Dimer acid structures. The dehydro‐dimer from methyl stearate and di‐tertiary‐butyl peroxide

Previous work by Sutton, and by Harrison, McCaleb and Wheeler has shown that methyl stearate is converted to dimers plus higher polymers by the action of di‐t‐butyl peroxide. The latter suggested that in the dimer, there was considerable linkage at carbon 2, the carbon ॅ‐ to the COOCH3 group, since the dimer ester was incompletely saponified by the usual procedures. Further proof of ॅ‐linkage is now presented. A fraction which is predominantly ॅ‐,ॅ艂‐linked dimer was isolated as the nonpolymeric cyclic anhydride (ॅ‐,ॅ艂‐dicetyl succinic anhydride) by molecular distillation from the linear polymeric non‐ॅ‐linked polyanhydride. The isolated cyclic anhydride appeared identical with a synthetic ॅ‐,ॅ艂‐dicetyl succinic anhydride, whose synthesis is described. The original dimer ester, the ॅ‐,ॅ艂‐cyclic anhydride fraction, and the dimethyl ester derived from it, were examined by mass spectrometry. The expected mol wt were confirmed by the parent ion peaks. Fragmentation patterns indicated appreciable ॅ‐linkage in the original dimer ester, and almost exclusive ॅ‐linkage in the ester from the isolated cyclic anhydride. Aside from preference for the ॅ‐position, joining appears to be randomly distributed.