Rubber and collagen-fiber blends for studless winter tire applications.

Rubber and collagen-fiber blends for studless winter tire applications

The all season tire which primarily requires mud and snow performance for use in winter has been developed mainly in the USA. In Japan, tires which primarily require comparable ice performance with studded tires have been developed in this field, but we have not accomplished satisfying performance of a so-called studless tire. Now it is said that the average ice performance of commercial studless tires is 90 index relative to 100 for studded tire in official reports. So it is considered to be a big effect to improve the ice performance more than 10%.

Generally, it has been referred that the friction of rubber consists of adhesion loss, deformation (hysteresis) loss and cohesion loss components. In addition, the principal factors affecting rubber friction according to the relationship to adhesion loss are interface shear (bond) strength and effective contact area[1] from the fundamental studies on the friction mechanism of rubber in the past[1-3]. Also, regarding the friction of rubber on ice, many studies have been carried out based on the thinking of this friction mechanism. But, an evaluation having accuracy and reproducibility is difficult because of the complicated crystal structure[4,5] and surface state[5] of ice, and the strong dependence on temperature [5-7] which is especially the influence of a liquid like layer on ice at relatively high temperature.

As to the studies of mechanism of friction of rubber on ice, it has been reported that the relationship between Schallamach wave appearing and high friction[7,8] and at relatively low temperature or high sliding velocity, the hysteresis loss for nondirect source affecting friction[9], on the other hand the viscoelastic properties directly affecting friction at low sliding velocity[10], and the softness of rubber compound affecting high frictions[8,9,11] were discovered. And next, as to the studies of the deference of friction ice between rubber compounds, it has been reported that the polymers with lower glass transition temperatures affecting high friction[9,12,13], silica affecting high friction compared with carbon black[7,11,14] and lately the nitrogenous materials in natural rubber affecting high friction[15] were discovered. However, these studies have been the analyses and evaluations in relatively defined rubber compounds, so from now we must complete further study, including the investigation of test method.

In this article, first, we practiced the measurements of skid resistances on ice and on polytetrafluoroethylene (PTFE) to analyze each factors of effective contact area and hysteresis loss affecting the friction of rubber compound on ice.

Second, we took notice that the very high polarity of collagen, consisting of amino group, carboxyl group and active hydrogen, gives high adhesion under the hydrophilic state, the collagen shows a very unique devitrification phenomenon which has been said to show a softening at the temperature of approximately 0 [degree] C[16], and also the collagen has micro and complicated fiber shape, receiving an idea from the ecology of animals in cold climates. And we tried to investigate whether an interface bond strength other than the effective contact area and the hysteresis loss might exist for the factor affecting to the friction of rubber compound on ice at the same time with improving it by blending a leather powder consisting of collagen fiber to rubber.

Third, we investigated the various types of powder shaped fillers compared with leather powder, and also some kinds of polymers compared with S-SBR. Finally, we investigated whether a rubber compound blended with leather powder is effective for studdless tire tread in an actual tire test on ice.

Experimental

Materials

The leather used in our investigation is the true skin of fiber network layer and nipple layer near the surface of cow body, and consists of about 80wt% of collagen protein. The leather powder mixed in rubber compound was for filler produced by grinding this leather into fine pieces, and has the average particle sixe of 100 [micro]meter. The leather powder and other powders mixed in rubber compounds for filler are described and classified in table 1. The average particle sizes of every powders are defined in the region of 30 [micro]meter 100 [micro]meter expecting wool powder and polyester powder which have the fiber lengths of 0.4mm and 0.5mm. Solution styrene butadiene rubber (S-SBR) made by Asahi Chemical Industry Co. has the microstructure of 10% bound styrene, 35% cis bond, 52% trans bond and 13% vinyl bond in polybutadiene, and shows a glass-transition temperature of -74 [degrees] C.

Compounding

Rubber compounds for laboratory properties were prepared in 2 liter internal mixer and cured at 160 [degrees] C for 20 min. Rubber compound formulations utilized are shown in table 2. Rubber compounds for preparing treads for tire tests were mixed in a 20 liter internal mixer.

