April, 1031
COMMUNICATIONS TO THE EDITOR
889
tered glass wick in an atmosphere of dried air until 600 cc. of water had been condensed in each case. After careful purification the water from kerosene had a density seven parts per million greater than that of ordinary water while the benzene water was about eight parts per million heavier. Water obtained from the combustion of honey was four parts per million heavier than FRICKCHEMICAL LABORATORY HUGHS. TAYLOR that of ordinary water. These data are comparPRINCETON LNIVEFSITY P. W . SELWOOD able to the results of Washburn and Smith, who PRINCETON, N. J . found that water from the combined hydrogen RECEIVED MARCH21, 1934 in the willow tree was between five and six parts per million more dense than ordinary water. I THE NATURAL SEPARATION OF THE ISOTOPES hope to investigate natural products of an animal OF HYDROGEN origin in the near future. Sir: The data for kerosene may shed some light on The densities of water obtained from the At- the origin of petroleum. lantic Ocean,’ from the Dead Sea and the Great DEPARTMENT OF CHEMISTRY MALCOLM DOLE UNIVERSITY Salt Lake and from the hydrated mineral rasorite NORTHWESTERN EVANSTON, ILLINOIS are greater than that of fresh water by several RECEIVED MARCH21, 1934 parts per million, thus indicating a natural isotopic fractionation of water presumably by distillation.2 The discoveries of Lewis and of ACCELERATED DIFFUSION IN DYE SOLUTIONS Taylor, Swingle, Eyring and Frost3of the inability Sir : of living organisms to live or to grow in deuterium In a recent paper [THISJOURNAL, 56, 52 oxide have suggested the possibility of a natural (1934)], McBain and Dawson give striking illusselection of protium in preference to deuterium trations of accelerated diffusion, of electrolytes in the animal a,nd vegetable kingdom. On the in aqueous solution, through a porous glass memother hand, dilute solutions of deuterium oxide brane. We wish to draw attention to another seem to stimulate physiological a ~ t i v i t y ,a~ type of accelerated diffusion which has recently fact which would lead one to predict that deu- been observed in this Laboratory. Our obterium could possibly be concentrated physiologi- servations are for the free diffusion of azo dyes in cally. Washburn and Smith5 have found that the presence of electrolytes as determined by the in the case of a growing willow tree an isotopic Furth [Kolloid-Z., 41, 300 (1927)] micro-diffractionation of hydrogen in the direction of a fusion method. This method is particularly preferential selection of deuterium does occur. adaptable to observation of effects arising from Before this interesting work of Washburn and rapid motion of color ions, and gives results with Smith was known to me, I had begun an investi- an average experimental variation of less than 5%. gation of the relative abundance of deuterium in When an aqueous solution of the sodium salt of organic compounds by burning the compounds p - sulfobenzene - azo - benzene - azo - 6 - benzoylin air or oxygen, condensing the resulting water amino-1-naphthol-3-sulfonic acid and sodium vapor and measuring the density of the water chloride is allowed to stand for a month a t 2j0, after purification. Kerosene from the Oklahoma the rate of diffusion of the color ion is markedly oil fields and Mallinckrodt thiophene-free ben- affected by the presence of sodium chloride in the zene obtained in the destructive distillation of solution into which diffusion occurs. The results coal were separately burnt a t the surface of a sin- are summarized in Table I. (1) E. S . Gilfillan, Jr., THISJOURNAL, 66, 406 (1934). Numerous attempts were made to carry out (2) Bur. Standards Tech. News Bull. 196, August, 1933. diffusion from dye solutions containing 1.0 g./l. 66, , 3503 (1933); H. S. Taylor, (3) G.N. Lewis, THIS J O U R N A L W. W. Swingle, Henry Eyring and A . A . Frost, J. Chem. Physics, 1, of sodium chloride, into sodium chloride solutions 751 (1933). varying from 0.75-1.0 g./L concentration. In (4) T. C. Barnes, THISJ O U R N A L 66, , 4332 (1933); A m . J . Bol., 20, 681 (1933), S. L.hIeyer, S c i e m e , 79, 210 (1934); 0. W. Richards, every instance, measurement of the diffusion A m . J . B o f . ,209, 679 (1938). constant proved impossible owing to an accel(6) E. W. Washburn and E. R . Smith, Sciencc, 79, 188 (1934). from the work of La Mer, Eichelberger and Urey [ibid., 56, 248 (1034)] on the freezing points of isotopic mixtures. Experience in the concentration of DzO from tap water has showp that the ratio DzO/IIzO in natural water is about 1 in 5 or G thousand, in agreement with the mass spectrographic study of Bleakney and Gould [Plzys. Rev., 41, 265 (1933)l.
