(2) Baddiley, J., Thain, E. M.,J . Chem. SOC.1951, 2253. (3) Biochemical Institutes Studies, IV, University of Texas, Publ. 5109 (1951). (4) Blair, J. A,, Graham, J., J . Biochem. 56,286 (1954). (5) Block, R. J., Durrum, E. L., Zweig, G., "Manual of Paper Chromatography and Paaer Electroahoresis." DD. 398409, Academic Press, New 'Yo& 1958. ( 6 ) Bowden, J. P., Peterson, W. H., J . Bzol. Chem. 178,533 (1949). ( 7 ) Brown, F., Blaxter ,K. L.. Chem. & Ind. (London) 1951. 633. (8) Brown, J. k,ANAL. CHEM.25, 774 (19.53).
(9j B&m, J. A., March, I f . hl., Ibid., 24, 1952 (1952). (10) Crammer, J. L., Xuture 161. 349 . (i948). (11) Crokaert, R., Arch. Intern. Physiol. 56, 189 (1948). (12) Dalgiesh, C. E., J . Biochem. 5 2 , 3 (1952). (13) Datta, S. P., Overell, B. G., StackDunne, AI., Yuture 164, 673 (1949). (14) Edviards, C. H., Gadsden, E. L., Edwards, G. A , , J . Chromatog. 2 , 188 (19.59). \ _ _ _ _
(15) Eggitt, P. W. R., Ward, L. D., Sci. Food Agr. 4, 176 (1953).
(16) Fountain, J. R., Hutchison, D. J., Waring, G. B., Burchenal, J. H., Proc. SOC. Exptl. Biol. Med. 83, 369 (1953). (17) Green, J., Marunkiewicz, S., Kature 176, 1172 (1955). (18) Guerillot, J., Guerillot-Vinet, il., Delmas, L., Compt. rend. 235, 1295 (1952). (19) Hais, I. M.,Pecakova, L., Suture 163, (68 (1949). (20) Hanes, C. S., Isherwood, F. A , Ibzd , 164, 1107 (1949). (21) Harrison, J. S., Analyst 76, 77 11951). (22) Heyndrickx, A,, J . Am. Pharm. Assoc., Sci. Ed. 42, 680 (1953). (23) Holman, W. I. AT,, J . Biochem. 56, 513 (1954). (24) Huebner. H. F.. Nature 167. 119 Lin, P. H., J . Am. Toc. 75,2971 (1953). icek, E., Reddi, K. K., Suture 5 (1951). trer, E., Lederer, >I., "Chromatography," pp. 256-7, Elsevier, Amsterdam. 1953. (28) Malyoth, G., Stein, H. W.,Biochem. Z. 323, 265 (1952). (29) Mapson, L. W., Partridge, S. &I., .\ atu 'e 164, 479 (1949).
. -.
r r
(30) Miyaki, K., Momiyama, H., Hayashi, M., J . Pharm. SOC. J a p a n 72, 688 (19521.
(32) Rossi, C. A., Boll.
SOC. ital. sper. 26, 1563 (1950). (34) Siliprandi, D., Siliprandi, N., Biochim. et Biophys. Acta 14,52 (1954). (35) Snyder, J. I., Wender, S. H., Arch. Biochem. Biophys. 46,465 (1953). (36) Viscontini, M., Bonetti, G., Ebnother, C., Karrer, P., Helu. Chim. \ - - ,
19.51\ --, 1 2 R 4 f,----,.
Arfn3A
--""I
( 3'i ) Weygand, F., Arkiv
Kemi 3, 11 (1950). i Si3) Winsten, W.A., Eigen, E., J . Biol. Chem. 17:7, 989 (1949). (39) Ibid., I181, 109 (1949). (40) Woodr,uff, - - - -H. B., Foster, J. C., Ibid., 183,569 (IYSU). (41) Wright, L. D., Cresson, E. L., Driscoll. C. A.. Proc. SOC.Exwtl. Biol. N e d . 86; 480 (1954). (42) Yagi, K., J.Biochem. (Japan)38,161 (1951). RECEIVEDfor review March 1, 1960. hccepted June 21 1960. In,
.
