Review of vulcanization chemistry.

The chemical nature of a polymer determines the useful application characteristics of a vulcanizate. In addition, the nature (type and concentration) of the crosslinks also determines application and performance characteristics. Vulcanization is the transformation of an elastomer from a `plastic,' `formable' material into an `elastic' material by the formation of a three-dimensional network of polymer chains chemically bonded to one another. This transformation can occur through various chemical mechanisms such as nucleophilic substitution, endlinking, addition or condensation, free radical coupling and ring opening reactions, among others. The chemically reactive sites available on the polymer determine the chemistry employed.

The strength and dynamic mechanical properties of the vulcanizate depend not only on the nature of the polymer chain itself, but also is proportional to the number of network supporting chains in the resultant network. The crosslink density determines the number of network supporting chains; a network supporting chain is that chain between two junction points. The hardness and modulus of a vulcanizate increase with increasing crosslink density. Tear strength, fatigue life, toughness and tensile strength initially increase, reach a maximum and then decrease with increasing crosslink density. Hysteretic properties and set characteristics decrease with increasing crosslink density. More detailed information regarding the physical properties of vulcanizates can be found in other sources (ref. 1).

Rheometers measure the vulcanization characteristics of a compound. The rheometer shows the increase in crosslink density (as measured by an increase in torque) as a function of time. The rheometer applies an oscillating torsional strain to a sample under pressure at elevated temperature and measures the increase in torque resulting with an increase in crosslink density. Vulcanization characteristics such as time to the onset of cure, ts2; minimum torque, MI; maximum rate of cure, MxR; maximum torque, Mh; and torque reversion are easily characterized. This information is useful in designing compounds for various applications and processing characteristics. MI relates to processability characteristics such as extrudability. The onset of cure, ts2, helps determine the process safety. (Another useful instrument is the Mooney viscometer, used to measure the Mooney scorch time.) The maximum rate of vulcanization, or t90 (time to reach 90% of the maximum cure, or [t.sub.x], where x is x% of full cure) can be used to determine the press time required during production. Mh will relate to the tensile, fatigue, tear, hysteretic and set characteristics of the compound.

Introduction to sulfur vulcanization

Sulfur vulcanization is technologically the most important chemistry employed in the production of diene elastomer vulcanizates. Each elastomer shows differences in various aspects of its vulcanization chemistry. These differences arise from the physical and chemical nature of the elastomer under consideration. Two recent reviews discuss in detail the early work that has led to the prevailing theories on vulcanization: Chapman and Porter summarize the chemistry of sulfur vulcanization in natural rubber (ref. 2) and Kresja and Koenig cover sulfur vulcanization in various elastomers (ref. 3). Recently, quantitative structure activity relationship studies have shown strong correlations between structures representing a `representative' sulfurating agent and scorch delay and cure kinetics. The studies suggest a polar ionic mechanism accounts well for sulfur vulcanization (ref. 4).

Several classes of compounds behave as accelerators in sulfur vulcanization. Table 1 shows various commercially available materials useful either as accelerators or activators (secondary accelerators) in sulfur vulcanization. A feature common to vulcanization accelerators is some form of a tautomerizable double bond, most accelerators contain the -N=C-S-H functionality. This is the common structural unit found in all of the 2-mercapto-substituted nitrogen heterocyclic accelerators known today. The delayed action precursors, 2-mercaptobenzothiazole disulfide and sulfenamides of 2-mercaptobenzothiazole decompose, generating structures containing this N=C-S- functionality.

