High temperature curing and high heat resistance compounding.

The rubber industry is continually striving to improve productivity and reduce unit manufacturing costs while maintaining a high level of product quality and performance.

Historically, the vulcanization process has restricted productivity improvements in many rubber manufacturing plants. Extended time at elevated temperature is required to provide a sufficient state of vulcanization throughout a rubber article. An obvious way to reduce vulcanization cycle time is by increasing the vulcanization temperatures, since the rate of sulfur vulcanization increases as the reaction temperature is elevated. However, the state of vulcanization is adversely affected with increasing temperatures. This is manifested in lower physical properties and product performance.

When sulfur vulcanized natural rubber compounds are exposed to a thermal aging environment, significant changes in physical properties and performance characteristics are observed. These changes are directly related to modifications of the original crosslink structure. Decomposition reactions tend to predominate and thus lead to reduction in crosslink density and physical properties as observed during extended cure and when using higher curing temperatures.

Over the years the rubber industry has developed several compounding approaches to address the changes in crosslink structure during thermal aging. As with many formulation changes in rubber compounding, there is a compromise that must be made when attempting to improve one performance characteristic. For example, improving the thermal stability of vulcanized natural rubber compounding by reducing the sulfur content of the crosslink through the use of the more efficient vulcanization systems will reduce dynamic performance properties such as fatigue resistance. The challenge is to define a way to improve thermal stability while maintaining dynamic performance characteristics.

High temperature cure compounding

Over the years, three approaches have been developed to address cure cycle reduction:

* Optimization of existing molding cycle;

* curative adjustments for faster vulcanization rates;

* increasing curing temperatures.

Effect of higher curing temperature

As illustrated in figure 1, the vulcanization reaction follows the classical law of thermodynamics. For each 10 degrees] C increase in temperature, there is a doubling of the rate of reaction. Optimum rheometer cure times, t90 values, as a measure of the rate of vulcanization, decrease as the curing temperature increases. Unfortunately, as curing temperature increases, a reduction in physical properties is observed.

[Figure 1 ILLUSTRATION OMITTED]

The reduction in properties observed at the elevated curing temperatures could be due to changes in the polymer backbone or could be associated with the nature and density of the crosslinks. However, C.T. Loo (refs. 1 and 2) has shown that chain scissions do not occur to any measurable extent even at high curing temperatures. Other possible changes in the polymer itself, such as isomerization and cyclization reactions, do increase with temperature, but these reactions only become significant at temperatures well in excess of those discussed in this article.

It would appear, therefore, that the decrease in properties such as modulus and tensile strength are due to changes in the type of crosslinks produced and to a fall in crosslink density. This is supported by further studies of Loo (ref. 2) on the effects of curing temperatures on crosslink density.

In a conventional natural rubber compound cured two hours at 140 [degrees] C (figure 2), the majority of the crosslinks are polysulfidic. However, after 20 minutes cure at 180 [degrees]C (figure 3), only di- and a preponderance of monosulfidic crosslinks are found.

[Figures 2 and 3 ILLUSTRATION OMITTED]

Therefore, when rubber compounds are overcured at high temperature, it can be concluded that:

* Crosslink density is drastically reduced;

* essentially monosulfidic crosslinks are obtained;

* there is a large increase in main chain modifications.

Of these factors, the most important in determining initial physical properties is the crosslink density.

It has been shown (ref. 3) that for long cure times, the crosslink density of a rubber vulcanizate is given by:

X = k1 * f (curative level)/(k1 + k2), where k1 is the kinetic constant for the formation of stable crosslinks and k2 is the kinetic constant for the formation of unstable crosslinks.

As the curing temperature is increased, the number of unstable crosslinks (X") increases faster than the number of stable ones (X'). This means that the value of k2 increases more rapidly than k1 so that the value of k1/(k1 + k2) falls. Consequently, to maintain crosslink density at elevated cure temperatures, it is necessary to reduce the value of k2 or to increase curative levels.

The formation of unstable crosslinks can be controlled by the use of more efficient curing systems, where k2 is reduced compared to k1. Using this approach, the decrease in crosslink density is reduced but not eliminated.

