0% found this document useful (0 votes)
101 views5 pages

Nicholson 1965

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
101 views5 pages

Nicholson 1965

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

minimum in the T C vs. concentration ACKNOWLEDGMENT H. Desty, ed.

, Academic Press, New


isotherm (21)-correctly describes the York, 1957.
The authors thank Floyd Fredricks (12) Keulemans, A. I. M., KFantes, A,,
data obtained. However, it seems un- Rijnders, G. W. A., Anal. Chim. Acta
likely that so many compounds would for obtaining some of the data. 1 6 , 2 9 (1957).
show maxima or minima in nitrogen but (13) Littlewood, A. B., “Gas Chroma-
none of them would do so in Ar or CO,, tography,” pp. 329-33, ilcademic
LITERATURE CITED
Press, hew York, 1962.
while they do give positive and negative (14) Madison, J. J., AXAL. CHEY. 30,
responses in these gases. (1) Bennett, L. A., Vines, R. G., J .
1859 (1958).
One possible alternative explanation Chem. Phys. 23, 1587 (1955). (15) hIessner, A. E., Rosie, D. &I.,
is that a concentration gradient is (, 2 .) Bohemen. J.. Purnell, J. H., J . A p..
pl.
Argabright, P. A., Ibid., 31, 230 (1959).
Chem. 8 , 433 (’1958). ’ (16) Panson, A. G., Adams, L. lI., J .
established in the cell as a result of Gas Chromatog. 2 , 164 (1964).
thermal diffusion between the cell wall (3) Dal Piogare, S., Juvet, R. S., Jr., 117) Pauschmann., H.., 2. Anal. Chem.
“Gas Liquid Chromatography,” p. 192, 203. 16 11964).
and the hotter filament. This could Interscience, New York, 1962;(
result in a transfer cf thermal energy (4) Dreisbach, R., compiler, Physical (18) Purcell, J. ’E., EXtre, L. S.,J . Gas
Properties of Chemical Compounds,” Chromatog. 3 , 69 (1965).
across the gradient, the so-called Dufour (19) Purnell, H., .‘Gas Chromatography,”
effect (’7). If so, it should be possible 3 Vols., American Chemical Society, D. 286, Wilev. New York, 1962.
Washington, D. C., 1955, 1959, 1961. (26) Rothman,” ‘ A . J., Bromlep, L. A,,
to eliminate this effect by increasing (5) Hansen, R. S., Frost, R. R., Murphy,
the tu1 bulence in the cell. Preliminary J. A., J . Phys. Chem. 68, 2028 (1964). Ind. Ena. Chem. 47. 899 11955).
(6) Harvey, D., Norgan, G. O., in (21) Schmiuch, L. J.,’ Dinerstein; R. A,,
investigation with such a cell indicates ANAL.CHEM.32, 343 (1960).
that this is so. “Vapor Phase Chromatography-Lon- (22) Smith, B. I)., Bowden, W. W.,
don, 1956,” p. 74, D. H. Desty, ed. Ibid., 36, 82 (1964).
The lack of W peaks and the low cost Academic Press, New York, 1957
of CO, suggest that it would be a better ( 7 ) Hirschfelder, J. O., Curtiss, C. F., (23) SDencer. H. M.,J . Am. Chem. SOC.
carrier gas than nitiogen for preparative Bird, R. B., “Molecular Theory of (24) ’Stuve,~K.,in “Gas Chromatography-
work (and possibly even some analytical Gases and Liquids,” p. 522, Wiley, 1958,” p. 178, I). H. Desty, ed., Aca-
New York, 1954. demic Press, X e w Ynrk. l Q 5 .
work). Some workers have reported (8) Hoffmann, E. G., ANAL. CHE%f.34,
collecting samples in CO, by freezing it 1216 (1962). (25) l‘erzele, AI., J .
(9) Hornstein, I., Croae, P., Ibid., 37, (1964).
to dry ice (9). Also, the wide variation (26) Williams, A. F., Murray, W. J.,
170 (1965). Talanta 10, 937 (1963).
in response values (positive and (10) Jamieson, C. R., J . Chromatog. 3,
negative) might be useful for qualitative 464. 494 11960): 4. 420 11060): \ ,, 8.
~~ -, RECEIVEDfor review April 7, 1965.
analysis of pure compounds. Further iii‘(i962): i S , i 6 0 (i964). Accepted July 22, 1965. Pre-ented at
(11) Keppler, J. G., Dijkstra, G., Schols, the Pittsburgh Conference on Analytical
work with COz and other carrier gases of J. A., in “T‘apour Phase Chroma- Chemistry and Applied Spectroscopy,
low T C is planned. tography-London, 1956,” p. 222, D. March 3, 1965.

