вход по аккаунту


Inverse Polarography and Voltammetry New Methods for Trace Analysis.

код для вставкиСкачать
Inverse Polarography and Voltammetry: New Methods for Trace Analysis
Inverse polarographic and voltammetric methods are surveyed. The electrochemical determination procedure is carried out by current reversal after electrolytic concentration on a working
electrode. Analyses in the nanogram (ng) region (10-9 g.) are possible because the sensitivity
is increased by a factor of 100 to I000 in comparison with the usual polarographic and
voltammetric methods. These methods are, therefore, of particular interestin trace analysis.
I. General
Polarographic methods are valuable analytical tools in
the region of microgram quantities. Sensitivities greater
than 0.2 to 1 pg/ml are generally not attainable. However, particularly sensitive polarographic and voltammetric methods have recently been described [l].
The present paper describes only those methods in which
the ion to be determined (e.g. by recording a currentvoltage curve and determining the diffusion current) is
electrolytically preconcentrated. If, after pre-concentration on the working electrode, it is possible to employ a
reversible electrochemicalprocess for the determination
of the element, a current increase proportional to the
concentration factor is obtained. The concentration
factor depends upon the conditions during electrolysis
and the relationship of the volume of the aqueous
solution to the volume (or the surface) of the working
The principle of inverse voltammetry is described by
means of an example in Figure 1: In the reduction of
I Pya
Recordinodirection -D
- 0.5
O-lyq &d/ml
Electrolysis voltage
/Time of electroiysis:5minj
Figure 1. Principle of inverse voltammetry; stationary mercury
electrodes (5 rnrnz), 0.06 V/sec., 0.1 N HCI.
Cd ion at the stationary mercury electrode (e.g., a
hanging mercury drop), peak currents are developed [2]
which are proportional to the concentration of Cd2+ in
the aqueous solution.
[ I ] M.v.Stackelberg and H. Schmidt, Angew.Chen.71,508 (1959).
[2] J. E. B. Randles, Trans. Faraday SOC.44, 327 (1947).
is. = K0.D1/2n3/~ccv1/2
D = Diffusion coefficient in mercury or water, c = Concentration in aqueous solution or in amalgam, v = Rate of
voltage change (Visec), n = Valence change, K = Constant,
0 = Electrode surface area.
Since the diffusion coefficients are of similar order of
magnitude and since the equation is also approximately
valid for the amalgams, the increase in the peak current
is explained by the increase in c.
In this technique the determination takes place when the
direction of both the current and diffusion is reversed. The
expression “inverse polarography and voltammetry” [*I
may therefore be considered justified.
lOyg C d h i
This peak current is indicated in the upper portion of
Figure 1. If the cadmium is electrolytically concentrated
(at -1.2 V) on the mercury prior to plotting of the
current-voltage curve, then peak currents derived from
the oxidation of the cadmium in the mercury are
obtained by recording the current-voltage curve in the
reverse direction. In an electrolysis of appropriate
duration, the concentration of cadmium in mercury is
considerably higher than that 01 tne cadmium ion in the
aqueous solution. According to Randles [2] and Sevtik
[3] the value of such peak currents is given by the
following equation:
The presently known methods for electrolytic concentration in inverse polarography and voltammetry are
shown in Table 1. The concentration process consists of
either (a), the electrolytic deposition of a metal on a
solid metallic electrode or (b), the production of an
amalgam at a mercury electrode, as indicated above.
A third possibility (c) is electrolytic oxidation of the
metal of the working electrode in the presence of an
anion, which together with the metal ions generated
forms a compound which is not readily soluble. This
compound is then deposited on the electrode surface in
[3] A . SevEik,Collect.czechoslov. chem. Commun.23, 349 (1948).
[*I In Anglo-American literature the term “strippingtechnique”
is often used. Concerning the term “chronoamperometry”see
[5]; concerning the term “voltammetry” see also [4]. The term
“inversepolarography’’hasbeen used repeatedly in the literature;
it is assumed to have originated with Hickling et al. [32].
[4] P. Deluhuy: New Instrumental Methods in Electrochemistry.
Interscience Publ., New York - London 1954, p. 1 1 .
[5] P. Delahay, G. Charlot, and H . A . Laitinen, J. elektroanalyt.
Chem. I , 425 (1959).
Angew. Chem. internat. Edit. / Yo/. 1 (1962) / No. 4
Table 1. Basic principles of inverse polarography and voltammetry
determination ;
procedure in
A) M@+
Me+ -+Me0
B) Men+
Me+ + Me0
C) An-
AMe0 -+
procedure of
inverse polarography and
On metal
Me+ + Me0
Me0 + Me’
O n mercury
electrode: Me
-+ Me0 (Hg)
Me0 (Hg) + Me+
On metal (also
A-+ Meo-+MeA
MeA -+ Me0 +A(cathodic)
proportion to the concentration of the anion. The
subsequent voltage reversal (cathodic) causes a reduction of the metal ion contained in the deposit. Peak
currents, whose values are dependent on the amount
and, therefore, on the concentration of the deposited
metal ion, are thereby generated.
The concentration of the ion on the working electrode
can again be considerably increased by proper selection
of the electrolysis conditions. An example of this method
is given below for the determination of iodide by means
of a silver electrode. (See also Figs. 10 and 11).
II. Equipment
The only factor common to all methods is the electrolytic concentration of ions on the working electrodes. The
concentrated substances can be subsequently determined
according to the well-known polarographic and voltammetric methods (Table 2).
With the exception of VII, all the methods can also be
used for determination of the ion in aqueous sdution.
The method described under VII is, in reality, a modification of method 11. Under certain conditions (slow
rate of voltage change, use of platinum electrodes or of
amalgamated wire-electrodes), the electrolytically deposited substance dissolves completely within the first
voltage sweep. The area under the current curve is then
proportional to the quantity of electricity necessary for
the redissolution of the deposit and, therefore, to the
quantity of deposited material. The second column
indicates the current-voltage-time function used for the
determination,while the third column shows a schematic
representation of the corresponding voltammetric curve.
