Aug., 1937
THEDENSITIESOF MIXTURES OF LIGHTAND HEAVY WATER
1483
3. Iodine in alkaline media has been shown 4. The bacteriological and roentgenological to produce azo- and nitroso-benzenesulfon- properties of these substances are under invesamides. These products are of bacteriological tigation. interest. NEWYORK,N. Y. RECRIVED MAY29, 1937
[CONTRIBUTION FROM THE hBORATORIES OF THE
ROCKEFELLER INSTITUTE FOR MEDICAL RESEARCH]
The Densities of Mixtures of Light and Heavy Water BY L. G . LONGSWORTH In the preparation of some Hz0-D20 mixtures for use in a series of transference measurements with such mixtures as solvents, it was observed that the mole fraction of DzO in the solvent computed from the weights of HzO and DzO deviated by as much as 0.5% from the value obtained with the aid of the relation Nnto = 9.377 A S
-
1.01
sz
(1)
in which
This equation is due to Lewis and Luten' and modified by Baker and La Mer.2 It seemed desirable therefore to redetermine the densities of these mixtures as a function of the mole fraction, ND,o, of deuterium oxide. The starting material was a sample of heavy water having a specific gravity of 1.10700. Accepting the value of 1.107903 as the specific gravity of pure DzO the value of NDs0for this sample may be taken tentatively as 0.10700/ 0.10790 = 0.9916s. Weighed quantities of this and ordinary water were mixed and the density of the resulting liquid was determined. This process was repeated until N D ~ O had been decreased to 0.2 in steps of 0.2. The density measurements were made with the aid of a pycnometer similar to that described by Smith and Wojciechowski.* The volume of the pycnometer was 8.5 ml. and the capillary neck, marked with a single graduation, had an internal diameter of 1.4 mm. The pycnometer was filled with the aid of a hollow stainless steel needle. The pycnometer was then placed in a thermostat a t 25.00' and the position of the meniscus relative to the graduation was observed to 0.001 cm. with (1) Lewis and Luten, TEISJOUXNAL, 66,6061 (1933). (2) a k e r and La Mer, J . Chem. Phys., 8,408 (1935). (3) Selwud, Taylor, Hipple and Bleakney, THIS JOURNAL, 67, 442 (1935) (4) Smith and Wejciechowski, Rocamhi Chcm., 16, 104 (1936)
a traveling microscope. Weighings were then made with a duplicate pycnometer as counterpoise. An air density of 0.0012 was used in the reduction of the weights to vacuum. The results are given in Table I. This table also includes the density of D20-free water and the value of ND,O for natural water from the work of Johnston5 and Tronstad, Nordhagen and B n n 6 The values of AS, column 4,are, in contrast to those of equation (l), referred to the density of DzOSfree water. The atomic weights used were 0 = 16.0000, H = 1.00756, D = 2.01309, as given by Urey and Rittenberg.' The oxygen isotope ratio was assumed to be normal in all HzO-D~Omixtures. It will be noted that the atomic weight of hydrogen has been given a correction for the deuterium normally present. TABLE I
THEDENSITIES AND MOLAL VOLUMES OF MIXTURES OF HsO AND D& 1
2
3
4
5
6
I'
d:'
- l.*,
Material AS AScalcd. ml. NDtO DrOa 1.00000 1.10466 0.10792 0.10790 0.0000 Starting material 0.99166 1.10376 ,10702 .lo701 .mol Dilution 1 .a2358 1.08570 ,08891 ,08892 ,000.3 .61023 1.06279 .06593 Dilution 2 ,06593 .OW3 .40243 1.04044 .04351 Dilution 3 .04351 .OOO1 Dilution 4 ,20192 1.01884 .02183 ,02185 .m .OOO17a 0.997074 Natural water' .oo801, .00001r .oooO HlOO'V' .ooooo .99705a . OOOOO .moo
It is of considerable interest that H20-D20 mixtures form a perfect solution, almost within the experimental error, as tested by the absence of a volume change on mixing. Identifying HtO and DzO with the subscripts 1 and 2, respectively, a perfect solution conforms to the relation8 Va
N1vl
+ NZVZ
(2)
in which Va is the volume of a mole of solution (6) Johnston, T H WJOUXNAL, 67, 484 (1935). (6) Tronstad, Nordhagen and Brun, Nature, 186, 615 (1935). (7) Urey and Rittenberg, I. Chcm. Phys., I,137 (1933). (8) J . H . Hildebrend, "Solubility of Non-Electrolytes," Reinhold Publishing Corp., New York, 1936,2d editiou, p 58