Cou Iometric Determi nation ot t u ropium and Ytterbium at Controlled Potential EDWARD N. WISE and EDWARD J. COKAL Chemistry Deparfment, University of Arizonu, Tucson, Ariz.
b Europium and ytterbium were determined individually b y the one-electron reduction of their trivalent ions in methanolic , tetraethylammonium bromide, with a mean absolute error of less than 0.05 peq. in the sample range o f 0.5 to 20 peq. The reduction of ytterbium was induced b y the reduction of europium, so that only their sum could b e determined b y reduction in samples containing both elements. In such cases, europium was determined in a separate aliquot b y coulometric oxidation in 0.1 M hydrochloric acid, and ytterbium was found b y difference.
T
of controlled POtential coulometry to the determination of the rare earths had led to the development o'f a n excellent method for the determination of europium ( I O ) . This involves reduction of the trivalent europium in dilute HCl, followed by qumtitstive coulometric oxidation of the divalent ion. However, the extension of this method to the determination of the remaining rare earths is not feasible because of the rapid oxidation of their dipositive ions in reactions of the type HE APPLICATION
'13
+ H20
+
XI+'
+
1/2H2
+ OH(1)
+
-
+
~ + 2 H+ 11+3 l / 2 ~ z (2) hlthough reduction of europium(II1) in aqueous ammonium chloride is nearly quantitative when proper background correction is made ( I S ) , the use of a less easily reduced solvent would be expected to yield reductions in which these interfering reactions do not occur. We have found absolute methanol to be a satisfactory solvent for the determination of europium and ytterbium by the one-electron reduction of the trivalent ions. Europium alone in a supporting electrolyte consisting of 0.1M tetraethylammonium bromide (TEABr) in methanol is reduced n-ith very nearly 1 0 0 ~ ccurrent efficiency, with the final electrolysis current approximately that due to the supporting electrolyte. Ytterbium(I1) reacts with the methanol in a manner similar to the reaction with water described above, thus giving less than 1007, current efficiency. However, the rate of this reaction is sufficiently small that a satisfactory correction for its effect can be made. When both europium and ytterbium occur in the same sample, only their sum can be determined by reduction, since the reduction of ytterbium is induced by the reduction of europium. A method of correcting for this induced reduction has been devised (9)$ but
the magnitude of the correction is too great to allow reasonable accuracy in the results for the individual metals. Better results were obtained by determination of the sum of ytterbium and europium at a potential sufficient to cause quantitative reduction of both elements, and determination of the europium in a separate sample by the coulometric oxidation method developed by Shults (10). Suitable control potentials for europium and ytterbium were determined polarographically in the methanolTEABr electrolyte. Half-wave POtentials were approximately 0.0 and -0.78 volt us. the methanolic silversilver bromide electrode. The n-aves appeared to be well formed, but a log i/id-i us. E plot indicated that the reduction of ytterbium in this medium is very irreversible, the reciprocal of the slope being 0.071 rather than the expected 0.059. Preliminary electrolysis and polarographic experiments indicate that the reduction potentials of europium(II1) and ytterbium(II1) in this system do not differ by more than 0.13 volt. A similar effect may be operative in aqueous solutions and could account for the discrepancy between the reported values for the standard POtentials of the Yb-f2-Yb+3electrode VOL. 32, NO. 1 1 , OCTOBER 1960
1417
(6, 12). Further study of this reaction
is in progress in this laboratory. Any potential more negative than
B
WA c