1) Recording Apparatus and Voltage Source
Generally, commercial equipment can be used. With methods
11,111, IV, and VII it is necessary to be able to record rapidly
occurring current-voltage curves (11, VII), voltage-time
curves (IV), or current-time curves (111). Suitable recorders
with fast response time (a maximum of 0.25 sec for full scale
deflection) are available and can be used in conjunction with
suitable amplifiers. For even higher recording speeds, very
sensitive, fast-response galvanometers with photographic
Angew Chem. internat. Edit. I Vol. I (1962) No. 4
recording are required. An installation of this type has been
successfully used for several purposes [12]. Oscilloscopes
with long-persistence phosphor screens are also used [16,18].
The so called “Blauschreiber” 1261 may well find many
applications in analytical methods for the recording of rapid
transient phenomena, such as observed in chronopotentiometry with amalgam electrodes. Kohlrausch cylinders or
rotary potentiometers, driven by synchronous motors are
usually employed for the generation of variable voltages for
the recording of current-voltage curves (method 11). Electronic voltages sources produce steady voltages which are not
subject to disturbances from mechanical contacts even if the
voltage change is very rapid. Their principle of operation is
based on the use of a well-insulated capacitor which is
charged through a large resistor from a constant-voltage
source (27-29). The amplification of the input voltage
through a cathode amplifier results in a voltage source with
a very small internal resistance. (See Fig. 2 for block circuit
120 v
Figure 2. Schematic diagram of an electronic voltage source.
R = 2 to 25 megohms; C = 16 microfarads; AV/At = 0.08 to 2.0V/sec.
When only the initial portion of the input curve is considered,
the voltage change is seen to be virtually linear. By varying
the load resistance one can obtain various voltage change
rates. Such electronic voltage generators have recently been
incorporated into conventional polarographs [30]. These are
the only usable instruments for applications in which the
voltage changes rapidly (such as in method 10 and also
where high sensitivities are required. A polarograph which
is particularly suitable for inverse voltammetry (method 11)
will soon become commercially available [3 11.
2) Electrodes and Electrolytic Cells
Stationary electrodes with constant surface area are used
almost exclusively, while dropping electrodes are used only
a) D r o p p i n g Amalgam Electrode
While it is analytically of little value, this electrode is used
for various polarographic investigations [6]. The amalgams
may be produced by electrolysis of aqueous solutions, with
the mercury of the drop electrode serving as the cathode.
[6] N. H. Furman and W. Ch. Cooper, J. Amer. chem. SOC.720
5567 (1950).
[71 A. G. Stromberg and A . A . Pyshkina, Trudy Kom. Anal.
Khim; Akad. Nauk. S.S.S.R. 7, 136 (1956).
[81 J. G. Nikelly and W.D.Cooke, Analytic. Chem.29, 933 (1957).
[9] R. D. de Mars and I. Shain, Analytic. Chem. 29, 1825 (1957).
[lo] W. Kemula, Z . Kublik, and S. Glodowski, J. electroanalyt.
Chem. I, 91 (1959).
[I I ] W . Kemula, Proc. Intern. Symp. Microchemistry (Birmingham 1958). Pergamon Press, London 1960, p. 281.
[I21 R. Neeb, Z. analyt. Chem. 171, 321 (1959).
[13] W. Kemula, E. Rakowska, and 2.Kublik, J. electroanalyt.
Chem. I, 205 (1959).
[14] W. Kemula and Z . Kublik, Analyt. chim. Acta 18, 104 (1958).
[15] G. Mamantov, P. Papoff, and P. Delahay, J. Amer. chem.
SOC.79, 4034 (1957).
[16] G. Mamantov,P. Papoff, and P. Delahay, Analyt. chim. Acta
18, 81 (1958).
[17] R. Neeb, Z. analyt. Chem., in the press.
Table 2. Analytical methods in inverse polarography and voltammetry
Curve developed
Calibration curvi
Term used
i = f(E)
( E - constant]
i = f(E)
“linear sweep
(E = linearly
i = f(r)
(E = const.)
E = f(t)
(i = const.)
a) Potential-step
b) Coulometry
a) c N it
b) c
r15, 161
Current step
[galvanostatic method)
C - 7
dE = f(E)
i s q . w . = f(E)
Method I
Classical constant voltage polarography : The average
current (i) found on the mercury drop is measured as a
function of the constant voltage (E). The voltage applied
is varied so slowly that “i“ is measured at a virtually
constant potential.
Method I1
The applied voltage (E) is changed at a rapid, constant rate
of about 0.3 V/sec. Because of the depletion of the material
on the electrode surface and the lack of sufficient supply
by diffusion, peak currents develop which are proportional
to the concentration (isp).
Method 111 The current at a stationary electrode is measured as a
function of time at constant potential (in this case more
positive than the deposition potential). Either the area
below the ijt curve (a) or the current (it) after a specific
elapsed time (t) are measured for the evaluation.
[21, 221
Method IV After application of a current (if (decomposition current
for amalgams) the potential (E) of the working electrode is
measured as a function of time. The transition time (the
period of stationary potential) is concentration dependent.
Method V
An alternating sinusoidal voltage (E) is applied at a
constant current to the mercury electrode. The oscillograph
registers the time derivative of the electrode potential
as a function of (E).
Method VI A square-wave alternating current is superimposed on the
current of the usual polarography. The amplitude (isq.
wave) of the constant voltage is measured at the pulse end
where the capacitance current has practically reached zero.
This is necessary to eliminate the effect of the capacitance
Method VII Similiar to method I1 (also see text).
[IS] R. Kulvodu, Collect. czechoslov. chem. Commun. 22, 139,
1251 T. L. Murple and L. B. Rogers, Analyt. chim. Acta l I , 574
( 1957).
( 1 954).
[I91R. Kulvodu, Analyt. chim. Acta 18, 132 (1958).
(201 R . Kulvodu, Proc. Intern. Sympos. Microchem. (Birmingham
1958). Pergamon Press, London 1960, p. 288.
1211 G. C. Barker, A. E. R. E. Report C/R 1563 (1957).
(221G.C.Barker, Analyt. chim. Acta 18, 118 (1958).
(231S . S.Lord, R. C. O’Neill, and L. B. Rogers, Analytic. Chem.
24, 209 (1952).
(241K. W.Gurdiner andL.B.Rogers,Analytic.Chem.25,1393(1953).