Theory and Application of Cyclic Voltammetry


f m Measurement of Electrode Reaction Kinetics
RICHARD S. NICHOLSON
Chemistry Department, Michigan State University, East Lansing, Mich.

b The theory of cyclic voltammetry these determine electrode reversibility surface. Under these conditions the
has been extended to include electron indirectly b y measuring the apparent separation of cathodic and anodic peak
transfer reactions which are described standard rate constant for electron potentials is about 60/n mv., and the
b y the electrochernical absolute rate transfer from only cathodic (or anodic) reaction is reversible.
equation. Results of theoretical calcu- polarization. I n a few cases both Clearly for this case no kinetic in-
lations made it possible to use cyclic cathodic and anodic polarization give formation about the electron transfer
voltammetry to measure standard rate consistent results ( I S ) . Many of the reaction can be obtained. However, if
constants for electron transfer. Thus, relaxation techniques developed for fast frequency is increased sufficiently, a
a system which appears reversible reactions have the disadvantage that point may be reached at which the
a t one frequency may b e made to small amplitude perturbations are used, kinetics of electron transfer become
exhibit kinetic behavior a t higher and consequently differences or changes competitive with the rate of potential
frequencies, as indicated b y increased in mechanisms are not easily detected. change. Under these conditions it may
separation of cathodic and anodic A method which overcomes this dis- be possible to study the kinetics of the
peak potentials. The standard rate advantage and a t the same time gives electrode reaction, and the separation of
constant for electron transfer is de- a direct estimate of reversibility is peak potentials should be a measure
termined from this peak potential cyclic triangular wave voltammetry. of the standard rate constant for elec-
separation and frequency. The Thus, the presence of homogeneous tron transfer. Thus, a t least in princi-
method provides an extremely rapid reactions in the mechanism is readily ple, one can estimate standard rate
and simple way to evaluate electrode detected, and interpretation of results constants simply by observing cyclic
kinetics. The reduction of cadmium usually is simple. il direct estimate of polarograms on the oscilloscope, and
is used as an illustration. electrode reversibility is provided, be- then increasing the triangular wave
cause the potentials at which oxidation frequency until the separation of peak
and reduction occur are observed potentials becomes greater than 60/n
D URING RECENT’ years a number of
methods have been developed
for the measurement of electrode reac-
directly. For example, a t low fre-
quencies it may be possible with a given
system that electrochemical equilibrium
mv. The standard rate constant then
should be a calculable function of fre-
quency at this peak potential separation.
tion kinetics. I n one sense, some of always is maintained a t the electrode Unfortunately, there is no a priori way

VOL. 37, NO. 1 1 , OCTOBER 1965 0 1351


advantages. First, it reduces by one
0.4 n l the number of independent variables in
the theoretical calculations. Second, i t
has the experimental advantage of not
requiring preparation of amalgams of
known concentration for the study of
reductions a t mercury electrodes. There E is the electrode potential; Eo
The method described can be applied is the standard potential; k , is the
to systems in which homogeneous standard rate constant at E = E O ;
chemical reactions precede or follow LY is the transfer coefficient; and the

