close

Вход

Забыли?

вход по аккаунту

?

Modern Methods for the Study of Electrode Processes.

код для вставкиСкачать
uE
@%I
Volume 13 Number 11
November 1974
Pages 683 - 750
International Edition in English
Modern Methods for the Study of Electrode Processes
By Joachim Heitbaum and Wolf Vielstich"]
The investigation of electrode processes calls for the use of methods that yield information
on the reaction mechanism and the reaction rate, on the nature and stability of intermediates,
and on adsorption processes at the electrode. It is no longer enough to record current-voltage
curves or charging curves. In addition to the potentiostatic scan method, other methods that
have proved useful for the investigation of electrode kinetics are the rotating disc electrode
in particular, and to a smaller extent AC polarography. Intermediafes can be identified with
rotating double electrodes and by ESR spectroscopy or mass spectrometry. Adsorption processes
on electrode surfaces can be studied by optical methods or by radioactive tracer methods.
1. Introduction
In the study of electrode processes, there are three phenomena
in particular that are of interest to the scientist. These are
the electrode kinetics (mechanism and rate of an electrochemical process), the occurrence of intermediates (free radicals,
radical-ions), and adsorption (of the starting substance and
of intermediates) on the electrode surface.
For experimental investigations, various methods are available, depending on the nature of the effect to be studied.
Apart from potentiostatic and galvanostatic pulse methods,
other methods that have proved useful for the study of electrode kinetics are primarily the potential scan method, the
rotating disc method, and modern polarographic methods
(AC polarography); intermediates of an electrochemical reaction can be detected and identified with the rotating double
electrode (ring-disc or ring-ring) and by optical methods (reflection spectroscopy, electroluminescence). Free-radical intermediates, particularly in organic reactions, can be studied by
ESR spectroscopy. Adsorption on the electrode surface or
in the electrolytic double layer can be investigated by optical
methods (absorption and reflection spectrosc0py)and by tracer
methods (use of radioactively labeled substances), as well as
by the conventional AC measurement.
[*] Dr J. Heitbaum a n d Prof. Dr. W. Vielstich
Institut fur Physikalische Chemie der Universitat
53 Bonn, Wegelerstrase 12 (Germany)
Angr-'. Chem. inrrrnat. Edir.
1 Vof.
13 ( 1 9 7 4 ) / No. I I
These methods and their possible combinations will be discussed below. Experimental details and results will be described only in so far as this seems necessary for an understanding of the measuring principle and of the scope of the methods.
2. The Potentiostatic Scan Method
2.1. Measuring Principle
The essential feature of the potentiostatic scan method is
a linear variation of the potential of the working electrode
between two reversal points (Fig. 1). For most investigations
the reversal points chosen should preferably be the potentials
Fig. 1. Variation of potential with time o n t h e measuring electrode in t h e
potentiostatic scan method; potential scanning rate between a few mV/s
a n d 1 Vis. cp = potential, r = time.
of anodic oxygen and cathodic hydrogen evolution, respectively. Electrode poisons are thus constantly removed by oxidation or reduction and the electrode is reproducibly activated.
683
Figure 2 shows the block circuit diagram for the application
of the potentiostatic scan method. To achieve a strictly linear
variation of the potential of the test electrode, the potential
difference between the working electrode and a reference elec-
tion, one obtains the current-potential diagram shown in
Figure 4, curve a. Since there are no electrochemically active
substances in the electrolyte, one observes the build-up and
breakdown of chemisorbed hydrogen and oxygen films, the
“surface-layer diagram”. The nature of the surface-layer diagram depends very strongly on the electrode material used,
and to a smaller extent on the electrolyte solution[3! The
surface-layer diagram for a smooth gold electrode in H2S0,
(Fig. 4, curve b) shows that the coverage of gold with hydrogen,
unlike that of platinum, is very low, and that coverage with
oxygen begins sharply only at + 1250mV. Investigations on
gold can therefore be carried out over a wide range of potentials
without interference from surface-layer currents. (For detailed
discussions of surface-layer diagrams see [ ’ 31.)
Fig. 2. Experimental device for the potentiostatic scan method. 2 =electrolysis
cell, V = test electrode, G =counter electrode, B = reference electrode,
D = triangular potential wave generator, P=potent~ostat, 0 =osclllograph,
W = resistance. L = Luggin capillary. U , ,= ci ,,,,m,,,ll, C’,= U,,,,
trode (Uactual)
must be constantly compared, by means of
an electronic potentiostat, with the voltage produced by a
triangular potential wave generator (Unomina,).
With a time
constant of T < 10-’s, the potentiostat equalizes the actual
voltage and the nominal voltage by passing a current through
the working electrode and the counterelectrode. This current
is measured as a function of the electrode potential. The
potential scanning rate of the triangular voltage generator
can be varied between a few mV/s and 1 V/s. An experimental
arrangement of this type was first used by Knorr and Will[’],
for investigation ofhydrogen and oxygen films on noble metals.
F i g 4. Surface-layer diagrams in 1 N H 2 S 0 , at room temperature. 100 niV s
N 2 flushing a ) on platinum. bl on gold. q=potential, j=current density.
If an electrochemically active substance is added to the electrolyte, the surface-layer currents and the current-potential
characteristics are superimposed. This is shown in Figure
5 for the anodic oxidation of NH3 in 1 M KOH on platinum.
The surface-layer currents can often be eliminated by a suitable
choice of the potential scanning rate. With slow potential
Fig. 3. Electrochemical cell for the application of the potential scan method
M =working electrode, G =counter electrode, 5 =hydrogen reference electrode, F,, FI, F, = sintered glass plates. L = Luggin capillary, T = thermometer,
W = water ducts t o thermostat jacket.
Figure 3 shows an example of a thermostatically controllable
electrolysis cell. The working electrode M can be flushed
with gas through a sintered glass plate F1.The counterelectrode and the hydrogen reference electrode B are separated
from the working-electrode electrolyte compartment by sintered glass separators F3 and F 2 respectively. (For further
details of the electrolysis cell see 12J.)
At smooth platinum in pure sulfuric acid, if the potential
ofthe working electrode is allowed to sweep the range between
the start of hydrogen evolution and the start of oxygen evolu684
I
0.2
0.4
0.6
0.8
1.2
1.L
1.6
Fig. 5. Current-potential diagram a ) on smooth platinum in 1 M K O H with
10 M NH,OH. 2 0 T , 100mV/s. N 2 flushing; for comparison b) surface-layer
diagram on smooth platinum in I M KOH. q=potential, ;=current density.
Angel?. Ckem. inttvuZ.
Ed]?. 1 Vol. 13 ( 1 9 7 4 ) 1 No. 1 1
a1
I
prcsent case. from thc inhibiting effcct of two kinds of chcniisol-bed oxygen.
2.2. Net Diagrams
-
Experimental separation of the surface-layer currents from
the current potential characteristic of an electrochemically
active substance is impossible if the part of the current due
to the ekctrochemicdl reaction of the active substance is low.
As was shown in Figure 5. the interpretation of the electrochemical spectruni is very difficult in such a case. since i t is
very hard to distinguish between the influence of the surface
layer formation and the electrochemical reaction. In cases
of this type it seems reasonable to determine the point-by-point
difference between the current potential curves obtained in
the presence and in the absence of the electrochemically active
substance.
II
!
