Vibration isolation characteristics, fatigue properties of chemically modified solution polymerized rubber blended with NR.

Vibration isolation characteristics, fatigue properties of chemically modified solution polymerized rubber blended with NR

The performance of automobiles has improved year after year, particularly the riding comfort of passenger cars, which is largely dependent upon engine mount and tire performance.

During the early 1980s active development was undertaken on new polymers for tires to provide a better balance between low fuel consumption and safety. The development began with the introduction of vinyl groups to polybutadiene (refs. 1 and 2). Also improved SBRs were developed with controlled styrene content; vinyl groups on the butadiene portion; their sequence distribution of these groups; and the molecular weight distribution (refs. 3 and 4). A technique was developed whereby solution polymers were chemically modified by adding a polar compound, such as 4,4'-bis(diethyl amino)benzophenone (EAB), to the end of the molecular chains. It was reported previously (ref. 5) that these elastomers have a good balance of properties for tire tread use. This chemical modification has been applied to both polybutadiene (BR), and solution polymerized styrene butadiene rubbers (S-SBR).

Vibration isolation characteristics of the chemically modified rubbers were also reported at the 135th Rubber Division meeting (ref. 6).

In this article, dynamic properties of the chemically modified SBR blended with NR are discussed, and also the effects of fatigue testing as well as the dynamic properties of chemically modified BR blended with NR are reported.

Basic characteristics of rubber engine mounts

The rubber engine mount has three basic functions. The first is to support the weight of the engine upon the chassis of the automobile, thus the rubber must have a proper range of static spring constant Ks. The second is to isolate the transmission of engine noise to the body. The third is to dampen road shocks and/or quickly stop engine shake caused by resonance when idling.

The relationship between the three functions of a rubber mount and their basic dynamic characteristics are given in table 1. The second and third functions are especially important and depend primarily upon the visco-elastic properties of the base polymer. The rubber must satisfy the following requirement to fully comply with these two functions.

First, so as to isolate engine noise transmission to the body, the stiffness ratio of the elastomer, that is the ratio of the dynamic spring constant at high frequency (Kd at 100-200 Hz) to the static spring constant (Ks), must be sufficiently low.

Second, the loss-factor of the rubber compound at low frequency (tan [Delta] at 5-30 Hz) must be high so as to dampen the engine shake quickly.

These two contradictory properties are in a trade-off relationship, as may be seen in the transmissibility equation (1) (ref. 7). (1) [Mathematical Expression Omitted] This equation was derived using the linear visco-elasticity theory of a single degree of freedom vibratory system shown in figure 1.

Needless to say, in actual fact the dynamics of an engine mount are not those of a simple linear system, especially with carbon black filled compounds. It has been often reported in the literature (ref. 8) that the well known Payne effect has established the non-linearity of strain amplitude dependence of dynamic properties.

The relationship between these two properties in conventional elastomers are illustrated in figure 2. Plots of tan [Delta] of various rubbers versus their stiffness ratios are linear. The arrow on figure 2 indicates the area of desirable properties in this region.

We reported previously that the chemically modified solution-polymerized rubbers exhibit more desirable dynamic properties for use in engine mounts, compared to conventional rubbers (ref. 6).

The fatigue behavior, as well as the dynamic properties, are also very important characteristics for engine mount rubbers. Thus the effect of a fatigue test on dynamic properties for the modified rubber blended with NR was studied.

Experimental

Materials

The rubbers employed in this study are mainly solution polymerized styrene-butadiene and polybutadiene rubbers whose molecular characteristics are shown in table 2.

Samples 1 to 4 in table 2 are S-SBR with chemical modification to their molecular chain ends. Their Tg were adjusted to about -30 [degrees] C by varying the micro-structure of the butadiene portion and the styrene contents. Sample 5, prepared for comparison, has a little lower Tg than samples 1 to 4.

Sample 6 is a lithium catalyzed, low vinyl polybutadiene with chemical modification to the molecular chain ends. Sample 7 is a standard cobalt catalyzed high cis-1,4 BR.

Test samples for the dynamic tests using the formulation in table 3, were mixed in a laboratory internal mixer and press cured at 150 [degrees] C.

