The Vulcanization of Rubber with Sulfur
Abstract E. H. Farmer compiled a review of vulcanization which was published in 1946 as an important chapter in Volume II of “Advances in Colloid Science”. About two years later, Normann Bergem published a review under the same title. These authors dealt with the subject from rather fundamental points of view, but, because of more recent work from many laboratories and the increasing importance of elastomers, a fresh survey seems warranted. Accordingly, the present review aims to bring a part of the subject up-to-date. The viewpoint taken differs from those of Farmer and Bergem in some respects which seem rather important. For example, emphasis will be placed on the various rates of reactions including those of the scorch and reversion periods and on the importance of possible intermediates such as hydrogen sulfide and suitably terminated sulfur chains. The kinetic theory of elastomeric behavior constitutes the background against which any current treatment of vulcanization should be developed. Actually, vulcanization is a part of this background. Thus, Busse considered the following conditions necessary for the existence of rubberlike elasticity: (1) the presence of long chain molecules with freely rotating links, (2) the presence of weak secondary forces between molecules, and (3) the crosslinking of molecules into a three dimensional network. The three conditions taken together, at least in present day thinking, imply that long range elasticity results from the kinetic activity of segments which resembles yet differs from the kinetic activity of gas molecules. A segment is composed at most of only a few monomer units and is terminated at either end with another segment, a crosslink or an endgroup. When terminated with an endgroup, its kinetic activity may even detract from network behavior and then it is a defect. When terminated by a crosslink or another segment it is an integral part of the network and as such can contribute to the overall elasticity. It is more or less obvious that segments terminated by crosslinks and the crosslinks themselves should display an activity different from that of segments distant from crosslinks. Crosslinks, though relatively massive, are expected to move in a limited way (microBrownian) as a result of the kinetic activity of nearby segments. Crosslinks are necessary but are thus a sort of network defect. Those of lower functionality probably can contribute beneficially to the overall kinetic activity.