RESULTS A N D DISCUSSION
The circuit consists of two main parts: a constant current source and a potential switch. The constant current source, based on operational amplifier 3, is conventional (6, 8 ) . The current through the cell is given by the voltage appearing a t the output of amplifier 2 divided by the total resistance a t the input of amplifier 3. The current through the cell remains constant, up to the voltage limits of the operational amplifier, regardless of the resistance and voltage changes in the cell during the electrolysis. -4mplifier 1 is wired as a potential switch or detector with hysteresis ( 7 ) . The output of amplifier 1 is either +50 volts or -50 volts depending upon the voltage a t the input terminals. A portion of the output voltage, e,, is fed back to the input through P1, and a rzference voltage, eref,is also added to this input terminal (the reference input) with P2. Switching occurs when the voltage on the other input terminal, e, is eref e, (the sign of e, changes as the sign of output voltage changes). For example, if erei is +1.0 volt, and e, is 0.5 volt, switching occurs a t 0.5 and 1.5 volts. When e is positive with respect to the reference input terminal, the output voltage 1s -50 volts, e, is -0.5 volt, and the potential on the reference input terminal is 0.5 volt. Switching occurs
+
when e drops just below 0.5 volt. Now the output voltage is +50 volts, e, is f0.5 volt, and the potential on the reference input terminal is 1.5 volts. Switching will next occur when c is 1.5 volts. This type of switch has also been used recently in a triangular sweep generator (10). The output voltage of amplifier 1 is changed to a strictly constant k 6 . 3 volts by the double Zener diode. This voltage is inverted and forms the input voltage to the constant current source. P 3 allows the ratio of anodic to cathodic current to be varied. The input voltage, e, is the potential of the reference electrode with respect to the working electrode, connected through a follower amplifier. For a system with no kinetic complications-e.g. cadmium ion in potassium nitrate-the first transition time was 62 msecond and the subsequent times were even less. This measurement could not have been made using the previously described circuit (S), since the switching time would have been an appreciable part of the measured transition times. The calculated times were in the predicted ratios (S), 1.00, 0.33, 0.59, etc. This circuit has been used successfully to study the current reversal and cyclic chronopotentiometry of systems following the E.C.E. mechanism, such as p-nitrosophenol, p-nitrophenol and onitrophenol and to study the reduction
of titanium(1V) in the presence of hydroxylamine; details of these studies will be presented elsewhere. LITERATURE CITED
(1) Bard, A. J., ANAL. CHEM.33, 11
(1961). (2) Bard, A. J., Herman, H. B., Third International Congress of Polarography, The Polarographic Society, Southhampton, England, July 1964. (3) Herman, H. B., Bard, A. J., ANAL. CHEM.35, 1121 (1963). (4) Herman, H. B., Bard, A. J., Ibid., 36, 971 (1964). (5) Ibid., p. 510. Malmstadt, H. V., Enk:: C. G., (6"Electronics for Scientists, p. 369, Benjamin, New York, 1963. ( 7 ) G. A. Philbrick Researches. Inc.. Boston, Mass., "Applications hlanual for :tal Plug-In Computing Nihers," p. 19, 1957. sillev, C. N., J. Chem. Educ. 39, 162). (9) Smith, L). E., ANAL.CHEM.35, 610 (19631. (10) Weir, W. D., Enke, C. G., Rev. Sci. Instr. 35, 833 (1964). HARVEY B. HERMAN' ALLEXJ. BARD DeDartment of Chemist'rv Thk University of Texas' Austin, Texas 78712 WORKsupported by the Robert A. Welch Foundation. 1 Present address, Department of Chemistry, University of Georgia, Athens, Ga. 30601.
