Crosslinking sulfur vulcanisation

Most EPDM applications require crosslinking except when used as an impact modifier for PP, polystyrene (PS) and polyamides or as an oil additive, e.g., as viscosity index improver or dispersant. Most commonly, accelerated sulfur vulcanisation is used for the crosslinking of EPDM. As a result of the low amount of unsaturation in EPDM (< 1 mole/ kg versus NR -15 mole/kg), sulfur vulcanisation of EPDM is rather slow and a relatively large amount of accelerators is needed. Because of the low polarity of EPDM the solubility of polar accelerators is limited, often resulting in low effectivity and/or blooming. Typically, up to 5 different accelerators are used in EPDM formulations. As for other rubbers environmental issues, such as nitrosamine formation and may be in the future the presence of zinc, are prompting the development of new accelerator systems. 

Recently, it was shown using FT-IR that the decrease of vulcanisation rate and final crosslink density of sulfur vulcanised NR upon increasing silica content may be related to increased absorption of zinc stearate onto the silica surface [69]. 

Schotman and co-workers tentatively assigned the new Raman peaks at 1625 and 1592 cm 1 observed during sulfur vulcanisation of squalene, to the formation of conjugated dienes and trienes, respectively [70]. When vulcanisation was carried out in the presence of l,3-di(citraconimidomethyl)benzene, this resulted in a reduced intensity of these two new peaks, corroborating that conjugated dienes and trienes, formed as a result of reversion, react with the diimide. Obviously, the diimide is not an anti-reversion agent in the sense that it prevents reversion, but it is in the sense that it repairs crosslinks when reversion has occurred. 

The mechanism of the accelerated sulfur vulcanisation of EPDM is probably similar to that of the highly unsaturated polydiene rubbers. The vulcanisation of EPDM has been studied with emphasis on the cure behaviour and mechanical and elastic properties of the crosslinked EPDM. Hardly any spectroscopic studies on the crosslinking chemistry of EPDM have been published, not only because of the problems discussed in Section 6.1.3 but also because of the low amount of unsaturation of EPDM relative to the sensitivity of the analytical techniques. For instance, high-temperature magic-angle spinning solid-state 13C NMR spectroscopy of crosslinked EPDM just allows the identification of the rubber type, but spectroscopic evidence for the presence of crosslinks is not found [72]. 

Only two spectroscopic studies on sulfur vulcanisation of EPDM by Fujimoto and coworkers are available [73-74], Using attenuated total reflectance (ATR) IR spectroscopy they showed that during sulfur/TMTD/MBT/ZnO/stearic acid vulcanization, the C=C bands at 3035, 966 and 870 cm 1 of the residual unsaturations of the EPDM third monomers, DCPD, 1,4-hexadiene (HD) and 5-methylidene-2-norbornene (MNB), respectively, decreased in intensity as a function of time at 140 and 150 °C. The relative decrease in intensity was shown to correlate with the increase in crosslink density. In Sections 6.2.2.2 and 6.2.2.3 it will be shown that this decrease of intensity should not be interpreted as a loss of unsaturation during sulfur vulcanisation of EPDM. 

After these early studies an extensive FT-Raman study [77] was performed to bridge the gap between the low-molecular-weight ENBH model vulcanisation studies and the vulcanisation studies using high-molecular-weight EPDM. These studies will be presented in detail. First, a series of low-molecular-weight dialkenylsulfides will be discussed in order to determine the effect of sulfur vulcanisation on Raman spectra of olefins. Subsequently, the attachment of the sulfur crosslinks at the allylic positions, the conversion of ENB, the length of sulfur crosslinks and the network structure will be addressed for unfilled sulfur vulcanisates of amorphous EPDM. Some preliminary network structure/ properties relationships will also be presented. 

The analytical techniques discussed previously can be used to study the EPDM network as such or its formation in time as well as to determine relationships between the network structure and the properties of the vulcanisates. In a preliminary approach some typical vulcanised EPDM properties, i.e., hardness, tensile strength, elongation at break and tear strength, have been plotted as a function of chemical crosslink density (Figure 6.6). The latter is either determined directly via 1H NMR relaxation time measurements or calculated from the FT-Raman ENB conversion (Table 6.3). It is concluded that for these unfilled, sulfur-vulcanised, amorphous EPDM, the chemical crosslink density is the main parameter determining the vulcanisate properties. It is beyond the purpose of this review to discuss these relationships in a more detailed and theoretical way. 