0
1000
500
[mVI
'p
1500
'A
500
0
rn
'p
1000
CmVI
1500
Fig. 6. Current-potential diagram o n smooth platinum u i t h O-. flushing
Development of a n O-. diffusion limiting curl-ent with slow potential
s c a n n i n g ( 3 0 m V , s ) :top: electrode fluyhed with N-.: bottom. electrode flushed
with 02.h ) Fast scanning ( I V s): top: electrode flushed with N,: bottom
trode flushed with O-.. 'p= potential. j = c u r r e n t dcnsity.
21)
E
Y
scanning and a high rate of the electrode reaction, the surfacelayer current makes only a negligible contribution to the
total current. This is illustrated in Figure 6 a for the reduction
of oxygen on smooth platinum in H,SO,. A cathodic diffusion
limiting current develops. With fast potential scanning the
surface-layer current is no longer negligible (Figure 6b). The
build-upand breakdown of the hydrogen layer and the reduction of oxygen overlap. It can be seen from the diagram
that the two processes proceed independently of each
other''. 51.
In many cases the addition of an electrochemically active
substance to the electrolyte leads, not simply to a limiting
current as in Figure 6a, but to a series of anodic or cathodic
current maxima, i.e. to a n electrochemical spectrum. As an
example. Figure 7 shows thc current potential diagram of
ethanol in H zSOJon smooth platinum. There are three anodic
current maxima. Since the electrode was flushed with nitrogen
during the recording of this spectrum, the current maxima
are not due to a decrease of the electrochemically active
substance o n the electrode, but result from the inhibiting
adsorption of reactants and/or intermediates or, as in the
rn
cp C V I
-
Fig 8. a)Ciirrciit-potciiti~ildiagram oil smooth plalinurn in 1 %f K O l i + I f )
N2HI. 20 C. 1(H) m V F. N1 flushing (curve I ) . for comparison the \iii-kicclaycr diagram in I M K O H (cqrve 2). b) Net diagram on smooth platinuni
in 1 M KOH + 10 ' M N-.H,. the diagram was plotted from t h e point-h)point differcnces hetueen c l i n e i a n d curve Z in a ) . cp= potential. /=current
density.
XI
m
0
0.5
[p
-
1,o
CVI
1.5
Fig. 7 Curr-en-potential diagram of cthanol o n smooth platinum i n I N
H,SO,. 350 m V s. N-. flushing q = potential, / = c u r r e n t density
Anyow. C'hPm. inrrrnur. EdiI.
1 Val. 13 ( I 9 7 4 1 / N o . I 1
T o illustrate this, Figure 8a shows the current potential curve
on smooth platinum in pure KOH (curve2)andin KOH + lo-"
M hydrazine (curve 1). Subtraction of one curve from the
other gives the net diagram shown in Figure 8 b for the anodic
oxidation of hydrazine on platinum[']. After an initial elimination of hydrogen (the sharp current peak in the anodic sweep
685
is due to nascent hydrogen), a diffusion limiting current is
established at the electrode. A slight decrease in the current
is observed at potentials above 1 V, due to the inhibitory
effect of the adsorbed oxygen.
Fig. 10. Surface-layer diagram of a platinum-gold alloy (Pt :Au=40:60) in
1 N HISO&.100mV/s, N 2 flushing. cp=potential, j=current density.
obtained with rotation than without in such a case. This
is shown in Figure 1 1 for the oxidation of methanol in H 2 S 0 4
on smooth platinum.
Fig. 9 Net diagram for the superposition of the oxidation of ammonia
and of hydrazine on smooth platinum in 1 M K O H + 1 0 - 3 M N H 4 0 H
+ 1 0 - 4 M N2H1. 20°C 100mV/s, N 2 flushing; a ) mathematical summation
of the net curves for N 2 H 4 and N H 4 0 H ; b) experimental sum curve for
the simultaneous presence of both substances in the electrolyte. cp = potential,
j=current density.
This technique of deriving net diagrams is very suitable for
the investigation of the mutual effects of simultaneous electrode
reactions. This will be illustrated for the superimposed oxidation of ammonia and of hydrazine on smooth
The net diagrams for hydrazine and for ammonia are first
constructed separately by the procedure described above, and
the calculated sum curve is obtained. This calculated sum
curve corresponds to the case where the two electrochemical
processes take place together without inhibition. The experimental sum curve is obtained by recording the current potential diagram for the simultaneous oxidation of NH 3 and N2H4.
Figure 9 shows both the calculated and the experimental
sum curves. The close agreement between curves a and b
shows that hydrazine and ammonia are oxidized independently
of each other over the entire range of potentials.
2.3. Special Applications of the Potentiostatic Scan Method
Since the surface layer diagrams on different electrode metals
are very different from one another, the potentiostatic method
can be used for the identification ofelectrode metals. In particular, the composition of the electrode surface can-bedetermined
in the case of alloys[’1. This is particularly successful when
the potentials of the oxygen reduction peaks of the alloy
components are as far apart as possible. Figure 10 shows
the surface layer diagram ofa platinum gold alloy. The oxygen
chemisorbed on gold is clearly reduced more readily than
that chemisorbed on platinum.
By recording current potential diagrams on rotating electrodes,
it is very easy to detect whether intermediates involved in
the charge transfer reaction pass into solution. On a rotating
electrode the intermediates formed are transported more
rapidly into the bulk of the solution by the flow on the
electrode surface; a smaller current should therefore be
686
2?
I
I
rn
0
I
I
0.4
I
I
I
-
0.8
cpcv1
I
I
1.2
Fig. 1 I . Current-potential diagram on smooth platinum in 1 N H,SO, +
1 M C H I O H recorded a ) on a rotating electrode (75 Hz); b) without rotation;
100mV/s. q=potential; i=current.
The current maxima are distinctly higher without rotation
of the electrode (curve a) than on the rotating electrode (curve
b). The soluble intermediates are formaldehyde and formic
acid. The magnitude of the effect depends on the potential
scanning rate. A fast potential variation accelerates the electrode reaction so strongly that only small quantities of the
intermediates are transported away into the solution. In this
case there is only a slight difference between the diagrams
with and without rotation.
3. Rotating Electrodes
3.1. Measuring Principle
The problem in studies on electrode kinetics often lies in
the elimination of transport phenomena. L e ~ i c h ~in* ~1942
therefore suggested the use of a rotating disc as the measuring
electrode. In a practical device it consists of a metal disc
sunk into the center of the front area of a plastic cylinder
(Figure 12) which rotates in the electrolyte solution. As the
cylinder rotates, the solution is sucked along its axis of rotation
and thrown radially across the front area, as illustrated in
Angew. Chem. infernaf.Edit.
1 Vol. 13 ( 1 9 7 4 ) / No. I 1
I
1.6
Figure 13. This leads to the establishment of a constant Prandtl
flow boundary layer Spr over the entire front area of the
cylinder, and hence, according to the equation
l i t ' PM
Ill
to a concentration profile that is independent of the radius
of the electrode (F,= Nernst diffusion layer, D = diffusion coefficient, v = kinematic viscosity). Since
one obtains a uniform current density j over the disc electrode
(z=charge number, F = Faraday constant, co = concentration).
Pure mass transport leads to the establishment of a diffusion
limiting current jgr,whose magnitude depends on the speed
~~]
of rotation o of the electrode. According to L e ~ i c h and
Gregorj, and Riddifordl'o], the equation
4 stI0
I
I
N
N
I
I
I
6 364
r-
m
is valid in this case.