Table 1 - classes of vulcanization accelerators and activators

zClass                           Speed

Amine-aldehyde                   Slow
(condensation products)
Guanidines                       Medium
Benzothiazoles                   Semi-fast
Sulfenamides and sulfenimides    Fast, delayed action
Dithiophosphates and xanthates   Fast
Thiurams                         Very fast
Dithiocarbamates                 Very fast

zClass                           Acronyms

Amine-aldehyde
(condensation products)
Guanidines                       DPG, DOTG
Benzothiazoles                   MBT, MBTS
Sulfenamides and sulfenimides    CBS, TBBS, MBS, DIBS, TBSI
Dithiophosphates and xanthates   ZBDP
Thiurams                         TMTD, TMTM, TETD
Dithiocarbamates                 ZMDC, ABDC

A typical rubber vulcanizate will contain various ingredients in addition to the sulfur and accelerator. An example of natural rubber vulcanizate prepared using a conventional cure system is given in table 2. The rates of vulcanization and states of cure depend on the type of polymer, the amounts and type(s) of sulfur, accelerator and activator(s) (e.g. stearic acid, zinc oxide and/or secondary accelerators such as DPG or TMTD, etc.) used. The time to the onset of cure varies primarily with the class of accelerator used, but also with the amounts of sulfur and accelerator. Variations in results of differing formulations suggest that each component plays an important role in determining the rate and state of the cure of the compound.

Table 2 - composition of a typical rubber vulcanizate

Ingredient                Parts in formulation
Natural rubber                    100
N-330 black                        50
Oil                               3.0
Stearic acid                      2.0
Zinc oxide                        5.0
Antidegradent                     2.0
Sulfur                            2.4
Accelerator (e.g. TBBS)           0.6

Sulfenamides and sulfenimides are special classes of accelerators that provide for a long delay period before the onset of the network formation. Sulfenamides and sulfenimides are commercially important classes of accelerators. Delayed action and fast cure characteristics are important in the preparation of large components made of rubber, such as tires. Large items require a great deal of processing to prepare the final form. Once in the final form and in the curing press, vulcanization should commence rapidly to allow for high productivity. The mechanical shaping and forming processes involve mixing, calendering and extrusion operations, each of which produces considerable heating due to the viscous nature of the rubber compound. The delayed action provided by the sulfenamide and sulfenimide accelerators allows time for processing before the onset of vulcanization.

The mechanism of sulfur vulcanization

It is agreed that the accelerator and activators react to generate an active accelerator complex. This complex, then, interacts with sulfur, a sulfur donor and other activators to prepare the active sulfurating agent. The active sulfurating agent then reacts at the allylic sites of the polymer to form the rubber bound intermediate. This rubber bound intermediate can then react with another rubber bound intermediate or with another polymer chain to generate a crosslink. This rather complex series of reactions is summarized schematically in figure 2 (ref. 1). In this scheme, two intermediates are formed and postulated to be the active sulfurating agent. Structure A is the accelerator polysulfide (or corresponding zinc complex) and structure B is the zinc complex of an accelerator polythiolate. As of 1988, according to Chapman and Porter (ref. 1), "the question of whether A or B is the active sulfurating agent in zinc-containing systems is the major unsolved problem of sulfur vulcanization chemistry." Still other intermediates have been proposed (ref. 5).

[Figure 2 ILLUSTRATION OMITTED]

While all of the intermediates formed during sulfur vulcanization are difficult to isolate and quantify, structures derived from the accelerator with the key structure `2-mercaptobenzothiazole' can be isolated and studied. Sullivan (ref. 6) and coworkers used high performance liquid chromatographic techniques to follow the metabolism of various intermediates during the course of sulfur vulcanization accelerated by N-t-butylbenzothiazole-2-sulfenamide (TBBS) and N-t-butylbenzothiazole-2-sulfenimide (TBSI). In experiments using various sulfur to accelerator ratios, 2-mercaptobenzothiazole moieties in the form of sulfenamide or sulfenimide, disulfide, trisulfide and polysulfides up to hexasulfides were detected (i.e., structures of type A). From the start of the heating cycle until the onset of cure (as determined by the rheometer [t.sub.2]), these extracted intermediates totaled the equivalent amount of 2-mercaptobenzothiazole in the formulation. This result suggests that the formation of a polymer-bound-accelerator polysulfide intermediate did not occur to a significant extent to this point ([t.sub.2]). The maximum concentration of the polysulfide species occurred near the midpoint in time from the start of the experiment to the onset of vulcanization or 1/2 of [t.sub.2] (figure 3). From this point on, the concentration of MBT continually increased. The measurement of MBT, however, represents all forms of MBT, including the accelerator polythiolate zinc complexes of the type B. The onset of cure, [t.sub.2], appears to coincide with the depletion of all of the accelerator polysulfides and TBBS. Figure 3 shows these results graphically. These results corroborate the results of Campbell and Wise in experiments of sulfur vulcanization accelerated with MBTS (ref. 7). (Interestingly, free sulfur is present up to the time to reach maximum modulus.)