Table 2 - standard compound formulation

Polymer (variable)        100
N339 carbon black    Variable
Filler (variable)    Variable
Process oil                50
ZnO                         2
Stearic acid                2
Wax                         2
Antiozonant 6C            1.5
Sulfur                    2.1
CBS                       1.2
DPG                       0.2

Measurement

Dynamic viscoelastic properties of rubber compound and leather powder blended rubber compounds were measured with a DVE Rheospectoler made by Rheology Co. under simple tension mode at the frequency of 100Hz, 15% static strain and 2.5% dynamic strain under the temperature range from 20 [degrees] C to -30 [degrees] C, and leather was also measured at the frequencies of 1, 10 and 100Hz, 15% static strain and 0.3% dynamic strain under the wide temperature range from 20 [degrees] C to -70 [degrees] C, the specimen measures 0.5mm thick, 5mm wide and 20mm long.

Ice skid resistances of rubber compounds were measured by a British Portable Skid Tester (BPST) set up in the small room controlled accurately at each temperature from 0 [degree] C to -20 [degrees] C. In this case, we prepared the thin ice of 0.5mm by freezing distilled water dynamically, repeating to slide the test piece of standard rubber compound using BPST at a temperature of -20 [degrees] C because of getting the smooth ice surface with strong and uniform crystal structure. The ice skid number measured was corrected by one of the standard rubber compounds for the reference specimen which was measured every ten specimens because the ice skid resistance varies with time.

Skid resistances of rubber compounds were also measured on the smooth surface of PTFE by BPST at each temperature from 0 [degree] C to -20 [degrees] C to evaluate the skid resistance on dry condition. Thin PTFE surface has the friction coefficient of 0.15. In this measurement we polished the PTFE surface with a cotton cloth every measuring to prevent the experimental error of skid number by the dew condensation of water on the PTFE surface. PTFE skid number of BPST was also measured at each temperature from 0 [degree] C to -20 [degrees] C on its surface covered with the solution of 5% calcium chloride to evaluate the skid resistance on wet condition. Vulcanized specimens were buffed before measuring by grinder on their surface, and measured for both ice and PTFE skid resistance because of the appearance on surface of powder shaped fillers in rubber compounds and so the correspondence with an actual running tire tread.

Ice grip tests of tires were practiced using studless tires of size 165SR13, which have the same construction but the different tread rubber compounds. Evaluations were made by the stopping distance when a car running on ice test track at the speed of 30km/h was braked.

Observations of every powder shaped fillers and their mixed rubber compounds were practices using a JSM-T220A SEM. Specimens were coated with gold by vacuum evaporation after cutting by a knife and observed at the magnifications of 50, 200 and 1,000 times.

Results and discussion

Generally, the friction of rubber can be expressed in terms of its components[1-3]:

(1) F = Fa + Fh + Fc where F = measured friction; Fa = adhesion loss; Fh = hysteresis loss; and Fc = cohesion loss. In the case of friction of rubber on ice, the component of Fc could not be considered to exist in equation (1) because of low friction coefficient, and it has been referred that the principal factors affecting rubber friction according to the relationship to Fa are effective contact area and interface bond strength[1]. So we can transform equation (1) into equation (2):

(2) F = Fco + Fh + Fbo + W where Fco = effective contact area; Fbo = interface bond strength; and W = residual friction factor.

Ice skid and PTFE skid

We measured ice skid number and PTFE skid number of BPST to analyze each factors of effective contact area and hysteresis loss affecting the friction of a rubber compound on ice. Figure 1 shows its results. Ice skid number of BPST is very low at the temperature of 0 [degree] C, but increases rapidly depending on lowering temperature. PTFE skid number of BPST, on the other hand, is very high at the temperature of 0 [degree] C, but decreases depending on lowering temperature, and these two skid numbers show almost the same value at the temperature of -20 [degrees] C. This means that both PTFE and ice surfaces are the same surface state at -20 [degrees] C which are considered to be the complete dry state.

Next, we covered on the surface of PTFE with 5% [CaCl.sub.2] solution layer which does not freeze still at -20 [degrees] C, and measured PTFE skid number of BPST under the wet state. Thereupon, the skid number at 0 [degree] C decreases extremely and shows the almost same value with the case of ice surface, but this value hardly changes depending on lowering temperature. This means that PTFE surface in the case of wet state covered with [CaCl.sub.2] solution is the same surface state with ice surface at 0 [degree] C. Namely, we treat a PTFE in place of an ice because the PTFE surface shows the same surface state with the ice surface under the both complete dry state and complete wet state. And we can consider that the value subtracted PTFE + [CaCl.sub.2] solution skid number from ice skid number is equivalent to the reciprocal of the thickness of liquid like layer on ice surface, regarding the changing from wet state to dry state on ice surface by the temperature changing from 0 [degree] C to 20 [degrees] C.