IT
'
rl
Fig. 14. Measuring apparatus with rotating electrode. in cross section. A .
B = current collectors for disc and ring electrode, respectively, D = revolution
counter, E =electrode, M =motor, St =steel plate, A1 =aluminum discs.
d,
I
II
/,,,///,r,///////,'
1
3.2. Possible Applications of the Rotating Disc Electrode
-r
3.2.1. Precision Determination of Diffusion Coefficients
Fig. 12. Flow on the front area of a rotating cylinder with inset disc electrode
(shaded); !.=axis of rotation, r=radius.
I f the rate of an electrochemical reaction is determined exclusively by mass transport (the discharge step is extremely fast)
a diffusion current is set up at the electrode and increases,
according to eq. ( I ) , in proportion to
According to eq.
(1 ), therefore, the diffusion coefficient D can be determined
if the concentration co is known, or the concentration c"
can be determined if the diffusion coefficient is known, from
a plot of j against
fi.
6.
Gostish-Michelci'c: Vielstich, and Heindrichd' 2] have used this
method for the precision determination of the trace diffusion
coefficient of the proton in KCI. This involves, not the diffusion
of the proton, but the rate of charge transfer in protonic
solution (hopping mechanism ricr hydrogen bonds). Figure
15 shows the measured diffusion limiting currents for several
rotation speeds between 31 and 183 Hz with a smooth platinum
disc electrode. The slope of the straight lines in the plot
of j g r against
gives the apparent diffusion coefficient of
the proton with an accuracy of *0.3%,. The value found
for 25°C was D1,.=74.6x 10-6k0.3%.The accuracy of the
method is therefore better by a factor of 5 to 10 than that
of the usual methods (polarography, diffusion after Cottrel).
1/0
I
Fig. 13. Flow lines on the front area of a rotating cylinder; a ) section,
b) elevation.
Figure 14 shows a section through the apparatus, with the
electrolysis cell[' '1. The motor connected to the cylinder allows
rotation speeds of between a few Hz and a few hundred
Hz. The rotation speed is recorded by means of a light beam
and a perforated disc, and is adjustable with an accuracy
of 0.1 Hz.
Angrw. Chrm. infrmat. Edit. J Vol. 13 ( I 9 7 4 )
1 No.
ll
3.2.2. Determination of the Characteristic Data of a ChargeTransfer Reaction
Before the current potential curve passes over into the limiting
current determined purely by mass transport, both the convective diffusion and the charge-transfer are rate-determining
for the reaction. Measurement of the current as a function
of the rotation speed o in this potential range yields informa687
Jahn and Viefs~ick~131
have tested this method on the system
Fe"/Fe"
in HCIOJ. The good fit of the experimental points
with the connecting straight fines in Figure 16 is a measure
oftheaccuracyofthemethod. Theyfounda valueof0.23 Alcm'
for the standard exchange current density of the system indicated; the transfer coefficient was 0.63.
[Hrl
183
I? 1
159
146
134
124
115
105
92
83
72
61
51
41
31
0,8
0.6
0.4
-
0.2
I
0
'p
.
-0.2
-0.4
3.2.3. Determination of the Rate Constants of a Preceding
Chemical Reaction
It was explained on the basis of eq. ( 1 ) that for pure mass
transport the diffusion limiting current on the rotating electrode increases in proportion to
However, if the electrochemical reaction is preceded by an inhibited reaction step
(adsorption on the electrode, chemical reaction), derivations
from linearity are found at high rotation speeds, as indicated
in Figure 17a.
fi.
i
L
-0.6
-0.8
[VI
Fig 15. Current-potential curves (diffusion limiting curt-cnts) o n ii rotating
plalinum clectrode as a function of thc speed of rotation in I M KCI +
3 x 10 "d HCI. v = p o t e n t i a l . !=current
tion about the rate of the electrode reaction if the influence
of mass transport is eliminated by extrapolation to o + x .
Jahri and Vielsrichl' 31 have derived the following equation
for this case:
121
where I,, is the pure charge-transfer current corresponding
to a given overpotential q. , j can
~ be determined from the
intercept on the axis in a plot of l / j against l / f i for a
fixed q (Fig. 16). A series of pairs of values ( j ~q,) then
yields the pure charge-transfer current potential curve, from
which the transfer coefficient a and the exchange current
density j o can be deduced.
6-
igr
-
Fig. 17. a ) Limiting currents on a rotating electrode plotted against I a:
curve 1. with purc mass transport: curve 2. deviation with kinetic inhibition.
b) /.,.'I! (o against /<,,: curve 3. with pure mass transport; curve 4. with kinetic
inhibition. ro=speed of rotation. 1". =reaction l i m i t ~ n gcurrent density.
~
A typical example is the evolution of hydrogen from a weak
acid HA, where the protons that react in the charge-transfer
reaction are formed by a preceding dissociation on the electrode surface:
The rate of the preceding reaction (i. e. the dissociation rate
of the weak acid in the above example) can be found from
the deviation of the current from the proportionality to /,&
that would exist without reaction inhibition.
According to Kourecki and Lerich["] and to Viefstich and
Jahn["], the following relation is valid in this case:
Fig. 16. Plot of I l l j l against I l k & for thc determination of the charge transfer
current density in ( t h e parameter is the overpotential q) for the system
I ( l - ' M Fe'. 10 ' W F e " ' In L M HCIO, on smooth platinum. oi=speed
o f rotation. j = c u r r e n t density.
688
where jgris the measured reaction limiting current density
and ,j9i,d,rfis the limiting current density due to the mass
against jp.,
transport. According to eq. (4), a plot of j&'&
as in Figure 17b, gives a horizontal straight line for pure
mass transport. In the case of reaction inhibition one obtains
a straight line with a negativegradient,from which the dissociation rate constant kd can be calculated according to eq. (4).
D, v, C ~ O and
,
the equilibrium constant k d / k , must naturally
be known from thermodynamic measurements or conductivity
measurements.
Gostish-Mihef?ic' and Vie/stich[' have used this method to
determine thedissociation constants of acetic and formic acids.
Figure 18 shows j q r f i as a function of jerfor the system
Aniirw Cham rntrrnaf. Edit.
J Vol. 13 11974)
No. 1 1
0.073
being so slow that the condition for eq. ( l ) , i. Y. the establishment of a steady state hydrodynamic flow, is not violated.
In this case
L.
'
For given wb ' and Am' and measured Ai. it is possible
and to comto calculate the pure transport component
pare it with the current i found without modulation.
0
m
0,s
1.0
igr
CmAl
1.5
-A
Fig. 18. ;,,,)l (I) I.\ l o , plotted Tor the system acetic acidisodium acetate;
2.73 x 10 ' M HAc + 3 x 10 M NaAc + 1 M KCi, measured on a rotating
platinum disc electrode. pH = 5.66. /,,,=reaction limiting current deiihity.
ru=speed of rotation.
'
acetic acid/sodium acetate from measurements on a smooth
platinum disc electrode. The value found for the dissociation
rate constant from the slope of the straight line is
k d = 5.4 x 10ss I . The diffusion coefficient of acetic acid,
D = 1.2 x 10scm2/s,can also be found from the intercept of
the straight line on the ordinate. A similar experiment gave
a dissociation rate constant kd = 4.9 x 10' s I for formic acid.
Owing to the higher rate of dissociation of formic acid, this
constant was obtainable only by the use of rotation speeds
of a few hundred Hz.
