Sept., 1944
THEVANADIC-VANADOUS OXIDATION-REDUCTION POTENTIAL log
&/UH
= -0.44
ui/uH = 0.36
We cannot evaluate quantitatively the influence Of the presence Of the and vo+ions On the activity coefficients of the hydrogen ion and therefore cannot compute the concentration of the hydrogen ion produced by hydrolysis or the degree of hydrolysis and the r4ative amounts of the V+++ and VO+ ions. However, the data indicate that the concentration of the V+++ must be many times that of the VO+ ion. Some attempt has been made to compute hy’+++
1573
drolysis constants from these data but the comDuted “constants” showed a svstematic trend. iresumably due to the unreliabiliiy of the assump: tions which were made as to the activity cients. These computations are therefore not published at this time.
Summary PH measurements have been made a t 25’ on solutions of vanadyl sulfate, voso4, and on vanadic sulfate, V2(SO&, which show that these solutions are acid owing to hydrolysis. CAMBRIDGE, MASSACHUSETTSRECEIVED J U N E 13, 1944
[CONTRIBUTION FROM THE MALLINCKRODT CHEMICAL LABORATORY OF HARVARD UNIVERSITY ]
Electrochemical Studies on Vanadium Salts. III. The Vanadic-Vanadous Oxidation-Reduction Potential’ B Y GRINNELL JONES AND JOHN HENRYCOLVIN
The earlier papers of this seriesla describe using a platinum or other solid metal electrode measurements of the vanadyl (1V)-vanadic (111) owing to the evolution of hydrogen but that the oxidation-reduction potential, and of the hy- difficulty was diminished by using a mercury drolysis of vanadyl sulfate and vanadic sulfate electrode. Unfortunately they used dBerent solutions. In the present paper the work is ex- concentrations of sulfuric acid in their vanadium tended by similar methods to the measurement of solutions and in their sulfuric acid, mercurous the vanadic (111)-vanadous (11) oxidation-re- sulfate, mercury reference electrodes thereby induction potential. troducing a liquid junction potential for which The history of the researches on the vanadic- they make no correction. Latimers gives V++ = V + + + e, Eo = +0.20 vanadous potential closely parallels that for the vanadyl-vanadic potential. The earliest in- with a reference to the above work of Foerster and vestigator was Rutter12 who studied the cell Bottcher as his authority. Experimental Procedures Pt, ‘/2VZ(SO4)a VSO‘ H~SO (0.25 ~ molar), + Potential measurements were made on the cell HzSOl (0.25 molar), Hg2SO4, Hg. The sum of the divalent and trivalent vanadtum was kept a t 0.1 Hg, ‘/zVz(SO4)3 (63) HzSOa (GI), + vso4 (C2) molar but the ratio varied from equality to a ratio of 1 :3. His measurements in which divalent and H2S04 (cJ, Hg2S04, Hg at total vanadium contrivalent vanadium were both 0.05 molar give the centrations ranging from 0.005 to 0.141 c in soluvalue EO = +0.210 referred to the Nernst zero tions of sulfuric acid of concentrations 0.05, 0.1, 0.2,0.5 and 1.0 c. Solutions of different vanadium point of the potential scale. Abegg, Auerbach and Luther (1911) in their concentrations in sulfuric acid of a particular concompilation of standard potentials give +0.2 for centration constitute a series. this potential and this figure is giveti unchanged Preparation of Materials by Drucker3in his second supplement to this work. Vanadous Sulfate (Divalent Vanadium Sulfate) VSO,. Gerke’ in his compilation for the “International 7H20.-78.5 g. of c . P. vanadium pentuxide was suspended in 25 cc. of concentrated sulfuric acid diluted with about H2O (1) = 300 Critical Tables’’ (1929) gives VSO, cc. of water and sulfur dioxide passed through the ‘/2(vo)2so4 2H+ ‘/2SOi e ; Eo = 0.21. solution until the vanadium pentoxide was completely reFoerster and Bottcher6 have made a few meas- duced to vanadyl sulfate as indicated by the clear blue urements similar to those of Rutter. They point solution. A small amount of vanadium pentoxide was then to use all the excess of sulfur dioxide and the clear out that variable potentials were obtained when added blue solution of vanadyl sulfate, with practically no excess
+
+
+
+
+
+
+
+
+
(1) Original manuscript received November 12, 1943. (1) (a) Grinnell Jones and John H. Colvin, Tars JOUXNAL, 66, 1683 (1944); (b) Grinnell Jones and Wendell A. Ray, ibid., 66,1571 (1944). (2) T. F. Rutter, Z . anorg. Chem., 62, 373 (1907). (3) R. Abegg, F. Auerbach and R. Luther, “Messtangen elektrornotorische Krllfte galvanische Ketten,” W. Knapp, Halle, 1911, p. 204; second supplement to the above by C. Drucker, Verlag Chemie, Berlin, 1029, p. 222. (4) “International Critical Tables,” 1929, Vol. VI, p. 332. (5) F. Foerstcr and F. Bbttcher, 2. phyrik. Chrm., 161A, 321 (1030).