-0.3 volt was suitable for the determination of europium, while reduction of ytterbium was most satisfactory a t - 1.2 volts, us. the Ag-AgBr electrode. Hydrogen ion and water interfere with the determination of ytterbium, causing high and erratic results. Therefore, all water and acid must be removed from the samples and the components of the supporting electrolyte. The consistently low continuous background currents obtained in this work, on the order microfaraday per second of 1.5 X (15 pa.), were the result of purification of the reagents and efficient removal of the trace of oxygen present in the argon used in purging the system. Another factor essential to the establishment of low background currents is the maintenance of a slight positive pressure in the electrolysis compartment t o ensure the absence of atmospheric oxygen. Tetraethylammonium bromide was selected as the supporting electrolyte because of its very negative decomposition potential which we found to be -0.2 volt us. Ag-AgBr, solubility in methanol, and relative ease of purification. The value for the decomposition potential obtained by Ehlers and Sease (S), approximately -2.4 volts, is probably due to the effect of a trace of water in the solvent on the -4g-AgBr electrode. Neodymium and samarium appear to be slowly reduced at -1.2 volts, and thus interfere with the ytterbium determination. The interference resembles a continuous faradaic background current which is proportional to the interfering ion concenttation. When the concentration ratio of neodymium to ytterbium is about 3 to 1, the continuous current due to neodymium reduction is equal to the kinetic current of the ytterbium determination. The samarium interference is much smaller than that of the neodymium. Similar coreductions were noted by Marsh (8, 11) during the preparation of europium materials free from other rare earths. APPARATUS AND REAGENTS
A potentiostat-integrator similar to that described by Booman (1) was used for all quantitative work. Integrator readout was by a Fluke Model 801 differential voltmeter. A Varian G-10 recorder was used for continuous recording of the current-time curve. Stirring was by a Heller GT-21 constant speed laboratory mixer, used a t 550 r.p.m. Electrolysis times were measured with a Lab-Chron electric stop clock. A Kin-Tel Model 204.4 electronic galvanometer was occasionally used for determination of electrolysis currents. Cell. The construction of the cell is shown in Figure 1. It could be made in slightly modified form from a 1418
0
ANALYTICAL CHEMISTRY
Figure 1. Cell for controlled potential coulometry A.
Mercury seal stirrer halide reference electrode C. Argon inlet D. Silver working electrode E. Neoprene plug F. Mercury inlet and cathode connection G. Drain
B. Silver-silver
Sargent S-29311 Polarograph cell by the addition of a mercury seal stirrer to the upper part, and an anode chamber and drain to the solution cup. The top of the cell shown was made from a 1/8-inch sheet of Lucite attached to the glass parts by EpoxiPatch cement. Glass wool packed in the working anode chamber greatly reduced its volume and the attendant problem of solution transfer across the sintered-glass disk due to ambient temperature changes. Both working and reference Ag-AgBr electrodes contained deaerated 0.1M TEABr as electrolyte. The reference electrolyte was saturated with silver bromide by the addition of a drop of methanolic silver nitrate solution to the electrode compartment. I n addition to the parts shown in the figure, a gas vent tube and a dropping mercury electrode were attached through the cell top. The dropping mercury electrode is not needed in the present work, but is convenient for other investigations. Cell constants are on the order of 0.5 to 0.6 min.-* when the solution volume is 5 ml. Reagents. Mallinckrodt analytical reagent grade methanol was refluxed for 2 hours with an iodine crystal and sufficient magnesium turnings to react with 3y0 water. The product was distilled and the constant boiling fraction a t 64.5' C. taken for use.'-It was stored in glass, since storage in polyethylene for extended periods resulted in deterioration of the alcohol. Eastman No. 1516 tetraethylammonium bromide was dissolved in the minimum amount of 1to 3 methanol-acetone, an equal volume of hot acetone added