Test procedures

The dynamic tests were made by uniaxial compression of a cylindrical test piece 25 mm in diameter and 25 mm in height (JIS N 1 type). The test apparatus was a dynamic servo type KC-V (Saginomiya Seisakusho) with a maximum dynamic force capacity of [+ or -] 2,000 Kg, maximum amplitude of [+ or -] 25 mm, and a frequency range of 5 to 1,000 Hz.

The dynamic properties were obtained by measuring the dynamic spring constant Kd and loss-factor tan [Delta]. The Kd was measured at a frequency of 100 Hz, static pre-compression strain of 8% (2.0 mm), dynamic compression strain amplitude of [+ or -] 0.2% ([+ or -] 0.05 mm) and a temperature of 25 [degrees] C. The tan [Delta] was measured at a frequency of 15 Hz, static pre-compression strain of 8% (2.0 mm), dynamic compression strain amplitude of [+ or -]2.0% ([+ or -] 0.5 mm) and a temperature of 25 [degrees] C.

The static spring constant Ks was obtained from the compression modulus at a static strain of 1 to 2 mm using the same samples used for the dynamic tests.

The strain amplitude dependence of the dynamic properties were obtained by measuring the shear modulus G' at a strain range from [+ or -] 0.2% to [+ or -] 20% and a frequency of 30 Hz using a cylindrical test piece 25 mm in diameter and 10 mm in height (JIS S2 type) on the dynamic servo.

The cyclic fatigue tests were carried out by the dynamic servo at a shear strain of [+ or -] 30% ([+ or -] 3 mm), a frequency of 10 Hz and a temperature of 25 [degrees] C.

The temperature dependence of the dynamic properties were measured by a dynamic analyzer type RDA (Rheometrics, Inc.) at a shear strain of [+ or -] 0.5%, at a temperature range from - 90 [degrees] C to + 100 [degrees] C and a frequency of 15 Hz.

Results and discussion

The effects of micro-structure of S-SBR in NR blends on dynamic properties

The physical and dynamic properties of the S-SBR vulcanizates are shown in table 4. The relationship between the stiffness ratio and tan [Delta] characteristics of rubbers designed for engine mounts are illustrated in figure 3. In this figure, the conventional unmodified rubbers, indicated by [diamond], are plotted on the solid line. The chemically modified rubbers, indicated by [circle], follow the broken line in this figure.

It may be seen that the chemically modified rubbers are clearly shifting into the desirable direction indicated in figure 2, that is a lower stiffness ratio and higher tan [Delta] relative to the conventional rubbers.

It was reported previously (ref. 6) that this shift was caused by the decrease in the strain amplitude dependence of the dynamic spring constant Kd due to the more stable carbon black dispersion and by the increased interaction between the rubber and carbon black.

The relationship between tan [Delta] and Tg of S-SBR samples number 1 to 5 is shown in figure 4. It is evident that the tan [Delta] value at room temperature increased with increased Tg.

The physical and dynamic properties of the vulcanizates of the NR/SBR (60/40) blends are shown in table 5. The relationship between the stiffness ratio and tan [Delta] characteristics of the NR/SBR blends are illustrated in figure 5. In this figure, the conventional unmodified rubbers, indicated by [diamond], are plotted on the solid line. The chemically modified rubbers, indicated by [circle], and the NR/SBR blends, indicated by *, follow the broken line.

Although the tan [Delta] values of S-SBR increased with increased Tg in figure 5, some tan [Delta] values of the NR/S-SBR blends exhibited the behavior that did not conform with the additive effect of those blends ratios. In particular, the NR blend in sample 4 showed a very low stiffness ratio and tan [Delta] value.

To clarify this characteristic behavior of NR/S-SBR, their dynamic property dependence on temperature was examined. The temperature dependence of dynamic-shear modulus G [prime] of NR/S-SBR-1 blend is shown in figure 6 in comparison with 100% NR and S-SBR-1. The NR/S-SBR-1 blend has two transition points, and this result suggests that the NR/SBR-1 blend is an immiscible blend.

The temperature dependence of the dynamic-shear modulus G [prime] of NR/S-SBR-4 blend is shown in figure 7 in comparison with 100% NR and S-SBR-4. In comparison to compound 1, the NR/S-SBR-4 blend had a single transition point at an intermediate range between those of 100% NR and S-SBR-4. In this case, it is evident that the NR/S-SBR-4 blend is miscible.

