Improved black sidewall compound performance using precipitated silica.

The tire is a composite of many rubbery components, each of which serves a specific and unique function. The black sidewall is the outer surface of the tire that protects the casing against weathering. It is formulated for resistance to weathering, ozone, abrasion, tear, radial and circumferential cracking and for good fatigue life (ref. 1). Chemical antiozonants are added to the sidewall at levels tested for effectiveness under both static and dynamic conditions. However, the anntiozonant is continually depleted from the sidewall surface by reaction with ozone (refs. 2-5), and by physical mechanisms such as curb scuffing and washing. In addition, the practical chemical antiozonant for rubber are staining materials, and therefore can only be used in limited amounts. N,N'-disubstituted-paraphenylenediamines are the most effective antiozonants, particularly the alkyl, aryl substituted (refs. 5-7). Tire appearance is important, particularly for passenger tires, and surface discoloration is a particular problem of using N,N'-para-phenylenediamines.

Use of a reinforcing precipitated silica is studied:

* To evaluate the effectiveness of silica in extending the durability of a model tire black sidewall compound;

* to determine if a delayed release interaction (ref. 8) exists between silica and the antiozonant based on, for example, hydrogen bonding interactions of the silica surface silanol O-H groups with the N-H functional groups of the paraphenylenediamine; and

* to explore the potential of using silica to limit the surface discoloration that results on the sidewall compound from use of the antiozonant N-1,3-dimethylbutyl, N'-phenyl-paraphenylenediamine (HPPD) upon exposure to ozone (ref. 7). Statistically designed compounding experiments varying carbon black, silica and antiozonant levels were performed in order to establish the effect that silica use has on the durability properties of the black sidewall compound and to determine which parameters affect the tear, cut growth and ozone age resistance of the compound.

Experimental

Materials

The rubber compound selected for study was based on a radial passenger tire black sidewall recipe in the Vanderbilt Rubber Handbook (ref. 9). Two ingredients were not used in order to simplify the recipe to observe silica effects: Vanplast R, a neutralized sulfonate of high molecular weight in a paraffin oil; and Agerite Resin D, a polymerized 1,2-dihydro-2,2,4-trimethylquinoline antioxidant. Three additional modifications were made: 1 phr of Sunolite 240 wax was used in place of 3 phr Vanwax H Special, a petroleum wax; 4 phr of Santoflex 13 antiozonant (HPPD) was used instead of the 2 phr specified; and N-330 carbon black was used at 46 phr in place of the specified 50 in order to incorporate the silica filler. Three materials were studied as variables in a rotatable central composite design with star points (ref. 10) (Box-Wilson design [ref. 11]: Antiozonant (2.3 - 5.7 phr) was used centering at 4 phr; carbon black (39.3 - 52.7 phr) was used centering at 46 phr; and Hi-Sil 243LD silica (0-10 phr) was used centering at 5 phr. Fifteen different compounds specified by the design were prepared, with the center point studied on three replications. A compound with 50 phr carbon black, 0 phr silica and 4 phr antiozonant was also studied. Table 1 gives the recipe.

Methods

All compounds were mixed in an internal mixer in the order listed in table 1 with sulfur and accelerator added in a second internal mix step. Compounds were milled into sheets and press cured at 150 degrees C to [T.sub.90] + 5 minutes (mold lag), which was approximately 18 minutes. Cure curves were obtained using a moving die rheometer.

