Silane coupling agent effect on the fatigue life of suspension bushing compounds.

The silica/silane reinforcement system has been used in vehicle tire compounds to reduce rolling resistance and improve wet traction for some time. The focus of this article is the fatigue cracking and cycle life of elastomeric bushing materials. Elastomeric bushings are molded onto steel pins and mold bonded or post-bonded to an outer can. These are inserted into the suspension or arm to absorb energy given from the road to the vehicle body. Noise, vibration and harshness (NVH) are important factors affecting the satisfaction of the vehicle owners, and elastomeric bushings play an important role in determining NVH. Noise and vibration sources in a vehicle are many and various. The key noise and vibration concerns involve the dynamic stiffness of the bushing as the bushing experiences a complex combination of compression and torsion. The NVH of a bushing is largely concerned with displacement, and the stiffness behavior for a suspension bushing is relative to frequency. So, the inputs to the vehicle's suspension tend to be random irregularities, and all of these input factors cause the bushing to crack and fail. For NVH requirements, most suspension bushings are made of natural rubber with sulfur vulcanization. The objective of this study is to understand how silane coupling affects the fatigue life or crack cycle of a bushing compound.

The coupling agents used in the experiments were bistriethoxysilypropyl tetrasulfide (TESPT), disulfide silane (TESPD) and mercaptosilane (MS). Combinations of a high surface area silica and a high surface area, high structure carbon black, N234, were used with the same curing system. The materials used in the study are currently used typical bushing compounds with a three stage mixing procedure, as described in table 1. For this mixing procedure, the heat treatment time and mixing temperature have been stated in many published articles. The typical formulations used are listed in table 2. The two sets of formulations from table 2 and table 3 show the increased silica and the three types of silane coupling agent used for the experiments. They have similar hardness for the comparison of their dynamic properties, which was carried out on an Instron dynamic tester 8502 using a 20 x 20 mm cylindrical solid molded rubber specimen.

The fatigue cycle for the cracking test was also carried out on an 8502 dynamic tester. The specimens used in the experiments were molded DeMattia pads of 125 mm length. The travel of the moving grip was 50 mm, amplitude 30 mm, and frequency was set to 2 Hz. The cracks were checked with a 2x magnifier.

This trial combined high surface activity carbon black (N234) and high surface silica (HiSil 243LD). The unique surface characteristics and morphology of the silica and carbon domains are expected to lead to a step improvement in the wet traction, rolling resistance and abrasion resistance compromise over a conventional filler, and have been approved in practice. This article, focusing on the system's two domains used in suspension bushing compounds, will illustrate their improvement on the life cycle by the fatigue cracking test.

Results and discussion

Figure 2 shows the stress-strain results for the table 2 compounds with increasing amounts of silica, but without using a coupling agent. Figure 3 shows the stress-strain of the table 3 compounds with the three different silanes, mercapto (MS), tetrasulfide (TESPT) and disulfide (TESPD). Figure 1 shows the rheometer charts for the compounds with the three different coupling silanes.

[FIGURES 1-3 OMITTED]

When there was no coupling agent used (table 2), as the loading of silica was increased, the tear increased (see table 4), but at the highest loading, tear strength did not continue to increase. This is because the increase in silanol groups contributes to the formation of the filler-filler network attraction, but has no bonding of the silica-polymer, only a physical filler-filler interaction. This also showed in the results of their stress-strain behaviors in figure 2; compound 3 had increased silica from compound 2 (30 phr vs. 16 phr), and its stress-strain curve is lower.

However, in contrast to figure 2, when silane coupling is introduced it can improve the compound strength, tearing and behavior of stress-strain properties. This silane first reacts with the silica during the mixing cycles; then in the vulcanization process, the silane couples the silica and polymer. The different silanes used in the same compound have different stress-strain behavior, as can be seen in figure 3. The curves in figure 3 and in figure 1 (rheometer) also show that mercapto-silane (the highest torque) is the most active because it possesses the shortest alkyl chain of the silanes. So it gave higher results for tensile, tear (table 4) and stress-strain behavior.

The following show the dynamic properties of the three types of silane coupling agent with comparison to the uncoupled compound 3. In contrast to the case above, when the coupling agents are used in the compound, all of the static stiffness data are higher than the uncoupled compound 3 (figures 4 and 5). This is due to the coupling agent, attributed to the reaction with silica and polymer, forming some chemical bonds between silica and rubber molecules. The data also showed that the different silanes have different dynamic properties. The mercapto (3a) and tetrasulfide (3c) gave higher static stiffness, and the dynamic stiffness [K.sup.*] also coincides. The deformation testing was at a set frequency of 10 hz and amplitude of 0.4 mm.