It is evident that there is a significant difference in behavior of miscibility between the two S-SBR samples having the same Tg. This is caused by the difference in vinyl contents of the butadiene portion and the styrene contents, as shown in table 2. It was reported (refs. 10 and 11) in the literature that high vinyl polybutadiene rubbers are miscible with NR over a wide range of blend ratios. Since S-SBR-4 has a high vinyl content and lower styrene content than S-SBR-1, it was concluded that the S-SBR-4 was more miscible with NR than S-SBR-1.

The characteristic behavior of dynamic properties of the NR/S-SBR blends shown in figure 5 can be explained by comparing the G [prime] values of NR/S-SBR blends at room temperature in figures 6 and 7. In the case of an immiscible blend (the NR/ S-SBR-1), the G [prime] value at room temperature is higher than that of the 100% S-SBR-1. On the other hand, the G [prime] value of the miscible blend (the NR/S-SBR-4) is a little lower than that of the 100% S-SBR-4.

The fatigue behavior of modified BR with NR blends

The fatigue test samples of BR/NR blends were prepared using the formulation shown in table 6. The tensile properties and dynamic characteristics of the original samples are shown in table 7. Strain amplitude dependence of dynamic shear modulus (G [prime]) and loss factor (tan [Delta]) of the original samples are illustrated in figure 8 (a) and (b). The chemically modified BR/NR blend, indicated by O, is plotted on the solid line, and the conventional unmodified BR (high cis-1,4 BR)/NR blend, indicated by [diamond], follows the broken line.

It can be seen that the dynamic shear modulus and tan [Delta] of the chemically modified BR blend is less dependent upon the strain amplitude than those of the unmodified BR blend. These results conform to the work of Boonstra and Medalia (ref. 9), in which the strain amplitude dependence on dynamic shear modulus changes as the mixing time of an SBR with a fixed carbon black content is extended. It was explained by Payne as being due to the better carbon black dispersion with longer mixing time (ref. 8).

On the other hand, our work was carried out at a constant mixing time and fixed carbon black contents. The chemically modified BR caused the decrease in strain amplitude dependence of dynamic shear modulus (G [prime]) and tan [Delta] due to the dispersion stabilization effect of the carbon black in the compounds.

Changes of dynamic shear modulus (G [prime]) with rest time, after the fatigue test of 10,000 cycles shear strain, are shown in figure 9. The dynamic shear modulus (G [prime]) of unmodified BR blend rose immediately after the fatigue test during the initial rest time, whereas the rise of G [prime] for the modified BR blend was relatively small. These results may be explained as follows:

The unmodified rubber blends form "a transient structure" or "quasi-network" of carbon black particles in the rubber matrix, which are broken by the cyclic shear strain of the fatigue test. However, when the cyclic shear strain is removed, the "transient structure" is reformed immediately.

In the case of the modified rubber blends, the structural change of the carbon black particles is less than in the case of the unmodified rubber blend because of the dispersion stabilization effect of carbon black due to the chemical modification.

The different values of the dynamic shear modulus at [+ or -] 0.5% strain amplitude ([G [prime].sub.0.5]) versus those at [+ or -] 20% ([G [prime].sub 20]) measured one hour after the fatigue test for BR/NR blends, are plotted in figure 10 against the number of strain cycles given by extending the fatigue test.

The difference between the limiting value at low strain amplitude ([G [prime].sub.0]) and the limiting value at high strain amplitude (G [prime] [infinity]) may be referred to as [Delta] G [prime], andis well known as Payne's "filler-filler linkages effect."

The [Delta] G [prime] ([G [prime].sub.0.5] - [G [prime].sub.20]) of the chemically modified BR blend always showed a lower value than that of the unmodified BR blend over the complete range of the fatigue test. These results suggest that there is very little change in the state of carbon black dispersion during the fatigue test for the modified BR blend. On the other hand, in the case of the unmodified BR blend, the [Delta]G [prime] maintains a higher value throughout the fatigue test, and it is suggested that the "transient structure" or "quasi-network" of carbon black particles does not disappear after the fatigue test, even though the structure is momentarily broken by the cyclic shear strain of the fatigue test.