1261 Rhode und Schwarz GmbH., Miinchen, Germany.
[27] E. C. Snowden and H. T. Puge, Analytic. Chem. 22, 969
1281 K . Err: Intern. Polarogr. Congress, Prague, 95, 111, 762.
1291 S. Weidmunn, Thesis, Universittit Bern 1942.
1301 M. T. Kelley, D. J. Fisher, and H. C. Jones, Analytic. Chem.
32, 1262 (1960).
[31]Analytic. Chem. 33, No. 3, 70A (1961).
Angew. Chem. internat. Edit. I Vol. I 11962) 1 No. 4
Figure 3 illustrates the apparatus in which the electrolytic
cell is so constructed that the generated amalgam may be
used to fill the drop-producing capillary under a nitrogen
When the current-voltage curves are recorded, anodic diffusion currents are generated which are proportional to the
concentration of the metal in the amalgam. In the presence
of several metals with differing half-wave potentials, separate
diffusion currents are generated. The dropping amalgam
electrode is of interest in trace analysis only where the diffusion currents, developed from the anodic dissolution (stripping) of the amalgams, are greater than those obtained in an
aqueous solution, even at the highest sensitivities. This can
be accomplished by the use of minimum amounts of mercury
in the electrolysis. In this case, however, the concentration
factor is dependent not only on the relationship of the mercury to the volume of the aqueous solution, but also on the
duration of the electrolysis. Small amounts of common metals
They improve the mechanical stability of the system, especially when the solution is vigorously stirred.
Figure 4. Simi mercury 3p electrode: a) Salt bridge in reference
electrode; b) Mercury drop electrode; c) TeEon spoon; d) Polarographic capillary; e) Nz inlet. (From: J . W. Ross, R. D. DeMars, and
J. Shain, Anal. Chem. 28, 1169 (19561.)
A small mercury meniscus, partially emerging from a
capillary of 1 to 3 mm diameter, has also been proposed as
an electrode [S]. The electrode surface can be conveniently
renewed by the removal of the drop. If the diameter of the
capillary is reduced, the quantity of mercury emerging from
the capillary can be controlled. In this manner, reproducible
and easily renewed mercury drops are produced.
Figure 3. Apparatw for production of dropping amalgam electrodes.
a) Cell with mercury electrode for production of the amalgam;
b) Vessel to "catch" the amalgam in nitrogen atmosphere:
c) Drop capillary. (From: N . H. Furman and W. C. Cooper, J. Amer.
chem. SOC.72, 5668, (1950).)
can be determined in the presence of higher concentrations
of noble metals by means of the amalgam drop electrodes.
This is possible, because, contrary to the case in aqueous
solution, the common metal constituents generate their
diffusion currents first. It is, however, possible that greater
amounts of noble metals, which have been carried into the
amalgam during electrolysis, may lead to interference during
the anodic dissolution [6,32]. Stromberg [7]reported on the
application of the amalgam drop electrode to the analysis
of slags and technical amalgams. Determinations of 10-3 %
of lead in bismuth and 10-2 % zinc in cadmium, as well as
10-3 % cadmium in lead can be made.
Barker [21] permits the mercury to flow from a capillary
of 0.5 mm in diameter with a 0.05 mm diameter neck.
The capillary is connected to a mercury reservoir. The
mercury flow is interrupted by means of a stopcock or a
needle valve (Fig. 5 ) after a definite time (30 sec).
Mercury drops, whose size is reproducibleto & 2 %,were
produced. A similar electrode has been recently used
for the determination of C1- [34]. Kemulu et al. [lo, 11,
13,141 utilize a screw-type needle piston which forces
well-defined mercury drops from capillaries of 0.1 or
0.2 mm in diameter (Fig. 6) [*I.
b) Stationary Mercury Electrode
This is the type most commonly used at present. The stationary electrode has almost all the advantages of the
dropping mercury electrode but it is important that reproducibility of the surface area can be assured. Hickling [32]
utilizes a large mercury electrode similar in form to the one
used for electrolytic deposition (Hg cathode). The simplest
form [9,15,33] of small, stationary electrodes consists of
one (or several) drops from a drop electrode which are
suspended from gold or platinum wires, after having been
caught with a small spoon (Fig. 4).
Electrodes of this type consistently deliver mercury drops
with a constant surface area. Electrodes in which the mercury drops are not suspended from a platinum wire but
collected in a small dish [12], operate in a similar manner.
[321 A. Hickling, J. Maxwell, and J. V. Shennan, Analyt. chim.
Acta 14, 287 (1956).
[33] H . Gerischer, 2. physik. Chem. 202, 302 (1953).
m l l
Figure 5. Stationary mercury electrode according to Barker. a) Needle
valve; b) Neck (0.05 mm. 0 ) ; c) Capillary (1 mm. 0 ) .
(From: G. C.Barker, AERE Report C/R 1563 (1957).)
A 90" turn of the piston produces mercury droplets
with 0.02 cm2 of surface area. To avoid variations of
electrode size due to the effects of temperature (ther[341 R . G. Ball, D. L. Manning, and 0 .Menis, Analytic. Chem. 32,
621 (1961).
['I The Kemula electrode is manufactured by the Radiometer
Co., Copenhagen, Denmark. A similar model is available from
Metrohm A.G. Herisau, Switzerland.
mometer effect), both the mercury and the electrolytic
cell must be placed in a constant temperature bath.
I- I
c) Mercury-coated metal electrodes
(‘‘am a 1g a m ate d e 1 e c t r o de s ”)
Platinum wires of 0.1 to 0.3 mm. in diameter, fused into glass
are most commonly used. Gold and silver wire is also
employed. Rogers et al. 124,251 were the first to use such
electrodes for inverse voltammetry. The small amounts of
mercury on the electrode (20 to 100 pg), depending on the
amalgamation conditions and the electrode size) result in
very high amalgam concentrations during electrolysis and,
therefore, in high sensitivity. For best results, platinum
electrodes should be re-coated with mercury through electrolysis of a mercury-salt solution, prior to each determination
[12]. The cleaning and pre-treatment of the electrode exerts a
great influence upon the reproducibility of the results [35].
Figure 6. Hanging drop mercury electrode according to Kemula.
a) To reference electrode; b) Vent; c) Magnetic stirrer; d) Screw needle;
e) Capillary (0.018 cm. 0). (From: W.Kemula, 2.Kublik, and S. Giadowski, J. Electroanal. Chem. I , 92 (1959).)
