This action might not be possible to undo. Are you sure you want to continue?

Andrzej Lasia

**Département de chimie, Université de Sherbrooke, Sherbrooke Québec, J1K 2R1
**

alasia@courrier.usherb.ca A. Lasia, Electrochemical Impedance Spectroscopy and Its Applications, Modern Aspects of Electrochemistry, B. E. Conway, J. Bockris, and R.E. White, Edts., Kluwer Academic/Plenum Publishers, New York, 1999, Vol. 32, p. 143-248.

Table of content I. 1. Introduction..............................................................................................................................4 Response of Electrical Circuits............................................................................................4 (i) Arbitrary Input Signal ......................................................................................................4 (ii) Alternating Voltage (av) Input Signal..........................................................................6 2. Impedance of Electrical Circuits..........................................................................................7 (i) Series R-C Circuit ............................................................................................................8 (ii) Parallel R-C Circuit......................................................................................................8 (iii) Series: Rs + Parallel R-C Circuit..................................................................................9 3. Interpretation of the Complex Plane and Bode Plots...........................................................9 II. 1. 2. 3. 4. 5. Impedance measurements ......................................................................................................10 Ac Bridges .........................................................................................................................10 Lissajous Curves ................................................................................................................11 Phase Sensitive Detection (PSD) .......................................................................................11 Frequency Response Analyzers .........................................................................................12 Fast Fourier Transform (FFT)............................................................................................13 (i) Pulse Perturbation ..........................................................................................................14 (ii) Noise Perturbation .....................................................................................................14 (iii) Sum of Sine Waves....................................................................................................14

III. Impedance of faradaic reactions in the presence of diffusion................................................15 1. 2. 3. 4. 5. 6. The Ideally Polarizable Electrode......................................................................................15 Semi-Infinite Linear Diffusion ..........................................................................................15 Spherical Diffusion ............................................................................................................20 Cylindrical Electrodes........................................................................................................22 Disk Electrodes ..................................................................................................................23 Finite-Length Diffusion .....................................................................................................23 (i) Transmissive Boundary .................................................................................................24 (ii) Reflective Boundary ..................................................................................................25 7. Analysis of Impedance Data in the Case of Semi-Infinite Diffusion: Determination of the Kinetic Parameters .....................................................................................................................26 (i) Randles’ Analysis64,65,67 .................................................................................................26 (ii) De Levie-Husovsky Analysis ....................................................................................27 (iii) Analysis of cot ϕ........................................................................................................27 (iv) Sluyter's Analysis.......................................................................................................28 IV. Impedance of A faradaic reaction involving adsorption of reacting species .........................29 1. 2. 3. 4. 5. V. Faradaic Reaction Involving One Adsorbed Species.........................................................29 Impedance Plots in the Case of One Adsorbed Species ....................................................31 Faradaic Impedance in the Case Involving Two Adsorbed Species..................................34 Impedance Plots in the Case of Two Adsorbed Species....................................................36 Faradaic Impedance for the Process Involving Three or More Adsorbed Species............36 Impedance of solid electrodes................................................................................................37

3 1. 2. 3. 4. Frequency Dispersion and Electrode Roughness...............................................................37 Constant Phase Element.....................................................................................................37 Fractal Model .....................................................................................................................40 Porous Electrode Model.....................................................................................................42 (i) Porous Electrodes in the Absence of Internal Diffusion................................................42 (a) De Levie’s treatment..............................................................................................43 (b) Rigorous treatment.................................................................................................45 (ii) Porous Electrodes in the Presence of Axial Diffusion...............................................46 (iii) Other Pore Geometries...............................................................................................49 5. Generalized Warburg Element...........................................................................................49 1. 2. 3. 1. 2. Linearity, Causality, Stability, Finiteness ..........................................................................50 Kramers-Kronig Transforms..............................................................................................51 Non-Stationary Impedances...............................................................................................53 Selection of the Model .......................................................................................................54 CNLS Approximations ......................................................................................................56 (i) CNLS Method................................................................................................................56 (ii) Statistical Weights .....................................................................................................57 (iii) AC Modeling Programs .............................................................................................58

VI. Conditions for "good" impedances ........................................................................................50

VII. Modeling of experimental data ..............................................................................................54

VIII.Instrumental limitations .........................................................................................................58 IX. Conclusion .............................................................................................................................60

but the reader may consult numerous books on the subject. Analysis of the system response contains information about the interface.4 I. later. the name impedance spectroscopy was later adopted. containing mathematical development of the fundamental equations. (i) . EIS is a very sensitive technique and it must be used with great care. INTRODUCTION Electrochemical Impedance Spectroscopy (EIS) or ac impedance methods have seen tremendous increase in popularity in recent years.200 in earlier years and ~500 citations per year of "electrochemical impedance". thus. Knowledge of the Laplace transform technique is assumed. It is a complementary technique and other methods must also be used to elucidate the interfacial processes. Response of Electrical Circuits Arbitrary Input Signal Application of an electrical perturbation (current. These measurements are carried out at different ac frequencies and. EIS studies the system response to the application of a periodic small amplitude ac signal. to an ac signal.8-17 specific books18. where Q is the charge stored in a capacitor. the system response to an arbitrary perturbation and. When the same potential is applied to the series connection of the resistance R and capacitance C.6. First. frequently. 1. Initially applied to the determination of the double-layer capacitance1-4 and in ac polarography. potential) to an electrical circuit causes the appearance of a response. the total potential difference is a sum of potential drops on each element. This may be connected with the fact that existing reviews on EIS are very often difficult to understand by non-specialists and. Besides. let us consider application of an arbitrary (but known) potential E(t) to a resistance R. the following equation is obtained: (1) Q( t ) 1t E ( t ) = i( t ) R + = i( t ) R + ∫ i( t )dt C C0 This equation may be solved using either Laplace transform or differentiation techniques.5-7 they are now applied to the characterization of electrode processes and complex interfaces. Although the term "impedance" may include also nonelectrochemical measurements and "electrochemical impedance" may not include all the electrochemical studies. it is not always well understood.19 on EIS. 32-34 Differentiation gives: di ( t ) i ( t ) 1 dE ( t ) (2) + = dt RC R dt which may be solved for known E(t) using standard methods for differential equations. EIS is now described in the general books on electrochemistry. the popularity of this technique cannot be denied. will be presented. The Chemical Abstract database shows ~1. Taking into account that for a capacitance E(t) = Q(t) /C. In this chapter.500 citations per year of the term "impedance" since 1993 and ~1. The purpose of this review is to fill this gap by presenting a modern and relatively complete review of the subject of electrochemical impedance spectroscopy. The current i(t) is given as: i(t) = E(t)/R. and there are also numerous articles and reviews. It should be stressed that EIS cannot give all the answers.20-31 It became very popular in the research and applied chemistry. However. its structure and reactions taking place there. they do not show the complete mathematical developments of equations connecting the impedance with the physico-chemical parameters.

and inductance. Ω. For the series connection of a resistance.i(0 ). In this case: 1 E ( s) (6) Z ( s) = =R+ i ( s) sC The inverse of impedance is called admittance. the parameter s may be complex. (6). the capacitance as 1/sC and the inductance as sL. E(s)/i(s) is expressed in the units of resistance. E0. 35 and 36. the current is described by eqn. and taking i = 0 at t = 0. according to: L [f(t)] = f ( s) = F (s) = ∫ f (t ) exp( − st )dt 0 ∞ (3) The Laplace transform is often used in solution of differential and integral equations. They are transfer functions which transform one signal. one gets: E0 E 1 (9) i ( s) = = 0 s( R + 1 / sC ) R s + 1 / RC Inverse transform of (9) gives the current relaxation versus time: . L. is applied at time zero to a series connection of R and C. 18. Y(s) = 1/Z(s). and is called impedance.e. (5). but in this chapter only the real transform will be considered. Knowledge of the system impedance allows for an easy solution of the problem. Direct application of the Laplace transform to eqn. Some other transfer functions are discussed in refs.g. s = ν +jω. Addition of impedances is analogous to the addition of resistances. f ( s) or F(s).g. and capacitance. e. In general. when a constant voltage. Z(s). In general. eqn. one obtains the current response in the Laplace space: i ( s ) = E ( s ) / ( R + sL ) (8) In both cases considered above the system impedance consists of the sum of two terms. current. is a sum of the contributions of these two elements: resistance. one can write contributions to the total impedance corresponding to the resistance as R. R. where j = −1 . Taking into account that the Laplace transform of a constant L (E0) = E0 /s. s = v. into another. 1/sC. taking into account that L ( ∫ i ( t ) dt ) = i(s)/s. It should be noticed that the impedance of a series connection of a resistance and capacitance. Both are called immittances.5 The Laplace transform is an integral transform in which a function of time f(t) is transformed into a new function of a parameter s called frequency. the total potential difference consists of the potential drop on both elements: di ( t ) (7) E (t ) = i (t )R + L dt + Taking into account that L [di(t)/dt] = s i(s) . e. For example. (1). corresponding to two elements: resistance and capacitance or inductance. R. applied voltage. gives: 0 t E ( s) = i ( s ) R + which leads to: i ( s) sC (4) 1 ⎞ ⎛ (5) i ( s) = E ( s) / ⎜ R + ⎟ sC ⎠ ⎝ The ratio of the Laplace transforms of potential and current. i.

in refs. More complex examples are presented.g. ω = 2πf is the angular frequency.g. (5) gives: Eω Eω 1 1 1 (13) i ( s) = 2 0 2 = 0 2 2 R s + ω s + 1 / RC s + ω R + 1 / sC Distribution into simple fractions leads to: ⎡ 2 ⎤ E0 s ω ω ω 1 (14) + − i( s ) = ⎢ω 2 2 + (1 / RC ) 2 ] 2 RC s 2 + ω 2 RC s + 1 / RC ⎥ R [ω s +ω ⎣ ⎦ and the inverse Laplace transform. e. the differential equations describing the system are of the second or higher order and the impedances obtained are the second or higher order functions of s. more than one capacitance or inductance in the circuit. In a similar way other problems of transient system response may be solved. In the Laplace space the equations (e. It should be added that an arbitrary signal may be applied to the system and if the Laplace transforms of the potential and current are determined. which allows numerical Laplace transformation to be carried out and the system impedance obtained. In the cases involving more time constants. and f is the av signal frequency.g. (8) which.. e. e. E(t). i. taking L -1 [s/(s2+ω2) = cos ωt. i (t ) = Alternating Voltage (av) Input Signal In the EIS we are interested in the system response to the application of a sinusoidal signal. Taking into account that the Laplace transform of the sine function L [sin(ωt)] = ω/(s2 + ω2)]. after substitution of E(s) = E0 /s. (10) and (12)) and analysis in the frequency space s allows for the determination of the system parameters. This analysis is especially important when an ideal potential step cannot be applied to the system because of the band-width limitations of the potentiostat. This problem may be solved in different ways. First. gives: E0 ω ω ⎤ ⎡ 2 i( t ) = ⎢ω sin( ωt ) + RC cos( ωt ) − RC exp( − t / RC )⎥ 2 + (1 / RC ) 2 ] ⎣ ⎦ R[ω (15) (ii) .e. where E0 is the signal amplitude. 33-34. eqns. let us consider application of an av signal to a series R-C connection.37 In this case it is sufficient to know i(t) and the real value of the potential applied to the electrodes by the potentiostat. Similarly.g. (9) and (11)) are much simpler than those in the time space (e. gives: E0 E E ⎛1 1 1 ⎞ (11) i ( s) = = 0 = 0⎜ + ⎟ s( R + sL) L s( s + R / L ) R ⎝s s + R / L⎠ Inverse transform gives the time dependence of the current: E ⎡ ⎛ Rt ⎞ ⎤ (12) i ( t ) = 0 ⎢1 − exp⎜ − ⎟ ⎥ R ⎣ L ⎠⎦ ⎝ The current starts at zero as the inductance constitutes infinite resistance at t = 0 and it increases to E0/R as the effect of inductance becomes negligible in the steady-state condition. application of the potential step to a series connection of R and L produces response given by eqn. by numerical transform calculations. current initially equals E0/R and it decreases to zero as the capacitance is charged to the potential difference E0. the system impedance is determined.6 E0 (10) exp( − t / RC ) R The result obtained shows that after the application of the potential step.: E = E0 sin(ωt). use of eqn. eqns.g.