3.2.4. Separation of Kinetic and Transport-Controlled Reactions
If two electrochemical reactions proceed simultaneously in
the potential range studied, the current components can be
separated if one of the reactions is transport-controlled while
the rate of the other is kinetically controlled.
According to eq. ( I ) , to a change Awl in the speed of rotation
of a rotating electrode there corresponds a current signal
Ai, which is determined exclusively by the transport-controlled
component of the current. Miller["] has therefore suggested
the superposition of a periodic change Awl ' (e.y. sinusoidal
or rectangular) o n a fixed value 08 ', this periodic change
As an example, Figure 19 shows the superposition of a copper
deposition controlled by convective diffusion and the evolution
of hydrogen. the kinetics of which are rate-determining at gold
electrodes in particular. The splitting of the current signal
by Ai in the case of the copper deposition can be seen in
curve bmod.No Ai corresponding to the frequency modulation
is observed in the evolution of hydrogen only (curve amod.).
3.3. The Rotating Double Electrode and Its Applications
The possibilities of a rotating system can be considerably
extended if instead of a single electrode one uses several
concentric electrodes (disc-ring, ring-ring). A substance formed
on the inner electrode is then transported by the radial flow
to the outer electrode, where it may undergo a further electrochemical reaction. The narrower the electrodes and the thinner
the insulating layer between them, the greater is the sensitivity.
Thus with a double ring electrode having a diameter of IOmm,
an electrode width of 1 mm, and an interelectrode distance
of 0.1 mm, intermediates having a lifetime of l O - ' s can be
detected at 400Hz.
With the aid of this method, Humunn and Viri~richi"~were
able to show that formate is formed as an intermediate in
the alkaline electrolyte during the anodic oxidation of carbon
monoxide to carbonate. I t was not possible to show this
directly with a double electrode, since CO and formate react
in the same potential range on most electrodes and the oxidation of formate is blocked by carbon monoxide. The detection
of formate as a n intermediate in the oxidation of CO was
made possible only by the use of an outer electrode ring
of palladium, an eiectrocatalyst that is substantially selective
for formate.
It can be seen from the foregoing that the double electrode
is particularly suitable for the study of short-lived intermediates. However, a measuring arrangement of this type also
allows the detection of homogeneous chemical reactions placed
between two electrochemical steps. For example, if Fe3L is
reduced to Fez' on the inner electrode the Fe 2- can be
reoxidized on the outer electrode if the potential chosen is
correspondingly positive. The ratio of the reduction current
to the oxidation current is known as the collection efficiency
N . It gives the percentage of the substance formed on the
inner electrode that reaches the outer electrode. N depends
only on the electrode geometry selected" '1.
I
-08
rn
I
I
I
-04
0
I
I
08
04
'p [mVGKE
1
I
I
12
16
Fig. 19. Cathodic current potential diagram on gold at 26.hHz. potential
scanning rate 5 mVjs (after [16]). a ) in I M HISO,: am%,,,) a s a with superimposed sinusoidal frequency modulation; b) in I M H,SO, + 2M CuSO+:
b,, )as b with superimposed frequency modulation v = potcntlal, i=current.
A n g w . Chcm. inrrmat. Edit.
/ Vol. 13 ( 1974) N o . 1 1
If a n oxidizing agent, e.y. H z 0 2 , is added to the electrolyte,
part of the Fez' formed on the inner electrode is oxidized
by the H z 0 2 in a homogeneous chemical reaction and is
lost with regard to detection on the outer electrode, and
the collection efficiency accordingly decreases. The longer the
time spent by the Fez' between the two electrodes, the greater
is the decrease. One thus obtains a collection efficiency that
689
depends on the rotation speed. The rate constant of the homogeneous chemical reaction can be calculated from the variation
of N with o at a given H z O z on cent ration"^^.
a1
4. AC Polarography
In conventional AC polarography, which was developed 30
years ago, the impedance of an electrochemical cell is measured
under polarographic conditions, a low-amplitude ( < 1OmV)
sinusoidal alternating current being superimposed on the direct current, which increases linearly with time (saw-tooth).
As a function of the linearly increasing electrode potential
one obtains an AC signal that corresponds, to a first approximation, to the first derivative of the normal polarographic
wave (Fig. 20), i.e. the potential of the current maximum
agrees with the half-wave potential of the DC polarogram.
This is not, in fact, a simple derivative of the DC curve.
as is shown, P.Y., by the fact that the irregularities of the
polarogram (drop peaks, maxima) are not differentiated at
the same time.
Fig. 20. Schematic representation a ) of a polarographic D C step and b)
the corresponding AC peak. E = = D C potential. I = and I . = D C and AC
currents respectively: E , = half-wave potential. E , = potential at current maximum.
E,CVI
L
I
I
I
-0.550
- 0.650
E, C V I
+
Fig. 22. Second harmonic of an AC polarogram in 1 M NazSOl + 3 x 10.’ M
Cd” (after [24]), amplitude 5 mV, frequency 80 Hz, potential scanning rate
25 mVjrnin; a ) “conventional” harmonic of a n AC polarogram, b) harmonic
with phase-selective recording. E = D C potential. I = alternating current.
~
-
Theoretical considerations by Bauer er aI.1201and by Breyer
and Bauer[*l]have shown that the observed AC amplitude
and the corresponding real and imaginary impedance components depend on the rate constants and transfer coefficients
of the reaction steps. A n analysis of the AC polarogram should
therefore yield information about the kinetics of the electrode
reaction.
-0.650
-0550
E= C V I
A
Fig. 21 AC polarogram in I M N a z S 0 4 + 3 x I O - ’ M C d * + (after [24]),
AC amplitude 5 mV, frequency 320 Hz. Potential scanning rate 25 mV/min.
E = D C potential, I . =alternating current.
690
The accuracy of the measuring method is limited by the ohmic
resistance of the cell and by the capacitive resistance of the
double layer. The ohmic voltage drop can be largely eliminated
by the introduction of a reference electrode (cf. the potential
scan method), which is not usual in conventional polarography.
The capacitive impedance component, on the other hand,
can be reduced only by the use of a low AC frequency. In
any case, the capacitive impedance is greater in AC polarography than in DC polarography, and the sensitivity is accordingly
poorer in the AC method. Conventional AC polarography
Angrw.
C h m . infernat. Edit. / Vol. 13 ( 1 9 7 4 ) / No 1 I
has therefore never really gained acceptance for kinetic investigat ions.
Since the early 196O's, a new AC polarographic method has
been developed, in which instead of the fundamental curve
(Fig. 21). the second (or a higher) harmonic is measured with
the aid of a frequency
The capacitive current of
the charge reversal of the double layer is largely free from
harmonics, and the capacitive impedance is significantly lower
in this type of measurement than in conventional AC polarography.
The second harmonic of an AC polarogram generally has
two current maxima (Fig. 22a), which, with phase-selective
measurement, have quantitatively the appearance of the second
derivative of the DC polarogram (Fig. 22b). The principle
of the phase-selective detector is based on a vectorial comparison of the measured voltage with a control voltage of the
same frequency. The detector may be, r . 8.. a phase-dependent
re~tifier1'~Ithat passes only the component of the measured
voltage that is in phase with the control voltage. The real
and imaginary components of an alternating current can thus
be measured directly with such an arrangement, and can
then be used to find the impedance components. The phaseselective measurement therefore yields additional information
about the nature of the measured (ohmic or frequency-dependent) resistance components, so that a double-layer capacity
component that is still present in the second harmonic can also
be eliminated.
k
x &
x,
OX + ! l e e
-
R
K = k , / k z = z . ( - ~ -). K = I . I r , = l x l O - ' ~ ~ ' (-. -.): K=O.l.