Another variation of the stationary mercury electrode
[**I, in which a silicone-coated capillary (0.5 mm
diameter) is inserted from the bottom into the electrolytic cell is given in Fig. 7.
d) Pure metal electrodes
Up to now pure metal electrodes have been used only rarely
for the determination of metals. The deposition and stripping
of small amounts of metals cause greater difficulties with
solid electrodes than with mercury electrodes. NichoZson [37]
reports on the determination of nickel (10-8 M) with platinum
or gold electrodes, and on the deposition and stripping of Ag,
Pb and Cu (10-7 to 10-9 M) with a platinum electrode [38].
Also, Ag down to 10-10 g. has been determined with a platinum
electrode in 20 microliter microcells using the procedure
of method VII [23] (Table 2). Small amounts of mercury
have been determined after deposition on a platinum electrode by the peak currents developed through current reversal
(Method 11, Table 2) [391. The highest currents peaks were
obtained after electrolysis at -0.1 V from 0.1 M acids (HNOs,
HC104). The stripping potential lies at approximately +0.4 V
(Fig. 8).
Figure 7. Stationary mercury electrode with micrometer burette (right;
full view). a) Electrolysis cell; b) Capillary; c) Current connection;
d) di and dz stopcocks; e) Glass injection capillary of the microburette;
f) Connection to reference electrode. (From: R. Neeb; Z. anal. Chem.
180, 163 (1961).)
This aids in the stabilization of the drop during intensive
stirring of the solution [35]. A mercury drop is forced
out means of a microburette (piston burette with a
micrometer screw). Valve No. 1 is closed after the drop
is formed. The capillary now contains only a minute
amount of mercury, rendering the entire system less
susceptibleto temperature changes. A reservoir, attached
to the side, permits the refilling of the electrode system.
Better control over the quantities of mercury to be
delivered is obtained if the glass piston of the microburette is replaced with a steel needle [36] which can be
moved by means of the micrometer screw. The measurement of small quantities can thus be improved.
Kalvoda [18,19] describes a larger stationary mercury
electrode, applicable to inverse oscillopolarography.
Approximately 0.03 ml of mercury are contained in a
cup-shaped enlargement of a silicone-coated, bent glass
tube of 5 111111. diameter. Additional stirring is possible
through vibration of the mercury [20].
[**I GlastechnischeWerkstPtte G.m.b.H., Rheinallee 28, Mainz,
[351 R. Neeb, Z. analyt. Chem. 180, 161 (1961).
[36] P. F. Scholunder, Science (Washington) 95, 177 (1942).
18 - 1
+OLO v
Figure 8. Mercury peak currents on a platinum electrode, 0.50 mg
mercury/ml;0.1 N HC104 (10 ml); electrolysis period 5 min; Elcctrolysis voltage - 0.1 V.
Interference by dissolved organic materials is considerable.
If the water and acids are stored in polyethylene containers,
the generation of peak currents is completely prevented.
The extreme purity of the solutions required in these methods
limits their range of applicability in practical analytical
work. Solid metal electrodes are more readily usable when
the phenomena occurring during the anodic pre-electrolysis
(C, Table 1) are utilized for preconcentration. In this manner,
small amounts of iodide, for example, may be determined by
means of a silver electrode [40]. The AgI, formed at +0.18 V
(vs. calomel) during electrolysis, is determined either by
cathodic stripping with constant potential (method 111,
Table 2, see also Fig. 9) or by voltammetric recording of the
stripping currents (Fig. 10). In the first case, the current is
automatically integrated by means of an analog computer
integrator circuit.
[37] M . M. Nicholson, Analytic. Chem. 32, 1058 (1960).
[38] M. M. Nicholson, J. Amer. chem. SOC. 79, 7 (1957).
[39] R. Neeb, unpublished.
1401 J. Shuin and S. P. Perone, Analytic. Chem. 33, 325 (1961).
Angew. Chem: internat. Edit. 1 Vol. I (1962) / NO. 4
In general, one should determine the geometry 01 the stirrer,
the electrode position and the volume of the solution which
result in accurate and reproducibleconditions for electrolysis.
t [SET]
Figure 9. Determination of iodide with a silver electrode: Stripping at
constant potential; Stripping potential: --0.40 V (YS. total calomel
electrode). Iodide concentration and electrolysis time:
a) Blank value; h) 4 x 10-6 M, 10 min.; c) 4 x 10-5 M, 5 min.;
d) 4x 10-4 M, 2 min. (From: I . Shnin and S. P . Perone, Anal. Chem. 33,
325 (1961).)
Figure 10. Determination of iodide with a silver electrode: Voltammetry
with linearly variable potential. Iodide concentration and electrolysis
time: a) 4 x 10-8 M, 30 min.; h) Blank value; c) 4 x 10-6 M, 10 min.
(From: I. Shah and S. P. Perone, Anal. Chem. 33, 325 (1961).)
m. Determination
1) Electrolysis and Base Solution
The accuracy attainable depends upon the reproducibility of the electrolytic concentration. The invariant
positioning of the electrode and of the stirring mechanism must therefore be carefully considered in the
design and construction of the electrolytic cell. For all
sensitive determinations stirring of the solution is
necessary in order to increase the amount of material
deposited during electrolysis. This may be accomplished
by means of magnetic stirrers (Fig. 6), by a stirrer
introduced from the top [12] or by circulation of the
solution [21]. Bubbling of gas [37], or the use of the
hydrogen evolved in the electrolysis when stronger
negative potentials are encountered [181, vibration [20],
as well as rotation of the electrodes [23,24] are also
occasionally suggested for stirring the electrolyte.
The extent of the influence of small variations in the mechanical conditions of the electrolysis (i. e. stirrer-electrode
distance, r.p.m. of the stirrer etc.) on the amount of material
deposited during electrolysis must be determined in all cases.
If possible, critical conditions should be avoided. It is
imperative that the volume of the solution remain constant.
1 Vol. I
[rn] f [min]
Figure 1 1 . The dependence of stripping currents on time of electrolysis
114 ng Tl/ml; 0.02 M “Titriplex” solution. Volume: 1 ml.
Electrode: amalgamated platinum wire ( 0 . 3 5~ mm.).
Electrolysis voltage: -1.1 V.