[Figure 3 ILLUSTRATION OMITTED]

In parallel NMR spectroscopic experiments on samples prepared for the same experiments, Krejsa and Koenig (ref. 8) report no detectable sulfurization of the polymer until a time very close to the onset of crosslinking, [t.sub.2]. Krejsa (ref. 9) concluded that the MBT formation was a result of sulfurization of the polymer backbone by the accelerator polysulfides of type A. However, the NMR sulfurization measurements dispute this conclusion, since MBT formation precedes sulfurization and the accelerator polysulfides are nearly depleted at the onset of cure. This evidence suggests that the active sulfurating species is not the accelerator polysulfides, but is likely to be the accelerator polythiolate or accelerator thiolate species.

Coran postulated a simplified four-step reaction scheme to explain the kinetics of delayed action sulfur vulcanization (ref. 10). This simplified scheme adequately accounted for the kinetics of crosslink formation. Recently, Coran has expanded the simple four-step scheme to a five-step scheme. The fifth step generalizes competitive decomposition reactions accounting for changes in state of cure and rates of vulcanization observed in vulcanizates cured at different temperatures (ref. 11).

Based on the accelerator metabolism results reported by Campbell and Wise (ref. 7), and Sullivan (ref. 5) et al., we can outline a sequence of generalized reactions which fit into a four step scheme suggested by Coran (ref. 10) (modified here to include network maturation and reversion). These reactions can then be used to characterize sulfur vulcanization accelerated by sulfenimides and sulfenamides.

Initiation

The initial reaction involves the decomposition of a portion of the sulfenamide or sulfenimide accelerator. This step is likely to involve a reductive or hydrolytic (or both) cleavage of the sulfenamide or sulfenimide -S-N-bond(s). The reducing agent may be elemental sulfur, thiolate, an electron rich antidegradant or moieties on the surface of carbon black. It may be mediated by zinc ion. The initiation reaction effectively liberates a quantity of amine and MBT. The amine, thus liberated, can react with [S.sub.8] to generate an activated sulfur capable of either insertion or reduction reactions. The net changes observed in this step are a slight reduction in free sulfur, sulfenamide or sulfenimide, and the formation of MBT and amine. The small but detectable reduction in sulfur concentration has been reported independently by Scheele, Wise and Campbell, and recently again by Sullivan et al.

Induction

Sulfur exchange reactions predominate in this step. The sulfur exchange reactions involve the small amount of sulfur (observed in the induction step) and the accelerator disulfides, sulfenimides and sulfenamides. These reactions result in the depletion of a large portion of the sulfenamide with the formation of accelerator polysulfides and disulfides. Maximum concentration of accelerator polysulfides marks the end of the induction period.

Activation

Accelerator polysulfides and disulfides formed in the induction step are depleted, liberating, most likely, MBTSx. These accelerator `poly-sulfur-thiolates' are likely to exist as zinc complexes. The end of the activation period is marked by the nearly complete disappearance of the sulfenamide accelerator.

Sulfurization and crosslinking

After the depletion of all the sulfenamide, di- and polysulfides, the polymer is sulfurated and crosslinked simultaneously.

