UNIVALENT OXYGEN. PREPARATION AND OXIDATION OF MESITOL
Journal of the American Chemical Society
When Gomberg isolated triophenyl-methyl and produced evidence that other radicals are capable of independent existence,' he revived the hopes expressed by Dumas, Kolbe and Frankland, and gave a new impetus to research in organic chemistry. At the present time several compounds of trivalent carbon are known; but with the exception of Kohler's triphenyl-indy12 they all belong to the triarylmethyl class. The search for new types of radicals, and especially those representing unsaturated conditions
... aturated conditions in elements other than carbon has not been very successful except in the discovery of compounds of divalent nitrogen derived from hydrazine and carbazole. Wieland found that tetra-anisyl hydrazine in benzene has an abnormally low molecular weight; develops a color which deviates from Beer's dilution law; and forms an addition product with nitric oxide. One or two other substituted hydrazines displayed similar proper tie^.^ Branch and Smith obtained a derivative of carbazole which dissociates into radicals of divalent n i t r~g e n .~ Porter and Borgstrom prepared tetraphenyl-diarsine5 and corresponding derivatives of phosphorus and although these products absorbed oxygen from the air and exhibited other evidences of unsaturation, their properties could be accounted for without assuming any measurable dissociation. A lead compound bis (lead triphenyl), (CsH&Pb-Pb(CaHb)3, acquires a color when heated but gives no other evidence of change in structure, and the corresponding derivative of silicon does not dissociate at aL6 An attempt to isolate an oxygen radical was made by Pummerer,' who oxidized several naphthol derivatives with potassium ferricyanide and other reagents, thus obtaining tautomeric products corresponding in composition to the original naphthols (or to the peroxides derived from them). Some of these bodies gave colored solutions deviating from Beer's law ; and the molecular weights were lower than the calculated values for peroxides. The many possible intramolecular rearrangements] however, made uncertain any deductions with respect to the structures of these oxidation products.