Using complex notation they may be described by: E = E 0 exp( jωt ) and i = i0 exp[ j (ωt + ϕ )] (21) These vectors rotate with a constant frequency ω and the phase-angle. It may be recalled that in complex notation: (20) Z ( jω ) =| Z |exp( jϕ ) =| Z |[cos(ϕ ) + j sin(ϕ )] Analysis of eqn. equals: (19) |Z| = ( Z ' ) 2 + ( Z " ) 2 = R 2 + (1 / ωC ) 2 and the phase-angle between the imaginary and real impedance equals ϕ ≡ arg( Z ) = atan(-1/ωRC). this is the imaginary Laplace transform. Impedance of Electrical Circuits In order to simplify the calculations of impedances.7 The third term in eqn. between potential and current. ϕ = arctan(1/ωRC). eqn. respectively. is different from that found between current and potential. (17). between them stays constant. as represented in Figure 1a. (17) indicates that the current represents a vector of the length i0 = E0 /|Z| which rotates with the frequency ω. eqn. (18). Z(jω). In the latter example the system impedance. The modulus of Z(jω). 2. by substitution: s = jω. Instead of showing rotating vectors in time space it is possible to present immobile Figure 1. It should be noticed that the sign of ϕ. (17). the result obtained for the periodic perturbation of an electrical circuit may be represented using complex notation. ϕ. eqn. Current and potential are rotating vectors in the time domain. The value |Z| has units of resistance. (6). The steady-state equation may be rearranged into a simpler form: E0 1 ⎤ ⎡ (16) i( t ) = ⎢sin( ωt ) + ωRC cos( ωt ) ⎥ ⎞⎣ ⎛ ⎦ 1 ⎟ R⎜1 + ⎜ ( ωRC) 2 ⎟ ⎠ ⎝ and by introducing tan ϕ = 1/ωRC the following form is found: E0 E (17) i( t ) = sin(ωt + ϕ ) = 0 sin( ωt + ϕ ) | Z| 1 R2 + ( ωC ) 2 where ϕ is the phase-angle between current and potential. eqn. It should be noted that the complex impedance Z(jω). may be represented as: (18) 1 1 Z ( jω ) ≡ Z = Z '+ jZ '' = R + = R− j jωC ωC and the real and imaginary parts of the impedance are: Z’ = R and Z” = -1/ωC. ^ . (15) corresponds to a transitory response observed just after application of the av signal and it decreases quickly to zero. may be obtained from Z(s). In fact. described above for the impedances. It is obvious that the current has the same frequency as the applied potential but is phase-shifted by the angle ϕ. it is the length of a vector obtained by addition of two perpendicular vectors: R and 1/ωC.

This circuit corresponds to an ideally polarized electrode in solution. a mercury electrode . the complex impedance may be written for any circuit by taking R for a resistance. where the initial phase shift of the potential was assumed to be zero. The complex plane plot is a plot of Z” versus Z’. The corresponding complex plane and Bode plots. The Nyquist plot shows a semicircle of radius R/2 and the center on the real axis. Several examples of this method are presented below.8 vectors in the frequency space. 1/jωC for a capacitance and jωL for an inductance. for the same values of R and C elements as used in the series R-C model above. In general. is the same as for the series and the parallel R-C circuits. that is. It should be stressed that for capacitive circuits the imaginary impedance is always negative and the imaginary admittance is positive. and applying Ohm’s and Kirchhoff’s laws to the connection of these elements. ~ ~ they are equal to E = E0 and I = I0 exp(jϕ). Z = R and ω → ∞ . as observed in the impedance Bode graph. The complex plane plots may also be obtained for admittances. The graph of log |Z| versus log ω. log |Z| (magnitude) and phase-angle. They are also shown in Figure 2. and the frequency at the semicircle maximum equal to: ω = 1/RC. (ii) that: Parallel R-C Circuit For the parallel R-C connection the total admittance equals: Y ( jω) = 1 / R + jωC such Z ( jω ) = (23) 1 R R jω R 2 C = = − 1 / R + jωC 1 + jωRC 1 + ω 2 R 2 C 2 1 + ω 2 R 2 C 2 ˆ ˆ There are two limits of the impedance: ω = 0. see Figure 1b. It consists of a straight line perpendicular to the real axis. Figure 2. the imaginary versus the real components. Figure 2d. Figure 2c.g.supporting electrolyte solution. versus log ω. Admittance for the series R-C connection equals: (i) Y ( jω ) = 1 1 j R = = + 1 Z ( jω ) R − j ⎛ 2 1 ⎞ R2 + 2 2 ωC⎜ R + 2 2 ⎟ ωC ⎝ ω C ω C ⎠ (22) It represents a semi-circle on the complex plane plot. C = 2×10-5 F) circuit is shown in Figure 2. contains one breakpoint or corner frequency. ϕ. These vectors are called phasors. A complex plane plot for a series connection R-C (R = 100 Ω. plotted for various frequencies. .e. Z = 0. The circuit’s characteristic breakpoint frequency (inverse of the characteristic time constant).j/ωC. This point corresponds to the system characteristic frequency ω = 1/RC = 500 s-1 or a time constant τ = RC = 0.002 s. Series R-C Circuit In the case of a series connection of the resistance and capacitance the impedance is given by: Z(jω) = R + 1/jωC = R . separated by the phase-angle ϕ. Other types of graphs are Bode plots i. The result may be represented graphically using two types of plots: complex plane (also known as Argand or Nyquist plots) and Bode plots. The phase-angle changes from 90o at low frequencies to 0 at high frequencies. are shown in Figure 3. e.

the Bode log |Z| plot shows that there are two breakpoints (bends). they do not show all details. Rs = 10 Ω and Cdl = 20 µF. shown in Figure 3 and Figure 4. Nyquist plots allow for an easy relation to the electrical model. (24) 1 1 / Rct + jωCdl The corresponding complex plane and Bode plots are also shown in Figure 4 for Rct = 100 Ω. Figure 3c. From eqn. and for ω → 0 Z → Rs + Rct. the contribution of each term in eqn.9 The complex plane admittance plot represents a straight line parallel to the imaginary axis. . Z ( jω ) = Rs + 3. (24). This equation may be rearranged into another form: ⎛R R C ⎞ (25) 1 + jω ⎜ s ct dl ⎟ ⎝ Rs + Rct ⎠ 1 + jωτ 2 Z = ( Rs + Rct ) = ( Rs + Rct ) 1 + jωτ1 1 + jω( Rct Cdl ) where τ1 and τ2 are the Bode characteristic time constants. However. Z → Rs and ϕ → 0. On the other hand Bode plots contain all the necessary information. log(|1+jωτ|) = 0 and when ωτ>>1.33 Let us consider the circuit shown in Figure 4a. (26) can be considered independently and then their sum may be easily obtained.e. which is similar to the impedance complex plane plot for the series R-C connection. The main difference between circuits in Figure 3 and Figure 4 is connected with the fact that in the latter circuit. For comparison. In addition. the admittance complex plane plot is also shown in Figure 4c. That is why Bode plots are mainly used in the circuit analysis. Each term log(|1+jωτ|) has two limits: when ωτ<<1. consisting of a series connection of the resistance Rs with the parallel connection of Rct-Cdl. (25) log(|Z|) is easily evaluated: log(| Z|) = log(| Rs + R ct |) + log(|1 + jωτ 2 |) − log(|1 + jωτ1|) (26) In order to construct asymptotic lines in the Bode magnitude plot. exactly the same Nyquist impedance plots. due to the presence of Rs. is given as: Figure 4. impedance of the circuit shown in Figure 4. Interpretation of the Complex Plane and Bode Plots Complex plane (Nyquist) plots are the most often used in the electrochemical literature because they allow for an easy prediction of the circuit elements. for example. ω<<1/τ. may be obtained for different values of the capacitance C. Nevertheless. The Bode magnitude plots may be easily predicted from the circuit impedance. The frequency corresponding to the maximum of Z’’ is still equal to ω = 1/RctCdl = 500 s-1. The only difference between them will be the fact that the points on the semicircle would correspond to different frequencies. Figure 3. Its impedance is presented by eqn. i. at ω → ∞. (iii) Series: Rs + Parallel R-C Circuit Finally.

corresponding to the maximum of the semicircle at Z’ = Rs + Rct/2 at ω = 1/RctCdl. principally in precise double-layer measurements. II. Although this method is slow. The plots of Z' and Z'' (or their logarithms) as a function of log ω are also sometimes shown in the literature. equals ω = 1658 s-1. 28. 19. The break-point frequencies in the Bode magnitude plot. 18. are ω1 = 1/τ1 = 500 s-1 and ω2 = 1/τ2 = 5500 s-1. in this case.39-41 . In this way Bode magnitude graphs may be constructed for other circuits. The ac bridge with potentiostatic control may also be used. to measure the electrode impedance in the presence of a faradaic reaction to determine the kinetics of electrode processes. The continuous line is the sum of the three asymptotes. It has been described in detail in refs. 26. The phase-angle is described by: ⎞ ⎛ ωR 2 Cdl ct ⎟ ⎜ ϕ = atan( Z ''/ Z ' ) = atan ⎜ ⎟ (27) ⎜ Rs + Rct + (ωRct Cdl ) 2 Rs ⎟ ⎠ ⎝ It can be shown that this function has a maximum at: RS + Rct 1 (28) ω= Rct Cdl Rs which. Ac Bridges This technique was the first used to measure the double-layer parameters (principally of the dropping mercury electrode) and. 30 and 31. 18. IMPEDANCE MEASUREMENTS Dc transient response of electrochemical systems is usually measured using potentiostats. later. 28 and 38. because bridge compensation must be carried out at each frequency manually.10 log(|1+jωτ|) = log τ +log ω. it is still used. The graphs corresponding to these lines are shown in Figure 5. The Bode phase-angle graph is shown in Figure 4e. Figure 5. 26. In the case of EIS an additional perturbation is added to the dc signal in order to obtain the frequency response of the system. 1. It should be noticed that the maximum of the phaseangle is different from the maximum of the imaginary part of the impedance. The system impedance may be measured using various techniques: 1) ac bridges 2) Lissajous curves 3) phase sensitive detection (PSD) 4) frequency response analysis (FRA) 5) fast Fourier transform (FFT) Because older techniques were described in detail in refs. The use of ac bridges provides a very good precision of measurements. which correspond to a straight line with a slope of one and intercept log ω = -log τ. Figure 4d and Figure 5. this chapter will be focused on the last three techniques.