Thus, it is necessary to increase curative levels. There are three basic ways of increasing curative levels:

* Increase sulfur;

* increase accelerator;

* increase accelerator and sulfur.

Increasing sulfur is unsatisfactory because it reduces cure efficiency and consequently k2 is increased. By increasing accelerator and sulfur level linearly, cure efficiency is unchanged. However, if accelerator level is increased at constant sulfur level, cure efficiency is increased, so that this approach should give the best retention of crosslink density as cure temperature increases.

Maintenance of vulcanizate properties at high curing temperatures

As indicated earlier, lower modulus compounds are obtained with increased cure temperatures due to a reduction of crosslink density. One way to increase the modulus is to use higher carbon black levels and this was the first route examined. Although the required modulus was achieved by increasing carbon black levels, the tensile strength was affected and resilience was greatly reduced.

The next attempt was to increase both accelerator and sulfur. This is shown in table 1.

Table 1 - increased accelerator and sulfur levels - NR/BR

Sulfur                           2.0     2.0             2.5
TBBS                            0.75    0.75             1.0
Curing temp., [degrees] C        140     170   [right arrow]
Curing time, mins.                70      50   [right arrow]

Physical properties
300% modulus, MPa               8.92     5.1            6.67
Tensile strength, Mpa          20.69   11.28           14.21
Elongation @ break, %            530     540             510

Sulfur                                 3.0               4.0
TBBS                                   1.2               1.5
Curing temp., [degrees] C    [right arrow]     [right arrow]
Curing time, mins.           [right arrow]     [right arrow]

Physical properties
300% modulus, MPa                     7.94              9.71
Tensile strength, Mpa                15.98             16.08
Elongation @ break, %                  520               440

As indicated, there was no difficulty in obtaining a modulus match between the low and high temperature cures by this means. However, only a relatively small improvement in tensile strength is obtained. This is probably due to the higher sulfur level used resulting in an increase in main chain modifications which weaken the chains.

A third method of retaining properties when curing at higher temperatures is to increase accelerator levels while maintaining sulfur constant. Figure 4 illustrates this approach and shows that a relatively good match of properties can be obtained.

[Figure 4 ILLUSTRATION OMITTED]

With SBR compounds, the effect of cure temperature and reversion are less severe than in natural rubber. Nevertheless, significant falls in state of cure are found when cure temperatures approach 200 [degrees] C. In this case, it is possible to maintain properties at a reasonably constant level by increasing accelerator and sulfur levels together. This is shown in table 2.

Table 2 - increased curative levels - SBR/BR

Sulfur                          2.0     2.5
CBS                             1.0    1.25
Curing temp., [degrees] C       170     205
Curing time, minutes             20      10

Physical properties
300% modulus, MPa              7.46    8.14
Tensile strength, MPa         18.83   17.66
Elongation @ break, %           500     540

Alternatively, maintaining sulfur constant and increasing accelerator level will also enable properties to be maintained as cure temperatures are increased. This is shown in table 3. By increasing accelerator levels alone, cure efficiency is increased, resulting in fewer main chain modifications and hence reduced marching modulus.

Table 3 - effect of sulfur/accelerator on physical
properties - SBR/BR

Sulfur                            2.0      2.0
TBBS                              1.0      1.0
Curing temp., [degrees] C         170      205
Curing time, mins.                 20       10

Physical properties
300% modulus, MPa                6.97     5.20
Tensile strength, MPa           18.54    15.69
Elongation @ break, %             570      630

Sulfur                                  2.0             2.0
TBBS                                    1.5             2.0
Curing temp., [degrees] C     [right arrow]   [right arrow]
Curing time, mins.            [right arrow]   [right arrow]

Physical properties
300% modulus, MPa                      6.67            7.55
Tensile strength, MPa                 16.77           16.97
Elongation @ break, %                   600             550

The effect of increased accelerator levels in a natural rubber compound is shown in table 4. The data shows that increasing the TBBS level from 0.6 to 1.0 phr produces little change in processing safety, a faster cure rate and 180 [degrees] C cured physical properties that are similar to those obtained at 160 [degrees] C cured compound. Further increase in TBBS from 1.0 to 1.4 phr gives further improvements.