Cured 15.22 x 1.27 x 0.32-cm (6 x 1/2 x 1/8-inch) compound strips were aged in an environmental chamber using the ASTM D3395 procedure (ref. 12) modified

with an on/off ozone cycle (ref. 13). Fifty parts per hundred million (pphm) ozone concentration, 40 degrees C and 0-25% elong-ation/relaxation in a 0.5 Hz cycle. Samples were preconditioned by cycling stretching at 40 degrees C temperature and 0 pphm ozone for 10 minutes. The modification of using a cycled on/off ozone procedure was developed through experimentation. The procedure selected for use is an ozone concentration of 50 pphm for four hours while the sample is being cyclically stretched and relaxed, followed by an ozone concentration of 0 pphm for two hours at zero percent strain while the temperature was maintained at 40 degrees C. A visual rating system was developed by testing 16 identical sample specimens of the same rubber compound removing a sample from the chamber at regular intervals until total failure occurred. Ten samples were then selected as representative of various stages of ozone degradation. Photographs under 10x magnification were obtained at 25% elongation and were used as standards. The original sample was assigned a value of one, a sample one day short of failure was assigned a value of ten, and a totally failed specimen (broken into two pieces) was assigned a value of twelve.

Diffusion of the antiozonant was studied in cured sidewall compound tensile sheets prepared with/without silica (0, 5 and 10 phr) and with/without antiozonant (0 and 4 phr) based on the formula in table 1, with N330 = 50 phr. Cured sheets were built into composites by plying together two different compounds of varying silica and antiozonant levels and by using steel plates clipped together to insure intimate contact of the rubber tensile sheet surfaces. Composites were oven aged at 100 degrees C for various time periods. Samples were prepared for analysis by peeling the composites into separate sheets, cooling to room temperature, dicing, Soxhlet extracting using chloroform and the solvent evaporated using a Buchi Rotovapor. The extract was analyzed by using Monsanto Method 13,555, which determines N-1,3-dimethylbutyl, N'-phenyl-para-phenylenediamine levels quantitatively by using gas chromatography. The following composites are examples of those tested:

* Compound 1 silica/antiozonant (phr) - 0/4; 0/4; 0/4;5/4; 10/4

* Compound 2 silica/antiozonant (phr) - 0/0, 5/0; 10/0; 0/0; 0/0

Antiozonant levels in each compound of composite were plotted against aging time at 100 degrees C.

All ozone-aged sidewall strips displayed a brown surface discoloration. Surfaces of both pieces of the failed sample specimens were washed with 10 mL dichloromethane sprayed from a syringe, the washings filtered through glass fibers to remove particulates, the glass fibers rinsed with 2 mL dichloromethane and the solvent evaporated to dryness by passing dry nitrogen gas over the filtrate. Ultraviolet-visible absorption spectra were obtained after volumetric dilution in acetonitrile and the discoloration quantified by using the absorbance at 400 nm since neither the antiozonant nor the solvent have appreciable absorption at this visible wavelength (ref. 7).

Results and discussion

Diffusion studies

Initial results indicated that a delayed release mechanism might be operative since graphs of the antiozonant level versus aging time at 100 degrees C showed that compounds with both sihca and the antiozonant had a slower rate of diffusion of the antiozonant into the ply without either the antiozonant or silica. However, data vary widely and the amount of total antiozonant extracted from the two plies of the composite was not a constant, decreasing with increasing age time. Follow-up studies improved experimental reproducibility by reducing the aging time periods from days (1, 4, 7 and 10) to hours (4, 8, 16 and 24) at 100 degrees C. These results showed that no delayed release of the antiozonant occurred since:

* The rates of antiozonant diffusion from compounds with silica were the same as those from carbon black compounds;

* the final antiozonant levels were the same in compounds with/without silica, regardless of whether the antiozonant was initially mixed into a compound containing carbon black or into a carbon black/silica-filled rubber stock.

Statistical compounding studies

The statistically designed compounding study identified improvements in rubber compound physical properties based on silica use in the tire black sidewall recipe shown in table 1. Plots of physical properties measured versus silica level show that silica use:

* Increases compound tear strength based on the molded groove tear test (figure 1);

* reduces cut growth based on the DeMattia flex fatigue test (figure 2); and

* does not appreciably affect the scorch ([T.sub.2]) and cure ([T.sub.90]) properties of the black sidewall compound based on moving die rheometer results at 150 degrees C (figure 3).

Figures 1-3 show that variations exist in measured properties at a particular silica level (i.e. 5 phr), which is thought to result from the different carbon black and/or antiozonant levels used in each compound. These variations indicate that carbon black and/or antiozonant also have an effect upon the measured tear, flex and/or cure properties.

