are used for low modulus applications; type B for high modulus. However, if we compare the area of small deformations, we see that characteristics of the compounds can change. At small deformation, the type B compound can be considered as the low modulus and the type A compound as the high modulus. The characteristics did not change when the gum rubber was changed from HNBR to NBR with the same type of filler system.

To test rubber properties at small deformations, a new device has been designed to measure stiffness in bending. This device consists of two metal parts, a yoke and a triangle anvil. Both metal parts have free rotational rollers and can be connected to any rubber tensometer. The anvil can fit into the triangular yoke space. To use the tester, a sample of rubber beam is inserted in the slit between the rollers, and the force required to bend the beam is measured between the two roller supports.

A typical deflection curve is shown in figure 6. There are two possible methods for calculating stiffness in bending. The first is to measure the slope of the tangent relative to the deflection curve, and calculate the modulus of elasticity in bending, using the following equation (ref. 2): (3) [E.sub.B] = [L.sup.3]m/[4wb.sup.3] Where:

[E.sub.B] - modulus of elasticity in bending N/m2 (psi)

L - support span, m (in)

w - width of the beam, m (in)

m - slope of the tangent to the deflection curve, N/m (lb/in)

The second method is to calculate the maximum bend stress or stiffness, using equation (ref. 3): (4) [M.sub.BS] = 3GL/[2wb.sup.2]

Where:

[M.sub.BS] - maximum bend stress or stiffness, N/m2

G - maximum load on the deflection curve, N (lb)

L - support span, m (in)

w - width of the beam, m (in)

b - thickness of the beam, m (in)

There is a linear correlation between the modulus of elasticity and the maximum stress. However, each compound shows its own slope. Three groups of curves can be distinguished. Each group characterizes compounds with different heat resistance. The higher the slope, the less resistant the compound is to aging conditions.

For operational conditions, it is necessary to calculate only one of the above mentioned characteristics, modulus of elasticity or maximum stress. For evaluation of the kinetics of heat resistance, we recommend using the maximum bend stress.

To calculate the heat resistance index using the kinetics of the stiffness measurement in bending, either equation (4) or (6) can be used: (5) [M.sub.BS]/[M.sub.BS][degree] = exp(bt) (6) [E.sub.B]/[E.sub.B][degree] = exp(bt) Where:

[M.sub.BS][degree], [M.sub.BS] - maximum bend stress or,

[E.sub.B, E.sub.B][degree] - modulus of elasticity before and after aging respectively, [N/m.sup.2] (PSI)

t - aging time, hours

b - regression coefficient 

In this case, the heat resistance index is quantified by the equation: (7) HRI = (1/b) x [10.sup.-2] Where:

"b" can be determined from equation (5) or (6).

Figure 7 shows the stiffness change for different rubber compounds in SAE-W30 oil at 135[degree]C. The test results obtained from the stiffness and elongation change exhibit the same order, but different magnitude of HRI.

The stiffness test's better reproducibility than elongation, and close resemblance to the deformation found in actual applications, make stiffness testing an effective method for evaluating rubber heat resistance.

Aeration effect

Any application of rubber parts is influenced by air oxidation. This statement is true even when the part is immersed in a liquid environment. In the case of rubber parts used in oil field applications, the interaction between rubber, oil and air becomes more complicated, because of additives in the base oil used to improve lubrication of metal parts. Previous publications (refs. 7 and 8) showed the importance of aeration of rubber performance in an oil environment. The concentration of oxygen dissolved in the oil can be calculated using the following formula: (8) [O.sub.2] = [gamma] * [P.sub.O2] Where:

[O.sub.2] - the concentration of oxygen, mol/liter

[P.sub.O2] - partial pressure of oxygen, Mpa

[gamma] - Henry coefficient, mol/liter x MPa For hydrocarbon liquids, [gamma] = 2.45 * [10.sup.-2] exp (3.8 RT) (ref. 9). It can be concluded from this equation that if the temperature increases, the amount of dissolved oxygen decreases.

In our study, we used the measurement of stiffness change test in oil with aeration. Figure 8 shows the influence of oil, air and oil plus aeration on HNBR compound. We can conclude that after long periods of immersion, the rubber stiffness increases significantly in oil-plus-aeration when compared to aging separately in oil or in air. Of course, the degree of oxidation would depend on the type of rubber, the type of oil and the temperature.

Figure 9 shows the stiffness change for different rubber compounds in SAE-30 oil at 135[degrees]C with aeration. Without going into a deep analysis of the influence of aeration on different rubber compounds, we can conclude that aeration did not affect FKM rubber, while at the same time it is very aggressive for NBR rubber. AEM, ACM and HNBR rubber have an intermediate position depending on compound formulation.

Figure 10 illustrates the correlation between the heat resistance index calculated fro the kinetics of stiffness change versus seal life. Comparison of the correlation coefficients calculated from the stiffness and elongation change shows a significantly higher correlation for stiffness measurement (0.932 versus 0.830 for elongation).