Futamura[17] expressed the total tire energy loss as DE"/[(E*).sup.n] + F of tread rubber compounds, where, D is a constant, F is a residual energy loss and n is a deformation index. And he concluded that wet traction was dominated by E". So, including the factors of E* of the rubber compound and the thickness of liquid like layer which changes depending on lowering temperature, we can propose equation (3) and equation (4) transformed from equation (2) for the factors affecting the ice skid number of BPST of the rubber compound. Here, we assume that interface bond strength does not occur, and equation (4) is the case of the assumption that E* and Lt act having interaction each other:

(3) BPST Ice - 1 = aFco(E*) + bFco(1/Lt) + cFh(E") + [W.sub.1]

(4) BPST Ice - 2 = dFco(E*) x Lt) + eFh(E") + [W.sub.2] Where, a, b, c, d, e = constants; Lt = the ratio of thickness of liquid like layer; and 1/Lt = BPST Ice - 1 = BPST PTFE + [CaCl.sub.2] solution

Table 3 shows constants, residual friction factors and correlation coefficients r calculated substituting 1/Lt, Lt and E*, E" at each temperature in figure 1 and figure 2 into equation (3) and equation (4). It is judged that both equations are proper because the residual friction factors are relatively small and the correlation coefficients are very high. And we can see that the ice skid number of BPST is affected by every three factors, but it is governed strongly by the thickness of liquid like layer on ice. So we tried to investigate some materials being able to improve the ice skid number of BPST, even if the liquid like layer exists on ice.

We can summarize as follows: By measuring the temperature dependence of BPST skid numbers of a general rubber compound on ice surface, on PTFE surface and on PTFE surface covered with [CaCl.sub.2] solution, ice skid number of BPST is proposed for the equations by complex modulus of the rubber compound and the thickness of liquid like layer on ice for the factors of effective contact area, and loss modulus of it for the factor of hysteresis loss. As a result, ice skid number of BPST has relatively a good correlation coefficient with them, and we can see that it is governed strongly by the thickness of liquid like layer on ice. So we tried to investigate some materials being able to improve the ice skid number of BPST, even if the liquid like layer exists on ice.

Leather

Many authors have also reported that the softness of rubber compound, that is effective contact area has a strong influence on high friction[8-10]. Then we measured the temperature dependence of complex modulus (E*) of the leather consisting of collagen fibers compared with the rubber compound of standard formulation for the first property of leather. Those measured at the frequency of 1, 10 and 100Hz are shown in figure 3. We can see an interesting characteristic of leather which has very small temperature dependence of complex modulus not depending on its frequencies compared with rubber compound.

There are different opinions by authors whether a hysteresis loss of rubber compound affects the friction on ice [9, 10]. Then we also measured the temperature dependence of loss modulus (E") of the leather consisting of collagen fibers and compared it with the rubber compound for the second property of leather. Those measured at the frequency of 1, 10 and 100Hz are shown in figure 4. We can see an interesting characteristic of leather which has very small temperature dependence also in loss modulus like the complex modulus.

The influence of high adhesion under the hydrophilic state by the high polarity of materials, that is the relationship between interface bond strength and rubber friction on ice, has almost never been reported in the past. We took notice of the high polarity of collagen molecule by amino group, carboxyl group and active hydrogen shown in figure 5 for the third property of leather, and tried to investigate the influence of interface bond strength on ice by leather. The wettabilities with water on both surfaces of the leather and rubber compound are shown by photographs and contact angles also in figure 5. We can see the better wettability of leather than rubber compound.

We can summarize as follows: As the result of evaluating the three properties of leather compared with rubber compound, we found the interesting properties that first, the temperature dependence of complex modulus related to effective contact area is very small not depending on its frequencies; second, the temperature dependence of loss modulus related to hysteresis loss is also very small like the complex modulus; third, the wettability by high polarity related to interface bond strength is high.