14M
fESSIE
e. TAYLOR
VOl. 58
Y. CANN AND ALICE
and ZJ the molal volume of a pure component. by the method of least squares gives CY = 9.2351 The actual molal volumes, 17, of the solutions and B 0.0327, and since these constants reare given by the relation produce the densitiea within the experimental uncertainties, as is shown by a comparison of colV (NiMi N&2)/d (8) in which M is a molar weight and d the observed umns 4 and 5 , their use is recommended. The author wishes to thank Dr. D. A. MacInnes density. The expansions on mixing, V - Va, recorded in column 6 of the table are zero almost for many helpful suggestions in connection with within the limits of precision of the density de- this work. terminations, i. e., a l X 10-6: Csnsequentily, Summary if this expansion is assumed to be negligible, The ctellsities at 25’ of a series of mixtures of equations (2) and (3) may be combined to give light and heavy water have been determined. the following simple relation between ND,o and Contrary to the conelusion of earlier workers AS the components mi% without appreciable change ND~O * &AS/(l - PAS) (4) in which of volume. This fact is used in obtaining a new formula relating the composition of the mixtures a = Mi/M,(l - di/di) -L 9.235 B [Mzdi/& Mi]/Mt(I - di/&) = 0.0309. to their densities. Evaluation of CY and j3 from the data of Table I NEWYORK,N. Y . RECEIVED JUNE 4, 1937 p
5
+
-
[CONTRIBUTION FROM THE b E M I C A L
LABORATORY OF SMITH
COLLEGE]
The Potential of the Ag(s), AgI(s), I- Electrode’ BY JSSSIS Y. CANNAND ALICEC. TAYLOR The purpose of this investigation was to determine, by means of electromotive force measurements, the potential of the Ag(s), .4gI(s), I- electrode. A search of the literature revealed that the results obtained for the value of this electrode were not concordant.
Method and Apparatus In this investigation the cell A d s ) , AgCl(s), m KCKaq.), m KI(aq.1, AgI(s), AgCs)
with flowing junction, Was measured. Two pi-es Of were used‘ (‘1 Of Machef# and Yeh3 and (2) that of C a m and Mueller,‘ ‘Wgested by the work Of Randall and In case w&s k, prevent the flow of liquid over the electrodes. The d l w w placed in the usual oil thermostat,
(1) T h e oxperimmtal part of this paper is a portion af a thesis submitted by A. C Taylor in partial fulfilment of the requirements for the dtgree 01 Master of Arts a t Smith Cdlege (a) (a) Noyes and Freed, THIS JOURNAL, 42, 476 (1920); (b) Gerke, d i d . , 44, 1684 (1922), (c) Pearce and Fortsch, t b t d , 46, ma m a 3 ) ; (d) H- and J ~ A ~ I zW bkYS8.k. ~, cheim ,ira,$63 (ioaa); (e) Owen, THISJOURNAL, 67, 1526 (1935) (3) MacInnes and Yeh, rbad ,46,26% (1981) (4) Cahn and Mhdltt, rbcd., H , 2&35 (1986) ( 5 ) Ran-! hmd Cam, ibtd , (Is, 889 (19aO). ( 8 ) ~ a r m b ~ i9t,a , 64, 210 (193%), Cane and LaRue, ,brd , 64, 3456 (1932).
regulated at 2 5 O , and measurements were made with a shielded Leeds and Northrup Type K potentiometer.
Materials All solutions were made up by weqght, moles per 1000 g, of water in vacuum, from high grade “malyzed” salts and cunductkity water. Care was t&en to remove aEl traces of oxygen by bubbliryg an inert gas through the solutions. The Ag(s), AgCl(9) spiral electrodes were made according to the method of R a n d d and Young.’
The Ag(s), AgI(s) spiral
were made first, e l e c ~ ~ a l l y foH6wing genwal method of RadalI =d Ybung,? &&.o1yzing the d1.s~spirals a 0.05d-d of hy&i0& acid, current of 0.005 ampere, for two hours; and, s e e oad, to 65f)o, the spirals myad
by two daerent
*
and nine part* wfi apaste Of One sitver oxide, for ten minutes in acconlance with sf oWeIL2e Both Sets Of S&Ond trpdes were protected, as much as possible, from light. After being washed in distilled water, each set was short-circuited together for some hours. Immediately before being introduced into the cell (7) RmdM and Young, cMd , 50,969 0928).
