5501
Nov. 5, 1955
NOTES
In order to increase the sensitivity of Procedure B, volumetric flasks as small as five ml. could be used for reaction of the oxalate with ceric ion. I n this case, all solutions would have to be carefully weighed, but an improvement in sensitivity by a further factor of ten should result. This would enable light intensities of the order of 1 X l O I 3 quanta/sec. to be measured to 5y0or better accuracy with a one-hour exposure. The molar extinction coefficient of the dye produced when malachite green leucocyanide is employed as an actinometergis 9.49 X lo4,and that of the ferrous-phenanthroline compound, lo formed after photolysis of potassium ferri-oxalate, is 1.1 X lo4. These extinction coefficients may be compared to 5.4 X lo3for ceric ion a t 3200 A. Thus by the spectrophotometric procedure the uranyl oxalate actinometer seems capable of a sensitivity within a factor of about twenty of that of the malachite green leucocyanide actinometer lo and within a factor of about two of that of the potassium ferrioxalate actinometer." It may be that by using a suitable, very highly colored oxidation-reduction indicator to reduce the excess ceric ion quantitatively one might further increase this sensitivity. Acknowledgment.-This work is a portion of studies in the general field of photochemical processes for which the authors gratefully acknowledge grants-in-aid to Northwestern University from the Public Health Service (Project G0375OC) and the Atomic Energy Commission (AT (11-1)-59, Project No. 4). We are also deeply indebted to Dr. Jack G. Calvert (The Ohio State Univ.) for helpful discussions and assistance in his laboratory.
the electromotive force of the cell: Pt, Hz; HC104(cl), NaC104 (c2); HClOd(cl), NaC104(cz), NazSOd(c3) ; Hz, Pt. Because of the high ionic strengths used in this cell the calculation of the liquid junction potential is not feasible, consequently an empirical approach was adopted. The proportionality constant, k , in the equation
(9) J. G. Calvert and H J. L. Rechen, THISJOURNAL, 74, 2101 (1952). (10) C A Parker, P r o c . Roy Soc (Londo,z), A220, 104 (1953).
THEDIVISIONOF PHYSICAL SCIEXCES UNIVERSITY OF CALIFORNIA RIVERSIDE, CALIFORNIA, ASD THECHEMICAL LABORATORIES NORTHWESTERN UNIVERSITY EVASSTON, ILLINOIS
f
relating the cell e.1n.f.) E , with the ratio of the hydrogen ion concentrations in the two cell compartments was obtained by titrating the perchloric acid in one-half of the cell with standardized sodium hydroxide which contained sufficient sodium perchlorate to maintain the ionic strength a t unity throughout the titration. Only a small fraction of the total perchloric acid was neutralized in each titration. No sodium sulfate was present in the determination of the values of the proportionality constant. I t was assumed that the low concentrations of sulfate and bisulfate made little contribution to the liquid junction potential in the titrations in which sodium sulfate was added to the perchloric acid-sodium perchlorate solutions. Values obtained for k were -0.0579 f 0.0002, -0.0461 =t0.0007 and -0.0175 f. 0.0004 for the 0.01, 0.10 and 1.0 M perchloric acid solutions, respectively. Values of the dissociation quotient for the bisulfate ion were obtained from the variation of the cell potential as a function of the sodium sulfate concentration by the considerations + f + [H]r = [HI, 10-E/k = [H]i Y
where the subscripts i and f denote the initial and final states, E is the cell e.1n.f. and k is the proportionality constant + +
[HSOa-Ir = [&I1 K - [HI, Y R is the dilution ratio, Vi/( Vi
is the solution volume. Then
K
BY EUGENEEICHLERA N D SHERMAN RABIDEAU RECEIVED h I A Y 26, 1955
I t is desirable to have values of the bisulfate dissociation quotient in solutions of unit ionic strength to permit comparisons of the relative stabilities of cations complexed by sulfate ion. A value of 0.24 was calculated by Rabideau and Lemons2 for the dissociation quotient of the bisulfate ion a t unit ionic strength and 25' from the solubility of calcium sulfate in acid-salt solutions. Zebroski, Alter and Heumann3 estimated the value of the acid quotient to be 0.084 f 0.020 a t an ionic strength of 2.0 from PH measurements of solutions of sodium sulfate, sodium perchlorate and perchloric acid. The procedure used in this work was to measure (1) This work was done under the auspices of the A.E.C. (2) S. W. Rabideau and J. F. Lemons, THISJOURNAL,73, 2895 (1951). (3) E. L . Zebroski, H. W. Alter and F. K. Heumann, i b i d . , 73,5646 (1951).