Several compounds showed notably increased resistance to ozone aging compared to the carbon black control. This was manifest as increased number of days to total specimen failure and by lower ozone crack ratings during aging. An example of the histograms obtained is shown in figure 4 for the design center point (46 phr carbon black, 5 phr silica, 4 phr antiozonant) and the 50 phr carbon black control. Table 2 compares data for the design center point and the corresponding 46 phr carbon black control compound. Computer analysis of these ozone aging data indicated that a reduction in carbon black levels would increase ozone age resistance.

Finally, an analysis was made of the relationship between the brown surface discoloration measured, based on absorbance of the surface washings at 400 nm, to the material variations and to the other physical properties measured. Analysis showed that surface discoloration only correlated with the antiozonant levels incorporated into the compound.

Follow-up compounding study

A simple follow-up study examined the effect of using 8 phr of silica in compounds having 50, 46 and 42 phr of carbon black to determine if silica should be used: In addition to, as a practical replacement of, or as a part-per-part replacement of N-330 carbon black. The compound with 42 phr carbon black and 8 phr silica was a predicted optimum compound for resistance to ozone aging. Results showed that partial or part-perpart replacement of carbon black with silica increased compound tear strength, reduced cut growth and increased ozone resistance, both the number of days to total failure and the interim ozone ratings. The part-per-part replacement of carbon black with silica afforded the best results. When silica is added to the compound containing 50 phr carbon black, tear strength and cut growth resistance are increased, but not resistance to ozone aging. Table 2 is a summary. Figure 5 shows histograms of the control and the best silica-conmining compound, the part-per-part replacement of carbon black with 8 phr of silica.

Finally, the observed improvement in ozone resistance for compounds with silica was again confirmed by repeat ozone testing of fresh specimens of selected compounds from the original statisticaily designed or follow-up studies. Figure 6 is a histogram of three of the silica-containing compounds that had better ozone resistance than the 50 phr carbon black control.

Statistical analysis

A detailed computer regression analysis was made for the variables and corresponding physical test results of the statistically designed and simple follow-up studies. Complex quadratic equations were generated based on initial design variables (silica, carbon black, antiozonant) for the ozone ratings after nine days (equation 1) and after 12 days (equation 2) of cycled dynamic ozone aging, and for tear (equation 3) and cut growth resistance (equation 4). Results are summarized in table 3 and include the regression equanons, coefficients of variation ([r.sup.2]) and the confidence factors. Surface contour diagrams were generated from these full order quadratic equations that visibly demonstrate the dependence of ozone resistance, tear resistance and cut growth resistance, respectively.

Surface contours in these diagrams (available upon request) also show:

* The predominant effect on compound age resistance is antiozonant concentration since increasing the antiozonant level from 3 phr to 5 phr significantly decreases the predicted ozone aging behavior for any combination of carbon black and silica; and

* use of silica can be beneficial at any antiozonant level since the contours of the three curves are essentially identical.

Also, it has been indicated that 3 phr antiozonant may be insufficient to protect the sidewall since high crack ratings are predicted as well as total failure of some specimens. To reduce ozone crack ratings equation 2 is consistent with either increasing the antiozonant level to 4 phr and above since no specimens are predicted to fail, or using silica as a replacement of carbon black.

Comparisons of surface contours show that: The antiozonant concentration (3-5 phr range) does not affect compound tear strength, increased use of carbon black linearly increases tear strength and increased use of silica exponentially increases tear strength. Thus, silica use is recommended in tire applications where increased tear strength of the sidewall is required.

The diagrams show the predicted surface contour for cut growth:

* Carbon black levels have no affect in this NR/BR model;

* increased antiozonant levels significantly increased the predicted cut growth; and

* silica use reduces cut growth.

