362 BELL SYSTEM TECHNICAL JOURNAL 



Composition of GR-S and Its Determination 



Given a piece of GR-S synthetic rubber, our first task from the standpoint 

 of determining its chemical composition is to separate the pure copolymer 

 which is responsible for the rubber-like properties from the non-rubber 

 constituents. The latter comprise soaps or other emulsifying agents, fatty 

 acids, salts, antioxidant and low molecular weight, non-rubbery butadiene- 

 styrene products to the extent of several percent. Some of these minor 

 ingredients, like the antioxidant, are essential whereas others play no 

 important role subsequent to polymerization. All, however, must be 

 separated from the copolymer before it can be properly evaluated. The 

 analysis for the non-rubber components after separation is fairly straight- 

 forward and standard and will not be gone into here. 



It has been found that the azeotrope of toluene and ethyl alcohol which 

 consists of approximately 30 parts by volume of toluene to 70 parts by 

 volume of alcohol is an excellent extractant for the non-rubber compounds 

 and hence may be used to effect a separation^' ^ The procedure for isolat- 

 ing the copolymer is simply to place a quantity, say 10 grams, of the GR-S 

 in an extraction thimble supported in an extraction flask as shown in Fig. 2. 

 Another, more rapid, procedure is to reflux the azeotrope over the rubber 

 for two hours, when extraction has been found to be essentially complete. 

 This method is now used in the Standard Specification for all GR-S. The 

 pure copolymer, left as residue, is the product to which we now turn our 

 attention. 



As has been mentioned, the ratio in which butadiene and styrene are 

 employed in the starting mixture does not determine either the ratio in 

 the whole copolymer at a given stage of reaction or the ratio present 

 in any given chain molecule of the copolymer. Therefore the starting ratio 

 cannot be reUed upon to control the composition of the final copoh-mer. 

 Experiments show that under certain process conditions large differences in 

 composition between different fractions of the copolymer do occur. Even 

 under the best conditions theoretical considerations predict that variations 

 must occur between molecules since the ratio of the reactants is continuously 

 changing during the reaction. 



Let us examine the chemistr>' of the process for a moment to try better to 

 understand why these variations are possible. WTien styrene {S) reacts 

 with itself polystyrene {Sx) is formed. Analogously polybutadiene {By) is 

 formed in the case of butadiene {B). In GR-S both styrene and butadiene 

 react to give a copolymer. 



When a quantity of styrene undergoes polymerization, a distribution 

 consisting of various numbers of long chain molecules of various lengths is 

 formed. Thus, if we start with Ni molecules of styrene, S, the polymeriza- 