electron transfer provided such reac- remaining terms have their usual
tions are either rapid or slow compared significance.
to the standard rate constant for elec- The potential in Equation 7 for the
-0
I , , , ,
tron transfer. The cases in which first scan (reduction) of the triangular
120 60 0 -60 -120 these conditions are not met are easily wave takes the form
-
(E E , , h mv. detected from the form of the experi-
E = E, - ut (8)
mental current-voltage curves (11).
Figure 1 . Cyclic polarograms showing The reduction of cadmium is used as and for the second scan (oxidation)
effect of charge transfer coefficient, a an illustration of the method developed
--- iC. = 0.5; CY = 0.7 here. Application to other systems is E = E, + ut - 2uX (9)
...... iC. = 0.5; a = 0.3 in progress and will be reported else-
Here, E, is the initial equilibrium poten-
where.
tial, v is dE/dt, and X is the period of
the triangular wave.
THEORY If Equations 7, 8, and 9 are combined
to make this correlation because effects the result is
of concentration polarization can be We assume the following mechanism:
treated only by mathematical analysis
kf
aco
Do - = k,(Co*/C~*)-"[Sx(t)]-* x
of the mass transport processes. Such ax
a n analysis is not available for the
O+ne*R I
kb
present case, and consequently applica- (CO, = 0 -
tions of these ideas using cyclic voltam- There k , and ka are heterogeneous rate
constants of the electron transfer, and ( C O * / ~ R * ) ~ A ( ~ o)) ~ R(10)
,
metry have been limited to simple
statements that a given system appears are assumed to be functions of potential T h e function Sx(t) is defined as
to be reversible or irreversible (7, IO). as expressed through the electrochem-
I n the study of electrode reaction ical absolute rate equation. The time t<X
mechanisms, it often is useful to be able dependence of electrode potential is the &(t) = (11)
form of an isosceles triangle. System I 2ax t>
to obtain experimentally a rapid esti-
mate of electron transfer rates. Be- is assumed to be initially in equilibrium,
Both 0 and R are to be soluble either where
cause this appeared to be possible with
cyclic voltammetry in the way just in the electrode or solution phase. a = nFv/RT (12)
described, we have attempted a theo- T o account for concentration polariza-
retical treatment of the problem which tion, diffusion to a plane electrode is Equation 10 is the final form of bound-
includes the effects of concentration assumed to be the only source of mass ary condition 6.
polarization. Our primary objective transport. Mathematical formulation By application of Laplace transform
was to determine if cyclic voltammetry of this problem follows methods the above boundary value
could be used to provide rapid and problem can be converted to the follow-
reasonably accurate determinations of aco
_- avo ing dimensionless linear integral equa-
standard rate constants for electron at - D O T ax tion with variable coefficients
transfer. Thus, the theoretical correla-
tions between peak potentials, standard X(Y) [?(CO*/cR*)sd(!/)I"=
rate constant, and rate of potential !4
scan have been emphasized, although
other correlations are possible. A seri-
ous limitation of this approach would
result if peak potential separations
depended also on the charge transfer
coefficient, C Y . This would require an
independent measure of C Y , and make the T o make Equation 13 dimensionless the
method of little use in terms of the following substitutions and changes of
stated objectives. However, through t2O,x=O variable were made:
proper selection of conditions, peak
potential separations become nearly Y = (Do/DR)"' (14)
independent of a .
Becauqe there is neither theoretical
y = at (15)
nor experimental advantage to consider-
ing more than one cycle of the applied x(y) = Do aco
a,/Co*l/?rao, (16)
triangular wave, theoretical calcula-
tions havt been limited to this case. Undefined terms have their usual
significance (11).
4 = y4s/l/*aDo (17)
ConsideratioIL also has been restricted
to the case of an situ generation of the With the electrochemical absolute For large values of $ (large k , or
reduced form of the couple under in- rate equation, Equation 6 can be trans- small v ) Equation 13 becomes independ-
vestigation. This actually has two formed t o ent of 4 and CY and reduces to the