Table  4 - accelerator adjustment in NR sulfenamide accelerator

                                             Control    Adj. A
Sulfur                                         2.5        2.5
TBBS                                           0.6        1.0

Rheometer data @ 160 [degrees] C
Min. torque                                    5.5        5.6
Max. torque                                   32.7       36.9
[t.sub.2], minutes                             3.0        3.2
[t.sub.90], minutes                            7.0        6.3

Rheometer data @ 180 [degrees] C
Min. torque                                    5.5        5.2
Max. torque                                   30.1       33.7
[t.sub.2], minutes                             1.0       1.05
[t.sub.90], minutes                           2.45       2.21

Stress-strain data cured @ 160 [degrees] C
Cure time, minutes                             7.0       6.30
Shore A hardness                              63.0       65.0
100% modulus, MPa                             2.34       2.96
300% modulus, MPa                             9.43      10.92
Ult. tensile, MPa                            20.60      19.91
Ult. elongation %                              550        505

Stress-strain data cured @ 180 [degrees] C
Cure time, seconds                           155.0      135.0
Shore A hardness                              58.0       61.0
100% modulus, MPa                             1.96       2.41
300% modulus, MPa                             8.20       9.54
Ult. tensile, MPa                            19.50      19.02
Ult. elongation %                              565        535

                                                  Adj. B
Sulfur                                              2.5
TBBS                                                1.4

Rheometer data @ 160 [degrees] C
Min. torque                                         5.6
Max. torque                                        38.9
[t.sub.2], minutes                                  3.4
[t.sub.90], minutes                                 5.9

Rheometer data @ 180 [degrees] C
Min. torque                                         5.4
Max. torque                                        36.2
[t.sub.2], minutes                                  1.1
[t.sub.90], minutes                                 2.3

Stress-strain data cured @ 160 [degrees] C
Cure time, minutes                                  5.9
Shore A hardness                                   67.0
100% modulus, MPa                                  3.17
300% modulus, MPa                                 11.57
Ult. tensile, MPa                                 20.84
Ult. elongation %                                   510

Stress-strain data cured @ 180 [degrees] C
Cure time, seconds                                130.0
Shore A hardness                                   62.0
100% modulus, MPa                                  2.62
300% modulus, MPa                                 10.16
Ult. tensile, MPa                                 19.36
Ult. elongation %                                   510

By conducting a series of experiments utilizing the appropriate time/temperature profile for a specific molding operation, one should be able to construct an accelerator concentration/cure temperatures correlation as illustrated in figure 5.

[Figure 5 ILLUSTRATION OMITTED]

From this curve, several time/temperature conditions could be chosen and accelerator concentrations can then be optimized for each temperature. In this way, the relationship between cure temperature and accelerator concentration required to maintain properties can be obtained.

High heat resistance compounding

It is well known that the thermal (reversion) and thermal oxidative aging resistance of rubber compounds can be improved through the use of efficient vulcanization (EV) systems. EV systems are produced by the use of sulfur donors as partial or total replacement of elemental sulfur or by the use of very high ratios of accelerator to sulfur. These vulcanization systems produce vulcanizates in which a high portion of the crosslinks are monosulfidic and disulfidic with minimal' modification of the main chain by sulfurization. For example, with natural rubber compounds, an efficiently vulcanized system would provide excellent resistance to reversion and oxidative aging. These compounds exhibit very poor initial fatigue properties, but are very effective in static applications environments.

Efforts made to overcome the EV cure deficiency have led to the development of the so-called "semi-efficient" (semi-EV) vulcanization system. The semi-EV systems are obtained by the use of intermediate accelerator to sulfur ratios or partial replacement of sulfur with a sulfur donor. When compared to the conventional high sulfur and low accelerator system, the semi-EV system also provides excellent resistance to reversion and oxidative aging, but with much improved fatigue properties as compared to the EV system.

The aging effects of semi-EV systems based on the use of high accelerator and low sulfur ratios is illustrated in figure 6.

[Figure 6 ILLUSTRATION OMITTED]

As with many formulation changes in rubber compounding, there is compromise that must be made when attempting to improve one performance characteristic. Improving the thermal stability of vulcanized natural rubber compounds by reducing the sulfur content of the crosslinks through the use of the more efficient vulcanization system will reduce dynamic performance properties such as fatigue.