[FIGURES 4-5 OMITTED]

If a cyclic deformation is applied to a rubber, the rubber has both viscous and elastic components, and the loss angle is the phase angle between these two components. Bushings and body mounts experience dynamic stresses, amplitude changes and combinations of both stress and strain during normal service conditions. The phase angle plays an important role in the performance.

Figures 6 and 7 show that all the coupled compounds have smaller loss angle than that of the uncoupled compound at different amplitudes and also at different deformations. But two silane coupling agents, tertrasulfide and disulfide, were close, and even had a bigger loss angle than that of 3 when at very small deformation. This is because, as strain is increased, the filler to filler aggregates would break down giving a rapid decrease in shear modulus G'. The value of the phase angle is different for different silanes, and the range from best to worst is: Mercapto silane--tetrasulfide--disulfide. The tetrasulfide and mercapto coincided with each other on the amplitude vs. loss angle.

[FIGURES 6-7 OMITTED]

The amplitude test set up was at a frequency of 10 Hz and displacement of 4 mm, with the amplitude variation from 1% to 9% of the sample height. When amplitude was applied to a rubber block and going higher, the stiffness of the rubber was going lower. But, as can be seen in figure 8, the silane coupled compounds all had higher stiffness than that of the uncoupled compound, and tetrasulfide silane showed the highest stiffness of all. When the amplitude continued to go higher, the tetrasulfide silane and mercapto silane tended to decrease their stiffness more rapidly, and eventually came close together.

[FIGURE 8 OMITTED]

Fatigue crack growth cycle test

The dynamic tester was used for the fatigue testing, and the set-up was at 2 Hz frequency and an amplitude of 30 mm. The crack initiation point was determined by checking for the cycles when the sample begins cracking. Figure 9 shows that, in the case where no silane agent was used, the cycle at the beginning of the crack dropped rapidly when silica was introduced. When 16 phr of silica was put into the black compound, the resulting initial crack cycle fell from 112 kc down to 36 kc, and with 30 phr of silica in the compound, it fell down to 12 kc. The following study was on the above compound 3, with 30 phr of silica, and three different silanes were added for the fatigue test.

[FIGURE 9 OMITTED]

All silane coupled silica compounds had improved fatigue cracking resistance, and different coupling agents had different results. Figure 10 shows that mercapto silane gave the highest fatigue cycles, followed by disulfide silane and then the tetrasulfide silane. This silane coupling agent effect on the cracking resistance can be explained from the following filler to filler and rubber reaction network:

[FIGURE 10 OMITTED]

The silane coupling agents have two different end groups. Tetrasulfide and disulfide silanes have ethoxy groups at one end and the sulfur group on the other. During the mixing process, the ethoxy groups react with the silanol groups on the surface of silica. The sulfur group (tetrasulfide or the disulfide) of each structure reacts, later during vulcanization, with rubber to give chemical bonding. The tetrasulfide silane has four sulfur atoms (Sx = is 3.8), while the disulfide silane has Sx = 2.2 (ref. 9). The longer the alkyl chain of silane, the bigger is the shielding of the silica. This results in greater inhibition of filler to filler network crosslinking. The tetrasulfide silane probably yields multi-sulfur bonds, and the disulfide may yield di-sulfur and mono-sulfur bonds.

[FIGURE 10 OMITTED]

When the mercapto silane reacts with silica and rubber, it has methoxy at one end and mercapto (-SH) at other end. The alkyl chain is shorter compared to tetrasulfide and disulfide silanes. The mercapto group is very active during vulcanization and would produce more mono-sulfur bonds. This phenomenon showed in figure 1 (the rheometer chart), and the dynamic properties with the mercapto silane include higher dynamic stiffness and lower loss angle.

For a suspension bushing, the NVH is mainly concerned with displacement, and the stiffness behavior relative to frequency. So, the ideal isolator will be stiff at low frequency and large displacement for durability, and will be compliant at high frequencies and low displacement. During a vehicle's operation, suspension isolators experience a complex combination of compression and torsion. These high stresses cause the bushings to generate extreme heat build-up. The heat build-up can cause bushing failure on the compressed side of the bushing, with cracks starting on the inside sleeve and propagating to the outside metal, and can lead to complete bushing failure. The heat build-up is very related to loss angle of the compound. Mercapto silane and tetrasulfide silane have smaller loss angle as the displacement or amplitude is changing.

