NOTES
May, 1950 Noting that the first term of each of the right hand members of equations (5)-(8) is constant
+ +
log K , = log K (E 1)kl (n 1)(Ly[21] 'y[12])P[1] (a
+
+
+ + 1 ) ( q 2+ a*,)!+
(9)
The a's of equation (9) are constant a t any given total ionic strength. Combining the constants of equation (9) log K , = k?
+ k a ~ ~ +i l
k4P1
(10)
2293
containing p1 and [PI] become negligible. Accordingly, log K , must be constant in the range below m cerous perchlorate. CHEMISTRY DIVISION NAVAL RADIOLOGICAL DEFENSELABORATORY MARCH14, 1949 SANFRANCISCO, CALIFORNIARECEIVED
Precise Measurement of Polarographic HalfWave Potentials
Noting that the ionic strength of an electrolyte is proportional to its molality
+ k,m[,] + k p ,
BY LOUISMEITES
The recent polarographic literature contains many examples of half-wave potential measureThe concentration of electrolyte 1 in the resin ments apparently accurate to * 5 'mv. or better. phase is proportional to the quantity that has been Because of the increasing interest in refined polaroabsorbed from the solution. Or graphic measurements, a method has been developed which yields values with a probable error of mt1] = W(m! - m l ) / R = W m z / R (12) about f 0 . 2 mv. where m! is the concentration of electrolyte 1 The residual current is measured a t 10-mv. inin the aqueous phase before equilibration; W = tervals over a range including the previously degrams of water involved in the equilibration; R = termined approximate value of the half-wave pograms of resin involved in the equilibration. tential in question, and also a t several points corConsequently responding to the plateau of the wave. The reducible material is added and its diffusion current log K , = k 2 + kSWmP/Rf ksml (13) secured from the average of several measurements Equation (13) can be tested by using the data on the plateau. The applied potential is then adof Boyd, Schubert and Adamson' (Table IV) justed so that the measured diffusion current is for sodium-hydrogen equilibrations a t constant roughly half that on the plateau, and the current total ionic strength. It was found that the is measured a t 1-mv. intervals on either side of this equation potential. After correction for the (interpolated) log K , = 0.413 - O.llOWmat/R (14) residual current a t this potential, the half-wave fits the data for samples 7-9 within an average potential (including, of course, the I R drop in the deviation of 0.3%. The data for samples 1-5 cell circuit) is most easily found by a short linear show an average deviation of 1.3% from the interpolation. The application of the method is illustrated by equation the data on the tetrammino-cadmium(I1) comlog K , = 0.137 + 0 . 1 1 4 W m ~ , + / R (15) plex given in Table I. The ammonia solution was If sample 4 were omitted from consideration, the standardized against recrystallized sodium tetraaverage deviation would be 0.2Oj,. (The average borate decahydrate v i a perchloric acid. Volumetdeviation obtained by Boyd, et aL,l without equipment was calibrated by standard methmaking the complete correction for mixed elec- ric ods. Polarographic measurements were made with trolyte activity coefficients, was 4.4%.) I t is known that (a12 a21) for sodium chloride TABLE I and hydrochloric acid mixtures is mall.^ When HALF-WAVEPOTESTIALOF CADMIUMIN 0.812 F AMthe concentrations of sodium chloride and hydroM O N I A - ~ . ~F ~POTASSIUM NITRATE chloric acid are not high, the product (a12 The cadmium concentration was 0.65 millimolar, Measazl)ml, does not vary sufficiently to cause sigurements were made at 25.00 * 0.01' E d . e . US. nificant change in the calculated values of K,. S. C. E., Microamperes For that reason, equations (14) and (15) are volt ir i l + ir accurate, notwithstanding the omission of the ml -0.770 0.522 variable. - .775 2.458 The observation of Tompkins and Mayer4 - .776 2.546 2.020 that K , is constant for exchange between < low4 - ,777 (0'526) 2.641 2.115 m cerous ion and 0.5 M ammonium ion can be - .778 2,727 explained on the basis of the considerations in this - .780 ,528 note. The concentration of cerous ion varied .950 .653 4.738 4.085} 4,086 m; however, through the range of 10-lo m to -1.000 .688 4.776 4.088 the total ionic strength of the solution did not - 2'020) 2.020 (0.001) = -0.7762 v. vary significantly. Accordingly, equation (9) E:/2 = -0.776 can be applied, with Ce(C104)3as electrolyte 1. idR = -0.7762 (2.034 X 10-6)(210) Because FI and [PI] are very small, the terms El/, = E ; / 2 2 log Km = k,
(11)
+
+
(i:;::
+
(4) Tompkins and Mayer, THISJOURNAL, 89, 2859 (1947).
+
= -0.7758 V.
NOTES
2294
a manual instrument, using a Leeds and Northrup Type HS galvanometer as current-measuring instrument. A diagram of the circuit will be presented elsewhere. All electrical components were calibrated, and the bridge potential was checked with a Type K potentiometer a t the time of use. The internal resistance of the cell under the conditions employed was found to be 210 ohms, while the additional resistance contributed by the galvanometer and its shunts was only 0.3 ohm. Plots of -&.e. against log ( i / ( i d - i)) for these and similar data indicate that the reduction is thermodynamically reversible. Independent diffusion current measurements, made with the same capillary] gave I = 3.205 * 0.009 in 0.10 F potassium nitrate and 3.85 * 0.01 in 0.10 F potassium nitrate-0.88 F ammonia. Then, by the fundamental equation for the reduction of a complex to the metal‘ 3.205 K f. (El/,), - EL/^). = 0.02957 log 3.85 fi
Vol. 72
given by Latimer,3
= -53,280 = -53,280 c d .