a sin( ωt + ϕ1 ) (29) where E1. is: E1 = E1.28.a is the signal amplitude and ϕ1 is the phase shift. when the frequency in eqn. is multiplied by a square-wave signal of the same angular frequency ω. Phase Sensitive Detection (PSD) Phase sensitive detection is used in lock-in amplifiers. 3. The square-wave signal may be represented as a Fourier series: (30) 4 ∞ 1 E2 = ∑ sin[ ( 2n + 1) ωt + ϕ 2 ] π n = 0 2n + 1 where n is an integer and the amplitude of the square signal is taken as unity. with increasing n. (30). harmonics. the obtained average signal: (33) 2E Average( E1E2 ) = 1 cos( ϕ1 − 3ϕ 2 ) 3π .g. Because of the frequency limitations and sensitivity to noise. 1/5. For example. although their influence is attenuated by 1/3.30 which may be used for the determination of the impedances. (29) is three times the reference frequency in eqn. Lissajous Curves Recording of the applied av potential and the resulting ac current on a twin-beam oscilloscope produces so-called Lissajous curves (in this case an ellipse)18. the average value of E1E2. The output signal is subsequently applied to a low-pass filter which averages the signal components having frequencies above the filter cut-off frequency. proportional to the ac current from the potentiostat. 1/7. It reaches a maximum when the phase difference of the two signals being mixed is zero.⎪ ⎩ ⎭ It contains one time independent component. and is proportional to the amplitude of the measured ac signal. Because the average value of periodic functions is equal to zero. noise). if they are present in the input signal (e. this technique is not currently used in electrochemical measurements. In this method the measured signal. equation (31).43 Only a general idea of these measurements will be presented here.. It produces a dc signal proportional to the amplitude. E1. equals: (32) 2E Average( E1E2 ) = 1 cos( ϕ1 − ϕ 2 ) π The disadvantage of the lock-in technique is that it retains contributions of the harmonic frequencies (2n+1)ωref.. etc.11 2. The resulting signal E1×E2 equals: ∞ 4 E1 sin (ωt + ϕ1 ) sin[ (2n + 1)ωt + ϕ 2 ] = E1E2 = ∑ n = 0 (2 n + 1)π = ∑ 2 E1 (31) cos[ − 2nωt + ϕ1 − (2 n + 1)ϕ 2 ] − cos[ (2n + 2)ωt + ϕ1 + (2 n + 1)ϕ 2 ] = n = 0 (2 n + 1)π ∞ { } ⎧ ⎫ ⎪cos(ϕ1 − ϕ 2 ) − cos( 2ωt + ϕ1 + ϕ 2 ) ⎪ ⎪ ⎪ 2E ⎪ 1 1 ⎪ = 1 ⎨+ cos( − 2ωt + ϕ1 − 3ϕ 2 ) − cos( 4ωt + ϕ1 + 3ϕ 2 ) ⎬ 3 π ⎪ 3 ⎪ 1 1 ⎪ ⎪ ⎪+ 5 cos( − 4ωt + ϕ1 − 5ϕ 2 ) − 5 cos( 6ωt + ϕ1 + 5ϕ 2 )+. which are interfaced with potentiostats. depending on the phase difference of two signals.42.

44 The measured signal. Frequency Response Analyzers Frequency response analyzers (FRA) are instruments which determine the frequency response of a measured system. Diard et al. a 1T Re( E1 ) = ∫ E1. Figure 7 shows the attenuation of the output signal as a function of frequency and the number of integration cycles N. that is correlation of sin(kωt + ϕ) with sin(ωt) or with cos(ωt) is equal to zero for k>1. 4. is multiplied by the sine and cosine of the reference signal of the same frequency and then integrated during one or more wave-periods: E1. Their functioning is different from that of lock-in amplifiers.47 as shown in Table 1. additionally. The advantage of the correlation process is also reduction of noise (of arbitrary frequency). Recently. with the reference signal shifted by π/2 to resolve the imaginary component. They derived theoretical expressions for the error in impedance measurements using the odd harmonic test criterion46.45 studied effects of electrochemical non-linearities on impedance measurements using a FRA. ϕ2 is equal to zero and the expressions become simplified. In twophase lock-in amplifiers. a 1T sin( ϕ1 ) ∫ E1. Measurements of the fundamental and third harmonic in the electrode response are sufficient to estimate the impedance error.5 (lower limit up to 10 Hz. Modern FRAs carry out all the computations digitally. its influence decreasing with the increase of the integration time. They are based on the correlation of the studied signal with the reference. FRAs have a wide frequency range (12 decades) and high precision. The measured signal is mixed with the reference square-wave signal of the same frequency and the resulting signal goes through a low-pass filter producing an average of all components. The phase shifter allows for precise adjustment of the reference phase in order to zero the phase difference ϕ1 . depending on the manufacturer) to ~105 Hz with a precision of 0.2%. eqn. a sin( ωt + ϕ1 ) cos( ωt ) dt = T0 2 (35) Such integration recovers the real and imaginary parts of the measured signal. It can also be shown that all the harmonics are strictly rejected. Figure 7. Lock-in amplifiers operate in the frequency range from 0. A schematic diagram of a lock-in amplifier is shown in Figure 6. . (29).ϕ2.12 is attenuated three times. a sin( ωt + ϕ1 ) sin( ωt ) dt = cos( ϕ1 ) (34) T0 2 and Im( E1 ) = E1.1 to 0. the measured signal is mixed with the reference signal to obtain the in-phase component and. Modern lock-in amplifiers are controlled by a microprocessor and permit automated measurements with autorange selection. If the reference signal is synchronized with the applied signal (they are both generated from the same source). Figure 6. A comparison of PSD and FRA was recently presented by Evans.

g. Another problem called aliasing is connected with the presence of the frequencies larger than one-half of the time domain sampling frequency.13 Table 1 Comparison of PSD and FRA Lock-in amplifier Advantages • Very sensitive • Effectively removes noise • Reduces harmonic distortion • Suppresses dc noise • Relatively low cost Disadvantages • Limited frequency range • Slower • Standalone readings difficult FRA Advantages • Faster analysis • Wide frequency range • Removes harmonic distortion • Direct output to external device • Easy standalone measurements Disadvantages • Higher cost • Limited noise removal • Limited sensitivity 5. s = ν +jω. the transformation parameter is complex. synchronization of the sampling time with the wave period is necessary. (36) involves integration to infinity.28. This problem is known as leakage. the system ac impedance may be obtained from: (37) F [ E ( t ) ] E ( jω ) Z ( jω ) = = F [i( t )] i ( jω ) where symbol F denotes the Fourier transform. eqn.50 First of all the Fourier transform defined by eqn. of potential and current. In general.49. where k is an integer. However. Therefore. That is if the data acquisition is terminated at other times than the multiple of the wave period. In some cases the highest frequencies may be filtered . This problem may be easily eliminated by assuring that the sampling frequency is greater than (or at least equal to) twice the highest frequency present in the measured signal.48 The number of points acquired must be equal to 2k.1(i) that the system impedance is defined as the ratio of Laplace transforms. e.48. (6). Taking the Fourier transform of the perturbation signal and that of the resulting signal allows determination of the transfer function. Certain proprieties of the FFT technique influence the obtained results. The imaginary Laplace transform: (36) ∞ F ( jω ) = ∫ f ( t ) e − jωt dt 0 is called the single sided Fourier transform. sharp discontinuities of the signal are introduced (because there is no continuity between the last and the first point of the signal) which causes frequency peak broadening. In practice only limited length data are transformed. Fast Fourier Transform (FFT) It has been shown in Section I. causing broadening of the computed frequency spectrum. that is a distribution of frequencies is obtained instead of discrete values. It may be minimized by increase of the data record acquired in the time domain. it will disappear when the acquisition time is equal exactly to an integer multiple of the wave repetition period. The fast Fourier transform (FFT) provides a fast and efficient algorithm of computation of the Fourier transform.

In general.52: 1) pulse. necessary to get information about the existing signal. is called the Nyquist sampling rate. the most often used perturbation signals are50.50. single frequency components obtained by the FFT have relatively low amplitudes and long data acquisition time is necessary to obtain reliable results. This arrangement is superior to other perturbation wave forms. This technique was introduced and extensively used by D. defined as: h(t) = 1 for t = 0 to T0 and h(t) = 0 elsewhere. that is it contains all the frequencies with the same amplitude K. However.51 that even low level noise disturbs significantly the obtained impedance spectra. maintaining the total amplitude at its initial level.53. It was shown50.49.51 Even low noise contamination of the measured signal leads to significantly disturbed impedance spectra. Popkirov and Schindler51 have demonstrated that by the appropriate selection of phases and amplitudes of the individual sinusoidal components the measured results may be improved. This minimum sampling frequency. the perturbing signal may have an arbitrary form. Figure 8. that is noise consisting of continuous spectrum of frequencies (or a computer generated pseudo-random white noise) may be used as a perturbation signal in practical impedance measurements. Smith and coworkers. where δ(t) is Dirac’s delta function. equals: H ( jω ) = K .50. It is obvious that the amplitude of the higher frequency signals is attenuated.14 out by a low-pass filter. signal phases may be optimized to minimize the observed peak to peak signal amplitude. Therefore.e. Fourier transform of such a function. Such a function cannot be realized in practice and must be substituted by a pulse of a short duration ∆t. Pulse Perturbation Fourier transform of an infinite short pulse function: h( t ) = Kδ ( t ) .50 All these frequencies are applied at the same time and the response to each frequency is found by the FFT. However. 2) noise. It should be mentioned the such technique is used in low frequency impedance analysis (below 10 Hz) in PAR 273 series potentiostats (software implemented). (ii) Sum of Sine Waves In this technique the perturbation signal is composed of a sum of selected sinusoids.50. and 3) sum of sine waves. thus increasing the power of the single (iii) .51. equals: H ( jω ) = ∫ h( t ) e − jωt dt = ∫ e − jωt dt = 0 0 ∞ T0 (i) 1 − e − jωT0 jω (38) Figure 8 presents the graph of the amplitude of |H(jω)| as a function of frequency.51 However. This also allows the amplitudes of the individual components to be increased by over 30%. in practice. frequency) space. only a limited frequency range may be studied because the higher frequency response is too small.E. First of all.54 The applied signal consists of a fundamental harmonic frequency f0 and a number of odd harmonics (2n+1)f0. Noise Perturbation White noise. This technique has been applied by Smith and coworkers to study electrode kinetics in the frequency range 10 to 500 Hz. such a function does not have uniform response in the Fourier (i.

we shall consider the case of an ideally polarizable electrode. 2.15 frequency components. Rs. A decrease of the standard deviation of the impedance by 25% was obtained in that way.2(i). An additional possibility is an optimization of the amplitudes. Figure 10. Analysis of such a circuit was presented in Section I. It is known that the response of electrochemical cells is different for different frequencies so that the response is weaker in the low frequency range and larger at high frequencies.55-61 Figure 9. that is they are selected in such a way that the response remains nearly constant. the electrodesolution interface and electrochemical reactions taking place on the electrode. it is possible to write the expression for the impedance for any mechanisms. An example of such optimization is displayed in Figure 10. higher noise is observed at low frequencies (for the same perturbation amplitude). therefore it may be applied to study impedances evolving with time (of course. III. impedance must be considered constant during the time of measurements). First. In this case the equivalent electrical model consists of the solution resistance. finally a finite-length diffusion. Semi-Infinite Linear Diffusion In general. The Ideally Polarizable Electrode An ideally polarizable electrode behaves as an ideal capacitor because there is no charge transfer across the solution-electrode boundary.g. For the reaction: (39) Ox + ne ¾ Red the current is described by: . Laplace transform of the applied perturbation and the obtained response to determine the impedance spectra. followed by semi-infinite diffusion in linear. spherical and cylindrical geometry and. in series with the double-layer capacitance.28. the response signal is much less sensitive to noise. It should be added that other types of analysis of the system responses were also used. 1. e. When the amplitudes of different frequencies are optimized. IMPEDANCE OF FARADAIC REACTIONS IN THE PRESENCE OF DIFFUSION Total electrode impedance consists of the contributions of the electrolyte. Cdl. The main advantage of the FFT technique is that the information is obtained quickly. Besides. The results of such optimization are shown in Figure 9. The procedure shown below is general and may be applied to other processes involving diffusion. The weakness of the FFT technique is that the response to individual frequencies is usually weaker than that when only one frequency is applied.