= [HIi (X - Y )
+
VNarSOI)j where
I/
= [h[r[S04=]r/[H%04]r=
{Y/(K
The Determination of the Bisulfate Dissociation Quotient from Potentiometric Measurements'
+
E = k log[HIz/[Hli
-
Y)ll[S04ll - [ h l ( R - 7))
The mean values of the bisulfate dissociation quotient obtained in perchlorate solutions of unit ionic strength varied with the acidity. The results are 0.095 f 0.002, 0.054 ZIZ 0.012 and 0.30 f 0.08 in 0.01, 0.10 and 1.0 M perchloric acid, respectively. The results of the titration with sodium sulfate in 0.01 M perchloric acid-0.99 M sodium perchlorate are given in Table I. Within the experimental error the values of K in 0.01 and 0.10 M perchloric TABLE I DETERMINATION OF BISULFATE DISSOCIATION QUOTIENT IN 0.01 ;If HC104-0.99 Lf NaC104 &li, moles/I. 0.01058
[Gn~i, moles/l. 2 . 8 9 7 X 10-4 5.479 1 . 0 3 5 x 10-2 1.999 2,697 3.666 4.722 5.463
E, volts X 10' y K 7 2 . 1 0.9971 0 . 0 9 6 0.9998 133.2 ,9947 ,9097 ,098 ,9995 2 6 0 . 3 ,9897 ,094 ,9990 4 5 5 . 4 ,9820 ,105 ,100 ,9986 641.9 ,9748 ,096 ,9981 898.9 ,9648 .094 .9543 ,9975 1174.3 ,092 ,9463 ,9971 1388.7
R
Mean
0 . 0 9 5 f0.002
5505
Vol. 77
NOTES
acid solutions are equal; however, i t appears that in spite of the fact that the ionic strength was held a t unity, the activity coefficient product is significantly different in molar perchloric acid as compared with the more dilute acid solutions. Experimental The electromotive force of the cells was measured t o the nearest 0.1 microvolt with a type K Leeds and Northrup potentiometer. The hydrogen used for the hydrogen electrodes was purified with alkaline permanganate, sulfuric acid, alkaline pyrogallol and heated copper. Mallinckrodt analytical reagent grade sodium sulfate was used. The sodium perchlorate was prepared by the reaction between C.P. sodium carbonate and perchloric acid. The perchloric acid was standardized against mercuric oxide. The temperature of the cells was maintained a t 25.00 f 0.05” by circulating thermostated water in the jacket of the cell container. Initially no sodium sulfate was present in the right compartment of the cell and the measured e.rn.f., which should have been zero, was generally less than 20 microvolts. Two hydrogen electrodes were used in each half of the cell and these were intercompared by means of a smitching arrangement. Standardized sodium sulfate solution was added in small increments from a weight buret. The solutions were stirred with magnetic stirrers and the cell e.m.f. was recorded after each addition of sodium sulfate.
Acknowledgment.-The authors wish to express their appreciation to Dr. J. F. Lemons for his assistance and interest in this problem. UNIVERSITYOF CALIFORSIA Los ALAMOSSCIESTIFIC LABORATORY L O S ALAMOS,NEW MEXICO
quantitatively precipitated by cupferron in 10% hydrochloric acid. Experimental.-Conventional polarographic techniques were employed, involving manual measurement with a Fisher Elecdropode. The polarographic cell-compartment was jacketed and thermostatically regulated a t 26.0 d~ 0.1 ’; a fritted disk and agar plug junction connected to a saturated caloniel reference electrode Distilled water and reagents of C.P. or reagent grade were employed. Oxygen removal n as accomplished by purging with tank nitrogen.