Maturation and reversion

After free sulfur is consumed from the reaction, the accelerator complex continues to react with polysulfidic crosslinks (as though it were elemental sulfur). The reactions include extracting sulfur from the crosslinks, cleaving crosslinks, formation of cyclic sulfide structures in the polymer backbone, formation of conjugated unsaturation in the polymer backbone and the precipitation of zinc sulfide.

Two previously described mechanisms involve the proposed active sulfurating species A and B in figure 4. According to Chapman and Porter (ref. 1), the zinc accelerator polythiolates, B, react in a concerted fashion to sulfurate the polymer and form crosslinks. Coran (ref. 12) speculated that the intermediate is the accelerator polysulfidezinc complex, A. Coran's mechanism is based on a zinc mediated pericyclic reaction or `ene' reaction. Generalized mechanisms for these two reactions are depicted in figures 5 and 6.

[Figures 4-6 ILLUSTRATION OMITTED]

Recent studies using MOPAC AM1 semiempirical quantum mechanical calculations and CODESSA QSAR software yield excellent correlations of molecular descriptors to onset of cure and maximum rate of vulcanization (ref. 4). The results support previously proposed mechanisms describing the origin of scorch delay for the delayed action fast curing sulfenamide accelerator class. In addition, the results support a carbanionic concerted mechanism for the sulfurization and crosslinking reactions. However, different from either of the previously suggested mechanisms is the nature of the complex and the proton acceptor in the reaction. In this study, the accelerator thiolate fragment is considered directly coordinated to the zinc ion, not through a `poly-sulfur' or `poly-sulfur-thiolate' chain. QSAR studies correlated molecular descriptors to the onset of cure ([ts.sub.2]) and maximum rate of vulcanization (of SBR). Four parameters gave a correlation coefficient of [r.sup.2] = 0.9667 for maximum rate of vulcanization. The four parameters are as follows (the sign of the coefficient is given in parenthesis):

* Max. electron-electron repulsion for the S-Zn bond (negative);

* Max. exchange energy for H-N bond (positive);

* Max. electron-electron repulsion for C-N bond (positive);

* Molecular surface area (negative).

These data support the flow of electrons from the S-Zn bond into the C-N bond in the rate-determining step. Hence, accelerator thiolate complexes having lower electron density in the S-Zn bond, higher electron density in the C-N bond and a weaker interaction between the ligand N and Zn, favor an increased rate of reaction. This is consistent with a sulfurating intermediate such as that shown in figure 7, in which the accelerator thiolate is bonded directly to the zinc atom. The nitrogen atom in the heterocyclic ring acts as the proton acceptor and the accelerator is eliminated from the complex as the thione with concommittant precipitation of zinc sulfide. Steric and electronic factors determine the net distribution of charge and characteristics of the accelerator zinc complex.

[Figure 7 ILLUSTRATION OMITTED]

In addition to the structural and electronic effects of the accelerator zinc complex, the polymer plays an important role in determining the kinetics of vulcanization. Several factors are important in determining the rate, including:

* Concentration of double bonds in the polymer;

* polarity of the polymer;

* number of allylic hydrogen atoms; and

* free volume of the polymer.

The concentration of double bonds is important since the rate of reaction depends upon the concentration of the reactants. Closely related to this is the number of allylic hydrogen atoms. Comparatively, polyisoprene has seven allylic hydrogens or reactive sites compared to four for butadiene based elastomers. The higher number of reactive sites naturally lowers the entropy of reaction and favors higher rates. This is a major reason why predominately saturated polymers, such as butyl and EPDM rubber, have slower rates of reaction. The polarity of the solvent also affects rates of reaction. Polar solvents favor polar-ionic reactions, ie., reaction rates are higher in polar solvents. Likewise, polar polymers such as nitrile and isoprene rubbers tend to have faster rates of reaction than SBR, butyl and EPDM rubbers. Finally, the vulcanization reactions are diffusion-controlled reactions. The zinc accelerator complex must diffuse through the elastomer to achieve high network densities. As discussed earlier, the QSAR studies pointed out a negative correlation between the surface area of the accelerator zinc complex and the maximum rate of vulcanization. The larger, higher surface complexes diffuse slower through the rubber, exhibiting a slower rate of reaction.