This linearization is a fundamental property of EIS. the current is a function of the potential and concentrations. is applied. and f = F/RT. When a small av perturbation signal.27. ∆E < 8/n mV6 peak-to peak.6. the oscillating potential and the concentrations may be written as: ~ ~ ~ ~ (41) ∆E = E exp( jωt ) . current and concentrations. therefore the amplitudes applied must be small. Higher harmonics analysis has also been described. CO. ∆CO and ∆CR only. respectively. ko and E0 are the standard rate constant and standard potential. ⎟ ⎜ ∂E∂C ⎟ ⎜ ∂E∂C ⎟ ⎝ O R⎠ ⎝ ⎝ O⎠ R⎠ It can be noticed that because eqn. and the symbol ∆ indicates a parameter oscillating periodically with time. In general. (40): (43) ∂i = − n 2 Ff αk f CO ( 0 ) + (1 − α )kbCR ( 0) ∂E [ ] ∂i ∂i = nFk f and = − nFk b ∂CR ∂CO (44) In order to find concentrations. In general. Because we are interested in the ac components of these parameters.62 The derivatives in eqn.e. ∆i = i exp( jωt ) ∆CO = CO exp( jωt ) and ∆CR = CR exp( jωt ) ~ ~ ~ ~ where E . we can solve equations for ∆E. n number of electrons. CO = CO. (40). Fick’s diffusion equation must be solved for ∆C. the current and concentrations oscillate around steady-state values: i = idc + ∆i . where the subscript dc indicates a parameter which changes only slowly with time i. i . and C R are the phasors of the voltage. (42) correspond to stationary conditions and may be obtained from eqn. either a steady-state term or the one that does not change with ω or its harmonics.16 i = nF[ k f CO (0) − kb CR (0)] (40) where kf and kb are the potential-dependent rate constants for the forward and backward reactions: kf = ko exp[-αnf(E-E0)] and kb = ko exp[(1-α)nf(E-E0)]. ∆E = E0 exp(jωt).. where n is the number of electrons exchanged in the reaction. eqn. and CR = CR . α is the transfer coefficient. (40) is linear with respect to the concentrations.7.. dc + ∆CR .. only first-order derivatives versus concentrations are different from zero. dc + ∆CO . ∆i. For small perturbations it is a good approximation to keep only the linear terms. CO(0) and CR(0) are the surface concentrations of the forms Ox and Red. and it may be represented as an infinite Taylor series: ⎛ ∂i ⎞ ⎛ ∂i ⎞ ⎛ ∂i ⎞ ∆i = ⎜ ⎟ ∆E + ⎜ ⎟ ∆CO + ⎜ ⎟ ∆C ⎝ ∂E ⎠ ⎝ ∂CO ⎠ ⎝ ∂C R ⎠ R 2 1 ⎛ ∂ 2i ⎞ 1 ⎛ ∂ 2i ⎞ 2 1⎛ ∂ i ⎞ 2 + ⎜ 2 ⎟ ( ∆E ) 2 + ⎜ 2 ⎟ ( ∆CO ) + ⎜ 2 ⎟ ( ∆C R ) ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ 2 ⎝ ∂E ⎠ 2 ⎝ ∂CO ⎠ 2 ⎝ ∂C R ⎠ (42) ⎛ ∂2i ⎞ ⎛ ∂2i ⎞ ⎛ ∂2i ⎞ ⎟ ∆E∆CO + ⎜ ⎟ ∆E∆C R + ⎜ +⎜ ⎜ ∂C ∂C ⎟ ∆CO ∆C R +. For semi-infinite linear diffusion the following equations must be solved: .

only a dc concentration gradient exists. (41) and (45) (46) ~ ∂∆CO / ∂t = jωCO exp( jωt ) eqn. (50). one gets: ~ ~ ~ i i ~ A = CO ( 0) = − and A' = C R (0) = nF jωDO nF jωDR Now. it is possible to make the substitutions into eqn. at x → ¼. conserving only linear terms: (54) . eqn. (42). dx nFDO (47) ~ ~ dC R i =− dx nFDR (48) (49) (50) ~ ~ dC O dC R DO + DR =0 dx dx x → ∞: ~ ~ C O → 0 and C R → 0 where. After further rearrangements one obtains: ~ ~ d 2 C R ⎛ jω ⎞ ~ d 2 CO ⎛ jω ⎞ ~ 2 ~ 2 ~ =⎜ =⎜ ⎟ CO = sO CO and ⎟ CR = sR CR 2 2 dx dx ⎝ DR ⎠ ⎝ DO ⎠ which have the following solutions: (51) ~ ~ CO ( x ) = A exp( − sO x ) + B exp( sO x ) and C R ( x ) = A' exp( − s R x ) + B' exp( s R x ) (52) For semi-infinite diffusion B and B’ are both equal to zero. In order to determine the constants A and A’. to fulfill the condition (50). (45) may be rearranged to: ~ ~ ~ ~ d 2 CO d2 CR and jω C R = DR jω C O = DO dx 2 dx 2 with the boundary conditions: ~ ~ dC O i x = 0: = . the condition at x = 0 must be considered: ~ ~ dC O i = −sO A = dx nFDO (53) and substituting sO and sR .17 ∂∆CO ∂ 2 ∆CO ∂∆CR ∂ 2 ∆CR = DO and = DR ∂t ∂x 2 ∂t ∂x 2 Taking into account eqn.

the surface concentrations are described by the Nernst equation: CO (0)/CR (0) = exp[nf(E-E0)] and eqns. (59) and (60) may be written as: ^ ^ ^ ⎛ RT 1 ⎜ ZW = ZW . Rct. which are contributions from ∂i / ∂Ci .O + ZW . R = Assuming that the process is dc reversible. are called impedances of mass transfer or semi-infinite Warburg impedance63-66. R = 2 2 + n F jω ⎝ DO CO ( 0 ) ⎞ 1 ⎟= DR CR ( 0 ) ⎠ 2 σ j ω (61) where σ is the mass transfer coefficient equal to the sum of the contributions of the forms Ox and Red: . consists of three terms: the first one comes from the derivative: ∂i / ∂E and is called charge-transfer resistance.O + ZW . and the two others. R Rct = RT 1 2 2 n F αk f CO ( 0) + (1 − α) k b C R ( 0) ^ ^ ^ ^ (57) (58) ^ RT ZW . ZW : Z f = Rct + ZW = Rct + ZW . Z f .O = 2 2 n F jωDO αk f CO ( 0 ) + (1 − α )kbCR ( 0 ) 1 [ kf ] (59) = ^ RT n2F 2 RT n2 F 2 jωDO {αC0 ( 0) + (1 − α )C R ( 0) exp[nf ( E − E 0 )]} exp[nf ( E − E 0 )] jωD R {αC0 ( 0 ) + (1 − α )C R ( 0 )exp[nf ( E − E 0 )]} (60) ZW .18 n2 F 2 ~ ~ i =− αk f CO ( 0 ) + (1 − α )k b C R ( 0 ) E − RT [ ] ~ ikf jωDO − ~ i kb jωD R (55) ~ ~ and the faradaic impedance equals Z f = − E / i (the negative sign arises from the assumed convention in which the cathodic current is positive): kf kb + 1+ ^ jωDO jωD R RT Zf = 2 2 n F αk f CO ( 0) + (1 − α )k b C R ( 0) (56) The total faradaic impedance.

Assuming that there is only the oxidized form initially in the solution (CO*).19 ⎡ RT 1 σ = σO + σ R = 2 2 + ⎢ n F 2 ⎢ DO CO ( 0 ) ⎣ Because 1 / ⎤ 1 ⎥ D R C R ( 0) ⎥ ⎦ (62) j = 1 / 2 (1 − j ) . The mass transfer impedance is called. eqn. the minimum of Rct is observed at the half-wave potential. in the case of semi-infinite linear diffusion. is equal to 0. (63) is obtained as before. the mass transfer impedance equals: ZW = 2 2 * n F CO jωDO ^ RT [1 + exp[nf ( E − E1/ 2 )]]2 exp[nf ( E − E1/ 2 )] (69) ⎡ nf ( E − E1/ 2 ) ⎤ 4 RT = 2 2 * cosh 2 ⎢ ⎥ 2 n F CO jωDO ⎢ ⎥ ⎣ ⎦ . R dCi i where. one gets: dE / dCO = RT / nFCO ( 0) and dE / dCR = − RT / nFCR ( 0) (65) After substitution of eqn. Substituting eqn. a semi-infinite Warburg impedance. (64) into (65). the surface concentrations may be estimated from11: (66) ξ * ξθ * CO ( 0 ) = CO and C R ( 0 ) = CO 1 + ξθ 1 + ξθ where ξ = (DO / DR )1/2 and θ = exp[nf(E-E0)] and ξθ = exp[nf(E-E1/2)]. (66) into (58) the charge-transfer resistance may be expressed as: RT RT 1 + exp[nf ( E − E1/ 2 )] 1 + exp[nf ( E − E1/ 2 )] Rct = 2 2 = 2 2 * * ξ n F kbCO n F k 0 CO ξ α exp[(1 − α )nf ( E − E1/ 2 )] and it has a minimum at: (67) α RT Es = E1/ 2 + ln nF 1 − α (68) For the processes for which the transfer coefficient. the mass transfer impedance may be written as: ( ) ^ (63) ZW = σω −1/ 2 − jσω −1/ 2 This equation may be also obtained directly by assuming that the charge-transfer reaction is reversible and calculating the mass transfer impedance from: ~ (64) ^ dE dE Ci ZW = = ∑ ~ di i = O. α.5. Similarly. from the Nernst equation.