A , = 1 x 10" s C i E = - E l
= difference between DC potential and halfwave, potential. I ( Z o ) r ) = alternating current.
(---):
~
doublelayerand ofthe formation and consumption ofreactants.
intermediates, and products on or in the vicinity of the electrode surface. In absorption spectroscopy, visible or U V light
is allowed to pass through a transparent electrode into the
electrolyte compartment, and the quantity of light absorbed
is determined. The light absorption is proportional to the
concentration of the absorbing substance in the electrolyte.
By varying the thickness of the layer of electrolyte, therefore,
one can determine the change in the concentration of the
absorbing substance as a function of distance from the electrode. Homogeneouschemical reactions that precede or follow
the electrode reaction and that proceed at some distance
from the electrode are thus also accessible to direct observation.
Conclusions regarding the electrode kinetics can be drawn
from the change in the absorption with time during an electrode reaction. In the simple case ofan electrochemical reaction
R+Ox + nee, resulting in the formation of the light-absorbing
species Ox which was not initially present in the rlectrolyk.
the absorption M can be calculated for the case of a diffusioninhibited reaction[2s':
In this equation, po is the molar absorption of the substance
Ox, x is the distance from the electrode, c$ is the initial
concentration of the substance R, and DR is the diffusion
constant of R. Eq. ( 5 ) states that in a diffusion-controlled
electrode reaction, the absorption increases in proportion to
However, if chemical reactions are coupled with the discharge step, one finds deviations from the fi-dependence
of eq. (5). The rate constant of the chemical reaction can
be determined from this deviation.
fi.
Srrojek and Kuwnna['" used this method in their studies
on the oxidation of o-toluidine in aqueous solutions. They
were able to show that the oxidation proceeds in accordance
with the scheme
R
s
K
m
-012
-006
E=-Ei/,
0
[VI
-
0 06
0 12
Fig. 23. Calculated second harmonic of a n AC polarogram (after 1241) foithe system
The shape of the first harmonic of an AC polarogram is
very strongly dependent on the kinetics of the charge transfer
and of the chemical reaction steps. Figure 23 shows this for
the case in which a preceding chemical reaction influences
the shape of the second h a r m ~ n i c [ ~ ~ l .
i~o
x
o\
LL
--->
S+ce
+ca
2s
t.iu an intermediate S and a subsequent synproportionation.
With the aid of absorption spectroscopy, Srrojrk and Kukvunu
were able to determine the rate constants of this subsequent
chemical reaction with an accuracy of about
5.2. Reflection Spectroscopy
Two fundamentally different methods are known in reflection
spectroscopy. In one case the light beam is allowed to reflect
several times (to increase the sensitivity) in the interior of
a plane-parallel ground transparent electrode, as illustrated
in Figure 24. If the angle of incidence of the light is greater
than the critical angle. the light is totally reflected. provided
5. Optical Methods
x
5.1. Absorption Spectroscopy
Optical methods are among the most promising in-sirm techniques (measurement on the electrode surface during the electrochemical reaction) for the study of the electrochemical
A I I ~ C WChrm.
.
intsrnaf. Edit. J Vol. 13 I19741 J No. 1 1
Fig. 24. Beam path
in
Electrode
Electrolyte
x
renectlon spectroscopy inside an electrode.
691
that no substances that influence the beam path are adsorbed
on the electrode surface. Theoretical considerations and practical experiments have shown that even with total reflection,
the reflected beam penetrates for a distance of a fraction
of a wavelength into the solution. The reflection coefficient
consequentlydependson the nature ofthe substances absorbed
on the electrode.
The method has already been repeatedly used, both with
and with IR lightr2’! IR light appears to be
particularly suitable for investigations on the surface state
of semiconductor electrodes. The conditions for the use of
the method in electrochemical studies are, on the one hand,
that apart from the substance under investigation there are
no other species (solvent, supportingelectrolyte) in the solution
that influence the beam path and, on the other hand, that
the absorption by the adsorbed molecules is sufficiently strong.
The latter condition restricts the method to investigations
on fairly large organic molecules; for smaller molecules the
sensitivity is insufficient for quantitative nieasurenients‘zyl.
x
\
Electrode
R
Electrolyte
/
x
Fig. 25. Beam path in reflection spectroscopy between two electrodes
The second method of reflection spectroscopy is used more
often. In this case the light beam is allowed to reflect several
times between two or more plane-parallel electrodes in the
interior of the solution, as indicated in Figure 25. Pfieth[301
recently described the theory and the possibilities of this
method.
Takarnura er ul.[31]
used a reflection cell of this type to study
the oxide formation and the adsorption of foreign metal ions
on gold electrodes. They used visible light and an electrode
arrangement that allowed 10-1 5 reflections. From the dependence of the reflection on the wavelength and on the electrode
potential, they were able to conclude that the change in the
reflectivity depends on effects of the electric field in the double
layer, on the concentration and the photoactivity of the
adsorbed substances, and on the structure of the metal surface.
The method thus seems very suitable for in-situ studies on
substances adsorbed on electrodes with visible and UV light
(IR light is strongly absorbed by water and other polar solvents,
and is therefore ruled out). If the wavelength dependence
ofthe adsorbed substance can be measured, information about
the nature of the specific adsorption is often obtainable (e.g.
d-orbital interactions).
5.3. Ellipsometry
For ellipsometry one uses only a single electrode, which is
irradiated with polarized light. From the change in the phase
and amplitude of the polarized light it is possible not only
to obtain general information about the adsorption on the
electrode surface but also, in particular, to determine the
thickness of a surface layer.
Figure 26 shows the principal device for ellipsometry. If one
starts, for example, with a plane, homogeneous metal surface
and electrochemically produces an oxide film on this electrode
by means of a suitable potential, the resulting change in the
692
refractive index of the surface layer very strongly influences
the polarization of the incident light wave. This polarization
change can be detected by adjusting the analyzer system ()1/4
plate and analyzer) to complete extinction (or minimum light
transmission) for the pure metal electrode and measuring
the quantity of light transmitted for the oxide-coated electrode
with the aid of a photomultiplier. (For further details see,
e. g., L.321.)
Fig. 26. Principal devlce of a n apparatus for elllpsomeiry. S = l i g h t source.
Q = L’4 plate, A=analyzer, D =detector.
W = electrode, E = electrolyte
C =collimator, P=polarizer.
The ellipsometric method has been used, e. g., for studies
of oxide formation on platinum[331and of the passivation
of nickel‘”] and
In the case of oxide formation on
platinum, the coating of the electrode with oxygen, with a
mean thickness of 0.2A, begins at 0.98 V against the normal
hydrogen electrode. The thickness of the oxide layer increases
linearly with the potential, and reaches a value of 6i4 at
about 1.6V, the potential at which the evolution of oxygen
begins. The mean layer thickness of 0.2A is interpreted as
being due to a partly formed monolayer.
Ellipsometric studies on the double layer have so far been
carried out in most cases at only one wavelength, and consequently yield only rather unspecific information about the
nature of the adsorbate layer. Measurements over a wide
range of wavelengths should provide information about the
interactions(natureofthe bonding) that occur in the formation
of monolayers on electrode surfaces.