1352 ANALYTICAL CHEMISTRY


calculated current-voltage curves could
be interpreted on the basis of only two 60 -
variables, \L and a. -
Results of Calculations. When + 80

becomes sufficiently large ( k , large, 100-


6 ,
or v small), current-voltage curves
calculated from Equation 13 are inde- 3 120-
4
pendent of the kinetic parameters $ and 140-
CY. For this case results are identical
to ones obtained previously where the 160-
I I I
electron transfer was assumed to be
Sernstian (11). This limit occurs when
$ > 7, which is in good agreement with
calculations of Matsuda and Ayabe (8).
When $ < 7 , the current-voltage
curves are dependent on both the values
(E - E,,h, rnv of + and a . For sufficiently small $
( k , very small or v very large) , the back
Figure 2. Cyclic polarograms showing
reaction for electron transfer is unimpor-
effect of charge transfer parameter, + tant, and the processes for oxidation
and reduction can be treated separately
.. .. . $$ == 0.25;
7.0; CY
CY
= 0.5
= 0.5 as the totally irreversible case. Thus, Figure 2 where curves for two values of
+ and CY = 0.5 are compared. The
results calculated from Equation 13 for
$ = 0.001 agree exactly with previous curve for $ = 7 corresponds (within
calculations in which electron transfer about 1%) to the reversible case and is
corresponding equation for reversible
(Nernstian) electron transfer (11). was assumed to be totally irreversible independent of CY. As $ decreases two
(11). This also is in agreement with effects can be distinguished from Figure
Likewise, as $ approaches zero, Equa-
calculations of Matsuda and Ayabe (8). 2. Both cathodic and anodic peaks
tion 13 approaches the case for totally
For intermediate values of $ (a region are lowered, and both waves appear
irreversible electron transfer (11). The
sometimes referred to as quasi-reversi- drawn out-Le., they occur over a
intermediate case, which is of interest
here, is described completely by Equa- ble), the form of the current-voltage wider potential range. Thus, the ca-
tion 13. curves depends markedly on the exact thodic, wave is shifted cathodically and
values of $ and cy. the anodic wave is shifted anodically.
Solution of Integral Equation.
The dependence -on CY is shown in This naturally results in an increased
Equation 13 cannot be solved an-
Figure 1 where curves for two values of separation of cathodic and anodic peak
alytically. Although a solution in
CY and $ = 0.5 are compared. Gener- potentials.
the form of a n infinite series can be
ally, the results are as expected, and CY The exact variation of peak potentials
obtained, series solutions for more
affects the symmetry of the curves in with $ is shown in Figure 3, and the
than one-half cycle become hopelessly
two ways. For cy < 0.5, the cathodic data used to const,ruct Figure 3 are in
complicated. T h e only practical ap-
proach, therefore, which for t h e peak is more rounded than the anodic Table I. I n Figure 3 A E p x n (n is
peak. This broadening also results in the number of electrons) is plotted as a
present case can be applied without
a lowering of peak height. For cy > 0.5, logarithmic function of $. Data for
loss of generality, is numerical evalua-
the converse holds. Figure 3 were obtained using CY = 0.5,
tion. We have used a method described
The second effect involves a slight but from the previous discussion the
previously (1I ) . Several other methods
displacement of the waves along the curve of Figure 3 is nearly independent
also are possible, and apparently the
method used by DeTtries and VanDalen potential axis. For example, as CY of CY. As mould be expected the curve
(6) involves less computational time decreases, the cathodic peak is displaced approaches linearity for small $ [see
for comparable acculacy ( 5 ) . cathodically. At least for the data used Equation for peak potential of irreversi-
Solutions of Equation 13 were inde- to construct Figure I , the effect is slight. ble waves (II)]. I n Ohis region AEp
Lloreover, as the cathodic peak shifts is markedly dependent on CY.
pendent of values of r(C,*/CR*) pro-
vided r(Co*/CE*) was greater than e6 5. cathodically with decreases of CY, the
This corresponds to the experimental anodic peak also shifts cathodically.
fact that if R is a b s m t initially and the Therefore, in terms of differences of peak
voltage scan is begun at potentials potentials, AEP, changes in CY tend to
Table I. Variation of Peak Potential
anodic of the pcllarographic wave, cancel. Thus, values of AEp for the Separations with Kinetic Parameters for
current-voltage curves are independent present case are nearly independent of
Cyclic Voltammetry
of the exact initial potential selected, cy in the range 0.3 < CY < 0.7. The
All calculations have been performed extent to which this holds depends on AEP X 7~,*
using r(Co*/CR*) = e 6 5 . the exact value of $. Saturally, as $ $" mv.
With this simplifiaation it was possi- increases the approximation becomes 20 61
ble to calculate numerically current- increasingly good. Thus, for $ = 0.5 7 63
6 64
voltage curves. These curves depended (AEp ca. 105/n mv.), there is about a 5 65
on three vaiiables, p, cy, and A. It has 5% variation of AEP for CY changing 4 66
been shown previously that h affects from 0.3 to 0.7. For $ < 0.5 the 3 68
variation is greater, and causes about a 2 72
only the anodic wave, and that there 1 84
the effect on anodic peak potentials is +
20% change of AEP for = 0.1. How- 0.75 92
quite small (11). This is particularly ever, for this case the curves have 0.5 105
the case when switching potentials close deviated considerably from the reversi- 0.35 121
ble case (AEPea. 200/n mv. compared 0.25 141
to the cathodic peak are avoided 0.1 212
[(Ex - E l , J n > 90 mv.]. All calcula- with 60/n mv. for the reversible case).
a See Equation 17.
tions reported here are based on a value The dependence on $ of theoretical * F o r CY = 0.5.
of (Ex - = 141/n mv. Thus, current-voltage curves is shown in