Non-sulfur vulcanization of unsaturated elastomers

Non-sulfur vulcanization of unsaturated elastomers encompasses several different mechanistic chemistries. Perhaps the most widespread is free-radical cure chemistry; other chemistries include resins, epoxidized rubber, maleimides, C nitroso compounds, among others. For the most part, sulfur vulcanizates provide for tough networks which show excellent tensile, tear and fatigue properties. Non-sulfur cures, on the other hand, often show better aging, lower compression set and higher resilience, but are often inferior to sulfur vulcanizates for tear and fatigue characteristics.

Resin cures

Resin cure chemistry is dominated by `ene' reactions (otherwise known as pericyclic or electrocyclic reactions). Resin cure chemistry is often compared to the familiar analogue in organic chemistry, the Diels-Aider reaction, in which a diene reacts with an olefin. This generic reaction mechanism, i.e., pericyclic or electrocyclic chemistry, is persistent in many of the reagents of this category. The reagents reacting by this mechanism include resins, maleimides and C-nitroso-compounds.

The generic requirement for this chemistry is the presence of a conjugated diene and an olefin. This allows for the generalized `2 [Pi] + 2 [Pi] + 2 [Pi]' pericyclic reaction. This reaction is favored by electron donating substituents attached to the diene and electron withdrawing groups attached to the olefin.

Vulcanization employing resin chemistry follows a similar pathway, but is more correctly a `2[Alpha] + 2[Pi] + 2 [Pi]' reaction. Wherein, 2 electrons involved in the reaction are derived from a C-H sigma bond. Common groups found in this chemistry are the substituted phenols and melamine derivatives. Figure 8 presents some example structures. The methylated methylol-melamines, phenol formaldehyde resins also provide suitable materials for resin cure chemistry.

[Figure 8 ILLUSTRATION OMITTED]

The resins are not in a suitable form to participate in this chemistry. These structures require activation by an elimination reaction. In the case of the halogenated phenolic resins, a dehydrohalogenation reaction (figure 9) will provide the suitable conjugated diene structure. The necessary dehydration, methanol elimination or dehydrohalogenation reactions are catalyzed by both Bronsted and Lewis acids (figure 9). Once the conjugated diene is formed, the `2[Sigma] + 2 [Pi] + 2 [Pi]' reaction proceeds. To complete the crosslink, a second dehydration occurs, followed by the cyclo-addition reaction (figure 10).

[Figures 9 and 10 ILLUSTRATION OMITTED]

Resin cures are used in components where high hardness is essential. This cure chemistry is often employed in combination with sulfur vulcanization. Typical components in tire compound employing resin cure include compounds requiring high hardness and low elongation, such as the bead, chafer, apex and wire skim compounds.

C-nitroso compounds react in a similar fashion (ref. 13). The resulting intermediate is an N hydroxy compound. These formulations often employ litharge. The litharge is implicated in converting the intermediate N-hydroxy compound the corresponding secondary amine (figure 11).

[Figure 11 ILLUSTRATION OMITTED]

Vulcanization with metal oxides

Metal oxides react with the halogenated elastomers, such as chloroprene and halobutyl rubber (figure 12). Crosslinking can be effected by the action of zinc oxide, magnesium oxide, litharge alone or in combination, and almost always with the aid of a solubilizing activator, such as stearic acid, zinc stearate or zinc octoate. The common feature of these elastomers is the presence of a halogen atom susceptible to nucleophilic substitution chemistry.

[Figure 12 ILLUSTRATION OMITTED]

The most active chemistry involves allylic substituted halogen atoms. In chloroprene, these groups are typically present at levels less than five percent. The reaction in this type of structure involves an addition-elimination reaction or a 1-3 substitution reaction (ref. 14). The oxygen of the metal oxide adds to the double bond, with subsequent elimination of the halogen ion and rearrangement of the double bond. The elimination of the CI may be assisted by zinc ion. Crosslinking is completed when this intermediate polymeric metal oxide reacts with another reactive site on the polymer of another polymeric metal oxide.

