1098 



THE BELL SYSTEM TECHNICAL JOURNAL, OCTOBER 1951 



ineffective collisions in the absence of an inhibitor and depends on chemical 

 structure) and on the efficiency of the antioxidant but, in some cases, par- 

 ticularly with Hquids, very remarkable protection is obtainable as shown in 

 Fig. 13. (The oxidation of the control sample is so fast at this high tempera- 

 ture that the autocatalytic period is not evident.) The effect is less in solids 

 but is still of great value. Antioxidants are of the greatest benefit where the 

 rate of initiation is low, a condition usually true of thermal oxidation. The 



POLYETHYLENE IN OXYGEN AT 150 °C 

 I g SAMPLES IN BOATS 5/ 8" X 4" 



Fig. 13-Polethylene in oxygen at 150° C, 1 g samples in boats f X V. 



reason is that since antioxidant is consumed^^- ^ in doing its job the rather 

 limited amounts which can be added from a practical point of view (usually 

 not over 1 or 2%) do not last long if the rate of chain formation is very high. 

 This explains the oft stated fact that an antioxidant is far more effective if 

 added before oxidation starts, than if added after oxidation has proceeded 

 for a while." In the latter case enough peroxide will have been produced to 

 overwhelm the antioxidant relatively quickly. This is also the explanation of 