Polarographic Determination of o-Phenylenediamine Increase in Sensitivity by Employing Lithium Iodide as Supporting Electrolyte SIR: Recently there has been considerable interest in the development of analytical methods for the determination of small amounts of o-phenylenediamine (1, 3, 6, 7 , 11). Interest in this compound arises from its widespread applications as a synthetic intermediate and anti-oxidant (1, 4 ) . An electrometric method, based on the catalytic polarographic prewave [thought to involve the reduction of a Ni complex which is regenerated in a cyclic manner (8, IO)] observed when Ni(I1) is reduced at a dropping mercury electrode in the presence of small quantities of o-phenylenediamine , has been reported ( 7 ) . The maximum sensitivity with acceptable precision was about 1 X lO-5M o-phenylenediamine. This paper reports the effects of both the nature and concentration of supporting electrolyte on the sensitivity of this indirect polarographic method. The sensitivity can be increased to allow o-phenylenediamine to be determined in
the 10-7M range with acceptable precision by using LiI in 0.1M concentration as the supporting electrolyte. EXPERIMENTAL
Apparatus. T h e dropping mercury electrode (D.M.E.) used in the analytical determination experiments had a drop time of 4.13 seconds at a height of 118.2 cm. of mercury in 0.1M LiI with no applied potential. Under these conditions the outflow of mercury was 0.816 mg. per second. The polarograms of Figure 1 were recorded using a D.M.E. with a drop time of 9.38 seconds in air free distilled water a t a height of 66.2 cm. with no applied potential; the outflow of mercury was 0.456 mgJsec. The actual drop time of the D.M.E. during the measurement was mechanically controlled a t 4.00 f 0.05 seconds by means of an electronically controlled hammer ( 2 ) . A saturated calomel electrode (S.C.E.) was used as the reference electrode, and its electrical contact with the sample solu-
tion in the polarograph cell was made through an agar-agar KC1 bridge. The polarograms were obtained with a Leeds & Sorthrup Type E Electrochemograph with no damping. A11 other experimental conditions and the preparation of reagents were the same as described previously ( 7 ) ,except Ca+2 ion was not added as a maximum suppressor ( 7 , 10); see below. The pH of the solutions was monitored by a Leeds & Northrup p H meter using miniature glass and calomel electrodes in the polarograph cell, throughout the polarographic measurement. RESULTS A N D DISCUSSION
Figure 1 shows the polarograms obtained for solutions of 5 x ~ o - ~ XiM (Clod)*. Curves A and C contain 1.OM N a h c buffer (pH = 5.5) and 6 x 10-1M Ca(.ic)* [same solution conditions employed in previously reported polarographic method for o-phenylenediamine ( 7 ) ] and curves B and D contain 0.1,21 LiI (pH = 7.1). Curves C and D conVOL. 37, NO. 4, APRIL 1965
* 591
Table I. Analysis for o-Phenylenediamine Solution contains 5 X 10-3M Nif2, 0.1.44 lithium iodide (pH = 6.5 i 0.1). NO. Found of (av.) Difference Taken detn. M X lo7 M x lo7 M x 107 4 1.0 1 . 0 - 0 . 1 to + 0 . 1
2.5 5.0
4
4 4 4
7.5
10.0
2.7 5.0 7.4 9.8
- 0 . 1 to + 0 . 4 - 0 . 1 to + 0 . 1 - 0 . 5 to 0 . 0 -0.8to + O . l
tain 1.6 X lO-'M o-phenylenediamine (note prewaves) and curves A and B were obtained in the absence of ophenylenediamine. Note the distinct difference in the characteristic shapes and potentials of the prewaves in the two different supporting electrolyte solutions. I n LiI the potential of the prewave is shifted to more positive potentials than in N a h c and has a very pronounced peak rather than a plateau. Also, the peak current of the prewave in LiI is considerably larger for a given concentration of a o-phenylenediamine than that in NaAc. This current enhancement in LiI is even greater when the o-phenylenediamine concentration is lower, lop6 to lO'M, when the maximum current is kinetic limited solely and not partially diffusion limited with respect to N i f 2 concentration as in the case above. Previous work (8) has shown that the prewave becomes diffusion limited rather than kinetic limited when the prewave height is approximately the same as the background Ni+2 limiting plateau. This is to be expected as the Ni+2, which participates in the cyclic catalytic regeneration reaction of the complex near the electrode surface, becomes depleted in the diffusion layer (8). Note, also, that the X i f 2 background wave in LiI (curve B) is shifted slightly positive of that in NaAc (curve A ) , b u t that this shift is not as great as the shift of the prewave potential [the positive shift of potentials in the presence of I- is thought to be a typical II. effect (a, 5 , 9) resulting from changes in the potential across the electrical double layer caused by the greater specific adsorption of