e. It should be stressed that the Warburg impedance cannot be represented by a connection of simple R and C elements because of the non-integer power of frequency (ω--1/2) and it constitutes a distributed element which can only be approximated by an infinite series of simple electrical elements. The mass transfer impedance is a vector containing real and imaginary components which are identical. based on nonlinear regression analysis was described recently. Figure 11. tf/td = 2σ2Cdl/Rct.g. (68). He has found that the complex non-linear least-squares regression is capable of extracting kinetic information from impedance measurements when the ratio of the charge-transfer process time constant tf = RctCdl to the diffusion process time constant td = Rct2/(2σ2). This is so called Randles model.20 The Warburg impedance has a minimum at E1/2. eqn. The dependence of the mass transfer and charge-transfer impedances on the electrode potential is displayed in Figure 12. (47) should be substituted by13.71 It allows for determination of the number of break-point frequencies on the Bode magnitude plot for an arbitrary mechanism and. In the other limiting case. (67) and (69). In this case a semicircle is observed on the complex plane plots. In this case eqn. On the complex plane plot it is a straight line with a slope of 1 and intercept Rct. when working with dropping or hanging mercury electrodes. A procedure for assessing nonlinearities in the Randles circuit. Spherical Diffusion This case arises. The faradaic impedance is shown in Figure 11b (dashed line). and E1/2.67 Figure 11b-d also shows complex plane and Bode plots for the total electrode impedance in the presence of slow charge-transfer kinetics. 3. the equivalent circuit corresponds to that in Figure 4.68 VanderNoot69 studied poorly separated faradaic and diffusional processes. Rs. in consequence.70 A general approach to multistep mechanisms involving soluble species in semi-infinite diffusion was presented recently by Harrington. In the case when the surface and bulk concentrations are the same. respectively. A mechanism including two successive electron transfer reactions was analyzed by Armstrong and Firman. is obtained on the complex plane plots. Let us consider semi-infinite diffusion to a sphere of radius r0 with both oxidized and reduced forms soluble in the solution. that is when the mass transfer impedance may be neglected. for the determination of the reaction mechanism and kinetics. Figure 12.. is lower than or equal to 30. in series with the parallel connection of the double-layer capacitance. The charge-transfer and mass transfer impedances have a minimum at Es. with a slope of 1.65.64. The total electrode impedance consists of the solution resistance. that is the phase-angle: '' ' ϕ = atan( ZW / ZW ) = atan( −1) = −450 . when the charge-transfer resistance is neglected (reversible case) a straight line.14: ~ ~ ⎞ ~ ~ ⎞ ⎛ 2~ ⎛ 2~ ∂ CO ∂CR ⎜ ∂ CO 2 ∂ CO ⎟ ⎜ ∂ CR 2 ∂CR ⎟ = DO ⎜ + and = DR ⎜ + ∂t r ∂r ⎟ ∂t r ∂r ⎟ (70) ⎟ ⎟ ⎜ ∂r 2 ⎜ ∂r 2 ⎠ ⎠ ⎝ ⎝ . according to eqns. and faradaic impedance. Cdl. Figure 11a.

On the Bode phase-angle graph a maximum is observed at low frequencies. (51). (64). by substitution ~ = rCO and ~ ~ = rC : v R ~ ~ d 2u d 2v 2 ~ 2 ~ = sO u and = sR v 2 2 dx dx (71) The solution is: ~ = A exp( − jω / D r ) and ~ = A exp( − jω / D r ) u v O R (72) ~ ~ ~ Taking into account that du / dr = CO + rdCR / dr and that at r = r0 (at the electrode surface) ~ ~ dCO / dr = i / nFDO the following solutions are obtained: ~ ~ ~ ~ i r0 i r0 and C R ( 0 ) = − CO ( 0) = − nFDO (1 + r0 jω / DO ) nFD R (1 + r0 jω / D R ) (73) The mass transfer impedance may be obtained from eqn. eqn.21 ~ u These equations may be rearranged into a simpler form. similar to the case of linear diffusion: ^ RT r0 ZW = ∑ 2 2 jω ⎞ i = O. R ∑ σ iω −1/ 2 1 + yi − j 1 + yi + 0. Spherical mass transfer causes formation of a depressed semicircle at low frequencies instead of linear behavior observed for linear semi-infinite diffusion. Figure 13. For very small electrodes (ultramicroelectrodes) or low frequencies the mass transfer impedances become negligible and the dc current becomes stationary. Assuming a reversible dc process one obtains.5 yi2 yi = 1 r0 2 Di ω (76) Influence of the nonlinearity of diffusion on the observed complex plane plots is shown in Figure 13. R n F Di Ci ( 0 ) ⎛ (74) ⎟ ⎜1 + r0 Di ⎠ ⎝ or (75) ZW = where ^ i = O. .

(64) one may get: i nF jωDO K1[ z 0 ] (82) ^ dE CO RT ZW = = 2 2 ~ * dCO n F DO CO i ~ r0 K 0 [ z 0 ] z 0 K1[ z 0 ] (83) The function in eqn. The solution is shown here for the oxidized form only: ~ ~ ⎞ ⎛ 2~ ∂ CO ∂ CO 1 ∂ CO ⎟ ⎜ = DO ⎜ + r ∂r ⎟ ∂t (77) ⎜ ∂r 2 ⎟ ⎝ ⎠ Rearrangement for the oscillating concentration. that is CO → 0 when r → 4. eqn. At the electrode surface. leads to: d 2 CO 1 d C O ⎛ jω ⎞ ~ + −⎜ ⎟ CO = 0 r dr ⎝ DO ⎠ dr 2 ~ ~ (78) Substitution of z = r (jω/DO)1/2 gives: d 2 CO 1 d CO ~ + − CO = 0 z dz dz 2 ~ ~ ~ (79) This is a modified Bessel equation of zero order with a general solution74: C O = AI 0 [z ] + BK 0 [z ] (80) where A and B are constants and I0 and K0 are zero-order modified Bessel functions. (46). In this case the diffusion equation is similar to that for spherical diffusion. the latter is simpler. Methods presented by both groups give the same results.22 4. using eqns. using eqn. (70).73 and Jacobsen and West74. Taking into ~ account semi-infinite diffusion conditions. Cylindrical Electrodes An example of cylindrical diffusion is a diffusion towards a conducting wire. The corresponding complex plane plots are shown in Figure . Solutions for cylindrical electrodes have been given by Fleischmann et al. (83) may be evaluated using Mathematica. leads to A = 0.72. Maple or specific subroutines for complex modified Bessel functions. however. r = r0: ~ d CO = dr where z0 = r0(jω/DO)1/2 and: jω i BK1[ z 0 ] = DO nFDO ~ (81) ~ B= Then.

The complex plane plots for the diffusion to a disk are shown in Figure 15 exhibit a flattened semicircle. J1 is the Bessel function of the first kind and first order. The first term in Z 'f corresponds to the charge-transfer resistance. Figure 15. for rotating disk electrodes. They obtained the following equations describing the faradaic impedance in the case of a slow charge transfer when only Ox is initially present in the solution. At low frequencies cylindrical diffusion produces a constant imaginary impedance component. 72. and tanΘ = 1/β2.23 14. In such cases. etc. l 2 =D/ω. 6. for a thin mercury layer deposited on surfaces. (84)-(85) were tabulated in ref. In this case the differential equation corresponds to the normal and radial diffusion (two-dimensional) to the electrode. Fleischmann and Pons72. observed for polymer electrodes. in eqn.72. Finite-Length Diffusion In many cases the diffusion is not semi-infinite. for example. 73 as functions Φ4 and Φ5 of (a2ω/D) and in a different form in ref.73. at low frequencies. Disk Electrodes The solution for disk electrodes was presented by Fleischmann et al. (52) parameters B and B’ are not equal to zero. its concentration being * CO : ⎡ ⎛ a ⎞ ⎤ cos(Θ / 2)dβ Z 'f = Rct + 2 2 1/ 2 1/ 2 * ∫ ⎢ J 1 ⎜ β ⎟ ⎥ n F DO ω CO 0 ⎣ ⎝ l ⎠ ⎦ β 1 + β 4 1/ 4 4 RT ∞ 2 ( ) (84) Z '' f ⎡ ⎛ a ⎞ ⎤ sin(Θ / 2)dβ = 2 2 1/ 2 1/ 2 * ∫ ⎢ J 1 ⎜ β ⎟ ⎥ n F DO ω CO 0 ⎣ ⎝ l ⎠ ⎦ β 1 + β 4 1/ 4 4 RT ∞ 2 ( ) (85) where a is the disk radius. The integrals in eqns. Two cases may be distinguished for finite-length diffusion depending on the condition at the boundary located at a distance l form the electrode: . 5. This case is. Figure 14.75 also considered diffusion to microring electrodes. At sufficiently high frequencies the results are similar to those for linear diffusion whereas. the impedance becomes real as for spherical electrodes.73.

observed in the case of conducting polymers. It is observed. This is the so-called conducting or transmissive boundary. for example for the rotating disk electrode. where the diffusion layer thickness is determined by the rotation rate.24 1) transfer of electroactive species is possible at x = l. (64) gives the mass transfer impedances: ^ ⎛ jω ⎞ ⎛ jω ⎞ RT RT tanh ⎜ tanh ⎜ ZW = 2 2 l⎟ + 2 2 l⎟ ⎝ DO ⎠ n F CR ( 0) jωDR ⎝ DR ⎠ n F CO ( 0 ) jωDO or. A’. Transmissive Boundary In order to determine the constants A. This is the so called reflecting boundary. At low frequencies the function tanh(x) ≅ x. but dC(l)/dx ≠ 0. and ZW becomes real and frequency independent: . And 2) no charge transfer is possible at x = l. and C(l) =0. and B’ in eqn. this becomes: ^ ⎛ jω ⎞ σ ZW = tanh ⎜ l ⎟ (1 − j ) ω ⎝ D ⎠ (90) (91) (92) The Warburg impedance is displayed in Figure 16 a and the total impedance in Figure 16b-d. assuming that the diffusion coefficients of O and R are the same. (52). boundary conditions must be used. that is dC(l)/dx = 0. At the electrode surface the concentration gradients are: ~ ~ ~ ~ (86) dCO dCR i i x = 0: = − sO A + sO B = and = − sR A'+ sR B' = − dx x = 0 nFDO dx x = 0 nFDR (i) ~ ~ x = l: CO = A exp( − sO l ) + B exp( sO l ) = 0 and CR = A'exp( − sR l ) + B'exp( sR l ) = 0 which leads to: ~ ~ exp( sO l ) exp( − sO l ) i i A=− and B = nF jωDO exp( sO l ) + exp( − sO l ) nF jωDO exp( sO l ) + exp( − sO l ) (87) (88) A' = ~ ~ exp( sR l ) exp( sR l ) i i and B' = − nF jωDR exp( sR l ) + exp( − sR l ) nF jωDR exp( sR l ) + exp( − sR l ) (89) The surface concentrations are: ~ ~ ⎛ jω ⎞ ⎛ jω ⎞ ~ ~ i i l ⎟ and CR ( 0) = l⎟ CO ( 0) = − tanh⎜ tanh⎜ nF jωDO nF jωDR ⎝ DO ⎠ ⎝ DR ⎠ and substitution to eqn. B.

the concentration gradient being equal to zero: ~ (95) dCO = − sO A exp( − sO l ) + sO B exp( sO l ) = 0 dx and ~ (96) dC R = − s R A'exp( − s R l ) + s R B'exp( s R l ) = 0 dx They give: ~ ~ (97) ⎛ jω ⎞ ⎛ jω ⎞ ~ i i ~ l⎟ l⎟ CO ( 0) = − coth ⎜ C R (0) = − coth⎜ nF jωDO nF jωDR ⎝ DO ⎠ ⎝ DR ⎠ and and (98) ⎛ jω ⎞ ⎛ jω ⎞ RT RT l⎟ l⎟ + 2 2 ZW = 2 2 coth ⎜ coth ⎜ n F CO ( 0) jωDO ⎝ DR ⎠ ⎝ DO ⎠ n F C R ( 0) jωD R or assuming equal diffusion coefficients: (99) ⎛ jω ⎞ σ ZW = coth ⎜ l ⎟ (1 − j ) ω ⎝ D ⎠ Figure 16. For these conditions the Warburg impedance corresponds to a series connection of the resistance RW = Z’W and the capacitance CW = l / 2 D σ . (ii) .25 2σl D Therefore the low frequency limit of the electrode impedance equals: ^ 2σl Z = Rs + Rct + D ZW (ω → 0) = ^ (93) (94) Reflective Boundary In this case the boundary conditions at x = 0 are the same as for eqn. At low frequencies coth(x) ≅ 1/x + x/3 and ZW becomes: (100) 2σl 2 Dσ ZW = − j ωl 3 D The imaginary part of the impedance goes to infinity and the real part to a constant value which indicates that no charge transfer occurs at low frequencies and the electrode behavior is purely capacitive. The corresponding Warburg and total electrode impedances are shown in Figure 16. (86) but at x = l they are different. The limiting value of the real part of the total cell impedance equals: (101) Z ' t = Rs + Rct + 2σl / 3 D The problem of finite length diffusion in spherical and cylindrical symmetry was solved by Jacobsen and West74.

