Abstract
The three principal approaches used in the structural resolution of sulfur vulcanizates of natural rubber are described and exemplified by their application to three systems: an unaccelerated sulfur system and two mercaptobenzothiazole (MBT)-accelerated sulfur systems, one of which makes very efficient use and the other inefficient use of the sulfur crosslinking. The ultimate aim of such studies is to relate vulcanizate structure to the physical properties and aging behavior of the vulcanizates during service. The first approach, which gives an overall measure of the structural complexity of the network, involves the determination of the crosslinking efficiency, (E), that is, the number of sulfur atoms combined in the network for each physically-effective, chemical crosslink formed. Crosslinking efficiencies of networks vary with increasing cure time from ca. 6-1.16 for the efficient-MBT system to 55-40 for the unaccelerated sulfur system. Between these extremes, it is found that variations of reactant concentrations and of temperature and time of cure lead to large variations in crosslinking efficiency and therefore of network complexity. Related work not reported here indicates that the latter features are also dependent on the type of accelerator used. More detailed information on network structure is obtained from analysis of reaction products of low molecular weight analogs of natural rubber with the vulcanizing agent and ancillary ingredients. Such studies reveal that the poor efficiency of the unaccelerated sulfur system is due to sulfur being combined wastefully in (i) long polysulfidic crosslinks of alkenyl t-alkyl structure, (ii) vicinal crosslinks which behave physically as one crosslink, and (iii) a large proportion of cyclic monosulfide groups which constitute an important modification of the main rubber chains. Conjugated triene groups, cis, trans-isomerized isoprene units, and main chain scission represent other possible types of modification. In contrast with this complex network, efficiently-cured MBT-accelerated sulfur vulcanizates contain, at early stages of cure, polysulfidic crosslinks of dialkenyl type which decrease with time to mainly monosulfide crosslinks; conjugated triene units are also present but little or no cyclic monosulfides, vicinal crosslinks, or main chain scission. Other accelerated sulfur vulcanizates possess structures intermediate between these two extremes, except that the crosslinks are generally of dialkenyl type and pendent groups terminated by accelerator residues constitute an additional possible type of modification. Knowledge from model olefin studies of the different types of sulfurated groups present in vulcanizates has led to the development of the third experimental approach, i.e. use of chemical reagents (‘chemical probes’) to determine specific groups. This approach is exemplified by reaction with actual vulcanizates of triphenylphosphine which removes sulfur atoms in excess of one or two from polysulfides; as cure proceeds in unaccelerated sulfur vulcanization, the proportion of combined sulfur present in cyclic sulfide groups rises from 75% to 95% while the average number of sulfur atoms in each crosslink unit falls from 12–13 to 2–4. Comparable data for the efficient MBT system confirm the essential simplicity of the network in this case.