These reactions can also be activated with sulfur and sulfur containing compounds. Ethylene thiourea was commonly used as an activator, however suspected carcinogenicity is limiting its use (figure 13). Thiocarbanalide is potentially suitable for a replacement chemical, and other reagents have been proposed (ref. 15). Activation by ethylene thiourea has been discussed previously (ref. 16). The activation is provided by the enhanced nucleophilicity of the sulfur atom. The metal oxide is still involved in the reaction, but in a different fashion. The first step is the formation of the isothiuronium halide. The second step, shown in figure 13, is the `hydrolysis' of this intermediate by the metal oxide, wherein the oxygen is supplied by the metal oxide. Finally, the resulting polymer-bound metal sulfide completes the crosslink by nucleophilic attack on another polymer chain.

Vulcanization of BIMS polymer-brominated poly(isobutylene-co-p-methylstyrene)

Metal oxide cure of brominated poly(isobutylene-co-p-methylstyrene), BIMS polymer, is unique to the halogenated elastomers. In this polymer, metal oxide cure proceeds via electrophilic aromatic substitution chemistry, not nucleophilic substitution chemistry. The reaction is catalyzed by the in situ formation of the Lewis acid catalyst zinc hydroxybromide (figure 14). This is thought to form from the reaction of advantitious water or by the action of stearic acid. The zinc hydroxybromide then catalyzes an electrophilic aromatic substitution reaction forming `diphenylmethyl' crosslinks.

[Figures 13-14 ILLUSTRATION OMITTED]

MBTS and sulfur are sometimes used in BIMS compounds. The MBTS decomposes to MBT by the action of zinc oxide and sulfur. Since MBT is a mono-functional nucleophile, it will react with the benzyl bromide sites forming nonreactive mercaptobenzothiazyl benzyl sulfides. MBT is thus a scavenger of reactive sites (figure 15). This chemistry modulates the degree of cure. It also serves to increase the scorch delay of BIMS elastomer. Thus, increasing the level of MBT or MBTS in a compound results in a lower state of cure for the BIMS elastomers.

[Figure 15 ILLUSTRATION OMITTED]

Bis(bunte) salts effect nucleophilic cure of BIMS elastomer. The reaction involves the hydrolysis of the bunte salt followed by the nucleophilic reaction of the bis(thiolate) with the benzyl bromides (figure 16). The bis thiolate reacts with the benzyl bromides groups to generate hexamethylene bis(benzyl) sulfide crosslinks.

[Figure 16 ILLUSTRATION OMITTED]

Bis(bunte) salts are also very effective in providing excellent cure characteristics in blends of BIMS with conventional diene elastomers (ref. 17). Bis(bunte) salts provide a uniform distribution of curatives between the phases of a blend of the dissimilar elastomers (ref. 18). Cure characteristics of NR/ BR/BIMS having a cure package of sulfenamides, sulfur, zinc oxide and a bis(bunte) salt provide rapid rates of vulcanization with excellent scorch delay characteristics (figure 17). And finally, BIMS can also be cured by bis(citraconamido-m-lene) (figure 18). This reagent is often used in combination with reactive resins and the cure mechanism has not been elucidated; however, it is believed to proceed via conventional resin chemistry. Table 3 provides a comparison of the three cure systems.

Table - 3

                Scorch delay      Rate              Extent of cure

Oxide/MBTS      Med.              Slow              Med. to low
Duralink HTS    Short             Very fast         High
                (with black)      (with black)
                Long              Fast
                (without black)   (without black)
Perkalink 900   Med.              Slow              Med.

[Figures 1, 17 and 18 ILLUSTRATION OMITTED]

Acknowledgements

"Low extractable lead stabilizers" is based on a paper given at the May, 1998 Rubber Division meeting.