Network density measurement

The previously presented kinetics of network density measurement could also be used to calculate heat resistance index (ref. 1). In testing rubber based on identical types of polymers, or the same comparing rubber in different aggressive media, equation (9) can be used. (9) Vr/Vr[degree] = exp ([bt.sup.1/2]) Where: Vr[degree]/Vr - volume fraction of polymer in the equilibrium swollen sample before and after aging. This could be calculated as follows: (10) VR = F * [M.sub.d/gamma.sub.P] / f * [M.sub.d/gamma.sub.P] + ([M.sub.s -M.sub.d])/[gamma.sub.s]

Where:

f - weight fraction of polymer in compound

[M.sub.d] - weight of sample after drying solvent out

[M.sub.s] - weight of swollen sample

[gamma.sub.p] - density of polymer

[gamma.sub.s] - density of solvent

If different polymers were used, the network density measurement could be carried out using the swelling technique and the Flory-Rehner equation (ref. 5): [Mathematical Expression Ommitted] Where:

N - number of the effective network chains (network density, chains [cm.sup.3])

[V.sub.r] - volume fraction of the rubber in the swollen sample

[V.sub.o] - molar volume of the solvent

f - functionality of the crosslinks

X - polymer-solvent interaction parameter

The problem in using this equation is connected with the determination of the polymer-solvent interaction parameter (X). This parameter at low network density depends also on [V.sub.r]. However, for industrial rubbers, which are characterized by a high crosslink density, this readjustment can be ignored. In this case "X" can be obtained for all rubber independent of type or amount of curatives. In this case, the kinetics of the network density change is described by the following

equation: (12) N/N[degree] = exp ([bt.sup.1/2]) Where:

N - network density after aging

N[degree] - network density before aging

The heat resistance index can be calculated as follows:

(13) HRI = (1/b)[10.sup.-1]

In order to determine b, the plot of 1n (N/N[degree]) or 1n ([V.sub.r]/ [V.sub.r[degree]]) versus [t.sup.1/2] can be used in the same way as described above.

Figure 11 shows the kinetics of network density for NBR rubber with 28% of acrylonitrile in SAE-W30 at 125[degrees]C. the value mu = 0.49 has been taken from the literature (ref. 6). The correlation coefficient for both curves in 0.998, but the heat resistance index is different.

Heat resistance index comparison

The different methods of heat resistance index were compared using measurements of:

* ultimate elongation change,

* stiffness change in bending at small deformations and

* the network density for NBR compound in synthetic Mobile 1 oil at different temperatures.

These values are presented in figure 12. Due to the different magnitudes of the heat resistance index, the comparison has been performed where the HRI determined at 107[degrees]C (225[degrees]F) has been taken as one.

Figure 13 shows that the heat resistance index determined using the kinetics of stiffness change at the small deformation is more sensitive to temperature change. However, crosslink density and ultimate elongation can also be used for calculating the heat resistance.

Summary

The determination of the heat resistance index and measurement of the kinetics of stiffness change at small deformations is an effective method for defining the resistance of rubber to aging in different aggressive media. The bending stiffness test is the most sensitive method for determining a rubber compound's ability to maintain its elastic properties.

PHOTO : Figure 1 - tensile change

PHOTO : Figure 2 - hardness change in oils (NBR @ 125[degrees]C)

PHOTO : Figure 3 - elongation change

PHOTO : Figure 4 - calculation of heat resistance index

PHOTO : Figure 5 - stress-strain comparison

PHOTO : Figure 6 - load-deflection curve in bending

PHOTO : Figure 7 - stiffness change in SAEw30 @ 135[degrees]C

PHOTO : Figure 8 - aeration influence on stiffness (HNBR @ 135C)

PHOTO : Figure 9 - stiffness modulus change in SAEw30 @ 135 * + aeration

PHOTO : Figure 10 - heat resistance index vs. seal life

PHOTO : Figure 11 - network density change for NBR in SAEw30 @ 125[degrees]C

PHOTO : Figure 12 - heat resistance calculations for NBR in Mobil 1

PHOTO : Figure 13 - heat resistance calculations for NBR in Mobil 1

References

[1.] B.N. Dinzburg, R.W. Keller, R. Bond, "Heat resistance evaluation for rubber compounds," Rubber World, vol. 197, NS, 28-37, 1988. [2.] A.R. Payne," In: technology of reinforcement of elastomers," London, Plastics and Rubber Institute, 1975. [3.] G.M. Bartenev, A.M. Kuchershi, Rubber Chem. Technology, 1972, vol. 45, p.71. [4.] A. Voet, J.C. Morawski, Rubber Chem. Technology, 1974, vol. 47, p.765. [5.] P.J. Flory, J. Rehner, J.J. Chem. Phys. 1943, 11, 512-521. [6.] C.C. Pectov, V.A. Shershnev, I.D. Gabibulaev, M.N. Rudneva, Kauchuck i Rezina, 1985, 7, 17 (Russ.). [7.] R.W. Keller, SAE technical paper, 871626, Milwaukee, Sept. 1987. [8.] I.A. Abou-Isa, H.E. Trexler, Rubber Chem. Technology, 1985, 58, 326-349. [9.] E.T. Denisov, G.I. Kovalenko, Okislenia i Stabilisacia Reaktivnih Topliv, Chemia, 1983, p.269 (Russ.).