The mechanism of peroxide crosslinking of elastomers is much less intricate than that of sulfur vulcanisation. Crosslinking is initiated by the thermal decomposition of a peroxide, which is the overall cure rate determining step. Next, the active radicals thus formed abstract hydrogen from elastomer chains to form macroradicals. Finally, crosslinking results either from the combination of two macroradicals or from the addition of a macroradical to an unsaturated moiety of another primary elastomer chain. 

For sulfur vulcanisation of EPDM it was shown that the relative ENB conversion (20 to 60%) is higher than often assumed. The absolute ENB conversion was shown to be governed by the vulcanisation recipe and to be independent of the EPDM type. For the ISO 4097 [82] recipe the average length of the sulfur crosslinks is 2.7 sulfur atoms. The number of converted ENB units per sulfur bridge is 2.0, indicating that crosslinks are formed predominantly. In a preliminary study it was shown that the mechanical properties of unfilled sulfur-vulcanised amorphous EPDM are determined by the chemical crosslink density. Clearly, these studies should be extended to other vulcanisation recipes and completely formulated compounds. Vulcanisation kinetics should be studied, preferably at different temperatures. 

There is wide variety of vulcanisation agents and methods available for crosslinking rubber materials including peroxide, radiation, urethane, amine-boranes, and sulfur compounds [20]. Because of its superior mechanical and elastic properties, ease in use, and low cost, sulfur vulcanisation is the most widely used. Although vulcanisation with sulfur alone is not practical compared to the accelerated sulfur vulcanisation in terms of the slower cure rate and inferior physical properties of the end products, many fundamental aspects can be learned from such a simply formulated vulcanisation system. The use of sulfur alone to cure NR is typically inefficient, i.e., requiring 45-55 sulfur atoms per crosslink [21], and tends to produce a large portion of intramolecular (cyclic) crosslinks. However, such ineffective crosslink structures are of interest in the understanding of complex nature of vulcanisation reactions. 

Model compounds based on 2-methyl-2-pentene were studied to supplement the 13C chemical shift assignments of the products from accelerated sulfur vulcanisation of NR. It is observed in the model compound data that it may not be possible to distinguish between a 13C NMR resonance which is due to disulfidic crosslinks and a peak due to pendent accelerator groups, while a large chemical shift difference ( 3 ppm) is observed for the monosulfidic bonds. The MBS-accelerated sample shows similar new resonances as seen in the TMTD accelerated systems. In this comparison however, the quantitative aspects of the data might be obscured due to the differences in the state of cure among the different formulations. 

The 13C chemical shifts were assigned in more detail for monosulfidic and polysulfidic crosslinks occurring in the accelerated sulfur vulcanisation of NR [18]. The NR was cured with a pure thiuram formulation (TMTD alone) in order to predominantly prepare monosulfidic bridges in the network. The distortionless enhancement by polarisation transfer (DEPT) experiments, in which the carbons with different level of protonation can be distinguished [22-24], were performed for the NR cured with extended levels of sulfur. Based on the DEPT results and previously reported model compound results [20], the chemical shifts of the resonances occurring in the spectra were assigned. 

Solid-state 13C MAS NMR has been applied for quantitative determination of crosslink density in accelerated sulfur-vulcanised NR [27]. The concepts used to calculate the crosslink density by 13C NMR are the same as the one mentioned above, but different resonances were used for the quantitative treatment based on the different assignments... 

The 13C NMR crosslink density results were compared with the crosslink density obtained by the mechanical measurements. In the determination of the crosslink density by mechanical methods, the contributions of the topological constraints on the results were neglected and the density was expressed as G/2RT. The 13C and mechanical-crosslink densities were obtained for both sulfur and dicumyl peroxide (DCP)-cured samples to ensure the effect of wasted crosslinks (pendent or intramolecular type sulfurisations), which are expected in the typical sulfur-vulcanisation of NR. In the major range of crosslink densities, the crosslink densities for those two systems are described by the same linear function with a slope of 1.0. Based on these observations, it is shown that the crosslink density of the sulfur-vulcanised NR as determined by 13C is identical with the true crosslink density, and the influence of the wasted or ineffective crosslinks (pendent and cyclic crosslinks) and chain ends is negligible. However, this conclusion seems to be only valid if the effect of topological constraints or entrapped entanglements on the mechanical modulus is negligible which is rarely the case in real systems. 

Vulcanisate structures of BR crosslinked with cyclic disulfides was studied by NMR.36 Using high resolution MAS techniques of DEPT, COSY, TOCSY, and HETCOR, the resulting spectra showed that crosslinking gave an addition product to the double bond and not the allylic structure found in typical sulfur vulcanisations. 