Discussion.-The polarographic behavior of uranium in its various oxidation states has been very thoroughly studied and is well sumniarized by Rodden.4 I n general, aqueous solutions of U(V1) in dilute ininera1 acid exhibit the reduction pattern shown on curve ABC of Fig. 1. A medium of 0.1 N sulfuric acid and 0.1 N potassium chloride was employed in the present study. Tirave A is primarily due to the one-electron reduction of U(V1) to V(V). However, the U (V) disproportionates W(V)
+U(V1) + U ( I Y )
(1)
to reform some U(1-I) and to give some U(1V); consequently, the wave is larger than required for a one-electron reduction. n’ave BC represents the two-electron reduction of U(V) to UiIII); the intermediate (IV) stage demarcating B from C appears only at higher acidities. Wave A is always larger than B, and C(IV to 111) is approximately equal to one-half of A plus B(V1 to IV). In (i Ai acid the B wave coalesces with A.
Uranium (111) Cupf errate’ BY CHARLESL. RULFSA N D PHILIPJ . ELYING RECEICED MAY9, 1953
Uranyl ion forms an insoluble cupferron salt in neutral solution; this compound, U 0 2 N H 4 ( C ~ p ) 3 , second, etheris insoluble in organic solvents.2 arid chloroform-extractable form appears to exist in acid media, but only in small proportions in very concentrated uranium solutions.2c Uranium(1V) forms brown U(Cup), which is insoluble in up to 8y0 sulfuric acid solution, but is quantitatively extractable into ethyl ether.lc Polarographic and ot.her studies in this Laboratory indicate the existence, also, of a very stable, ether-soluble uranium(II1) cupferrate. While i t ..-02 -04 -06 -08 -10 has not been recognized so far as is known, there E , VOLTS SCE is some independent evidence both that U(II1) does form a cupferrate and that this compound must be chloroforn3.- and ether-soluble. The reported work Fig. 1.-Polarographic reduction of uranium(V1) and of on cupferron extraction of U(IIr) has involved both cupferron in dilute mineral acid solution. the use of Jones reductors without a subsequent Kolthoff and Libertij have reported on the polaraeration step and extraction in the presence of liquid amalgams. The presence of some U(II1) might ographic behavior of cupferron. The large reducbe expected in either case and quantitative extrac- tion wave which occurs a t inore negative potentials Grimaldi3 is one in alkaline media, starts from about -O.S to -0.9 tion of the uranium is of the few authors to indicate that U(III), as well as v. us. S.C.E. in 0.1 li’ acid, as shown by curve D of U(IV), cupferrate is involved and that U(II1) is Fig. 1. When a solution of IrnM uranyl ion in the pres(1) This investigation was supported in part by t h e .4ir Force Camence of 10 m d l cupferrate w3s examined, curve bridge Research Center under contract with t h e Engineering Research EFGH was obtained. The maximum a t F is Institute of t h e University of Michigan. ( 2 ) (a) 0. Baudisch and R. F u r s t , Bey., SO, 324 (1917); (b) K . H. largely removed in the presence of 0.005%- gelatin. Furman and D. R. A-orton, Atomic Energy Commission, Report It7ave E represents the reduction of L(T’1) to hIDDC 1623 (1947); ( c ) N. H. Furman, W. B. Mason and 3 . S . Pekola, & + -+ * -
_---
VP.
Anal. Chem., 21, 1326 (1049). (3) F. S. Grimaldi in “Collected Papers o n XIethoris d Analysis for Uranium and Thorium,” C . S. Geol Survey Dull. lOW, 1‘3 (1054).
(4) C . J. Rodden, “.4nalytical Chemistry of the RIanhattan Project,” hIcCraw--Hill, New York, N. Y . , 1950, p p . 596-611. ( 5 ) I. AI. Kolthoff and A . Liberti, T H I S JOURN?I., 7 0 , 1885 (1048).