Randles’ Analysis64. on the electrode potential. Figure 17 presents the dependence of the total admittances of a process limited by the diffusion of electroactive species to and from the electrode and the kinetics of the charge transfer process. Proper rearrangement leads to: ξ kb σ exp[nf [E − E1/ 2 ] = Rct 1 + exp[nf [E − E1/ 2 ] 2DO (104) . ac voltammetry or ac polarography is usually used and the impedance curves are determined from a series of ac voltammetric curves registered at different frequencies.65. to maintain Rs constant). This may be possible when this distance is large or the solution in the cell is exchanged without changing the electrode configuration. the real and imaginary components of the faradaic impedance are plotted against ω-1/2. Figure 17. 7.67 The experimentally measured ac current or total admittances are functions of the electrode potential. The dependence of Rct on potential allows the determination of the standard rate constant and the transfer coefficient. that is from potentials more positive and more negative than the peak potential. keeping the same distance between the working electrode and the tip of the Luggin capillary (i. Information on the kinetics of electrode process is included in the faradaic impedance.e.65 ^ (102) 1 1 =Yf = ^ − jωCdl ^ Zf Z t − Rs (i) It should be kept in mind that for the calculation of the impedances from admittances the following equation must be used: (103) Y '' Y' Z'= and Z '' = − 2 2 2 2 ( Y ') + ( Y '') (Y ') + (Y ' ') The parameters Rs and Cdl must be determined in a separate experiment in a solution containing supporting electrolyte only. They form two parallel lines with slopes of σ and intercepts of Z 'f = Rct and zero.26 Analysis of Impedance Data in the Case of Semi-Infinite Diffusion: Determination of the Kinetic Parameters In the case of the charge transfer to diffusing species. Figure 18. It may be simply determined from the total electrode impedance:64. The other possibility is to extrapolate the admittance (or impedance) from the range where the faradaic impedance is negligible. In order to eliminate influence of the depolarizer concentration one can also evaluate the ratio of the slope to intercept of the Randles' plot. Then. The methods of analysis of such curves are described below.

From eqns.27 from which the rate constant as a function of the electrode potential may be evaluated. (58). The difference between the potential of the maximum of . (57). It has a maximum at the potential Es described by eqn. (67) and (69) one may get: (109) Z 'f R cot ϕ = '' = 1 + ct ω1/ 2 σ Zf (iii) or after substitution: (1− α )/ 2 α / 2 1/ 2 DR ω cot ϕ = 1 + 2 DO (110) k0 exp[(1 − α ) f ( E − E1/ 2 )]{1 + exp[− nf ( E − E1/ 2 )] It is clear that cot ϕ depends linearly on ω1/2 and on electrode potential. De Levie-Husovsky Analysis De Levie and Husovsky76 have proposed method based on analysis of faradaic admittances. Analysis of cot ϕ Another type of determination of kinetic parameters is based on the determination of the phase-angle of the faradaic impedance. The Faradaic admittance may be easily determined from the total impedance: ^ (106) 1 Yf = ^ − jωCdl Z t − Rs The ratio of the imaginary to real faradaic admittances equals: Y f'' ζ = ' Yf 1− ζ where ζ= kf 2ωDO + kb = 2ωDR kf 2ωDO (ii) (107) (1 + exp[nf ( E − E1/ 2 )]) (108) From the dependence of log [ζ/{1+exp[nf(E .E1/2)]}] versus E the forward rate constant is easily determined. At the reversible half-wave potential this ratio gives directly the standard rate constant: (105) ⎛ σ ⎞ 2ξ α k0 = ⎜ ⎟ ⎝ Rct ⎠ E 2 DO 1/ 2 Figure 18.

79 . illustrated in Figure 20.28 cot ϕ and E1/2 allows for estimation of the transfer coefficient α. The potential dependence of cot ϕ is shown in Figure 19. (iv) Sluyter's Analysis Complex plane plots obtained in the case of a slow charge transfer with semi-infinite diffusion were presented in Figure 11a. They represent a semicircle (at high frequencies) followed by a straight line. General equations describing total real and imaginary impedance were analyzed by Sluyters and coworkers26.78. Figure 20. The total electrode impedance is given as: ^ 1 Z t = Rs + 1 + jωCdl (113) ^ Zf where the faradaic impedance is described by eqn. gives access to the standard rate constant k0. (57). Figure 19.27. The maximal value of cot ϕ is described by: [cot ϕ]max = 1 + α (2D1−α DR )1/ 2 O ⎡⎛ α ⎞ − α ⎛ α ⎞ β ⎤ k 0 ⎢⎜ ⎟ +⎜ ⎟ ⎥ ⎝ β⎠ ⎥ ⎢⎝ β ⎠ ⎦ ⎣ ω1/ 2 (111) while at the half-wave potential it is given by: ⎛ D1− α D α ⎞ R⎟ cot ϕ ] E = 1+ ⎜ O [ ⎟ ⎜ 1/ 2 2 ⎠ ⎝ 1/ 2 ω1/ 2 k0 (112) Analysis of cot ϕ as a function of ω1/2. This equation leads to rather complicated expressions for the real and imaginary parts of the total impedance: Z ' = Rs + and (σω1/ 2Cdl + 1) Rct + σω −1/ 2 2 2 + ω 2Cdl Rct + σω ( −1/ 2 2 ) (114) .

In this Section reactions involving adsorbed species in the absence of diffusion limitations will be presented. reactions of charge transfer to diffusing species in solution were considered.β ) f ( E − E1 ) 0 0 − k1 ΓB e 1 v1 = k1 Γs a A e 1 and 0 0 (119) -β f ( E − E2 ) (1. Although in simple cases the process parameters may be obtained graphically. 1. IMPEDANCE OF A FARADAIC REACTION INVOLVING ADSORPTION OF REACTING SPECIES In Section III.29 )2 + σ2Cdl + σω −1/ 2 Z '' = 2 (σω1/ 2Cdl + 1)2 + ω 2Cdl ( Rct + σω −1/ 2 )2 ωCdl Rct + σω −1/ 2 ( (115) The graphical illustration of these equations is presented in Figure 11b. The rates of these reactions may be written. Faradaic Reaction Involving One Adsorbed Species Let us consider the following reactions: Asol + e ⇔ Bads k −1 k2 k1 (116) Bads + e ⇔ Csol k− 2 (117) where index sol denotes species in solution and ads adsorbed species. The latter condition means that the concentration gradient at the electrode surface is negligible. At the equilibrium potential. IV. the best way to analyze the impedances is.β 2 ) f ( E − E 2 ) 0 0 − k 2 Γs a C e v 2 = k 2 ΓB e 2 where k i0 are the standard rate constants of these two reactions.β f ( E − E1 ) (1. aA and aC are the surface concentrations of A and C (assumed as equal to the bulk concentrations) and Ei0 are the standard red-ox potentials of these reactions. by the complex nonlinear least-squares approximation technique. that is the concentrations in solution are large enough and/or currents low. Reactions involving one and two and more adsorbed species will be considered subsequently. βi are the symmetry coefficients. Eeq the net rates of both reactions are null and the following relations are obtained: e e 0 f ( E eq − E1 ) 0 f ( E eq − E 2 ) 0 = Γs0 a A / Γ A = (1 − Θ 0 )a A / Θ 0 0 0 = ΓB / Γs a C = Θ 0 / (1 − Θ 0 )a C (120) (121) . ΓA and Γs are the surface concentrations of the species A and of free adsorption sites. however. as: 0 0 (118) . assuming Langmuir adsorption isotherm for B. respectively. Rct and the Warburg coefficient σ. From such fits the following parameters may be obtained: Rs. Cdl.

eqn. k −1 e (1-β1 ) fη . which implies an additional condition: (128) k1k 2 =1 k −1k − 2 that is there are only three independent rate constants in the system. giving: (130) ⎡ ⎤ ⎛ ∂i ⎞ ⎛ ∂r ⎞ ⎛ ∂r ⎞ ⎛ ∂i ⎞ ∆i = ⎜ ⎟ ∆η + ⎜ ⎟ ∆Θ = F ⎢⎜ 0 ⎟ ∆η + ⎜ 0 ⎟ ∆Θ ⎥ ⎝ ∂Θ ⎠ η ⎝ ∂Θ ⎠ η ⎝ ∂η ⎠ Θ ⎢⎝ ∂η ⎠ Θ ⎥ ⎣ ⎦ . (120) and (121) into (118) and (119). In order to calculate the reaction impedance equation describing current i(η. (129). and surface concentration was introduced: Γi = Θ i Γ∞ and Γ∞ is the maximal surface concentration. k − 2 = k − 2 e (1-β 2 ) fη The total observed current is: i = F ( v1 + v 2 ) = Fr0 (127) At the equilibrium potential rates of reactions (2) and (3) are equal to zero. Θ.Θ). eqn. where η is the overpotential.Θ). (127). gives: β − (1−β1 ) (1-β1 ) fη 1−β 1−β 0 0 v1 = k1 Γ∞ a A 1 (1 − Θ 0 )−β1 Θ 01 (1 − Θ )e −β1 fη − k1 Γ∞ a A 1 (1 − Θ 0 )1−β1 Θ 0 Θe = k1(1 − Θ )e-β1 fη − k−1Θe(1-β1 ) fη → ← = k 1(1 − Θ ) − k −1 Θ (122) and 0 0 v 2 = k 2 Γ∞ a C2 (1 − Θ 0 ) β2 Θ 0 2 Θe −β2 fη − k 2 Γ∞ a C2 (1 − Θ 0 ) − (1−β2 ) Θ 0 β −β β 1−β 2 (1 − Θ)e (1-β1 ) fη = k 2 Θe -β1 fη − k −2 (1 − Θ)e (1-β1 ) fη = k 2 Θ − k −2 (1 − Θ) where the following new rate constants were introduced: → ← (123) k1 = 1− β ⎛ 0 k1 Γ∞ a A 1 ⎜ β Θ0 ⎞ 1 ⎟ ⎝ 1− Θ0 ⎠ β2 k −1 = 1− β ⎛ 1 − Θ 0 ⎞ 0 k1 Γ∞ a A 1 ⎜ ⎟ ⎝ Θ0 ⎠ 1− β1 (124) k2 = → β ⎛ 1 − Θ0 ⎞ 0 k 2 Γ∞ a C2 ⎜ ⎟ ⎝ Θ0 ⎠ k− 2 = β ⎛ 0 k 2 Γ∞ a C2 ⎜ Θ0 ⎞ ⎟ ⎝ 1 − Θ0 ⎠ 1− β 2 (125) ← → → (126) k 1 = k1e -β1 fη . and r1 (η.30 where index 0 indicates equilibrium conditions and a relation between surface coverage. k 2 = k 2 e -β 2 fη . Under steady-state conditions the rate of formation of adsorbed species B is the same as the rate of their consumption. therefore: (129) dΓ B dΘ σ1 dΘ = Γ∞ = = r1 = v1 − v 2 = 0 dt dt F dt where σ1 = F Γ∞ is the charge necessary for the total surface coverage by B. should be linearized. Introduction of eqns. and taking into account that: E − Ei0 = E − Eeq + Eeq − Ei0 = η + Eeq − Ei0 .