sulfuric acid, was filtered from the excess vanadium pentoxide through a sintered glass filter. Thevanadyl sulfate was reduced at a mercury cathode in a divided cell with the platinum anode inside a porous porcelain cup. I n going through the trivalent state considerable precipitate was formed which Gnally required the addition of about 15 cc. more of concentrated sulfuric acid. Continued reduction for about three or four days with 2 amperes re-
(6) W. M. Latimer, “Oxidation Potentials.” Prentiue-Hall, Inc.& New York, N. Y., 1038, pp. 243.296.
T h e same cell, including the parts, was used in this work stilted iti a klk, since the activity coefficient of the trivalent ion is more decreased by the increase in ionic strength due to the presence of acid than is the activity coefficient of the divalent ion. In a similar manner we may identify the intercept of these straight lines with the sum of all of the terms in equation (6) which are independent of the concentration of the vanadium ions giving
A definite numerical value can be assigned to the term from the work of Harned and Hamer as is explained in detail in the first paper of this series. Introducing (10) into (9) and rearranging gives
The values for E," are shown in Table IV, and as will be seen show a systematic trend with the con-
GRINNELL JONES
1578
GI
AND JOHN
HENRYCOLVIN
Vol. 66
E:
VI,
l.u 0.5 2 .1
.05
1 0 0 .5 .2 .t 05
1.0372 0.5lU0 ,2019 , 10061 , 0,5022
0.2821i
1 ,0372 0,5lO(i .2010 100til ,05022
0.2913 ,3890 ,2904 ,2861 ,2828
,2818 ,2822 ,2795 'Q-F
centration of the acid, varying at 25' from 0.2826 when c = 1.0 to 0.2755 when c = 0.05; and at 0' from 0.2013 to 0.2828 over the same range of concentration. This trend is due to the term RT/F In f;/j;. As explained in the first paper of this scries this term is expressed as a function of the ionic strength by the Huckel equation and the numerical values of the ionic strength evaluated by the use of Hamer's data and the value of the parameters in the Huckel equation and the value of ,& determined by the method of least squares with the results shown in Table V and in Fig. 3. This rnethod of determining nEi assumes that the plot of logfl against the square root of the ionic strenyth hris the theoretical limiting slope. Tlie results are as follows A t 23" E," = E:
+ I l l J I l ll?2:3l 2 2019 .IIAR83 .I ,10061 .lIJPiY .05 .0.5022 , li7a 0 extrapolated
1 11323 0.5828
I1 2!)13
.??sa
?!((I4
,1213
?Slil
.2YW
28?S
fl(
t 0 (I005 2 . 8 4 - 0 6 1 Y 2 04 OIJ1!1 2 6 7 llO!)l 240 - 01107 2 1 6
+ +-
. "3-1
+ ( I 2553
A\.. dev.
!3
.lAljl,l