Angew. Chem. internal. Edit.
The duration of electrolysis depends on the type and
shape of the electrode and the volume of the solution, as
well as on the concentration of the ions to be determined.
With a very short electrolysis period and especially with
an unstirred solution, the influence of the given conditions on the final result is noticeable. It is then important that the timing of the electrolysis and the
stripping cycles be maintained accurately. With small
electrodes, larger volumes of aqueous solution and short
electrolysis times, the stripping currents are almost
linear with the electrolysis period (depending on the
method selected - see Table 2). An increase in the
electrolysis period contributes to only a small increase
of the stripping currents [8,12] (see also Fig. 11). The
concentration of the element available on the electrode
for stripping (e.g. concentration of the amalgam) does
not increase noticeably because of the diffusion of the
metal into the mercury (in the case of large mercury
electrodes), or because of the depletion of the solution
(which is especially rapid when small volumes are used).
(1962)1 No. 4
The electrolysis periods vary from a few minutes to
several hours. Thirty minutes appears to be an optimum
period for high sensitivity with a stationary mercury
electrode (5 mm.2 of surface area, 10 ml of electrolyte
volume). In general, only a small portion of the element,
present in aqueous solution, is deposited during electrolysis. Only 0.25 % of the lead available in 25 ml of
solution was deposited on a stationary mercury electrode
(4.5 mm.2 of surface area) during a five-minute electrolysis period with stirring [8]. However, approximately
25 % of the lead is deposited on an amalgamated platinum electrode (4 to 5 mm.2) during thirty minutes with
a 1 ml cell [12]. Naturally, the solution is depleted more
rapidly when small sample-volumes are used. In contrast to polarography, therefore, the use of microcells
(0.01 to 0.05 ml) does not reduce the absolute amount
of determinable material.
The voltage used for electrolysis is normally 0.2 to 0.3 V
more negative than the polarographic half-wave potential. Even then (depending on the electrode, element
and the solution), maximum deposition rate may
occasionally not be attained, since in highly dilute
20 1
solutions this will occur only at a considerable overvoltage [41]. A thorough investigation of these conditions is suggested in each particular instance.
Figure 12 shows the slopes of the anodic voltammetric
peak currents (on an amalgamated platinum electrode)
for Pb, T1 and Bi in the electrolysis of a “Titriplex”
solution [12].
% of maximumpeak hight
,,,,- ---
.;; ,!:
‘., ’......Pb
Figure 12. The dependence of the stripping currents on the electrolysis
voltage. 0.10 mg Bi, TI and Pb in 0.02 M “Titriplex” solution.
(From: R. Neeb, Z. anal. Chem. 17, 326 (1959).)
According to this graph, the determination of thallium
by electrolysis at -1.0 V is possible even in the presence
of high concentrations of Pb and Bi, although one
would not expect this from the half-wave potentials
(Pb: -1.1 V, Bi: -0.70 V), at least in the case of Bi. (In
many cases, the anodic peak currents drop off again
when higher voltages are used for electrolysis.) Considerable separation is attained by selection of an
appropriate supporting solution, as in the case of polarography.
In addition to the variable capability for electrolytic deposition, the “stripping” action caused by different potentials also
permits a sharp separation of individual elements. This will
be discussed later.
Another example of the importance of the supporting solution
is given in the determination of lead in the presence of tin
[lo]: While both elements are deposited from a strong HCl
solution (1.5 N HCI) and even produce coinciding voltammet
ric stripping peaks, tetravalent tin is not deposited at all from
a weakly acid solution (PH = 2 to 3), thus making possible
the determination of lead in the presence of large quantities
of tin.
The stripping peaks of TI and Cd (half-wave potentials: -0.51
and -0.58 V, respectively) in 0.1 N KNO3 and in 1: 1 mixture
are barely separable [8]. In 0.05 N NH4CI/NH40H solution,
however, the two peaks can be separated, even in mixtures of
1 : 100. This is because the half-wave potentials, and therefore the peak potentials (see below), are at -0.51 and -0.77 V
and are consequently, well resolved.
The stripping methods, while highly sensitive, also are
subject to a number of possible errors. These are:
1) the contamination of the supporting solution, and
2) the limited shelf-life of dilute solutions.
Purification of the supporting solution (for example, by
extraction) is possible only when no material detrimental to
the analytical procedure is thereby introduced. Among the
impurities, which must be avoided, are not only. metallic
contaminants, but also, and primarily, organic materials
which may interfere by production of complexes, poisoning
of the electrodes, etc. It is therefore important to guard
against the presence of impurities with the same amount of
care that is applied to ascertaining the absence of the ion to
be determined in the solution blank. As an example, NaCl
solution extracted with dithizone cannot be utilized as a
t411 L. B. Rogers et al., J. electrochem. SOC.98, 447, 452, 457
working solution for the voltammetric determination of lead
[12] even if the dithizone and solvents are of the highest
purity [42]. The NaCl is re-usable only after evaporation and
combustion (methods which, in turn, expose the material to
contamination). Surface active compounds, even though
they may be intrinsically chemically inert, do interfere
particularly with AC methods (Method VI, Table 2), as well
as in work involving pure metal electrodes (see above).
Methods for the purification of the working solutions i.e.
in order to remove traces of surface active compounds by
means of activated charcoal, silica gel etc., are discussed in
reference [21]. In many cases the removal of metallic impurities should be possible by means of their electrolysis on a
mercury cathode.
The limited stability of highly dilute solutions necessitates the
fresh preparation of standard solutions from concentrated
master solutions. Quartz is the only material suitable for
vessel construction in high sensitivity work. Slightly acid
solutions of lead and cadmium (pH = 3) with metal concentrations of a few ng/ml.can be stored without deterioratjon
in quartz flasks for at least several days [43]. The use of
polyethylene containers in connection with highly sensitive
polarographic or voltammetric analytical methods is not
recommended. The organic materials which are almost always
released in noticeable quantities, often produce uncontrollable
interferences. Other, “purer”, plastics may be more applicable. If glass storage vessels are to be used they should be
filled with a portion of the solution to be stored and the
contents shaken for an extended period. When adsorption
equilibrium between the solution and the container wall is
reached, the vessel is refilled with fresh solution which is then
said to remain stable [9].