Leather powder blended rubber compound

The leather powder of 20, 40 and 60phr were each blended replacing carbon black in the rubber compound of standard formulation. In this case, the carbon black contents of each compound were controlled as each hardness at 23 [degrees] C became 55. It is to remove the influence by different hardness at 23 [degrees] C to ice skid number. The temperature dependence of ice skid number by BPST of these compounds is shown in figure 6. Generally, judging from the skid number of BPST, the temperature region which is considered to be slippery in actual tire use below the skid number of about 35 is from 0 [degree] C to -5 [degrees] C. And in the still lower temperature, its region is not considered to be a problem as to slippy performance of tire on ice because their skid numbers correspond to ones of wet asphalt road and more dry asphalt road. Namely, we consider that the most important performance required in the actual use of studless tire is to improve the ice skid number of BPST under the temperatures ranging from 0 [degree] C to -5 [degrees] C. In figure 6, 20phr leather powder blend shows the superior ice skid number under the wide temperature region from 0 [degree] C to -20 [degrees] C compared with rubber compound, and shows its more effective improvement rate from 10% to 7%, especially under the temperature region from 0 [degree] C to -5 [degrees] C required in actual tire use. On the other hand, 40phr and 60phr leather powder blends show the superior ice skid number under the lower temperature region from -10 [degrees] C to -20 [degrees] C, but the lower in ice number under the temperature region from 0 [degree] C to -5 [degrees] C compared with rubber compound.

We can propose equation (3) or equation (4) plus interface bond strength (Fbo) for the factors affecting the ice skid number of BPST of leather powder blended rubber compounds, considering the factor of Fbo as to these phenomena. That is to say, because of worse contribution by higher E* than good contribution by higher E" affecting the ice skid number of BPST about leather powder blends. And, not to be so high E* of 20phr leather powder blend and to be so high E* but relatively low E* under the lower temperature region of 40phr and 60phr leather powder blends compared with rubber compound, suggest the relation to these phenomena.

The surface states and their wettabilities with water of 20phr, 40phr and 60phr leather powder blends are shown compared with rubber compound in figure 7. The 20phr leather powder blend looks like higher possession of leather powder on the surface and has higher wettability for its lower content.

We can summarize as follows: Under the temperature region from 0 [degree] C to -5 [degrees] C which the liquid like layer exists on ice surface and is most slippery state in the actual use of studless tire, 20phr leather powder blend has the maximum in ice skid number of BPST and improves it by 10% 7% compared with rubber compound. We suppose that an interface bond strength exists other than effective contact area and hysteresis loss for the factor contributing this improvement.

Other fillers and polymers

Other various powder shaped fillers were evaluated to compare with leather powder. Photographs of these powders observed at the magnification of 200 times using scanning electron microscope (SEM) were taken and are available. Some proteins other than collagen, inorganic materials and synthetic materials including some fiber shaped fillers were blended by 20phr in the rubber compound of standard formulation, and ice skid numbers of BPST were measured. In this case, each hardness at 23 [degrees] C was controlled at 55 by carbon black contents. Figure 8 shows these results. Leather powder blended rubber compound does not show superior ice skid number at the temperature of -10 [degrees] C, but shows clearly superior ice skid number at the temperatures of 0 [degree] C and -5 [degrees] C which is most strongly required, compared with other fillers blended rubber compounds. In SEM photographs of the surfaces of these various filler blended rubber compounds (available from the authors) leather powder appears to have relatively strong bonding at the interface on rubber, and is seen clearly to have micro and complicated fiber shape in its particles compared with other fillers.

We evaluated ice skid number of BPST and reinforcing properties of each 20 phr leather powder blended rubber compounds using E-SBR or NR65/BR35 for polymer to compare with S-SBR. Figure 9 shows the results of ice skid number of BPST measured. The just combination blend of S-SBR and leather powder shows the superior ice skid number under the all temperature region from 0 [degree] C to -10 [degrees] C than other polymers and leather powder blends. Also, the combination blend of S-SBR and leather powder has the superior reinforcing properties of Mooney viscosity, Pico abrasion and stress-strain curve than other polymers and leather powder blends. In SEM photographs of the surface of these various polymer compounds, the S-SBR blend appears to have a stronger bonding on the interface between polymer and leather powder than other polymer blends.