This challenge to define a way of improving thermal stability while maintaining performance characteristics was met with the commercialization of hexamethylene-1,6-bisthiosulfate disodium salt, dihydrate (HTS). HTS is a chemical that promotes the formation of flexible hybrid crosslinks (ref. 4), (figure 7).

[Figure 7 ILLUSTRATION OMITTED]

By reducing the average sulfur chain length at the points of attachment to the polymer backbone, thermal stability is improved. Maintaining a long chain within the crosslink structure provides enhanced flexibility under dynamic conditions. The use of HTS eliminates the compromise between thermal aging and dynamic properties as illustrated in table 5.

Table 5 - effect of HTS on thermal aging and fatigue of NR

Sulfur                               2.5   2.5   1.20
TBBS                                 0.6   0.6   1.75
HTS                                   --   2.0     --

Retained 300% modulus (%)
10 x t' (90) @ 144 [degrees] C        75    91    105
5 x t' (90) @ 181 [degrees] C         56    68     69
t' (90) @ 181 [degrees] C             73    89     91
t' (90) @ 140 [degrees] C

Fatigue-to-failure (KC)
t' (90) @ 144 [degrees] C            197   241    123
10 x t' (90) @ 144 [degrees] C       127   169     90

Figure 8 further shows that with HTS, excellent oxidative aging resistance occurs without sacrificing fatigue properties. The addition of HTS (stock 4) to the semi-EV cure system (stock 3) improves both thermal and thermal oxidative aging while providing similar fatigue property to that of the control (stock 1).

[Figure 8 ILLUSTRATION OMITTED]

The data also show that the addition of HTS may affect processing and curing properties. However, previous work (ref. 5) indicates that optimization of HTS based cure systems can be achieved through statistically designed experiments, giving the compounder the flexibility to select vulcanization systems which best fit the performance requirements.

Recent developments

Continued developments in search of crosslink density stability for high temperatures cured and high heat resistant compounds led to the commercialization of 1,3 bis(citraconimidomethyl benzene), BCI-MX, in 1995. BCI-MX (figure 9) works by a crosslink compensation mechanism (ref. 6).

[Figure 9 ILLUSTRATION OMITTED]

It is a material that reacts to form stable crosslinks in sulfur vulcanized rubber by the compensation mechanism. It can substantially reduce the deterioration of rubber compound caused by reversion. These new crosslinks produce a vulcanized network which is resistant to overcure and provides improved high temperature performance. BCIMX is unique in that it only becomes active after crosslinks begin to revert, therefore compound designed properties are maintained.

A schematic representation showing the incorporation of BCI-MX crosslinks during the reversion process is shown in figure 10.

[Figure 10 ILLUSTRATION OMITTED]

The compensation of the loss of crosslinks during reversion is attained by the incorporation of thermally stable BCIMX crosslinks. The curing curves for a natural rubber compound vulcanizate with a conventional cure system are shown in figure 11. It is clear that the control compound, as expected, exhibits reversion during continued curing, while the compound with BCI-MX shows no reversion. Quantitatively, the control has 18% reversion and the BCI-MX has none.

[Figure 11 ILLUSTRATION OMITTED]

The effect of using this technique in high temperature curing is illustrated in table 6.

Table 6 - effect of BCI-MX on NR vulcanizates - high
cure temperature

Mixes                              01          02
                            (control)    (BCI-MX)
Cure temp., [degrees] C         150.0       160.0
Cure time, mins.                 60.0        30.0

Crosslink density(*)              3.6         3.7
% monosulfide                    30.0        35.0
% disulfide                      10.0        15.0
% polysulfide                    60.0        20.0
% C-C crosslinks                   --        30.0

(*) Crosslink density = [([2M.sub.c]Chem).sup.-1]x105 gmol/gRH

The data show that incorporation of BCI-MX allows the crosslink density at 160 [degrees] C to remain at a similar level as the conventional cure system vulcanized at 150 [degrees] C. This is being accomplished by the formation of C-C crosslinks which is in accordance with the model studies (ref. 7).