Figure 10.6 The distribution function of the chain length for two EPDM sulfur vulcanisates whose crosslink densities differ by a factor of about two and that for the physical mixture (50 50 mass%) of these vulcanisates [72]...

Figure 10.18 Dependency of the chemical crosslink density (1/2MC) on the actual vulcanisation time during accelerated sulfur vulcanisation of EPDM [179]. The line has been included to guide the eye. The vulcanisation temperature was 413 K (140 °C)...

These rubbers are based on atoms of silicon chains rather than carbon atoms. Their unique structure is responsible for their extreme temperature properties. The most common types of silicone rubbers are specfically polysilaxanes. The Si-O-Si bonds can rotate much more freely than the C-C bond or the C-O bond. So the silicone chain is much more flexible and less affected by temperature. Silicone rubber is vulcanised by the action of peroxides which crosslink the chains by abstracting hydrogen atoms from the methyl side groups, allowing the resulting free radicals to couple into a crosslink. Some varieties of polysiloxanes contain some vinyl methyl siloxane units, which permit sulfur vulcanisation at the double bonds. 

It should be recognised that appreciable shifts in properties are sometimes made possible by special compounding variations. For instance, the heat resistance of natural rubber vulcanisates may be improved considerably by variation of the vulcanising recipe. The normal sulfur vulcanisation system is capable of many variants which will govern the chemical nature of sulfur crosslinks, i.e., whether it is essentially a mono, di or polysulfide linkage. The nature of sulfur crosslinks can have considerable influence on the heat and chemical resistance of vulcanisates. 

The chains must be crosslinked to form a network (cf. Fig 7.16). In most elastomers containing double bonds, covalent bonds are introduced between chains. This can be done either with sulfur or polysulfide bonds (the well known sulfur vulcanisation of natural rubber is an example), or else by direct reactions between double bonds, initiated via decomposition of a peroxide additive into radicals. Double bonds already exist in the chemical structure of polyisoprene, polybutadiene and its copolymers. When this is not the case, as for silicones, ethylene-propylene copolymers and polyisobutylene, units are introduced by copolymerisation which have the property of conserving a double bond after incorporation into the chain. These double bonds can then be used for crosslinking. This is how Butyl rubber is made from polyisobutylene, by adding 2% isoprene. Butyl is a rubber with the remarkable property of being impermeable to air. It is used to line the interior of tyres with no inner tube. 

Structure of Sulfur Vulcanised Rubber and the Properties of Sulfur Crosslinks... 

Vulcanisation is the term used for the process in which the rubber molecules are lightly crosslinked in order to reduce plasticity and develop elasticity. It was originally applied to the use of sulfur for this purpose, but is now used for any similar process of cross-linking. Sulfur, though, remains the substance most widely used for this purpose. 

Thermoset polymers (sometimes called network polymers) can be formed from either monomers or low MW macromers that have a functionality of three or more (only one of the reagents requires this), or a pre-formed polymer by extensive crosslinking (also called curing or vulcanisation this latter term is only applied when sulfur is the vulcanising or crosslinking agent.) The crosslinks involve the formation of chemical bonds — covalent (e.g., carbon-carbon bonds) or ionic bonds. 

Raman and IR spectroscopic studies dealing with the qualitative and/or quantitative determination of rubber compounding ingredients, i.e., the elastomer itself [22, 26-31], fillers [32, 33], vulcanisation chemicals and other additives [34-37], are not included here. The same applies to studies dealing with the crosslinking of elastomers by means of chemicals other than sulfur or peroxide [38-41], self-crosslinking of elastomers blends [42-44], crystallisation (strain-induced) [45-48] and oxidation/ageing [49-53]. 

For all samples, except the vulcanisate of EPDM with 4.6 wt.% ENB and 1.5 phr sulfur, the ratio of the converted number of ENB moieties and the number of chemical crosslinks is around 2.0 ( 0.2). The statistical spread in the ratio of the converted number of ENB molecules and the number of chemical crosslinks is estimated to be 10-15% based on the error in the Raman and NMR data. Furthermore, there might be a systematic error in the crosslink density as determined by NMR, originating from the assumptions made in calculating the number of chemical crosslinks from the total number of crosslinks (chemical and entanglements). 

More precisely, the magnetic relaxation depends on the variable of gelation, i.e., the density of crosslinks, and is closely related to the modulus of elasticity, E, on the one hand and to the swelling ratio, Qm, on the other hand. Long polybutadiene chains are currently randomly crosslinked, using sulfur they can serve to illustrate the NMR approach to the characterisation of vulcanised polymers. It has been shown that the... 

---