VOL. 37, NO. 1 1 , OCTOBER 1965 1353


It should be noted that the theoretical and increase current capabilities, a
curves of Figure 2 could correspond Philbrick Model K2-BJ booster Table II. Determination with Cyclic
either to two different systems having amplifier was used in series with the Voltammetry of k, for Reduction of
different electron transfer rates-and potentiostat. The bandwidth of the
potentiostat was adjusted for the solu- Cadmium
observed a t the same scan rate- tion used to provide optimum rise- AEP
or to the same system observed at time and stability. To do this the v, x n, ke ‘,
two different scan rates. This is ideas of Booman and Holbrook were volt/sec. mv. $5 cm./sec.
exactly the behavior anticipated used ( I , 2 ) . A good quality commercial 48.0 94 0.70 0.25
in the discussion a t the beginning and potentiostat (e.g., Wenking Potentio- 60.0 98 0.61 0.25
provides the basis of using cyclic voltam- stat, Brinkmann Instruments, West- 90.0 108 0.48 0.24
metry to estimate electron transfer bury, 3. Y.) presumably could be used 120,o 115 0.41 0.23
rates. Figure 3 gives quantitatively in place of the potentiostat just de- a Determined from Figure 3; see
the relationship betm een peak poten- scribed. Equation 17.
Two different signal generators were b 2 X 10-4M cadmium sulfate, 1.OM
tials, scan rate, and electron transfer sodium sulfate.
used, and both proved adequate. One
rate. was constructed from operational ampli-
To use Figure 3 in this way a scan fiers and has been described previously
rate is selected which gives an experi- (10). The other was an Exact Model
mental A E p in the useful range of 255 function generator (Exact Elec- to k,, the diffusion coefficients of
Figure 3. From the experimental AEp tronics, Inc., Hillsboro, Ore.) , which Okinaka (12) were used. A value of (IL
x n, the corresponding value of Vl. is proved useful and versatile. equal to 0.25 also was used. Although
obtained from Figure 3. If scan rate is The detector was a Tektronix Model no effort was made to determine a
known, k , can be calculated from Equa- 536 oscilloscope provided with a Po- accurately, the value of 0.25 was esti-
tion 17. A4ctually,to apply Equation 17 laroid camera attachment (Tektronix mated by comparing with theory the
Type C-12). The vertical input was symmetry and shape of cathodic and
rigorously yo also must be known. always a Type D plug-in preamplifier.
However, except for the unusual case of anodic waves for cadmium. The calcu-
This could be operated in the differen- lated value of k, is fairly insensitive to
very large differences between DO and tial mode and permitted use of a floating
D E , the quantity y a is very near unity, load resistor in the potentiostatic circuit variations of a,as already mentioned.
regardless of CY. (10). To measure scan rates for the The values of k , in Table I1 agree
=ilthough the useful working range of function generator constructed from reasonably &-ellwith some other workers
Figure 3 is fairly restricted, this is not operational amplifiers, a Type T plug-in (14). The disagreement with others
a serious limitation, especially if rough preamplifier was used for the horizontal (12, 14) is not entirely explained. The
input of the oscilloscope. With the apparent rate constant does change some
estimates of k , are all that is required. Exact function generator, scan rates
