ion -C-CH=CH+ -CH-'CH ' CH-) |-I ) Trimer -CHCH=-CH here that, in addition to those involving more reactive radicals, hydrogen atom abstraction from dimer via attack by allylic oleate radicals is also energetically feasible, the transfer being driven by the favorable difference in energy of secondary and tertiary allylic radicals.) Observed yields of products are also in agreement with the proposed reaction mechanisms. On the assumption that y-radiation yields of H(G 2.9)9 and of HO(G 2.4) are approximately correct at the slightly alkaline pH involved here (21), and that these reactive fragments of the solvent only either add to the double bond of the oleate solute molecules or abstract hydrogen from them, resulting ultimately in either case in the production of an equivalent quantity of allylic oleate radicals, then, on the basis of the observed yields of stearate and of hydroxystearate,1 the mechanism outlined above predicts an over-all loss 1It is of interest to note here that the yield of stearate is about 30% of the primary yield of reducing fragments from water radiolysis, whereas only about 16% of the hydroxyl radicals initially produced are isolated in the form of hydroxystearate. The significance of this differentiation should be tempered in light of the observed stability of stearate and of the unknown validity of the assumption that no appreciable amount of hydroxystearate, once formed, is converted to other products by secondary reactions. The apparent addition of either type of 182 This content downloaded from 157.55.39.17 on Fri, 02 Sep 2016 05:30:57 UTC All use subject to http://about.jstor.org/terms y-IRRADIATION OF POTASSIUM OLEATE of oleate amounting to G -6.55 and a dimer yield of G 2.65. Again, discrepancies between these figures and those actually observed (-5.8 and 1.8) are in the direction expected, on the basis of the proposed sparing of oleate at the expense of accumulating dimer. This view is also supported by the nature of the observed variation of yields as a function of dose of radiation (Table I). Although the results obtained in the present study are in accord with this admittedly oversimplified picture of the sorts of events which should reasonably be anticipated, certain structural features of the various individual substances comprising the dimer fraction are not. Thus, although diene dimer makes up somewhat more than half the total material in this fraction, the amount of moresaturated components is far too great to be explained by invoking processes analogous to those resulting in conversion of oleate to stearate. In addition to there being no obvious reason for believing that the dimers might be markedly more attractive than oleate itself as objectives of Haddition, the failure of stearate yield to be depressed at higher doses (see Table I) indicates clearly that the presence of greater amounts of dimer has no exceptional influence on the amount of Hadding to oleate. Also unexplained is the observation that the proportions of various types of dimer are not those expected on the basis of random combination of allylic oleate radicals. These apparently anomalous observations may possibly be explained in terms of consequences of the fact that the concentrations involved (1% by weight = 0.0355 M) are considerably above the critical micellar concentration (about 0.001 M) of potassium oleate at 26?C (22). Inasmuch as the solute therefore occurs to an appreciable (probably large) extent as micellar aggregates, it would appear reasonable to propose that concerted processes such as are presented here are involved in the formation of monoene dimer.