All the optical methods described so far can be used to detect
the existence of an adsorbate and to measure its film thickness
and the kinetics of the adsorption, but do not allow the
identification of an unknown adsorbate.
5.4. Photoelectrochemistry and Electroluminescence
Photoelectrochemistry is based on the Becquerel effect, which
was discovered in 1839. Becquerel found that a current flows
between two unequally illuminated platinum electrodes in
an acidic electrolyte. The different levels of illumination of
the electrodes evidently lead to the development of a potential
difference. which causes the current to flow. It therefore seems
reasonable to assign a “photopotentia1” to an electrode, this
being defined simply as the potential of the illuminated electrode with respect to the potential of the unilluminated electrode:
(P*
= (PL
- (PD
cp* is the photopotential, (PL is the potential of the illuminated
electrode, and (pD is the potential of the unilluminated electrode. The change in the potential of the electrode may be
due to a series of primary and secondary effects.
Primary effects: 1 , Production of excited molecules by light
absorption; 2, photodissociationasa result of light absorption;
Angrw. Cht,m. internat. Edit.
/ Val. 13
(1974)
/ Nu. / I
3, absorption of light by the metal, non-metal, or semiconductor. with consequent photoionization, photoemission, or
change in the population of valence and conduction bands.
Secondary effects: 1. Fluorescence or phosphorescence; 2, photochemical reactions; 3, energy-transfer reactions; 4, other
non-rodia tivede-excitation modes (inelastic collisions);5, thermal effects (rise in temperature of the illuminated electrode).
This list shows that photoelectrochemistry allows the investigation of numerous effects by measurement of the photopotential or the photocurrent as a function of the exposure to
light. The method seems promising in particular for the study
of excited molecules and free radicals in organic electrode
reactions and for investigations on the oxide film on solid
electrodes (for discussion see 13'I).
In the measurements, the strength of the homogeneous magnetic field is varied linearly and the energy absorbed is measured.
This gives a n absorption curve as shown in Figure 27a, with
a maximum at the point of resonance. However, the first
derivative is recorded in most cases (Fig. 27b). Interaction
of the electron spins with the magnetic moments of the atomic
nuclei of the molecules (protons and nitrogen in the case
oforganic substances; "C and ''0 have no magnetic moment)
leads to a splitting of the resonance line (hyperfine structure).
The number of hyperfine structure lines can be considerable
in the case of organic molecules. As an example, Figure 28
shows the ESR spectrum of the nitrobenzene radical-anion,
with 54 hyperfine structure lines.
If a molecule or free radical in an excited state is formed
in an electrochemical process, it may pass to the ground
state with emission of light. This effect, which is known as
electroluminescence, is shown in particular by aromatic hydroc a r b o n ~ ' ~ ' In
! certain circumstances it allows the identification
of a n intermediate, and is thus an important addition to
the range of methods for the investigation of reaction
mechanisms.
6. ESR Spectroscopy
The measuring principle of electron spin resonance spectroscopy is based on the fact that free radicals d o not have
saturated electron spins and are therefore paramagnetic. In
a homogeneous magnetic field, the electron spin has two
possible states (parallel or antiparallel to the magnetic field),
whose energies are slightly different. The number of radicals
in the lower energy state is slightly greater than the number
in the higher state (Boltzmann distribution). If energy is supplied to the system in the form of electromagnetic radiation
(microwaves) whose energy is exactly equal to the energy
difference between the two states, radicals pass from the lower
energy level to the higher. This is opposed, however, by the
fact that radicals from the higher energy level relax to the
lower level as a result of interaction with the environment
(spin-lattice interaction). In the resonance case (injected energy=energy difference between the states) this leads to a continuous measurable absorption of the microwave radiation.
Fig. 28. ESR spectrum of the nitrobenzene radical-anion in aqueous alkali
solution (after 1371) T h e 54 lines observed correspond to the splitting d u e
to the "N nucleus. o n e proton ( p i i r o ) . a n d t w o groups of two equivalent
protons each f oriho a n d !iii,ru J
Analysis of the hyperfine structure splitting of the ESR spectrum with regard to the number, intensity, and spacing of
the lines yields information on the number and nature (H
or N ) of atomic nuclei that influence the electron spin. In
many cases the molecular structure of the radical can be
deduced. i. e. the radical can be identified.
The microwaves are injected by means of a cavity resonator
inside which the radicals either flow through a thin capillary
(external production of free radicals, Fig. 29a), or are directly
produced in a shallow cell (internal production of free radicals,
Fig. 29b). In the former case, not only is it possible to identify
-E
-E
v
I
Fig 27. i i ) Absorption .4 iis ii function of the field strength H. b) First
dcri\ative of the absorption curve d k d H ( H , = r e s o n a n c e field strength).
A
Fig. 29. Scheme of a n apparatus for ESR measurements (after [ 3 7 ] ) with
a ) external generation of free radicals, b) internal generation of free radicals.
A =working elec1rodc. B. C =connections to reference electrode and counterelectrode. D = niicrowabe resonator. E = hollow conductor coiinection.
ni;ignetic ficld normal to the plane of the drawing
693
the radical-ion, but its lifetime can also be determined by
variation of the distance from the point at which it is formed
to the measuring position. Moreover. if the flow capillary
is branched, a reagent solution may be added, and free-radical
reactions can thus also be studied (Fig. 30). The method
with internal production of the free radicals can be used
for the study of short-lived intermediates.
The radioactive tracer method has become a valuable aid
to the study of adsorption processes in organic electrochemistry. Thus Huckermunn et ~ 1 . 1 ” ~ studied the adsorption of
“C-labeled organic inhibitors in the corrosion of iron under
stationary conditions. Bockris ef ui. have published investigations on the adsorption of organic substances (aldehydes,
benzene, naphthalene, and n-decylamine) on platinum and
partly also on gold and nickel[40! They interpreted their results
by a model of competitive adsorption of organic substances
and of water.
8. Identification of Electrochemical Products by Mass
Spectrometry
rn
Fig. 30. Scheme of an apparatus with extcrnd production of free radicals
and mixing with a reagent solution before the ESR rneasuremcnt (after
[ 3 7 ] ) . A to E as in Fig. 29. F=storage vessel for reagent d u t i o n . G =mixing
chamber.
As far as electrochemical investigations are concerned, the
most important results of ESR measurements are the proof
of the presence of free radicals and the determination of their
molecular structure. I t is also possible to determine absolute
or relative concentrations; this is important for kinetics.
Further information can be obtained from the hyperfine structure splitting (energies of molecular orbitals) and from the
line width (solvation processes, ion-pair formation). ESR spectroscopy thus provides numerous data for electrochemical
investigations (for examples see ‘”1).
7. Radioactive Tracer Methods
In addition to the foregoing methods for the investigation
of the electrolytic double layer and the specific adsorption of
ions? intermediates, etc. on the electrode surface. radioactively labeled substances have frequently also been used for
this purpose in recent years. However, the radioactive tracer
method did not attain any importance until radioactive isotopes of practically all elements became available e.g as products from nuclear reactors.
The following is an example of a procedure that can be used
for the study of the adsorption of organic molecules on metal
electrodes. The electrode is immersed in an electrolyte containing the radioactively labeled form of the molecules to be
investigated. The adsorption begins immediately after the
immersion of the electrode, and an equilibrium is reached
after a time. The time taken to reach equilibrium depends
e.9. on the nature of the adsorbate and of the adsorbent,
the temperature the electrolyte concentration, the surface condition of the metal electrode, and its potential. By varying
one parameter and keeping the others constant, one can obtain
information about the nature and the kinetics of the adsorpti~nf~~’.