This is so because the wide range of with different supporting electrolytes.
could be dialed directly with an ac- For example, we find k , equal to about
scan rates normally available with curacy of =t2% and this proved to be
cyclic voltammetry allows simple experi- 0.6 cm./sec. in 2M perchloric acid.
a considerable advantage over the other
mental selection of any AEp for meas- signal generator. To measure peak The value in sulfuric acid is slightly
urements. potential separations a Type H plug-in higher. I n addition, electrodes are sub-
The upper limit of rate constants that preamplifier was used for the horizontal ject to an aging effect noted by Delahay
can be determined experimentally by input of the oscilloscope. The hori- ( 3 , 4 ) . We also observe this decrease of
the above procedure is discussed in a zontal axis then was driven by the cell apparent rate constants with time, but
potential, and peak potential separations the effect in our solutions is not as great
later section. could be measured with a n accuracy of as Delahay found (about a factor of two
about 1 2 mv. change in k , for the first 20 minutes
APPLICATION OF THEORY TO REDUCTION OF The cell assembly has been described
CADMIUM quiescence).
previously (10). The upper limit of rate constants that
T o evaluate the method described Materials. Solutions were prepared
from cadmium sulfate (Baker A. R.) can be determined by the above method
above a model system is required for and anhydrous sodium sulfate (Baker appears to be set by a combination of
which the mechanism and kinetic pa- A. R.). The concentrations were: cad- factors. First, in the measurement
rameters are well knonn. The extensive mium sulfate, ea. 2 x 10-4M1; sodium of peak potential separations, uncom-
investigations on cadmium, which have sulfate, 1.0 X , The exact cadmium pensated ohmic potential losses are a
been made previously using a variety of concentration was not determined. serious source of error. Thus, for the
techniques, appeared to make cadmium Xeasurements were made in a constant rapid scan rates required to study fast
a logical choice. However, the compila- temperature room a t ambient tempera- electrode reactions, peak currents be-
tion by Tanaka and Tamamushi of tures of 23’ to 25’ C. come fairly large and even relatively
kinetic data (14) shows a considerable small solution resistances can introduce
variation of both standard rate con- RESULTS A N D DISCUSSION serious error. Moreover, the effects of
stants and charge transfer coefficients uncompensated iR drop qualitatively
for cadmium when determined by Both one-cycle and multi-cycle are very similar to kinetic effects (9).
various workers. Nevertheless, cad- (steady state) experiments were run This is amply illustrated by Figure 3
mium was considered the best model and as would be expected the differ- where the dashed curve is a plot of data
system available, and therefore i t was ences between A E p for the two methods calculated previously for the effect of
used to test the theoretical calcula- were of the order of experimental error. iR losses on cyclic voltammetry (9).
tions. Nevertheless, data reported here are The definition of the horizontal axis ($)
for one-cycle experiments to conform to of Figure 3 for this case should be (9) :
theoretical calculations. For all experi-
EXPERIMENTAL