"Review of vulcanization chemistry" is based on a paper given at the May, 1998 Rubber Division meeting.

"A review of the fundamentals of crosslinking with peroxides" is based on a paper given at the May, 1998 Rubber Division meeting.

"Thermoplastic elastomers for automotive applications - past, present and future" is based on a paper given at the October, 1998 Rubber Division meeting.

Tech Service

Lead Chemicals, International Lead Zinc Research Organization, New York.

7. The Migration of Lead Stabilizers to the Environment, available from Halstab, Hammond, Ind.

8. R.F. Grossman, "Effects of lead stabilizer structure on polymer applications," SPE Retec, Cincinnati, Oct. 1997.

9. Determination by Hill Top Laboratories, Cincinnati, Ohio per 40CFR, Part 762.

10. TCLP analyses by Gabriel Laboratories, Highland, Indiana per 40CFR, Part 261, Appendix II.

References

(1.) Frederick R. Eirich, Ed. "Science and technology of rubber," Academic Press, 1978; L. Bateman ed., "The chemistry and physics of rubber-like substances," MacLaren, London, 1963; P.J. Flory, "Principles of polymer chemistry," Cornell Univ. Press, Ithaca, NY 1953; Aklonis MacKnight and Shen, "Introduction to polymer viscoelasticity," Wiley Interscience, 1972, J.E. Mark.

(2.) A. V. Chapman and M. Porter, in "Natural rubber science and technology," A.D. Roberts, Ed., Oxford University Press, Oxford, 1988, p. 511-620.

(3.) M.R. Kresja and J.L. Koenig, Rubber Chem. Technol, 66, 376 (1993).

(4.) F. Ignatz-Hoover, A. Katritzky, M. Karelson and V. Lobanov, Rubber Chem. and Technol., in press.

(5.) K.J. Van der Koi and J. Sherrit, Rubber Division, ACS Meeting, Cleveland, OH, October 1993.

(6.) A.B. Sullivan, C.J. Hann and G.H. Kuhls, Rubber Chem. Technol, 65, 488 (1992). C.J. Hann, A.B. Sullivan, B.C. Host and G.H. Kuhls, Jr., Rubber Chem. Technol, 67, 76 (1994).

(7.) R.H. Campbell and R.W. Wise, Rubber Chem. Technol., 37, 650 (1964).

(8.) M.R. Krejsa and J.L. Koenig, Rubber Chem. Technol., 65, 427 (1992).

(9.) M.R. Krejsa and J.L. Koenig, Rubber Chem. Technol., 67, 348 (1994).

(10.) A.Y. Coran, Rubber Chem. and Technol. 37, 689 (1964).

(11.) R. Ding, A.I. Leonov and A. E Coran, Rubber Chem. and Technol. 69, 81 (1996).

(12.) A.Y. Coran, Rubber Chem. Technol., 37, 689 (1964). A. Y. Coran, Rubber Chem. Technol., 38, 1 (1965).

(13.) J. Rehner and P.J. Flory, Rubber Chem and Technol., 19, 900 (1946); R.F. Martell and D.E. Smith, Rubber Chem. and Technol., 35, 141 (1962); A.B. Sullivan, J. Org. Chem., 2811 (1966), L.M. Gan and C.H. Chew, Rubber Chem. and Technol., 56, 883 (1983).

(14.) W. Hoffman, "Vulcanization and vulcanizing agents," Maclaren, London, 1967.

(15.) K. Mori and Y. Nakamura, Rubber Chemistry and Technol., 57, 34 (1984); H. Kato and J. Jufita, Rubber Chemistry and Technol., 65, 949 (1982).

(16.) R. Pariser, Kunstoffe, 50, 623 (1960).

(17.) N. Newman and F. Ignatz-Hoover, Rubber Division ACS, Cleveland, Ohio, October, 1995.

(18.) Diaz, Golanka, Rubber Division ACS, Cleveland, Ohio, October, 1995.3