It has been shown previously2 that ald-chlorimines (RCH=NCI) readily undergo thermal decomposition. Nitriles and hydrogen chloride have been qualitatively identified as primary products.In this paper is reported the synthesis of some new ald-chlorimines and the results of a study of the relative thermal stabilities of these and other ald-chlorimines. The products of thermal decomposition of certain of these compounds have been determined semi-quantitatively at temperatures where their decomposition is practically instantaneous. ExperimentalReagents.---Most of the aldehydes used in this work were Eastman or Kahlbaum products. Solids were recrystallized if melting points were low. Liquids were redistilled before use. Eastman 4-methylbenzaldehyde was purified through a crystalline addition product with phosphoric acid.8 2-Bromo-* and 3-bromoJ benzaldehydes, and 4nitrocinnamaldehydee were synthesized by methods described in the literature. 3-Nitrocinnamaldehyde was prepared by a method similar to that used for the preparation of the para isomer.Monochloramine was prepared as described previously,* by mixing equivalent amounts of molar solutions of sodium hypochlorite? and ammonium hydroxide containing crushed ice.Preparation of Ald-chlorimiies.-The following general method includes certain modifications of the one previously described.* Six grams of a liquid aldehyde was shaken with 400 cc. of freshly prepared monochlorimine solution containing about 100 g. of crushed ice. If a solid product formed it was coagulated by shaking and then filtered rapidly with suction. The product was washed free from alkali with cold water. If no solid had formed within a short time (five to twenty minutes) the oil was extracted (1) This paper is from portions of theses presented by A. G. Gillaspie and John W. LeMaistre in partial fulfilment of the requirements for the degree of Doctor of Philosophy in the Graduate School of Duke Umversity, (2) Hauser, Hauser and Gillaspie, THIS JOURNAL, 82, 4158 (1930). (3) Raikow and Schtarbanoa, Chem.-Zeit., 86, 1135 (1901). (4) Brady, Cosson and Roper, J . Chem. SOC., 127, 2429 (1925). The procedure n a s modified, 10 g. of chromium trioxide being dissolved in 30 g. of acetic anhydride instead of glacial acetic acid as recorded in this reference. ( 5 ) Einhorn arid Gemsheim. Ant& , 284, 141 (1894). (ti) Fecht, Ber., 40, 3893 (1907). (7) A convenient sourcc of hypochlorite was found to be "H. T. I$.." a calcium hypochlorite product of the Mathieson Alkali Works. An aqueous solution of "H. T. H." was treated with sodium carbonate, the precipitate allowed to settle, and the sodium hypochlorite siphoned off 'I-required. 567with ether or drawn off with a pipet. Throughout the procedure the reaction mixture should be kept a t 0".A modified procedure was used for solid aldehydes, the latter being dissolved in 30-40 cc. of ether and the solution added to 400 cc. of aqueous monochloramine reagent so that two layers formed. The mixture was then shaken vigorously. A precipitate formed almost immediately. Afte...

No abstract

No abstract

In connection with a study of the reactions of compounds of the type RCH=NX (where X is halogen, acetate, etc.) with bases1, it seemed of interest to investigate the reactions of these compounds with Grignard reagents. The results obtained with ald-chlorimines and certain Grignard reagents are reported in this paper.The characteristic group of ald-chlorimines, (-CH=NC1), should be capable of undergoing at least three different types of reaction, involving, (1) the elimination of hydrogen chloride, (2) a reaction in which the nitrogen-chlorine group is attacked (as in hydrolysis), or (3) addition to the carbon-nitrogen double bond. A previous study10 has shown that in the presence of certain bases, such as alcoholic alkali, only the first reaction occurs to an appreciable extent, giving in certain cases practically quantitative yields of nitrile. The present investigation shows that in the presence of Grignard reagents, the first two types of reaction occur, the second predominating; apparently, addition to the carbon-nitrogen double bond (third type) does not take place to an appreciable extent with these reagents.The reactions of ald-chlorimines with Grignard reagents has been carried out by slowly adding the latter to the chlorimine in ether solution. In this way secondary reactions of the Grignards with the products (e. g., nitriles) was minimized. The products obtained from the reactions of aldchlorimines with ethyl-, phenyl-, or p-chlorophenylmagnesium bromide may be accounted for by equations (I) and (II). In I, the Grignard reagent removes hydrogen chloride from the ald-chlorimine to form a nitrile and R'H, while in II, the Grignard reagent is chlorinated, giving R'Cl and a nitrogen-magnesium compound.

Although the addition of ethylene oxide to starch (3) and cellulose (2) has been reported, no instance of a reaction of this type with sugars of low molecular weight has been found in the literature. We have, therefore, studied the action of ethylene oxide on sucrose.Sucrose and ethylene oxide did not react appreciably in neutral aqueous solution or in liquid ethylene oxide but reacted readily a t room temperature in aqueous sodium hydroxide solution. Reactions were carried out with molar ratios of ethylene oxide to sucrose of 1 , 2 , 3 , 4 , and 11. In all cases the ethylene oxide was entirely converted to non-volatile products and only in the first case could any unchanged sucrose be recovered from the reaction mixtures. The products were gummy or glassy mixtures whose sweetness decreased with increasing ethylene oxide content. Attempts to separate these mixtures into their components were not very successful. One crystalline substance was isolated in small yield by utilizing as seeds a few crystals which appeared spontaneously in some of the mixtures. The composition of this material agreed essentially with that calculated for a bis-(j3-hydroxyethy1)sucrose. Figure 1 shows a photomicrograph of the crystals.The products of these reactions were gums which were soluble in mater, methanol, and pyridine and insoluble in ether, chloroform, and acetone. Their solubility in ethanol and dioxane increased with the proportion of ethylene oxide. They were hygroscopic in about the same degree as glycerol. They did not depress the surface tension of water except when the molar ratio of ethylene oxide to sucrose was 4 : l or over and then the effect was small. In this respect the addition products resembled sucrose (1).Since it did not appear feasible to isolate the individual components of the reaction mixtures, indirect evidence was obtained that these were addition products and not mixtures of sucrose with polyethylene glycols. The reacted mixtures had about the same observed optical rotations as the original sucrose solutions, but acid hydrolysis of the former gave final values of rotation that differed significantly from those obtained by inverting the sucrose solutions. These inverted solutions also differed from those of sucrose in their reaction with phenylhydrazine. As the ratio of ethylene oxide to sucrose increased, the yield of D-glucose phenylosazone decreased. When the molar ratio exceeded two to one, orange-red oils only were obtained rather than crystalline osazone. Further, the acetylation of the reaction products gave acetates whose saponification