694
Any investigation of the kinetics of an electrode reaction
must necessarily be incomplete until the intermediates and
products have been definitely identified. The measuring
methods described in the foregoing sections allow this identification only in exceptional cases, when the substances under
investigation are present in the dissolved state (double-ring
electrodes, electroluminescence, ESR spectroscopy). However.
the reaction products are often adsorbed on the electrode.
Their existence can be detected by optical methods, but identification has not been possible in the past.
Attempts have recently been made, therefore, to analyze
adsorbed products of an electrode reaction in the mass spectrometer. B r u ~ k e n s r e i n used
~ ~ ’ ~a microporous non-wetting electrode which is fitted as a cell wall with its hydrophobic side
directly on the inlet part of a mass spectrometer (Fig. 31).
Low molecular weight, readily volatizable reaction products
are extracted from the electrode by the vacuum on the mass
spectrometer side, and after ionization by electron impact
they are subjected to mass analysis. Since the electrochemical
cell is connected directly to the mass spectrometer, the product
distribution can be measured as a function of the electrode
potential, and the m/e signal can be measured as a function
of time at a constant potential. Detailed information about
the dependence of the reaction mechanism on the potential
and about the strength of the adsorptive bond can be obtained
in this way.
<[I>
llp3311
Fig 31. Schematic representation of an apparatus for the direct analysis
of clcctrochemical reaction products with thc mass spectrometcr (after 1411).
B =reference electrode. E =electronics, G =counterelectrode. M = manometer, MS=mass spectrometer. P= porous electrode. R =reservoir. S = pen
irecorder.
Another mass-spectrometric method for the identification of
adsorbed reaction products has been proposed by the present
authorsI4'! The electrode in this case is a metal foil 7 pm
thick, which is etched at its lower edge. After the electrochemical reaction. this foil is placed in a mass spectrometer; the
adsorbate is ionized by application of a very high electric
field strength and detected in the mass analyzer. The field
desorption method developed by Brckey (I[ crl.1""
has the
special advantage that the molecular peak is normally the
base peak. Since the starting substance and the reaction product are present together on the electrode. there is consequently
only a slight overlap of the mass spectra, so that a definite
identification of the desired substance is generally possible.
T o demonstrate the applicability of the suggested method,
the reaction product formed on anodic oxidation of I-butanol
in alkaline electrolyte at 700 mVRkjl.was investigated.
Modern electrochemical methods nowadays make it possible
to obtain accurate information about the kinetics of an electrode process. The rate-determining step of a reaction sequence
can often be indicated, and the rate constants of preceding
or subsequent chemical reactions can be measured with good
accuracy. In establishing a reaction mechanism, however, the
investigator must often resort to plausibility considerations,
since the identification of the intermediates and frequently
even of the end products of a reaction is possible only in
exceptional cases.
I t is not surprising, therefore, that the efforts of many electrochemists are being concentrated at present on the detection
of intermediates. This progress report has shown the successes
that have been achieved so far by the use of ESR spectroscopy
and mass spectroscopy. I t should be remembered, however.
that ESR spectroscopy is confined to measurements on relatively long-lived free radicals in solution, mainly in organic
solvents. and that mass spectroscopy can be used for the
analysis only of practically stable products. There is as yet
no sensitive in-situ method for the determination of short-lived
intermediates on or in the vicinity of the electrode surface
that would allow a rapid identification without influencing
the course of the reaction.
I t is clear that optical methods really offer the only possibility
of further assistance here, and the first steps in this direction
have already been taken. In the discussion of reflection spectroscopy and ellipsometry it was pointed out that variation
of the wavelength of the incident light could yield information
about the nature of the adsorbed substance. Fleischtiiurii? c't
al.'"J were recently able to identify surface layers by Raman
spectroscopy with laser light sources. Sieghuhn et I J / . ' ~ pro'~
posed the use of X-rays and electron diffraction.
b
I I
/ I
/ I t
752 nn
Irn321
9. Outlook
-
616058
mle
Fig. 32. Mass-spectromctric detection of the oxidation product of I-butanol
adsorbed o n the e1ectrode.a) Background spectrum a n d calibration spectrum
of t h c platinum emitter used (calibrated with acetone, mass numbers 5Y,
60. a n d 61); b ) mass spectrum of a solution of 0.473 M I-butanol tn 1 M
K O H Thc mass number 74 corresponds to the molecular peak of butanol
(molecular weight 74.17); c ) mass spectrum of the same solution as for
b ) after electrochcmical oxidation of the hutanol i i t + 7 0 0 r n V R t , l . T h e m i i s
numbers 71 and 72 correspond to the oxidation product in question.
The resultsareshown in Fig. 32. Curvea shows the background
spectrum of the mass spectrometer with the platinum foil
described above as the emitter. The spectrum of acetone,
which was used for calibration of the emitter in the mass
spectrometer, can be seen at the right-hand edge of the diagram.
Curve b shows the same mass range as curve a. In this case
the platinum foil was immersed in a solution of 1 M KOH
+ 0.473 M I-butanol without application of a potential before
it was placed in the mass spectrometer. The mass numbers
74 and 75 appear in the mass spectrum, 74 being the molecular
peak of butanol (molecular wight 74.12).
Finally, before curve c was recorded, the butanol was electrochemically oxidized for about 3 0 s at a potential of 700mVRHF.
In this case the mass spectrum contains the additional mass
numbers 7 I and 72, which correspond to the adsorbed oxidation product formed from the butanol by the elimination
of hydrogen.
Anqpw. ('hcrn. intcmar. Edir.
1 Vol I 3
( 1 9 7 4 ) / No. I I
The examples show that the development of optical methods
into analytical procedures for electrochemically produced
intermediates is being vigorously pursued at present. New
impetus for the study of electrode kinetics may be expected
from this quarter in the near future.
Received. January 17. 1974 [ A 23 IE]
G c r m a n verbion' Angew. Chem. Xh. 756 (19741
Translated by Express Translirtion Service. London
[I] f G U l l l a n d C. A. Kriorr. Z. Elektrochem. 6 4 . 258. 270 (1960)
W : L'iel\~ic.!i. Z. lnstrumentenkunde 71. 19 (1963).
[?I
[31 W. b i p ! \ ~ i c J i : Brennstoffcleniente Moderne Verfahren ztir elektrochemischcn Encrgiegewinnung. Verlag Chcniie. Weinhemi 1YO5.
141 See ref [ 3 ] . p. 121.
W c'relsric/i, Electrochim. Acta 17. 1829 (IY72)
[ 6 ] I@ C'ie/srir./r. L'. Liigd. M. Kuirrur, a n d J . Mrhrr, Entropie 2.31 (lY70).
[5] 8. D. S r r d and
[7] M
1969.
[XI
M Brrirr,r- Electrochemical Processes in Fuel Cells Springer. Berlin
I.: G. Lmich. Acta Physicochim. URSS 17. 257 (1942).
[Y] C: G. Lei-ich. Physicochemical Hydrodynamics. Prentice Hall. London
1962.
[lo] D. P. Gr c y or r a n d A . C . Rrdrliford. J. Chem Soc. 1YS6. 3756.