Above, it was shown that the mercapto silane coupling agent mainly formed single sulfur bonds, disulfide silane formed double sulfur bonds, and tetrasulfide was yielding polysulfur bonds. The bond energy for mono-sulfur is 285 Kj/mol, while for the double and poly sulfur it is below 268 Kj/mol.

These data showed that the mercaptosilane coupling agent possessed heat resistance, had good stress-strain response and higher tear resistance (table 4); so the mercapto silane coupled silica system for the NR sulfur cure compound had higher fatigue crack resistance.

Conclusion

From these test results, it is possible to show the following conclusions:

* To increase the cut growth crack resistance, the silane coupling agent with the mercapto group was superior to the other two silane groups.

* The tetrasulfide is better than disulfide, because both mercapto and tetrasulfide have smaller loss angle as amplitude and displacement change.

Table 1 - mixing procedure

Pass 1 - speed 50 rpm
  0 sec. - polymer
  30 sec. - black, silica, silane
  1 min. - zinc oxide, small amount chemicals
  2.5 min. - oil.
  Dump batch at 155[degrees]C.
Pass 2 - speed 50 rpm.
  Pass 1 compound (matured for 24 hrs).
  Dump at 150[degrees]C.
Pass 3 - speed 30 rpm.
  0 sec. - pass 2 batch.
  30 sec. - curatives.
Dump batch at 110[degrees]C.

Table 2 - formulations

                   1     2     3

NR               100   100   100
ZNO                5     5     5
Stearic acid     1.5   1.5   1.5
CB N234           30    16    16
HiSil 243 LD       3    16    30
Antidegradants   4.5   4.5   4.5
Oil                5     5     5
Sulfur           1.6   1.6   1.6
Accelerator      2.1   2.1   2.1

Table 3 - formulations

                    3    3A    3C    3D

NR                100   100   100   100
ZNO               3.5   3.5   3.5   3.5
Stearic acid      1.5   1.5   1.5   1.5
CB N234            16    16    16    16
Silica HS243       30    30    30    30
Antidegradant     4.5   4.5   4.5   4.5
Oil                 5     5     5     5
Sulfur            1.6   1.6   1.6   1.6
Accelerator       2.1   2.1   2.1   2.1
Mercapto silane     0     3     0     0
TESPT silane        0     0     3     0
TESPD silane        0     0     0     3

Table 4 - physical properties

                   1      2      3     3A     3C     3D

Tear (n/mm)       90    107    101    124    106     96
Durometer A       62     63     59     62     60     62
Tensile (mpa)   30.5   29.3   23.4   27.6   27.4   25.6
Elongation %     654    680    623    651    629    638
300% modulus    10.2   10.6    9.2   11.9   10.3    9.6

References

(1.) "Carbon-silica dual phase filler application to passenger tread compound," by Ping Zhang, Meng-Jiao Wang, Yokov Kutsovsky, Stephen Laube and Khaled Mahmud, Cabot, Rubber World, April 2002.

(2.) "Future vehicle noise, vibration and harshness requirements for elastomeric isolators," by R.S. Himes, North American Truck and David Griffiths, Ford Motor, Research & Vehicle Technology, February 2001.

(3.) "Silane coupling agents for enhanced silica performance," by John T. Byers, Degussa Corp., Rubber World, September 1998.

(4.) "Elast-O-Flux silanes boost processability, control," by Andrew Tisler, Rhein Chemic Corp., Rubber & Plastics News, September 2001.

(5.) "Silane-rubber coupling in sulfur, peroxide and metal of in-rubber carbon black dispersion," by M. Gerpacher, C.P. O'Farrell, H.H. Yang and L. Nikiel, Sid Richardson Carbon, Rubber World, April 1999.

(7.) "Response of elastomeric bushings to combined compression/shear loadings," by R.P. Kezar; Z. Tao, K.J. Patenaude, J.L. Mead and R.G. Stacer, University of Massachusetts Lowell, and G. Rodringuez, U.S. Army Research Laboratory, Rubber World, September 2003.

(8.) "Flexometer predicts heat generation," by Hiroshi Mouri, Brigestone., Rubber World, September 1996.

(9.) "Crosslink and reinforcement of silica/silane-filled rubber compounds," by Andre Hasse, Andre Wehmeier and Hans-Detlef Luginsland, Degussa, Rubber World, April 2004.