Apparatus. T h e circuit was es-


ments an initial potential of -0.370 + = l/(nF/RT)nFAX
volt us. S.C.E. was used, and the ampli- ( T ~ D O ) ’ ’ ~ C ~ *(18)
R,
sentially the same as described previ- tude of the triangular wave always was
ously (10). All measurements were 300 mv. Thus, the variation of peak potential
made with a three-electrode potentio- separation with scan rate is quite similar
static circuit, using a single Philbrick Results for reduction of 2 X 10-4M
>lode1 SK2-V operational amplifier cadmium in 1.022 sodium sulfate are for the two cases.
(G. A. Philbrick Researches, Inc. summarized in Table 11. T o convert The effect of ohmic potential losses
Boston, Mass.). To improve rise-time experimentally determined values of + can be minimized in three mays. First,

1354 e ANALYTICAL CHEMISTRY


relatively concent-rated electrolytes of measured by peak potential shifts on (4) Delahay, P., J . Chim. Pkys. 54,
high conductivity can be used to reduce moLing the Luggin probe more than 369 (1957 .
(5) DeVries, W. T., Free University,
the total cell resistance. Second, with 10 radii from the working electrode Amsterdom. The Xetherlands. orivate
I -

a three-electrode potentiostat place- agreed with values calculated from the communication, 1965.
ment of the Lugg-in capillary probe in conductivity of the solution (2). By (6) DeVries, W. T., VanDalen, E., J .
close proximity with the working elec- careful placement of the Luggin probe Electroanal. Chem. 8, 366 (1964).
(7) Galus, Z., Lee, H. Y., Adams, R. N.,
trode serves to compensate for a ma- the uncompensated resistance was esti- Zbid., 5 , 17 (1963).
jority of the total cell resistance [see mated to be 3 to 4 ohms ( 2 ) , so t h a t (8) hlatsuda, H., Ayabe, Y., 2. Elektro-
Iiooman and Holbrook for the exact maximum ohmic potential losses were chem. 59. 494 11955).
relat,ionship (2) 1. Third, relatively low of the order of 1 mv. (9) Nicholbon, R . S.,'ANAL.CHEM.37,
667 (1965).
concentrations of dectroactive material The upper limit of rate constants (10) Nicholson, R. S., Shain, I., Ibid.,
can be used to keep total cell currents measurable with cyclic voltammetry p. 190.
small. This latter approach is limited, therefore depends on several factors. (11) Zbid., 36, 706 (1964).
however, by charging current. Thus, a t The most important of these is probably (12) Okinaka, Y., Talanta 11, 203 (1964).
(13) Selis. S. >I.. J . Phus. Chem. 68.
high scan rates charging current may conductivity of the solution under ~ 2538 (1964). '
become a n appreciable fraction of the investigation. With electrolytes of rea- (14) Tanaka, X.,Tamamushi, R., Electro-
total cell cuirent, especially when lo^ sonably high conductivity, it should be chem. -4cta 9,963 (1964).
concentrations of electroactive mat'erial possible to measure rate constants as RECEIVEDfor review June 28, 1965.
are used. large as 1 to 5 cm./sec. Accepted August 3, 1965. Presented in
part at the Division of Analytical Chem-
For the cadmium system investigated istry, 150th XTeeting ACS, Atlantic City,
here a concentration of 2 to 5 X 10-4Jl LITERATURE CITED N. J., September 1965. Work supported
appeared to be optimum for low cell in part by the United States Army Re-
currents without excessive charging (1) Booman, G. L., Holbrook, W. B., search Office-Durham under Contract
ANAL.CHEM.37, 795 (1963). No. DA-31-12PARO-LL308. Other sup-
current contributions. hIasimum peak ( 2 ) Zbid., 35, 1793 (1963). port was received from the National Sci-
currents n-ere of the order of 0.2-0.3 ma. 13) Delahav, P., Trachtenberg, I., J . Am. ence Foundation under Grant KO. GP-
Total solution resistance (ea. 20 ohms) Chem. Si.' 80,2094 (1958).- 3830.