Processing/curing and physical properties of the control and the BCI-MX compounds are shown in tables 7 and 8.

Table 7 - processing and curing properties - NR high temperature

Mixes                                           01         02
                                         (control)   (BCI-MX)
Mooney scorch at 121 [degrees] C, min.          33         31
Rheometer MDR 2000
Temp., [degrees] C                             150        160
[t.sub.90], min.                                12          6
Table 8 - effect of physical properties - NR high temperature cure

Mixes                                       01               02
                                         (control)        (BCI-MX)
Cure time, [degrees] C min.             150     160         160
Cure time, min.                          60      30          30

Physical properties
Shore A hardness                         69      67          69
Tensile strength, MPa                  22.0    20.4        21.7
Elongation at break, %                  450     430         440
Modulus, MPa (300%)                      12      10          13
Tear strength, kN/m (crescent)           76      56          70
Abrasion, DIN, [mm.sup.3]               157     178         148

Increasing the cure temperature from 150 [degrees] C to 160 [degrees] C results in a slight reduction in scorch safety and significant reduction in rheometer t90 cure.

With the control system, increasing the cure temperature drastically reduces the modulus, tensile and tear strength. The addition of BCI-MX gives a relatively good match of physical properties.

The data suggest that initial physical properties are dominated by the crosslink density, with the crosslink distribution (table 6) being of less importance. A correlation between crosslink density and modulus is to be expected, but the relative insensitivity of tensile strength to crosslink distribution is less obvious as it has been proposed that a high proportion of polysulfidic crosslinks are necessary to achieve high tensile strength (ref. 7), although this finding has been disputed (refs. 8 and 9). The data presented in tables 7 and 8 show that BCI-MX compensates for the loss of crosslinks, thus allowing the maintenance of properties when curing at the higher temperature.

In addition to maintaining crosslink stability under high temperature cure conditions, BCI-MX also substantially reduces the deterioration of rubber compound physical properties caused by reversion.

An extended heat build-up test with a Goodrich Flexometer showed that the BCI-MX containing compound attained an equilibrium temperature in less than an hour and maintained that temperature for six hours without failure, while the control compound could not survive one hour.

Summary:

* Increased curing temperatures will result in significant reduction in cure time; the degree will depend on the magnitude of the temperature increase.

* Increased curing temperatures result in loss of state of cure or crosslink density.

* Maintaining vulcanizate properties at high curing temperatures may be achieved by increasing accelerator concentration at constant sulfur.

* Recent developments introduce two new approaches to improve rubber compound thermal stability to a greater extent than before through (i) formation of stable hybrid crosslinks by using hexamethylene-1,6-bisthiosulfate disodium salt, dihydrate (HTS) and (ii) crosslinks compensation by using 1,3 bis(citraconimidomethyl)benzene (BCIMX). HTS should be the system of choice for good reversion resistance and oxidative aging properties with improved tear or fatigue. BCI-MX should be selected where optimum reversion resistance is required combined with lower heat generation to ensure longer product service life.

In conclusion, increased productivity through high temperature cure and enhanced product stability through high heat resistance compounding can be realized without compromise on performance.

References:

[1.] J.R. Pyne, d. Inst. Rubber Ind. 7, 22 (1973).

[2.] C.M. Blow and C.T. Loo, d. Inst. Rubber Ind. 7, 205 (1973).

[3.] C.T. Loo, Polymer 15, 357 (1974).

[4.] D.G. Lloyd, European Rubber Journal, 27, 1988.

[5.] W.H. Helt, B.H. To and W.W. Paris, "Post vulcanization stabilization for natural rubber," presented at Ausplas `90, Sydney, Australia, 1990.

[6.] A.H. Hogt, A.G. Talma, R.F. de Block and R.N. Datta, U.S. Pat. 5,426,155, 1995.

[7.] R.N. Datta and M.S. Ivany, Rubber World, 212, 24, 1995.

[8.] L. Bateman, et al in the "Chemistry and physics of rubber-like substances," J. Willey, NY (1963).

[9.] J. Lal, Rubber Chemistry & Technology, 1970, 43, 664.

[10.] E.J. Blackman, E.B. McCall, Rubber Chemistry & Technology, 1971, 43, 651.