Knowing the faradaic impedance. may be determined using eqn. (127) and (129): → ← ← ⎧ → ⎫ A = Ff ⎨β 1 k 1 (1 − Θ) + (1 − β 1 ) k −1 Θ + β 2 k 2 Θ + (1 − β 2 ) k −2 (1 − Θ) ⎬ (136) ⎩ ⎭ B= → ← ⎞ F2 f ⎛ → ← ⎜ − k 1 − k −1 + k 2 + k −2 ⎟ × ⎠ σ1 ⎝ ← → ← ⎧ → ⎫ ⎨β 1 k 1 (1 − Θ) + (1 − β 1 ) k −1 Θ − β 2 k 2 Θ − (1 − β 2 ) k −2 (1 − Θ) ⎬ ⎩ ⎭ C= (137) → ← ⎞ F ⎛→ ← ⎜ k 1 + k −1 + k 2 + k −2 ⎟ (138) ⎠ σ1 ⎝ It is evident that parameters A and C are always positive and B may be positive or negative. The derivatives in eqn. (135) is the inverse of the charge-transfer resistance: A = 1/Rct. (133)-(134) gives the faradaic admittance as: F 2 ⎛ ∂r0 ⎞ ⎛ ∂r1 ⎞ ~ ⎜ ⎟ ⎜ ⎟ ^ σ1 ⎝ ∂Θ ⎠ η ⎝ ∂η ⎠ Θ ⎛ ∂r0 ⎞ B i Y f = − ~ = −F⎜ = A+ ⎟ − F ⎛ ∂r1 ⎞ ⎝ ∂η ⎠ Θ jω + C jω − ⎜ ⎟ η ⎝ ∂Θ ⎠ η σ1 (135) The first term in eqn. (122). the total electrode impedance.81 and Diard et al.2): ~ ~ ~ ~ (132) ∆i = i exp( jωt ) . (123).82 Taking into account that (see Section III. and 2. Bai and Conway84 . Z f . (130)-(131) to fundamental harmonic impedances was recently discussed by Darowicki80. depending on the values of rate constants. ∆ri = r i exp( jωt ) ∆η = η exp( jωt ) and ∆Θ = Θ exp( jωt ) one obtains: ⎛ ∂r ⎞ ~ ⎛ ∂r ⎞ ~ i ~ = r 0 = ⎜ 0 ⎟ η+ ⎜ 0 ⎟ Θ ⎝ ∂Θ ⎠ η ⎝ ∂η ⎠ Θ F ~ (133) and ~ ⎛ ∂r ⎞ ~ ⎛ ∂r ⎞ ~ σ1 jω Θ = ⎜ 1 ⎟ η+ ⎜ 1 ⎟ Θ ⎝ ∂Θ ⎠ η ⎝ ∂η ⎠ Θ F (134) ~ Elimination of Θ from eqns. Analysis of the complex plane plots in such a case was presented by Cao83. (135) may be easily evaluated from eqns. (135).31 ⎛ ∂r ⎞ σ1 d∆Θ ⎛ ∂r ⎞ = ∆r1 = ⎜ 1 ⎟ ∆η + ⎜ 1 ⎟ ∆Θ ⎝ ∂Θ ⎠ η ⎝ ∂η ⎠ Θ F dt (131) A model containing higher-order term contributions in eqns. (102). Impedance Plots in the Case of One Adsorbed Species The faradaic admittance of the reactions (116) and (117) is described by eqn.

. The complex plane plots are analogous to those shown in Figure 4 and represent one capacitive semicircle. The complete equivalent circuit in this case is represented in Figure 21. Equation (24) represents a series connection of the charge-transfer resistance with parallel connection of the resistance Ra and pseudocapacitance Ca. b) C . while the limit at zero frequency is called the polarization resistance. Rp: (142) lim Z = R and lim Z =R ω→ 0 2 ( f) p ω→∞ ( f) t In our case Rp = Rct +R ct |B|/ (C . Two general cases should be considered depending on the sign of the parameter B: I) B < 0 In this case the faradaic admittance may be written as: ^ B 1 Yf = − (139) R ct jω + C − − It changes from R ct1 at very high frequencies to R ct1 − B / C at very low frequencies. The faradaic impedance is described as: 2 ^ R ct B Ra 1 1 Zf = = R ct + = R ct + = R ct + (140) ^ 1 jω + C − R ct B 1 + jωRa Ca + jωCa Yf Ra where: 2 (141) R ct B 1 Ra = and Ca = 2 C − R ct B R ct B The limit of faradaic impedance at infinite frequency is also called the transfer impedance. The observed complex plane plots depend on the sign of the denominator of Ra. Figure 22. Rt.Rct |B| > 0 In this case all the elements are positive and the faradaic impedance represents one semicircle on the complex plane plots. a) C . The corresponding complex plane plots are presented in Figure 23. When A >> |B|/C the faradaic impedance is equal to Rct . When Ca >> Cdl the total impedance represents two semicircles. Figure 22.Rct |B| = 0 In this case the faradaic impedance is: ^ (143) 1 Z f = R ct − j ωCa which corresponds to a series connection of Rct and Cdl . Figure 21. see Figure 22.32 presented three-dimensional plots for such a reaction.Rct |B|) and Rt = Rct.

concentrations. however useful it may sound. 8). In fact. Figure 25. Figure 4. etc. II) B = 0 When B = 0 the faradaic impedance is real and equals Rct . diffusion coefficients. In this case Rp = RctRo/(Rct + Ro). III) B > 0 In this case the faradaic admittance is given by: ^ 1 1 1 B + = + Yf = (144) R ct jω + C R ct R o + jωL with (145) R o = C / B and L = 1 / B and the faradaic impedance by: ^ 1 Zf = (146) 1 1 + R ct R o + jωL which corresponds to the parallel connection of the charge-transfer resistance with series connection of the resistance Ro and inductance L. These plots are described by various equivalent circuits. It seems that the electrical representation of the faradaic impedance. The above analysis shows that in the simple case of one adsorbed intermediate (according to Langmuirian adsorption) various complex plane plots may be obtained. In fact . potential. Figure 24. which are only the electrical representations of the interfacial phenomena. c) C . all these parameters are highly nonlinear. The system may be described in a simpler way directly by the equations describing impedances or admittances (see also chap. inductances or resistances in the circuit (faradaic process). there are no real capacitances. transfer coefficients. In a system following Frumkin adsorption isotherm discontinuous impedances may be obtained. Besides. that is they depend on the electrode potential.85 determined conditions under which such a low frequency pseudo-inductive loop may be found. The equivalent circuit and the corresponding complex plane plots of faradaic and total impedances are shown in Figure 25.33 Figure 23.86 It should be added that for the system involving one adsorbed species described above there are two sets of kinetic parameters giving the same experimental curves87. is not necessary in the description of the system. Diard et al.Rct |B| < 0 In this case the parameter Ra is negative and the corresponding complex plane plots are displayed in Figure 24. One semicircle is observed in the complex plane plots. Figure 21b. depending on the relative values of the system parameters. These parameters originate from the behavior of the kinetic equations and they are functions of the rate constants.

From the condition at the equilibrium potential: v1 = v2 = v3 = 0 the following condition for the rate constants is obtained (cf. and rate constants k3 and k-3 are potential independent. 128): k1k 2 k 3 =1 (153) k − 1k − 2 k − 3 The charge is exchanged in reactions (1) and (2) only.88-91 The impedance of a more complex process involving coupling between adsorption and diffusion was studied by Armstrong and coworkers. Θ1 and Θ2. (154)-(156) and introduction of phasors gives: .92. oxygen evolution.95 Asol + e ⇔ Bads k −1 k2 k1 (147) Bads + e ⇔ Cads k− 2 (148) and Cads ⇔ Dsol k −3 k3 (149) The rates of eqns. The problem of identifiability and distinguishability of electrode processes was further studied by Bertier et al. (122) and (123): → ← (150) v1 = k 1 (1 − Θ1 − Θ 2 ) − k −1 Θ1 v 2 = k 2 Θ1 − k − 2 Θ 2 and → ← (151) v 3 = k 3Θ 2 − k − 3 (1 − Θ1 − Θ 2 ) (152) where Θ1 and Θ2 are the surface coverages by B and C. therefore the total current is given as: i = F ( v1 + v 2 ) = Fr0 (154) Mass balance for Θ1 and Θ2 gives.93 3. and β1 ↔ β2 leaves the same values of the dc current. Faradaic Impedance in the Case Involving Two Adsorbed Species Typical examples of processes involving two or more adsorbed species are reactions of corrosion or anodic dissolution of metals. the electrochemical reactions may be written as:84. etc. similarly to eqn.94. k-1 ↔ k-2. r1 and r2 are the functions of η. eqn. linearization of eqns. the charge-transfer resistance and the parameters B and C. similarly to eqns. respectively.34 permutation of the kinetic parameters: k1 ↔ k2. (147)-(149) may be expressed with respect to the equilibrium potential. (129): (155) σ1 dΘ1 = v1 − v 2 = r1 F dt (156) σ 2 dΘ 2 = v 2 − v 3 = r2 F dt Taking into account that r0. In the case of two adsorbed species B and C.