to the boiling solution, and the flask and contents allowed to cool to room temperature. The crystals were collected, washed twice with acetone, and dried a t 90" C. for 15 minutes. The product was stored over PZOs. Fresh solutions of TEABr (0.1M) in methanol were prepared daily, as some deterioration as evidenced by lowered decomposition potential was noticed with week-old solutions. Matheson 99.998% argon was passed over copper turnings in a tube furnace a t 500" C. and through a methanolfilled gas saturater before entering the electrolysis cell. The copper was regenerated when necessary by passing hydrogen through the system. Regenerated copper is much more efficient in removing oxygen than fresh copper turnings. Ytterbium, samarium, and neodymium oxides were obtained from the Ames Laboratory of the U. S. Atomic Energy Commission, and the ,europium oxide from Research Chemicals, Inc., Burbank, Calif. These were taken as 100.0% MgOs after ignition a t 900" C. overnight. Stock solutions of the chlorides were prepared by dissolving weighed amounts of the oxidesin 6N HC1 with warming, followed by transfer to volumetric flasks. Methanol stock solutions of the individual metals were made daily by evaporation of aliquots of the aqueous stock solutions in a hot water bath in a stream of air, and resolution in measured volumes of the purified methanol. Samples were taken for analysis by pipetting aliquots of the methanol solutions. Mallinckrodt analytical reagent grade and Metalsalts-Corp. mercury were used interchangeably with recovered mercury which had been purified by a Bethlehem Instruments Co. oxifier and gold adhesion filter. PROCEDURE
Introduce sufficient cathode mercury to reach within 2 mm. of the reference electrode and 5 ml. of supporting electrolyte into the cell and deaerate for 5 minutes a t an argon flow rate of about 0.15 cu. foot per hour and a stirring rate of 550 r.p.m. Pre-electrolyze a t -1.2 volts and record the final background current. Add the sample aliquot, which should be on the order of a few hundred microliters, and deaerate for an additional 3 minutes. Zero the integrator and electrolyze a t - 1.2 volts. When the electrolysis current drops to an apparently final constant value, record the integrator reading without interrupting the electrolysis. Continue to record the integrator reading a t 100second intervals until it is obvious that the current is constant. Determine the blank by repeating the above procedure with a like volume of the purified methanol in place of the sample solution. The sum of europium and ytterbium is found by application of background corrections given in the following section. The procedure for the determination of europium as described by Shults (10) was modified slightly to the follow-
Table I.
Eu, geq. Taken Found 0.91 0.86 0.91 0.89 0 91 0.86 0.91 0.89 2.32 2.26 2.31 2.26 2.32 2 26 2.26 2.26 9.05 9.06 9.08 9.06 9.09 9.06 9.09 9.06 9.10 9.06 22.64 22.66 22.64 22.66 22.64 22.63 22,56 22.64 Table II.
Error, Peq. -0.05 -0.02 -0.05 -0.02 + O . 06
+0.05 $0.06 =to.00 -0.01 +o. 02 + O . 03 $0.03 $0.04 $0.02 $0.02
Eu 0.76 0.76 5.66 5.66 5.66 5.66 5.66 5.66 5.66 5.66 9.06 9.06
peq. Taken Yb = 6.71 6.71 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 0.40 0.40
+
Total 7.47 7.47 10.66 10.66 10.66 10.66 10.66 10.66 10.66 10.66 9.46 9.46
Totalo 7.32 7.44 10.65 10.57 10.70 10.63 10.70 10.64 10.71 10.74 9.45 9.41
EuTZ
(3)
peq. Found - Eub = 0.72 0.73 5.58 5.58 5.68 5.68 5.69 5.71 5.68 5.70 8.99 9.01
Ybc 6.60 6.71 5.07 4.99 5.02 4.95 5.01 4.93 5.03 5.04 0.46 0.40
- peq EU -0.04 -0.03 -0.08
-0.08
$0.02 $0.02 $0.03 $0.05
Error Yb -0.11 10.00 +0.07 -0.01
+0.02 +0.04
-0.07 -0.05
$0.02
-.0.05 $0.01 -0.07 $.0.03 tO.04 $-0.06
10.00
By reduction. By oxidation. e By difference. 0
b
f0.01
-0.08
Ytterbium Determination by Reduction
Yb, peq. Taken, Found 0.37 0.34 0.33 0.37 0.37 0.33 0.37 0.34 0.73 0.74 0.73 0.72 0.73 0.74 0.73 0.72 0.81 0.83 0.82 0.81 0.82 0.81 4.03 4.04 4.03 4.01 8.05 8.11 7.99 8.05 20.14 20.17 20.14 20.1.5 20.i4 20. if 20.14 20.15 20.14 20.17 20.14 20.11
Analysis of Samples Containing Both Europium and Ytterbium
Table 111.
Europium Determination by Reduction
Error, wq. -0.03 -0.04 -0.04 -0.03 $0.01 -0.01 $0.01 -0.01 +0.02 $0.01 $0.01
+o. 01 -0.02 $0. 06
-0.06 +0.03
Eu+3
+e
and a continuous faradaic current. The blank is believed to represent mainly the faradaic impurity quantity of electricity associated with the oxygen dissolved in the methanolic sample aliquot. Experimental results of determinations of europium by reduction are given in Table I. Ytterbium reductions yielded final background currents of many times that found in the pre-electrolysis, The total integrated quantity of electricity was corrected by considering the total background to be the sum of the blank, a continuous faradaic current, and a kinetic current which is believed t o arise from the reaction.