Pola rog ra p hy of Lead(II) in Aqueous Hy d rof I u oric


Acid (1 to 12M) with a Dropping-Mercury
Electrodle of Teflon
HELEN P. RAAEN
Analytical Chemistry Division, Oak Ridge National Laboratory, Oak Ridge, Tenn.

b The polarographic behavior


Pb+2 (-0.03 to 0.8rnM) in aqueous
HF (1.00 to 12.00M) was studied with
of
T HE DROPPING-NERCURY electrode of
Teflon (DuPont) has been described
(8),and its operation in noncorroding
polarographic behavior of the ions of a
number of elements, including lead, in
0.1X NH4F-O.lM HF (4) and polaro-
a D.M.E. of Teflon. The polarograph- media was investigated (9, 10). A graphically determined molybdenum in
ically usable potential span for 12.00M review was given (8) of the polarography niobium-base alloys using 0 . 5 X HF-
HF is about +0.4 to -1.0 volt vs. -done before the D.3I.E. of Teflon was 0.5M H2S04 as supporting medium ( 5 ) .
S.C.E.; for 1 .OOM HF, the limit extends developed -of substances in liquid hy- With the D.M.E. of Teflon, the polaro-
to about - 1.2 volts. Within this span drofluoric acid and in aqueous acid graphic characteristics of P b + 2 in
only one polarographic wave for Pbt2 fluoride media. aqueous hydrofluoric acid solutions have
exists; it is for the polarographically The polarography of lead in acid now been studied in some detail. T h e
reversible 2-electron-change reduction fluoride media has been studied to a data given herein are the first polaro-
Pb+* -+ PbO. The E l l 2 values de- limited estent. West, Dean, and Breda graphic data reported for Pb+2 in
termined are -0.:380 (for 1.00M HF) ( 2 1 ) obtained a well-defined wave for aqueous solutions of reasonably high HF
fa -0.370 (for 141.00M HF) h 0.003 lead in 0.5X Na2F2 [qic]-O.Ol% gelatin concentration and that contain no other
volt vs. S.C.E.; the values are not solutions of p H 1.1 to 2.75 that con- added constituents-e.g., a salt to con-
affected by change in Pb+2 concen- tained considerable amounts of nitric tribute ionic strength or a maximum
tration. The wave is of excellent acid added to dissolve the lead fluoride suppressor.
form and is usable for both quali- precipitate; the solutions therefore con-
tative and qirantitcitive purposes. The tained lead nitrate and slightly dis- EXPERIMENTAL
relation of j d , andl also of ($/dt)maz, sociated hydrofluoric acid. The polarog-
Reagents. Standard Solutions of
to Pb+2 concentration is linear; plots raphy of lead, as PbF2,in liquid hydro- Pb+2 in Aqueous HF. A quantity of
of these variables pass through the fluoric acid was investigated briefly by lead fluoride, FbFz, (B & X purified
origin. With increase in HF concen- Clifford (2). Mesaric and Hume ( 7 ) grade) calculated to give 1 m X solu-
tration from 1.00 to 12.00M, no used polarography to study lead fluoride tion was added to a n aqueous solution
change occurs in the number of lead complexes and their solubilities; the of HF (1.00, 4.00, 8.00, or 12.00M).
waves, shape of the wave, polaro- solutions they studied were 1mJf in After maximum dissolution the solu-
-
graphic reversibility of the reduction, Pb+*, u p t o 0.7M in F-, and of 2 M tion was filtered to remove undissolved
PbF2. The Pb+* concentration of the
or n; the i d deceases slightly; and ionic strength adjusted by adding filtered solution was determined by a
the El/* becomes about 10 mv. more KaC104. With a D.M.E. of Teflon, separate analysis in which HF was
positive. Headridge and associates studied the removed from test portions by fuming

VOL. 37, NO. 1 1 , OCTOBER 1965 1355

You might also like