Vol. 61
214
COMMUNICATIONS T O T H E EDITOR STEREOISOMERIC o-HYDROXYBENZOPHENONE OXIMES
is accompanied by partial loss of reducing power, presumably due to the formation of an anhydroSir : D-altrose to the extent of 57yo a t equilibrium.” At that time we attempted to isolate the assumed We have found, in the extension of earlier experiments on the chemical behavior accompany- anhydro-altrose by oxidizing the altrose coming hydrogen bonding [Blatt, THIS JOURNAL, ponent of this equilibrium mixture (ca. 12 g.) with 60, 205 (1938) 1, that bonded o-hydroxybenzo- bromine water in the presence of barium benzoate phenone oximes, such as (I), readily form charac- [Hudson and Isbell, ibid., 51, 2225 (1929)], reteristic sparingly soluble copper derivatives, moving the benzoic acid, barium, and bromine whereas stereoisomeric non-bonded oximes such in the customary way, and the D-altronic acid as its calcium salt; the anhydro-altrose remained as (11) do not. and was concentrated to a sirup with the expected high levorotation. After three years in a desiccator the sirup began to crystallize, and from i t has been obtained 2.5 g. of a new compound. This substance, from a consideration of its comI I1 position, rotation, and lack of reducing action The reaction is extremely sensitive, and the toward Fehling’s solution, must be the desired differentiation between stereoisomers is sharp. anhydride of altrose, and accordingly is named With the oxime (I) the limit of sensitivity is D-altrosan. It is readily soluble in water and in between 0.0001 g. and 0.00001 g. of oxime per methyl alcohol, sparingly soluble in absolute cc. of solution, using an ethereal solution of the ethyl alcohol. It was recrystallized as colorless, oxime and shaking with an excess of saturated transparent prisms by allowing its methyl alcohol aqueous copper acetate. Taking the mean of solution to evaporate in a desiccator over granuthese two values, the limiting concentration of lar calcium chloride. The melting point of Dthe oxime is approximately 0.0001 molar. The altrosan in a capillary tube appeared to be about non-bonded isomer (11) forms no insoluble copper 13.5’) after considerable preliminary sintering; derivative a t a concentration as high as 0.025 g. however, a microscopic examination indicated per cc., which is some five hundred-fold greater that the probable melting point lies between 80 than the minimum concentration necessary with and 90°, the tiny melted particles being too the bonded isomer. These experiments open a viscous to coalesce until the higher temperature new approach to the problem of the structure is reached. The rotation of D-altrosan, [ a I z 0 D of chelate metallic derivatives; they will be 215’ in water (c, 1)) is close to the -220’ estireported in detail a t a later date, together with mated “from the rotation of D-altrose (f32’) and the conclusions t o which they lead. the value -98’ for the equilibrium mixture conHOWARD VXIVERSITY A . H. BLATT taining 43y0 D-altrose.” Heated for two and ~VASHIXGTON, D. C. one-half hours a t 98-100’, a solution of D-altrosan RECEIVED DECEMBER 16, 1938 in N hydrochloric acid (c, 1) became strongly reducing toward Fehling’s solution, and changed in rotation to [aI2O~ - 109.1’ ; calculated as altrose, CRYSTALLINE D-ALTROSAN’ this value becomes -98.2’, in perfect agreement Sir: with the rotation - 98.2’ previously reported for Several years ago i t was stated [Richtmyer and D-altrose heated with acid under the same conHudson, THISJOURNAL, 57, 1721 (1935)] that ditions. AnaZ.2 Calcd. for C~H1006:C, 44.44; “crystalline D-altrose is converted by hot hy- H, 6.22. Found: C, 44.46; H, 6.30. drochloric acid from [ a ] ” D 34 t o -98’; this The study of D-altrosan, and particularly the
+
(1) Publication authorized by the Surgeon General, E. S. Public Health Service.
(2) Through the kindness of Dr. W. T. Haskins of this Laboratory.