2) Solution process
Redissolution (“stripping”) of the deposited materials
follows the electrolytic concentration. Because of the
relatively high concentrations encountered, interferences, specific to inverse polarography and voltammetry,
are observed (see Sect. V). All known polarographic
and voltammetric as well as the microcoulometric and
potentiostatic methods are, in principle, applicable for
use with the stripping process. Voltammetric methods
(Method 11, Table 2, also called “linear sweep volt10v
Figure 13. Voltammetric curves for various metals. Hanging drop
mercury electrode. Electrolysis period: 3 min, 1.5 N HCI; dE/dt =
0.1 V/min; Concentration(mol/liter).
5.10-7 510-7
(From: W. Kemula, Z. Kublik, and S. Glodowski, .
I , 92 (1959).)
[42] G. Zwantscheff: Das Dithizon und seine Anwendungen in der
Mikro- und Spurenanalyse. Verlag Chemie GmbH., Weinheiml
Bergstr. 1958.
[431 R . Neeb, unpublished results.
Angew. Chem. internal. Edit.
Vol. I (1962)1 No. 4
ammetry") are, however, most commonly used. In the
latter case, application of a linearly variable voltage
generates peak currents sirniliar to those found in
cathode-ray polarography [2,44]. The peak potentials,
in this case, are comparable to the polarographic halfwave potentials [8]. When more than one metal is
present in the amalgam, several peaks will be generated
during the anodic stripping, providing that their potential difference is at least 0.1 V. The resolution of current
peaks is normally better than that encountered with
diffusion currents in the case of polarography (Fig. 13).
This resolution is a function of the rate of voltage
change, v(V/sec), and of the type of electrode used and
improves with smaller v, as well as with the use of
amalgamated platinurn electrodes instead of stationary
mercury electrodes [35,601.
The dependence on v(V/sec.), given by the Randles-Sevfik
equation (see above) for voltammetric methods, also
holds for inverse voltammetry in the case of stationary
mercury electrodes [8,35]. Amalgamated platinum electrodes, on the other hand, produce a more rapid increase
in peak currents with increasing (Fig.14). The circumstances regarding in the redissolution of metals deposited
on solid electrodes are apparently similar [37,39].
b r a 1
determination of Pb or T1 with amalgamated platinum
electrodes, positive temperature coefficients of 1.35 % were
determined for the over-all process over a temperature range
from20 "C to 40 "C[12]. In a microcoulometricdetermination
of Ni, a temperature increase of 25 O C leads to a threefold
increase in the current.
The accuracy attained in inverse polarography and
voltammetry will, in general, be lower than that obtained by the usual polarographic methods. The electrolytic preconcentration will always result in an additional
error. Standard deviations of 3 % to 5 % can, however,
be reached if the electrolysisis carried out carefully [35].
For the determination of very small concentrations,
this reduction in accuracy is relatively unimportant,
particularly because all the methods discussed here are
far superior to any others when applied to trace analysis. As indicated in Table 3, the analysis of 10-6 to 10-9 M
solutions is possible. This indicates a limit of detection
lower than 10-9 g., since 1 ml. samples are sufficient for
the determination. The sensitivity of these methods is
not normally limited by instrumentation, but rather by
the base current of the supporting solution which masks
the very small stripping currents. The application of
stationary electrodes now permits the application of
differential methods. These were proposed earlier for
polarographic analytical methods [45] but have rarely
been used because of the hard-to-overcome inherent
irregularities of two-drop electrodes.
A schematic diagram for a differential arrangement is
shown in Fig. 15. One of the cells contains the solution to be
analyzed, while the other contains only the supporting
solution. The electrodes in both cells are of identical size.
Both reference electrodes must have identical potentials and
equal internal resistance.
0 0.10 1120 0.30 0.C0 0.50 0.60 0.70 0.80
Figure 14. Dependence of the voltammetric peak currents on the rate
of voltage change (dE/dt). 0.085 mg Pb/ml; 0.2 M NaCl (pH = 3 to 4).
Volume: 10 ml; electrolysis period: 10 min. Electrolysis voltage -1.1V;
Electrodes: a) Stationary mercury electrode; b) Amalgamated platinum
electrode. (From: R. Neeb, 2.anal. Chem. 180, 166 (196l).)
When it is necessary t o determine a noble constituent,
in the overwhelming presence of a less noble constituent,
under conditions of unfavorableconcentration relations,
the less noble constituents may be stripped out from the
common deposit by the application of a constant
potential. The determination of the more noble constituent can then be carried out undisturbed [lo].
Figure 15. Schematic diagram of a differential arrangement. Z1 and
Z2 = cells; R1 and RZ= internal cell resistance; PI and Pz = potentiometers (G = galvanometer).
Small differences between the cells may be equalized by-means
of the potentiometers PI and P2.An additional advantage of
the carefully equalized differential system may be found in the
cancellation of currents generated by impurities in the
solution material.
3) Calibration curves
Assuming constant and reproducible electrolysis conditions, the various methods of determination shown
above yield calibration curves similar to those obtained
in the direct analysis of aqueous solutions without
electrolytic preconcentrafion, but with higher concentrations. The majority of the methods result in linear
calibration curves, so that extrapolation is possible.
The temperature dependence of the electrolysis as well as of
the stripping process require close temperature control of the
electrolysis cell for accurate analysis. In the voltammetric
[44] J. E. 3.Randles, Analyst 72, 301 (1947); G. F. Reynolds and
H. M . Davis, ibid. 78, 314 (1953).
Angew. Chem. internat. Edit. / Vol. I (1962)/ No. 4
Figure 16. Operation of the differential circuit. 0.02 mg Cd/ml; 0.1 N
(slightly acid) NaCl solution. Electrodes: amalgamated platinum;
a) Without differential circuit; b) With differential circuit.
[45] G.Semerano and L.Riccoboni, Gazz.chim.ital.72,297 (1942).
The demand for high purity reagents is therefore not too
severe. Figure 16 shows the method of operation of the
differential circuit. The initial steep increase of the base
current is compensated by the reference cell. Small peak
currents can now be more readily recognized and evaluated.
IV. Analytical Applications
Table 3 contains the presently known analytical applications of inverse polarography and voltammetry. At
least 15 metals and several anions can be detected, but
the full scope of these procedures is not yet exhausted.