I-
6)1'
The enhancement of the catalytic current of the prewave and also the improvement in the definition of the prewave (greater separation of prewave from background) suggest that the use of LiI as a supporting electrolyte would
592
ANALYTICAL CHEMISTRY
result in a considerably more sensitive and precise analysis of o-phenylenediamine than previously reported in N a h c supporting electrolyte ( 7 ) . A plot of prewave current height us. concentration of o-phenylenediamine in the 10-7-11 range in a O.1M LiI and 5 x l O - 3 M Ni+2 solution (pH = 6.5) resulted in a straight line. Table I shows some typical results of analysis of o-phenylenediamine solutions in this range using the same solution conditions. The analysis in this range had an average deviation of about 0.2 X 10-7M, indicating that the use of LiI as a supporting electrolyte resulted in about a 100-fold increase in the sensitivity (while not decreasing accuracy and precision) of this indirect polarographic method for o-phenylenediamine. Several different supporting electrolytes were investigated to determine which was best suited for the analysis. Both the anion and cation had an effect. The anions Ac-, NO3-, S04-3, CI-, and I- were tried and only I- gave any improvement over Ac-. 1111 others suppressed the prewave. The effect of the cations was less than that of the anions, but was still appreciable. The cations Li+, Na+, K+,Cs+, and Ca+2 were investigated and only Li+ gave any improvement over N a f as used previously. All others suppressed the prewave. Also the presence of Ca+2 ion [used in previous work ( 7 ) as a maximum suppressor] did slightly suppress the prewave when LiI was the supporting electrolyte. Thus Ca+2 was omitted in the present paper. The effect of concentration of the supporting electrolyte was also investigated. The maximum sensitivity of the prewave to o-phenylenediamine was obtained when the supporting electrolyte was rather dilute from 0.05 to 0.1M for all supporting electrolytes tested. Further dilution increased the sensitivity more, but migration currents began to be a problem below the 0.05M range. The suppression of both prewave and Ni+2 background wave (see curves A and B of Figure 1) with increasing concentration of supporting electrolyte is a typical $ effect (a). At present, the mechanism of the effects of the different anions and cations on the mechanism of the prewave are not fully understood, although $ effects must play an important role ( 2 , 5 ) . Further investigation is in progress on this problem. The m- and p - isomers of phenylenediamine did not give prewaves under the above experimental conditions and behaved in the same manner reported
Figure 1. Variation of Nif2 polarograms with addition of o-phenylenediamine and different supporting electrolytes Temperature 2 5 ' C. f 0.1 ', [Ni+'],5 X 1 O-4M A . [ o - p d a ] = 0, 1.OM NaAc, and 6 X 1 O - W Ca+*, ( p H = 5.5 f 0.1) E. [ o - p d a ] = 0,O.lM LI, ( p H = 7.0 f 0.1) C. [ o - p d a ] = 1.6 X lo-", 1.OM NaAc, and 6 X 1 0 3 - M C a + 2 (pH = 5.5 f 0.1) 13. [ o - p d a ] = 1.6 X lO-'M, 0.1M Lil, (pH = 7.0 f 0.1)
with N a h c supporting electrolyte ( 7 ) . The pH effects were also the same ( 7 ) . LITERATURE CITED
(1) Bryan, R. H., ANAL.CHEM.36, 1980 (1964). (2) Dandoy, J., Gierst, L., J . Electroanal. Chem. 2, 116 (1961). (3) Frieser, R . G., Scardaville, P. A,, ANAL.CHEM.32, 196 (1960). ( 4 ) Gaylor, 1.. F., Conrad, A . L., Landerl, J. H., Ibzd.,27, 310 (1955). (5) Gierst, L., "Trans. Symposium on Electrode Processes," Chap. 5 , p. 109 Wiley, New York, 1959. (6) Lee, H. Y., Adams, R. N., AXAL. CHEM.34, 1587 (1962). (7) Mark, H. B., Jr., Ibid.,36,940 (1964). (8)Mark, H. B., Jr., J . Electroanal. Chem. 7, 276 (1964). (9) Mark, H. B., Jr., Reilley, C. S . ,ANAL. CHEM.35, 195 (1963). (10) Mark, H. B., Jr., Reilley, C. N., J . Electroanal. Chem. 4, 189 (1962). (11) Parker, R . E., Adams, R. N., ANAL. CHEM.28, 828 (1956). LOWELL R . McCoy HARRYB. MARK,JR. Department of Chemistry The University.of Michigan Ann Arbor, Mich. SUPPORTED in part by the U. S. Army Research Office-Durham, Contract No. DA31-124-ARO-D-284 and by a grant from the Horace H. Rackam School of Graduate Studies of the University of Michigan.