We can summarize as follows: The rubber compound blended with leather powder in various powder shaped fillers shows the superior ice skid number at the temperature of 0 [degree] C and -5 [degrees] C which is most strongly required. And the just combination blend of leather powder and S-SBR shows the superior ice skid number and reinforcing properties compared with leather powder and E-SBR or NR65/BR35 blends. Namely, we can decide that the combination of leather powder and S-SBR blended compound is the most suitable for studless tire tread from these results.

Tire performance

Figure 10 shows the results evaluated whether a rubber compound blended with leather powder is effective for studless tire tread in an actual tire test on ice. In the treads of general rubber compounds, stopping distance index of tires on ice has a good correlation with tread hardness as shown in this figure. The tread of S-SBR compound blended with 20phr leather powder shifts toward better direction from this correlation line of treads of general rubber compounds. And we confirmed that this improvement of stopping distance index by 20phr leather powder blend is 10%, and almost the same degree with the improvement rate of ice skid number of BPST in laboratory.

Conclusions

* Regarding a general rubber compound, ice skid number of BPST was analyzed by its complex modulus and the thickness of liquid like layer on ice for the factors of effective contact area, and its loss modulus for the factor of hysteresis loss. Ice skid number of BPST had a relatively good correlation coefficient with them, and was strongly affected by the thickness of liquid like layer on ice.

So we tried to investigate some materials being able to improve the ice skid number of BPST, even if the liquid like layer exists on ice.

* We found that a leather consisting of collagen fiber has the interesting properties which are very small temperature dependences of complex modulus and loss modulus, and high wettability by high polarity.

* The 20phr leather powder blended rubber compound had the maximum ice skid number of BPST and improved it by 10% 7% compared with rubber compound at the temperature region from 0 [degree] C to -5 [degrees] C which has liquid like layer on ice and is most slippery state. We suppose that an interface bond strength exists other than effective contact area and hysteresis loss for the factors contributing this improvement.

* The just combination blend of leather powder and S-SBR showed the superior ice skid number of BPST and reinforcing properties.

* The tread of S-SBR compound blended with 20phr leather powder showed the improvement of stopping distance index by 10% in an actual tire test on ice and this was almost the same degree with the improvement rate of ice skid number of BPST in laboratory. [Tabular Data 1 Omitted] [Figures 1 to 6, 8 to 10 Omitted]

PHOTO : Figure 7 - SEM photographs and wettability of various contents of leather powder blended rubber compounds compared with carbon black filled rubber compound

References

[1]H.M. Kummer, "Unified theory of rubber and tire friction" Engineering Research Bulletin B-94, Th Pennsylvania State University, 1966. [2]D.F. Moore, "The friction and lubrication of elastomers," Pergamon Press. [3]A.G. Veith, "Tire traction," A comprehensive review. [4]K. Tsushima, Snow and Ice 41, 245 (1979). [5]H.G. Jellinek, J. Colloid Interface Sci. 25, 192 (1967). [6]A.D. Roberts and J.C. Richardson, Wear 67, 55 (1981). [7]A.D. Roberts, Tribology 14, 14 (1981). [8]A.D. Roberts, J. Adhesion 13, 77 (1981). [9]A. Ahagon, T. Kobayashi and M. Misawa, Presented at 131st ACS Rubber Division meeting, paper No. 95, Montreal, Quebec, Canada, May, 1987. [10]E. Southern and R.W. Walker, Polym. Sci. Technol. 5A, 223 (1974). [11]C.F. Derham, R. Newell and P.M. Swift, NR Technol. 19, 1 (1988). [12]N.S. Grace and G. Winter, Rubber World 126, 633 (1952). [13]A.C. Bassi, Rubber Chem. Technol. 38, 112 (1965). [14]M.P. Wagner, Rubber Chem. Technol. 49, 760 (1976). [15]A.D. Roberts, R. Pitrola and P. Barnes, NR Technol. 20, 1 (1989). [16]J.W. Chien, Biopolymers 11, 2015 (1972). [17]S. Futamura, Tire Sci. Technol. 18, 2 (1990).

II. Takino, K. Kamamizu, M. Komai, K. Hiramatsu and T. Okazaki, Toyo Tire & Rubber Co.