Fa(”,),++
pCd(NH&++
cal.
Previously Lingane had found -0.578 v. for the half-wave potential of cadmium in 0.1 F potassium nitrate4 and -0.81 v. in 1 F ammonia-1 F ammonium chloride. Neglecting, as here, correction for activity effects, Euler6 found 1 X for the dissociation constant of the ammino- complex from potentiometric measurements. The deviation in concentrated ammonia solutions is probably to be attributed to the formation of such complexes as Cd(NH3)s++. (3) W. M. Latimer, “The Oxidation States of the Elements and Their Potentials in Aqueous Solution,” Prentice-Hall, Inc., New York, N.Y . , 1938, p. 305. (4) J. J. Lingane, THISJOURNAL, 61, 2099 (1939). (5) J. J. Lingane, Ind. Eng. Chem., Anal. E d . , 15, 583 (1943). (6) H. Euler, Bcr., 36, 3400 (1903).
STERLING CHEMISTRY LABORATORY YALEUNIVERSITY NEWHAVEN,CONN. RECEIVED OCTOBER7, 1949
0.1183 log CNB, (1)
where the subscripts t and s refer to the amminoand aquo- complexes, respectively, K is the dissociation constant of the ammino- complex, and fc and fs the activity coefficients of the two complexes] which, for want of experimental data, we shall consider to be equal. Table I1 summarizes the measurements a t ammonia concentrations between 0.12 and 3.6 F. From these data and equation (l),one computes K = 1.26 (*0.02) X and, with the value of the cadmium-cadmium ion standard potential given by Harned and Fitzgeralde
The Preparation of Calcium D-Arabonate from Calcium 2-Keto-D-gluconate by Electrolytic Bromine Oxidation BY C. L. MEHLTRETTER, W. DVONCH AND C. E. RIST
Isbell and Frush and others1 have shown that aqueous solutions of D-glucose in the presence of calcium carbonate can be oxidized to calcium D-gluconate in nearly quantitative yield with bromine generated electrolytically. By prolonging the electrolysis so as to consume twice the current required for the conversion of D-glucose Cd 4“s + Cd(NH&++ 2e; Eo = 0.6060 v. to D-gluconic acid, Cook and Major2 were able to (N. H. E.) isolate calcium 5-keto-~-gluconate in small amount. TABLE I1 In a study of the oxidation of calcium DEFFECTOF AMMONIACONCENTRATION ON HALF-WAVE gluconate by electrolytic bromine whereby 4 POTENTIAL OF CADMIUM(II) IN 0.10 F POTASSIUM NI- faradays of electricity per mole of gluconate was TRATE used, we obtained calcium D-arabonate as well All measurements were made with ca. Od6 millimolar cad- as the calcium salts of 5-keto-~-gluconicand oxmium at 25.00 * 0.01 alic acids. The crude calcium D-arabonate was E l / , VS. S. C. E., (El/r)o - (B1/,)*f 0.1183 converted to ~-arabobenzimidazole,~ from which CNHI,F volt log CNHS the yield was calculated to be 9%. Electrolysis 0 -0.5777 ...... with 8 faradays per mole of calcium D-gluconate 0.1267 - ,6776 -0.2065 resulted in a 17% yield of calcium D-arabonate. - .7243 - ,2066 .310 The presence of 2-keto-~-gluconicacid could not ,2060 .433 - .7407 be detected in any of the oxidized solutions. - .7535 - ,2064 ,550 The transformation of calcium D-gluconate t o - .7758 - .2068 .812 arabonate suggested that calcium 2-keto-D-gluco- .2065 1.120 - .7899 nate is first formed and is then rapidly decarboxyl- .8033 - .2078 1.389 ated and oxidized to calcium D-arabonate. Honig - .8206 - ,2103 1.890 and Tempus4proposed such a mechanism for the - .2135 2.790 - .%41 hypobromite oxidation of D-glucose because - .8604 .2172 3.565 both 2-keto-~-g~uconic and D-arabonic acids were Mean ( C N H ~< 1.2 F) -0.2064 * 0.0002 (1) (a) Isbell and Frush, J . Research N a f . Bur. Standards, 6 , 1145
+
+
-
-
Further, with the standard free energy of
(as.)
(1) I. M. Rolthoff and J. J. Lingane, “Polarography,” Interscience Publishers, Inc., New York, N.Y., 1946, pp. 161-166. (2) H.S. Harned and M. E. Fitzgerald, THISJOURNAL, 68, 2624
(1936).
(1931); (b) Helwig (to Rohm and Haas Co.)U.S. Patent 1,937,273 (Nov. 28, 1933); (c) Fink and Sommers, Trans. Elccfrochemical Soc., 74, 625 (1938). (2) Cook and Major, TEIS JOURNAL, 67, 773 (1935). (3) Moore and Link, J. Biol. Chem., 133, 293 (1940). (4) Honig and Tempus, ELI., 67, 787 (1924).