+ MeOH
-
+
Yb+*
+0.03 +0.01 +0.03 -0.03
The kinetic quantity of electricity was estimated from the equation =
0.70i/t/
LITERATURE CITED
(1) Booman, G. L., ANAL. CHEM. 29,
i0.01
&k
Yb+3 OMe-
the background correction equation. Europium was determined in 3, second aliquot of the solution by oxidation, and the quantity of ytterbium was calculated by difference. Experimental results are given in Table 111. The over-all accuracy of the method is roughly t h a t of the more recently reported complexometric titration procedures for the individual rare earths (2, 4), and is superior to the existing polarographic methods (6, 7 ) with samples larger than about 2 peq. of rare earth metal per 5 ml. of solution. Accuracy of 0.27, is t o be expected with samples of 20 peq. or more.
+ l/*H2
(4)
(5)
213 (1957). (2) Bril, K. Y., Holzer, S., Rethy, B., Ibid.,31,1353 (1959). Ibid., (3) Ehlers, V. B., Sease, J. ]I7., 31, 16 (1959). (4) Fritz, J. S., Oliver, R. T., Pietrzyk, D. J., Zbid.,30,111 (1958). (5) Hibbits, J. O., Menke, M. R., Davis, W. F.. U. S. Atomic Enerev Comm. Apex. 405, 19 pp. (1958). (6) Laitinen, H. A., J . Am. Chem. SOC. 64, 1133 (1942). (7) Laitinen, H. A,, Taebel, W.A., ISD. ENG.CHEX.,ANAL.ED. 13, 825 (1941). (8) Marsh, J. K., J . Chen. SOC.1943, 531. (9) Meites, L., RIoros, S. A., ASAL. CHEM.31,23 (1959). (10) Shults, W. D., Zbid.,31, 1095 (1959). (11) Vickery, R. C., “Chemistry of the Lanthanons. Chan 8. Academic Press, Kew York. i953. (12) Kalteri, GI C., Pearce, D. W., J . Am. Chem. SOC.62,3330 (1940). (13) Wise, E. N., “Coulometric Analysis of EuroDium a t Controlled Potential,” Paper 91, Southwestern and Rocky hlountain Division of American Association for the Advancement of Science and the New Mexico Academy of Science, April 30, 1968. RECEIVED for review December 9, 1959. Accepted July 25, 1960. Division of Analytical Chemistry, 137th RIeeting, ACS, Cleveland, Ohio, April 1960. Work supported in part by the U. S. Atomic Energy Commission under Contract AT(11-1)-553. I ”
ing. The sample aliquot, in 0.1012’ HCl, is reduced a t -0.9 volt us. the AgAgCl electrode in a cell similar to that used in the work with methanol solutions. When the current appears to have reached a constant background value, the electrolysis is stopped and the potentiostat set for a n oxidation a t -0.1 volt. About a minute is allowed to elapse before the oxidation step is begun so that any coreduced ytterbium may be reoxidized by the acid. The electrolytic oxidation is then performed, the background current eventually dropping to a small constant negative value. The blank and background current are determined as in the above procedure. RESULTS AND DISCUSSION
Europium when alone was successfully determined by assuming that the total integrated current, after subtraction of the appropriate blank, corresponded to the reduction
where t, is the time required to reach the constant background current, estimated to the nearest 100 seconds from the current-time graph. The difference between the final background currents measured after the sample run and preelectrolysis, i f , is measured in microfaradays per second. The constant 0.70 was determined empirically from a series of experiments using amounts of ytterbium in the range 0.73 t o 20 peq. The correction of the kinetic current is also discussed by ihfeites and ?c.loros (9). Experimental results of determinations of ytterbium by reduction are given in Table 11. The sum of europium and ytterbium was determined in an aliquot of a sample solution by reducing them both at - 1.2 volts and calculating their sum by applying the same background correction given above for the reduction of ytterbium. The presence of europium did not change the value of the constant in
VOL. 32, NO. 11, OCTOBER 1960
0
1419