Table 3. Analytical Applications of Inverse Polarography and
18, 10-13,
46, 471
(0.1 M KzC03) and analyzed for heavy metals (down
to 10-5 or 10-6 % of Cd, Pb. T1, Bi and Cu). Thallium
can be determined in concentrations down to 10-6 % in
salts and in potassium minerals when dissolved and
added to a “Titriplex” solution [12]. For the analysis
of T1 in organic materials [48], a wet oxidation is
carried out followed by a bromide extraction. The excess ether is destroyed (H2S04-H202) and the thallium
is determined by means of voltammetry or chronopotentiometry after dissolving in the supporting solutions [17]. Concentrations of 10-6 to 10-8 % of T1 have
been thus determined and the determination of the
natural T1 content of various biological materials [49] is
In the analysis of potassium minerals [50] with high T1
content (10-5 %) the samples needed for the analysis were so
small, because of the sensitivity of the method, that true
averages were no ionger obtained. Table 4 indicates this for
sylvinite. Small amounts of the sample (50 mg.), broken from
the sample, as well as some larger powdered samples were
analyzed. The variations lie far outside experimental error;
they can only be explained by lack of homogeneity in the
samples. The difficulty, if not impossibility, of obtaining a
true average sample for microanalysis, is indicated by this
Table 4. Voltammetric determination of TI in sylvinitc
[8-10, 12, 131
[lo, 12, 131
66.5 mg
67.9 mg
59.9 mg
57.7 mg
61.1 rng
1. Q10-5
0.8 lo-,
0.94 10-S
Pb, Cu and Zn have also been determined in biological
materials [20] by (inverse) oscillopolarography. Furthermore,
work has been performed on the determination of Pb in
blood [HI. Triphenyltin acetate, important as a fungicide,
is, after extractive separation from the plant material 1521 and
mineralization of the extract, determined voltammetrically at a
stationary electrode; 0.2yg Sn/25ml. can be determined [53].
Cathodic stripping with a mercury electrode has been
proposed for the determination of halogens, particularly
chloride in uranium sulfate solutions [34]. After addition of citric acid (in order to form the uranium-citrate
complex), as little as 0.2 pg C1-/ml. can be determined.
V. Interferences
11,111, VII
134, 401
[a] See also Table 2.
For the analysis of highest purity zinc [lo], the material
is dissolved in 10 N HC1 and after dilution to approximately 1.5 N HCI, it is electrolyzed directly on a stationary mercury electrode (hanging mercury drop
electrode) at 0.8 V. Pb, Cd, Sn and In may be determined
in concentrations down to 10-6 %. Similarly, uranium
salt can be simply added to the supporting solution
[46] E. N . Vinogradova and G . V . Prokhorova, SavodskajaLaboratorija 26, 41 (1960).
[47] A. G. Stromberg and V . E. Gorodovykfz, Savodskala Laboratorija 26, 46 (1960).
Some of the possibilities for interference should be
mentioned. These are caused by the high concentrations
of elements to be determined after electrolytic concentration on the working electrode and have been
I481 H. Eschnauer and R. Neeb, 2.Lebensmittel-Unters. und -Forsch. 112, 275 (1960).
1491 W. Geilmann, K . Beyermann, K . H. Neeb, and R . Neeb, Biochem. 2.333,62 (1960).
[50] R . Neeb, unpublished results.
[51] 0. Rottovri-Kloubkovci, and R . Kalvoda, Pracovni Lekarstvi
12, 20 (1960) (Analyt. Abstr. 7,5349 (1960)).
[52] S. Gorbach and R. Bock, Z. analyt. Chem. 163,429 (1958).
[53]P. Nangniof and P . H. Martens, Analyt. chem. Acta 24, 216
Angew. Chem. internat. Edit.
Val. 1 (1962)
I NO.4
thoroughly investigated, particularly for mercury as the
electrode material. They normally reduce the stripping
currents, and may even inhibit the stripping action
completely. The three major causes for interference are:
1) the formation of intermetallic compounds
2) the change of diffusion conditions in amalgams of
high concentration, and
3) exchange reactions on the amalgam surface.
After electrolytic deposition on mercury, metals may
form intermetallic compounds either with the mercury
itself or especially with other deposited materials.
Only the formation of such compounds with mercury
itself, and in the case of, Co, Ni, and Fe [54], has been
described so far. Anode current peaks appear (up to
four from cobalt) upon voltammetric stripping of the
amalgams of the these elements. Sometimes these peaks
are more positive than would be expected from the
standard potentials of the metals. They are assumed to
be caused by the formation of intermetallic compounds
with mercury. Figure 17 shows these phenomena in
connection with nickel which produces a well-defined
current peak at t-0.22 V, particularly after the amalgam
has been aged.
pared with pure metal amalgams, but also by the
generation of new oxidation currents at other, usually
more positive positions. These are due to the redissolution of the intermetallic compounds.
Figure 18. Voltammetric curves from the dissolution of mixed Ni-Zn
amalgams. Hanging drop mercury electrode; 2 min electrolysis period
at -1.4 V; supporting solution 0.1 M KCI.
Curve 1
5.10-4 m
5.10-4 m
5.10-4 m
5.10-4 m
2.10-4 m
4.10-4 m
6.10-4 m
4.10-4 m
From: W. Kemula, 2. Galus. and 2. Kublik, Bull. Acad. Polon. Sci., Cl.
111 6, 661 (1958).
Figure 18 indicates the conditions during the determination
of zinc in the presence of nickel [57]. A reduction in the
peak height for zinc due to the presence of nickel becomes
evident, and at the same time a new current peak appears at
0.1 V, which is due to the oxidation of the NiZn compound.
11111 I
Figure 17. Voltammetric curves obtained from dissolution of nickelamalgams. 0.1 N KCI = 10-4 N NiClz, hanging-dropmercury electrode.
Electrolysis period: 1) 1 min.; 2) 2 min.; 3) 4 min.; 4) 6 min.; 5) 8 min.
From: W. Kernulu and Z . GalusBull. Acad.Polon. Sci., ClIII 7,732(1959).
This phenomenon may be useful in the determination of
the element even in 10-7 M solutions [54,55]. The
stripping of amalgams which are difficult to oxidize, is
apparently noticeably influenced by the composition of
the supporting solution. In KCNS solutions, nickel, for
example, is completely oxidized before mercury [56].