Thus, use of increasing antiozonant levels to achieve resistance to ozone aging has adverse effects upon cut growth resistance and surface discoloration of the sidewall. Use of silica in conjunction with use of increased levels of antiozonant affords increased resistance to ozone aging and to cut growth.

Finally, an equation was also generated for ozone crack rating after 12 days, but was based on the dependent physical property test results. Even though physical properties were not independent variables, the ability to predict aging behavior using significant properties may suggest mechanisms involving compound ozone resistance. This equation suggests that resistance to ozone aging is dependent on the degree of polymer crosslinking (G'), the ability of the compound to dissipate energy as heat (G"), the tear resistance of the compound and the dispersion of the fillers.

Summary

Use of silica in a tire black sidewall recipe significantly increases compound tear strength and cut growth resistance without significantly affecting scorch and cure time. The ozone age resistance of the compound is primarily dependent on antiozonant concentrations, but is also dependent upon the presence of silica. Increased antiozonant levels also result in increased surface discoloration and loss of cut growth resistance. Silica should be used as a part-per-part replacement for carbon black to the ca. 10 phr level. Five equations were generated based on regression analysis of the dam. Three equations were generated for ozone resistance, and one each for tear strength and cut growth resistance. Surface contours were generated from these equations. The equations indicate the complexity of rubber physical properties in general and ozone aging resistance in particular on compounding ingredients.

References

1. R.S Bhakuni, S.K. Mowdood. W.H. Waddell, I.S. Rai and D.L. Knight, "Tires" in Encycl. Poly. Sci. Eng. 16, 834 (1989).

2. R.W. Layer, Rubber Chem. Technol., 39, 1584 (1966).

3. S.D. Razumovskii and L.S. Batashova, Rubber Chem. Technol., 43, 1340 (1970).

4. R.P. Lattimer, E.R. Hooser, R.W. Layer and C.K. Rhee, Rubber Chem. Technol., 56, 431 (1983).

5. W.H. Waddell, K.A. Benzing, L.R. Evans and J.M. McMahon, Rubber Chem. Technol., 65, 411 (1992).

6. W.L. Cox, Rubber Chem. Technol., 32, 364 (1959).

7. O. Lorenz and C.R. Parks, Rubber Chem. Technol., 34, 816 (1961).

8. L.R. Evans, D.A. Benko, J.G. Gillick and W.H. Waddell. Rubber Chem. Technol., 65, 201 (1992).

9. W.H. Waddell, R.S. Bhakuni, W.W. Barbin and P.H. Sandstrom, "Pneumatic tire compounding" in The Vanderbilt Rubber Handbook, 13th ed., R.F. Ohm ed., R.T. Vanderbilt Co., Inc., Norwalk. 1990, pp. 595-611.

10. R.H. Meyers, "Response surface methodology," Allyn & Bacon, Inc., Boston, 1971.

11. J.S. Murray, Jr., "X-Stat: Statical experimental design/data analysis/nonlinear optimizations," version 1.1., John Wiley & Sons, Inc., New York, 1984, appendix VI.

12. ASTM D 3395, 1991 Annual Book of ASTM Standards, Section 9 Rubber, p. 551.

13. D.A. Benko, LR. Evans, J.G. Gillick. W.H. Waddell, B.A. Metz, B.F. Benton, G.E. Pickett and W.R. Kruman, U.S. Patent 4,895,884, January 23, 1990.

  Table 1 - model tire black sidewall formula
                                                       phr
Natural rubber (SMR 5)                                  50
Butadiene rubber (BR 1220)                              50
Carbon black (N-330)                                    46*
Silica (Hi-Sil 243LD)                                    5*
Processing oil (Sunthene 310)                           10
Vanplast R                                               -
Agerite Resin D                                          -
Stearic acid                                             2
Wax (Sunolite 240)                                       1
Antiozonant (Santoflex 13)                               4*
Antioxidant (Wingstay 100)                               1
Zinc oxide                                               3
Sulfur                                                   1.8
Accelerator (AMAX)                                       1
                                                       174.8
  * = center point of rotatable central composite design

[Tabular Data Omitted]