Conclusions

It was found that the dynamic properties of chemically modified S-SBR/NR blends may be attributed to their miscibility. In the case of an immiscible blend (NR/S-SBR-1), the G [prime] value of the blend is higher than that of 100% S-SBR-1), SBR-1, and the stiffness ratio of the blend increases. Conversely, the G' value of the miscible blend (NR/S-SBR-4) is a little lower than that of 100% S-SBR-4, and the stiffness ratio of the blend is lower in the desirable direction for an engine mount. The [Delta]G' value of the chemically modified BR blend was lower than that of unmodified BR blend over the whole range of the fatigue test. It was evident that there was little change in the state of carbon black dispersion during the fatigue test for the modified BR blend. [Tabular Data 2, 4 and 5 Omitted]

Table 1 - requirements of engine mount rubber materials

Basic functions         Quality requirements of
of engine mount         rubber mounts
Support the weight      Proper static spring constant
of the engine             Ks = 7   15x[10.sup.4] N/m
Isolate engine noise    Lower dynamic stiffness ratio
transmission              (Kd/Ks) at 100   200 Hz
Damp out road shock     Higher loss-factor
and shake when idling     (tan [Delta]) at 5   30 Hz

Table 3 - test formulation (wt. phr)

Polymer        100
Carbon black    40
Aromatic oil    15
ZnO              5
Stearic acid     2
Sulfur           2
CBS            1.2

Table 6 - test formulation (wt. phr)

NR              60
BR              40
Carbon black    50
Aromatic oil    15
ZnO              5
Stearic acid     2
Sulfur         1.5
CBS            1.5
TMTM           0.5

Table 7 - dynamic properties of NR/BR

Sample no.                            6       7
Tensile strength (MPa)               16.5    18.5
Elongation (%)                       410     390
Hardness (JIS)                        64      63
Static spring                        12.5    12.5

constant Ks x [10.sup.-4] (N/m)

Dynamic spring 21.0 17.8 constant Kd x [10.sup.-4] (N/m)

Stiffness ratio Kd/Ks                1.68    1.42
Loss factor (tan [Delta])           0.099   0.072

PHOTO : Figure 1 - vibration of a simple single-degree-of-freedom system

PHOTO : Figure 2 - dynamic stiffness ratio vs. tan [Delta] of conventional rubbers

PHOTO : Figure 3 - stiffness ratio vs. tan [Delta] of modified [circle] and unmodified [diamond] rubbers

PHOTO : Figure 4 - tan [Delta] vs. Tg of modified rubbers

PHOTO : Figure 5 - stiffness ratio vs. tan [Delta] of NR/S-SBR 60/40 blends

PHOTO : Figure 6 - temperature dependence of G [prime] for NR, S-SBR-1 and NR/S-SBR-1

PHOTO : Figure 7 - temperature dependence of G [prime] for NR, S-SBR-4 and NR/S-SBR-1

PHOTO : Figure 8 - strain amplitude dependence of G [prime], tan [Delta] (NR/modified BR [circle], NR/high cis BR P [Diamond])

PHOTO : Figure 9 - G [prime] at [+ or -] 0.2% strain amplitude vs. rest time after 10,000 fatigue cycles

PHOTO : Figure 10 - changes of [Delta] G [prime] by extending the fatigue tests

References

[1]S. Akita, A. Ueda and Y. Todani, Annual meeting of the Society of Rubber Industry Japan, (1981) A. Ueda, H. Watanabe and S. Akita, Int. Rubber Conference, IRC85, Kyoto, Japan, Oct. 15-18 (1985). [2]K.H. Nordsiek and K.M. Kiepert, Int. Rubber Conference, D4 (1981). [3]L.H. Krol, Int. Rubber Conference, A3 (1981). [4]K.H. Nordsiek, Kaut. u. Gummi Kunst., 38, 178 (1985). [5]K. Noguchi, A. Yoshioka, K.Komuro and A. Ueda, the Rubber Division, ACS, New York, April 8-11 (1986). [6]A. Ueda, H. Watanabe, T. Ohyama, The Rubber Division, ACS, Cincinnati, Ohio, Oct. 18-21 (1988). [7]P.K. Freakley and A.R. Payne, "Theory and practice of engineering with rubber," chapter 5, p.166, Applied Science Publishers, Ltd., London, (1978). [8]A.R. Payne and R.E. Whittaker, Rubber Chem. Technol., 44, 440 (1971). [9]B.B. Boonstra and A.I. Medalia, Rubber Age, 92 892 (1963). [10]A. Yoshioka et al., Pure and Apli. Chem., 58, 1697 (1986). [11]S. Akiyama, N. Kawahara and A. Ueda, Annual meeting of the Society of Polymer Science, Japan (1988).