The simultaneous deposition of several elements may
cause a reaction by producing intermetallic compounds.
The fact that they have been produced, is indicated not
only by a decrease of the oxidation currents as com[54] W. Kemula and 2.Galus, Bull. Acad. Polon. Sci., Cl I11 7,729
[55] W. Kemula, Z . Galus, and Z. Kublik, Nature (London) 182,
1228 (1958).
[56] J.T. Porter and W. D. Cooke, J. Amer. chem. SOC.77, 1481
Angew. Chem. internal. Edit.
Vol. I 119621 I NO.4
The following systems have been investigated voltammetrically by means of a hanging drop mercury electrode
Zn/Cu, Cu/Ni, Mn/Cu, Ni/Cr, Ni/Mn, Mn/Fe, Co/Ni,
As/&, Ni/Sn, Ni/Sb, Ag/Cd, Cd/Cu, and Au/Zn. The
interferences caused by the frequent formation of intermetallic compounds limit the applicability of these
methods. The interferences can, however, be often
eliminated by proper selectionof the supporting solutions
and electrode conditions, or, in certain cases, by prior
chemical separation. Another source of interference
through intermetallic compound formation occurs in
the use of F’t or Au wires for the suspension of the
mercury drops, because of the fact that both gold and
platinum are slightly soluble in mercury. It is therefore
possible for some of this material to migrate into the
mercury and form intermetallic compounds with the
deposited metals [59,61]. Thus, even as little as 0.001 %
gold in mercury prevents the oxidation of dilute zinc
amalgams because of the formation of ZnAu [61].
Smaller concentrations of gold are of no consequence.
[57] W. Kemula, Z . Galus, and Z . Kublik, Bull. Acad. Polon. Sci.,
C1. 111.6, 661 (1958).
[58] W. Kemula, in J. S. Longmuir: Advances in Polarography.
Pergamon Press, London 1960, vol. 1, p. 105.
[59] W. Kemula, Z . Kublik, and Z . Galus, Nature (London) 184,
56 (1959).
[60] W. Kemula, Z . Galus, and 2.Kublik, Bull. Acad. Polon. Sci.,
C1 I11 7, 723 (1959).
[61] W. Kemula, Z . Galus, and Z . Kublik, Bull. Acad. Polon Sci.,
C1. I11 7, 613 (1959).
Gold concentrations above 0.01 % [59] interfere in the
determination of Cd. The problems connected with
platinum are similiar [W].Mercury-coated platinum
electrodes are, in reality, dilute platinum amalgam
electrodes with a variable Pt content. Dissolved platinum
markedly interferes with the determination of antimony,
tin and zinc [a].Pb, Cd, Cu and TI do not, however,
form intermetallic compounds under these conditions
and can be determined without interference.
A short reference should be made to the importance of
voltammetric methods in the study of formation kinetics
and the composition of intermetallic compounds [58].
Variations in the diffusion conditions of elements of high
concentration in mercury have occasionally been observed
but have not so far been investigated in connection with the
possibility of interference. In particular, those elements which
react, with mercury such as Fe, Co and Ni (see above) could,
in high concentrations, cause interference with the diffusion
of even those elements which do not form intermetallic
compounds. In the oscillopolarographic determination of
copper, e.g. in the presence of much iron, substantially
smaller peaks are observed than in pure copper solutions [18].
After addition of a complexing agent to the supporting
solution to prevent the deposition of iron, the original peak
height was once more observed.
Reactions on the amalgam surface may well cause changes in
the surface concentration of the elements in the amalgam;
these, in turn, may cause a change in the stripping currents. In
inverse oscillopolarographic determinations of Cd and Zn in
uranium salts, peaks which are 20 % to 50 % lower than in
solutions of the tetravalent cation U4+ occur in U02*+
solutions. It is assumed that these are caused by the oxidation
of the amalgams by the uranium ion together with their
electrolytic oxidation [18]. When measuring the stripping
currents of mixed Zn-TI amalgams (as stationary electrodes)
considerably higher currents are obtained for the TI than for
the pure thallium-amalgam electrodes. This occurs because
the displacement reaction, 2 TI+ Zn -+2 TI Znz+, increases the surface concentration of the thallium, thus causing the
appearance of the higher diffusion currents [32].
Received, August, 10th. 1961
[A 187/24 IE]
A New Type of Converter for Ammonia Synthesis
A comparison of known types of converters used for the synthesis of ammonia, and an
evaluation of their respective advantages and disadvantages are given. A new type of converter is described, in which catalyst contact beds alternate with heat exchange regions.
This arrangement permits both close control of the operating temperature at or near opiimum conditions and utilization of the excess heat of reaction for preheating the reaction
For over 40 years, the production of ammonia has been
based on the Haber-Boschprocess developed in Germany
during World War I. Although the converters in use at
the present time are still based on the original principle
of the invention - combination of hydrogen and
nitrogen at high pressure in the presence of suitable
catalysts - the choice of a particular type of converter
depends on local conditions. In the last few years, a
fundamentally new converter has been developed at the
Gsterreichische Stickstoffwerke AG, Linz, Austria. It
has several advantages over other converters available at
present. Calculations made on electronic computers by
the engineering firm Haldor Topsse of Hellerup, Denmark, and by our own departments, have indicated a
high economic potential for this converters and patents
have been applied for in several industrial countries.
The most important types of converters known .today
may be classified as follows:
1) Continuous packed converters (Fig. 1).
2) Packed converters, subdivided into several sections:
the heat of formation of ammonia, liberated in each
packed stage, increases the gas temperature sharply;
the gas temperature is then reduced by direct injection
of a cold nitrogen-hydrogen mixture into the free spaces
between the packed stages (Quench converters, Fig. 2).
Fig. 1. Packed converter with a single continuous catalyst bed
a) Catalyst zone, b) Heat recovery section
Fig. 2. Packed converter with subdivided catalyst beds and injection of
cold gas into free spaces between catalyst zones (quench converter)
a) Catalyst zone, b) Heat recovery section, c) Cooling mils or tubes
Angew. Chem. internat. Edit.
1 VoI. 1
No. 4
Без категории
Размер файла
1 091 Кб
voltammetry, method, trace, analysis, new, inverse, polarography
Пожаловаться на содержимое документа