[ 1 I] G. Eir?7horii. A . Heiridridis. a n d W Vie/sric,k. Messtechnik Y. 224 (1968).
[ I21 B. L-'o.sriBr-Mi/i~,/["i(:Mi
r r d i . a n d A . l!~~iiir!ridi.\.Bcr. Bunsenges.
Phys. Chem. 76. 19 (1972)
1141 J .
rich. I . Electrochem. Soc. IOY. 849 (1962).
KoiiiwL!.and c' C Lciich. Zh. FIZ.Khim. 32. 1556(1958).
[ 151 H . Go.\riUr-.Mih~,/~?r'
a n d W Viristrcli, Ber. Bunsenges Pbys. Chem. 77.
476 (lY73)
695
[ 161 B.
Zlil/c,r.
.M. 1. Brlluc-uriw. and S /3ruckw\iwi. Anal. Chem. 44. 19x3
( 1977).
[ 171 C H . Huinuiiii a n d M: ~ ' i r / . s t k ~I / .i . Electroanal. Chem. 32, 459 ( I97 I ).
[ I 8 3 M: J ..llhrry and S. B r u c h , r e t i i . Trans. Faraday Soc. 62. 1915, 1920
(I966).
[ 191 G .
Tkiiithorn.
Dissertation. IJniversitit Bonn 1969.
/~ikuiii~iru
Ii. / c i / , u i i i i i r ( i . C i : , V I ~ J ~;ind
W . E. l i w y e r . I ZSth N;itional
Meeting. T h e Electrochem. Soc.. Nen York 1969 (Extcnded Ahatracta p.
1.311 I
IXX).
1321 F. G m ;itid K J L<,rr<,r.%. Ph!\ ('hem N F 77. 317 11972)
[ 3 3 ] 4. K N K d d . , I!. G ~ w ~ l i c i i tand
.
J . O'.$I. 50d,ri\, _I Electroanal.
Chciii. 8. 406 (1964).
[20] H . H Buiirr. L). L. Siiiirh. and P. J . Elriiiq J Amer Chem. Soc. 87,
2094 1 1960).
[34] .1. K . M. K ~ d d i . .I f G 8. K u o . and ./. O ' M . Biickri\. J C'hrm. Phys
47. 2246 ( I965).
[21] 8. Bi-arer and H . H . Buuur- Alternating Current Polarography a n d
Tensametry. Wiley. New York 1963.
[ 3 5 ] J L. Or0 and D. J . DrSiiirf. J . Electrochen?. Soc I l 3 . 1258 (1966).
[36] 7 K t r i t o i i u . Electro;tnal. Chem 1. I57 (1966).
1371 B. Kmrcviim/. Chetn. Ing. Tech. 4.7. I90 (1970)
[ 2 2 ] R. Nrrh. Z. Anal. Chem. 188. 4001 (1962).
1233 H Gwi.sdiei-. 2. Elektrochem. SH. 9 (1954).
[24] D. €. Sntirli. Electroanal. Chem I . I (1966).
[ 2 5 ] J . U: Srr-ojek a n d T Kuiwnu. J. Electroanal. Chem. 16, 471 (1968).
[26] W N . Huiiscii, 7:
Soc. X X . 1062 (1966).
K u i ~ o i i o ,a n d
K . .4. O.\feri-ouiiq. 1. Amer. Chem.
[27] B. S. Poiis, J . S. Morrcoii, L. 0.Wiizsfrowi, and H . B. ,\4ur/,, Jr.. Anal.
Chem. 3Y. 685 (1967).
[28] 4. H . K c w d and E. Yc.uyer, 19th CITCE Conference, Detroit 1968.
paper 308 (Extended Abstracts p. 195).
[29] A . H . Rrud a n d €. Ewyer. Appl. O p t 7. 451 (1968).
[30] W J . Pltrrh. Chem. Ing. Tech. 44, 221 (1977).
[iX]
N . 4 . Bulu\hocx a n d I : E . K u z u r i w r . Elcctroanal Chem. 3. 135 (1969).
[39] I .
~ W ~ L W ! Ya n d
N. Hdcvv1uiiii. J. Phys. Chem. 61. 1.123 11957)
1401 Sec. c q , J O M . Bociri., a n d D. 4. Siwih/t,.\. J. Electrochem. Soc
/ I / . 736(1964).
1411 S. Briiciurisieiti and J Cottiwtt. Diss. Faraday Soc.. i n prcss.
[42] J . Hrirhmtiri a n d CC L i d > f t d i . t o be published
1431 H . D. Bcr/,c.i- Field Ionisition M a \ \ Spectrometr? P c r p n i o n . O\forcl
1971. Angew. Chem. XI. 662 (1969): Angew. Chem. internill. Edit. 8 . 673
( 1969)
[44] ,%f. F / e i ~ c l i i i i ~ i i i iP.
i . J . Hriidru. a n d 4 .I
M < Q u t / / u i i . .I. C. S. Chcm.
Coniiii. 1973. 80.
[J5] C . (;L~III\. E . B i i d t w . S. S r w i w i . 7: Biwqiiiui-h. xnd K . Sirghohii. C o m m .
of the Uppsala Universtcy UUIP-817. April 1973.
Effects of Through-Bond Interaction
By Rolf Gleiter[*I
The effect of through-bond interaction is derived for 1,4-butanediyl as an example, and changes
caused by this effect in the ground state of the system are demonstrated. Spectroscopic methods
for the detection of the effect are discussed, and its chemical consequences are illustrated
by a number of examples.
1. Introduction
Until a few years ago the interaction of lone pairs of electrons
and of R orbitals situated more than 3a apart was assumed
to be very small. Examples of this situation are found in
1,4-diazabicyclo[2.2.2]octane (1 ), p-benzoquinone (2), anritricycI0[4.2.0.O~.~]octa-3,7-diene
( 3 ) ; and pyrazine ( 4 ) . The
assumption that these interactions are very small became
the subject of critical examination following calculations by
Hoffmann et a!. for the three isomeric diazabenzenes['] and
didehydrobenzenesf2! These calculations were stimulated by
experiments by Rees and S t ~ r r [on
~ lcycloadditions to 1,S-didehydronaphthalene (1,8-naphthalenediyl) and by investigations
by McKinney and Geske14] on the radical-cation ( 1 ) +.
[*] Prof. Dr. R. Gleiter
Institut fur Organische Chemie der Technischen Hochschule
61 Darmstadt, Petersenstrasse 15 (Germany)
696
The results of these calculations wilI be illustrated for the
three diazabenzenes (diazines) pyrazine ( 4 ) . pyrimidine ( 5 ) ,
and pyridazine (6). Ifthe two orbitals indicated on the nitrogen
atoms in ( 4 ) , I S ) , and (6) are denoted by n, and nb, we
can write down the two non-normalized, but symmetryadapted linear combinations
n + = (n, + nh)
n - = (n, - nh)
which are shown schematically in Figure 1. The n + linear
combination is symmetric (S) and the n- linear combination
antisymmetric (A) with respect to the mirror plane m.
If the interaction between n, and nb is very small, n + and
n - should have approximately equal energies (Fig. 2, center).
If the interaction is large, on the other hand, the energy
A n g c n . ('hfm. intrrnut Edit. f ./'/I
13 ( 1 9 7 4 ) / No. I /
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 249 Кб
Теги
electrode, stud, method, modern, processes
1/--страниц
Пожаловаться на содержимое документа