Showing papers in "Journal of the American Chemical Society in 1949"

A Spectrophotometric Investigation of the Interaction of Iodine with Aromatic Hydrocarbons

Abstract: Introduction The various colors of iodine in different solvents have attracted the attention of investigators for over half a century. Beckmann’ showed that the differences are not due to any change in the molecular weight of iodine and suggested solvation as the cause of brown color. Lachman2 supported this explanation, pointing out that the solvents which give a violet color are saturated, those which give brown colors are unsaturated. He stated that small additions of alcohol to a solution of iodine in chloroform suffice to shift the color stepwise from violet to brown. Hildebrand and Glascock3 found that when iodine and an alcohol were both added to a “violet” solvent, bromoform or ethylene bromide, the molal lowering of the freezing point was considerably less than additive, indicating combination of iodine with alcohol. Furthermore, they investigated colorimetrically the equilibria of iodine with ethyl alcohol, ethyl acetate and nitrobenzene dissolved together in a “violet” solvent such as carbon tetrachloride, chloroform or carbon disulfide, and found in each case an equilibrium constant corresponding to a 1 : 1 compound. They calculated the heat of combination in one case. There followed a long series of investigations of the general problem, well summarized by Gmelin4 and in a recent review by Kleinberg and Davidson.6 But the original division of iodine solutions into two or more distinct color classes became untenable with the work of Waentig, Groh’ and \\X7alkers which showed that the position of the absorption bands of iodine solutions in the visible region moved gradually from violet to brown, the shift being only small in the case of benzene in which the color of iodine is a red-violet. We have long been curious about the chemical basis for iodine solvates. It is not a question of dipole moment, as Walkers concluded, for we have recently shown rather conclusively that the iodine solubility, which would be altered by solvation effects, bears little or no relation to the dipole moment of the solvent mole~ule .~ The alcohols are extraordinary dipoles, and their effect is readily understandable, but the case of benzene seemed

Reduction of Steroid Ketones and other Carbonyl Compounds by Modified Wolff--Kishner Method

Abstract: According to Dutcher and Wintersteiner2 the reduction of C3-steroid ketones by the Wow-Kishner method under usual conditions gives mainly the corresponding Cs-epimeric carbinols with small amounts of the expected Cs-methylenic products. In the case of the a,P-unsaturated ketone, cholestenone, the reduction follows a still more complex course. In addition to the epimeric unsaturated carbinols and a small amount of the normal product, A4-cholestene, the saturated carbinols, a-coprosterol and P-cholestanol have also been isolated. The authors stated that the abnormal reduction can be explained by the assumption that the hydrazone or semicarbazone is partially hydrolyzed to free ketone, which is then reduced by sodium alcoholate to give the secondary alcohol. These possibilities are, however, eliminated by the modified WolffKishner r e d ~ c t i o n , ~ , ~ in which the water is evaporated during the heating period. Moreover, the excess of hydrazine hydrate used in this reaction should keep the concentration of the regenerated ketone a t a low level.2 In fact all the C3-steroid ketones so far investigated gave the normal C3-methylenic compounds by the modified Wolff-Kish20-diketopregnane (I) gave a reduced product melting a t 152-153' in which one of the two keto groups is unattacked. In view of the fact that the 20-keto compounds, such as A5-pregnen-3 (P)-ol20-one (IV) and its hydrogenation product, allopregnan-3(P)-ol-20-one (VI) can be transferred to A5-pregnan-3 (P)-ol (V) and 3,P-hydroxyallo-pregnane (VII), respectively, the reduced product from I may be formulated as 3,a-hydroxy-1 l-ketopreg-