35 ~ ⎡ ~ ⎛ ∂r ⎞ ~ ~ ⎤ ⎛ ∂r ⎞ ⎛ ∂r ⎞ η+ ⎜ 0 ⎟ i = F ⎢⎜ 0 ⎟ Θ1 + ⎜ 0 ⎟ Θ2 ⎥ ⎥ ⎢⎝ ∂η ⎠ Θ . The faradaic impedance may be obtained from eqn. Θ ⎝ ∂Θ 2 ⎠ η. Θ Rct 1 2 (161) ⎧ ⎛ ∂r0 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r0 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r2 ⎞ ⎫ ⎟⎜ ⎟⎜ ⎟ +⎜ ⎟⎜ ⎟⎜ ⎟⎪ ⎪− ⎜ F 3 ⎪ ⎝ ∂Θ1 ⎠ ⎝ ∂η ⎠ ⎝ ∂Θ 2 ⎠ ⎝ ∂Θ1 ⎠ ⎝ ∂Θ 2 ⎠ ⎝ ∂η ⎠ ⎪ B=− ⎨ ⎬ σ1σ 2 ⎪ ⎛ ∂r0 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r2 ⎞ ⎛ ∂r0 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r2 ⎞ ⎪ ⎪− ⎜ ∂Θ ⎟ ⎜ ∂Θ ⎟ ⎜ ∂η ⎟ + ⎜ ∂Θ ⎟ ⎜ ∂η ⎟ ⎜ ∂Θ ⎟ ⎪ ⎠ ⎝ 2 ⎠⎝ ⎠⎝ 1⎠⎭ ⎩ ⎝ 2 ⎠⎝ 1⎠⎝ (162) ⎡ 1 ⎛ ∂r ⎞ ⎛ ∂r ⎞ 1 ⎛ ∂r0 ⎞ ⎛ ∂r2 ⎞ ⎤ C = −F 2 ⎢ ⎜ 0 ⎟⎜ 1 ⎟ + ⎜ ⎟⎜ ⎟⎥ ⎣ σ1 ⎝ ∂Θ1 ⎠ ⎝ ∂η ⎠ σ 2 ⎝ ∂Θ 2 ⎠ ⎝ ∂η ⎠ ⎦ ⎡ 1 ⎛ ∂r ⎞ 1 ⎛ ∂r2 ⎞ ⎤ D = −F ⎢ ⎜ 1 ⎟ + ⎜ ⎟⎥ ⎣ σ1 ⎝ ∂Θ1 ⎠ σ 2 ⎝ ∂Θ 2 ⎠ ⎦ and (163) (164) ⎡⎛ ∂r1 ⎞ ⎛ ∂r2 ⎞ ⎛ ∂r1 ⎞ ⎛ ∂r2 ⎞ ⎤ ⎟⎜ ⎟ −⎜ ⎟⎜ ⎟⎥ (165) ⎢⎜ ⎣⎝ ∂Θ1 ⎠ ⎝ ∂Θ 2 ⎠ ⎝ ∂Θ 2 ⎠ ⎝ ∂Θ1 ⎠ ⎦ where the negative sign before the parameters A. Θ ⎝ ∂Θ 2 ⎠ η. (157)-(159) present a system of three equations with three unknowns: Θ1. Θ F ⎝ ∂Θ1 ⎠ η. (160) as: F2 E= σ1σ 2 . D and E are always positive and parameters D and E may be positive or negative. B and C originates from the current definition (positive current for reduction).Θ 1 2 2 1 (158) and ~ ⎛ ∂r ⎞ ~ ~ ⎛ ∂r ⎞ σ ~ ⎛ ∂r ⎞ jω 1 Θ 2 = ⎜ 2 ⎟ η+ ⎜ 2 ⎟ Θ1 + ⎜ 2 ⎟ Θ2 ⎝ ∂η ⎠ Θ . If the reactions (147)-(149) are written as oxidations this sign should be omitted. Θ 1 2 2 1 ⎦ ⎣ (157) ~ ⎛ ∂r ⎞ ~ ~ ⎛ ∂r ⎞ σ ~ ⎛ ∂r ⎞ jω 1 Θ1 = ⎜ 1 ⎟ η+ ⎜ 1 ⎟ Θ1 + ⎜ 1 ⎟ Θ2 ⎝ ∂η ⎠ Θ . Θ 1 2 2 1 ~ ~ (159) Eqns.Θ ⎝ ∂Θ 2 ⎠ η. Θ ⎝ ∂Θ1 ⎠ η.Θ F ⎝ ∂Θ1 ⎠ η. The faradaic admittance is determined as: i B + jωC Y f = ^ = ~ = A+ jωD − ω 2 + E Zf η ^ 1 ~ (160) where: A= ⎛ ∂r ⎞ 1 = −F⎜ 0 ⎟ ⎝ ∂η ⎠ Θ . Calculation of the derivatives show that parameters A. Θ2 and i / η .

(169) ⎛ ∂r1 ⎞ ~ ⎡⎛ ∂r1 ⎞ σ ⎤ ~ ⎛ ∂r ⎞ ~ ⎛ ∂r ⎞ ~ ⎟ − jω 1 ⎥ Θ1 + ⎜ 1 ⎟ Θ 2 + ⎜ 1 ⎟ Θ 3 = 0 ⎜ ⎟ η+ ⎢⎜ ⎝ ∂η ⎠ F⎦ ⎝ ∂Θ 3 ⎠ ⎝ ∂Θ 2 ⎠ ⎣⎝ ∂Θ1 ⎠ ⎛ ∂r2 ⎞ ~ ⎛ ∂r2 ⎞ ~ ⎡⎛ ∂r2 ⎞ σ ⎤ ~ ⎛ ∂r ⎞ ~ ⎟ Θ1 + ⎢⎜ ⎟ − jω 2 ⎥ Θ 2 + ⎜ 2 ⎟ Θ 3 = 0 ⎜ ⎟ η+ ⎜ ⎝ ∂Θ3⎠ ⎝ ∂η ⎠ F ⎦ ⎝ ∂Θ1 ⎠ ⎣⎝ ∂Θ 2 ⎠ and σ ⎤~ ⎛ ∂r3 ⎞ ~ ⎛ ∂r3 ⎞ ~ ⎛ ∂r3 ⎞ ~ ⎡⎛ ∂r ⎞ ⎟ Θ1 + ⎜ ⎟ Θ 2 + ⎢⎜ 2 ⎟ − jω 3 ⎥ Θ 3 = 0 ⎜ ⎟ η+ ⎜ ⎝ ∂η ⎠ F⎦ ⎝ ∂Θ1 ⎠ ⎝ ∂Θ 2 ⎠ ⎣⎝ ∂Θ3⎠ (170) (171) (172) . (167) there are 54 theoretically different cases of poles and zeros. Taking into account eqn. (167) are real or imaginary. the double-layer capacitance and solution resistance should be added to the faradaic impedance. Some examples complex plane plots of faradaic impedances are presented in Figure 26. They were considered systematically in ref. In order to obtain the total impedance. Its complex plane plots display different forms from two capacitive semicircles through various capacitive/inductive loops to two inductive loops. [ ] (166) 4.96 Assuming existence of three adsorbed species a system of equations similar to eqns. Impedance Plots in the Case of Two Adsorbed Species The second term in eqn. depending on the sign of parameters B and C and relative values of all the parameters94. 95. The faradaic impedance may be represented by many different equivalent circuits.36 ^ Zf = 1 B + jωC + A A 2 ( D + B / A) + jω( E + C / A) − ω 2 The polarization resistance is: Rp = Rct + B/[A2(D + B/A)]. (157)-(159) may be written: − ⎛ ∂r ⎞ ~ i ⎛ ∂r0 ⎞ ~ ⎛ ∂r0 ⎞ ~ ⎛ ∂r0 ⎞ ~ +⎜ ⎟ Θ1 + ⎜ ⎟ Θ 2 + ⎜ 0 ⎟ Θ3 = 0 ⎟ η+ ⎜ F ⎝ ∂η ⎠ ⎝ ∂Θ1 ⎠ ⎝ ∂Θ 2 ⎠ ⎝ ∂Θ 3 ⎠ ~ 5. more complicated cases may be considered. Such a case is often found in corrosion. (166) represents a second-order electrochemical impedance95 and its denominator may be expressed in the following form: 2 (167) 1 + jω 2ζ / ω − ω / ω n ( n ) where ωn is called the undamped natural frequency and ζ is the damping ratio of the system32 expressed as: 1 E+C/ A ω n = D + B / A and ζ = (168) 2 D+B/ A Depending on the value of the parameter ζ the poles of the second term of eqn. Faradaic Impedance for the Process Involving Three or More Adsorbed Species Similarly to the case of two adsorbed species presented above. Figure 26.

Figure 4a. (e. mercury.). and that in the presence of a faradaic reaction. (173) reduces to eqn. that is the observed impedances cannot be represented by connection of simple R-C-L elements. Frequency Dispersion and Electrode Roughness The general model of the ideally polarizable electrode presented in Section III. (175) represents a semicircle . various models used for explanation of solid electrode impedance behavior are presented. (17) and Figure 2).1. (see also eqn. Figure 41b. (174). When φ = 1 there is only one time constant in the system (no dispersion) and the eqn. Problems similar to those observed on ideally polarizable solid electrodes arise also in the presence of faradaic reactions at these electrodes. as in eqns. When there is only one relaxation time constant. (155)-(156).37 where Θi are surface coverages of adsorbed species. On solid electrodes98 deviations from the ideal behavior are often observed. are found experimentally on liquid electrodes. Eqn. Constant Phase Element Dispersion of the measured complex dielectric constant is known from dielectric relaxation experiments. Eqns. Crammer’s method). very complicated complex plane plots may be obtained. etc. ladder circuit).g. The Warburg impedance is an example of a distributed element. indium-gallium. Section III.2. amalgams. However. eqn. σi their charges necessary for monolayer coverage. A general model of a multistep mechanism involving adsorption and diffusion was recently given by Harrington. IMPEDANCE OF SOLID ELECTRODES 1. (173) simplifies to: (174) ε * −ε ∞ 1 = Cole and Cole described the observed distribution of relaxation times as: ε * −ε ∞ 1 = ε s − ε ∞ 1 + ( jωτ 0 )φ 99 εs − ε∞ 1 + jωτ 0 (175) where φ is a constant between 0 and 1. respectively.97 V. On ideally polarizable solid electrodes the electrical equivalent model cannot be usually represented (with except of monocrystalline electrodes in the absence of adsorption) as a series connection of the solution resistance and double-layer capacitance. a transmission line (see Section VI. which may be different for different mechanisms. and G(τ) is the time constants distribution function. Some examples are presented in ref. that is G(τ) = δ(τ-τ0). on solid electrodes a frequency dispersion is observed. 2. In the case of three adsorbed species. and ri are the corresponding mass balances and relations between adsorbed species. 96. (169)-(172) are solved using methods for the solution of the system of linear equations (e. Below.e.18 The complex dielectric constant ε* may be represented as: ε * −ε ∞ ∞ G(τ ) = ∫ dτ (173) ε s − ε ∞ 0 1 + jωτ where ε¼ and εs are the dielectric constants determined at ω → ¼ and ω → 0. through so-called distributed elements. The impedance of such systems may be approximated by an infinite series of parallel R-C circuits i.g. The impedances may often be represented by an equation without simple electrical representation.

This behavior can be explained by eqn. pure resistance for φ = 0 and pure inductance for φ = -1. In general. etc. Eqn. However. which causes coupling of the solution resistance with the surface capacitance and (ii) a capacitance dispersion of interfacial origin. Brug et al. eqn. Assuming that the electrode impedance may be expressed as a sum of the solution resistance and the impedance of the CPE element. The impedance of the ideally polarizable electrode may be represented as a series connection of the solution resistance and the double-layer capacitance. connected with slow adsorption of ions and chemical inhomogeneities of the surface. Figure 28a. An example of such a distribution function is shown in Figure 27.100 presented a method for estimation of the average double-layer capacitance. For φ = 1 the distribution function becomes the Dirac's delta function.100 It was supposed101 that such a distribution may arise from: (i) a microscopic roughness caused by scratches. infinite Warburg impedance for φ = 0. from the value of T. (177) may also be written as: Z CPE = ^ 1 Tω φ [ cos(φπ / 2) − j sin( φπ / 2)] (178) and it represents a "leaking" capacitor. Figure 28a. the total impedance may be expressed as: ⎡ ⎤ ⎡ ⎤ 1 1 ⎢1 + ⎥ ⎥ = Rs Z = Rs + = Rs ⎢1 + φ φ φ⎥ ⎢ ( jωτ 0 ) ⎦ ⎢ ( jω ) (TRs ) ⎥ ( jω ) T ⎣ ⎦ ⎣ ^ 1 (179) . CPE. on the complex plane plots. C dl . Its impedance is given by: ^ (177) 1 Z CPE = T ( jω )φ where T is a constant in F cm-2 sφ-1 and φ is related to the angle of rotation of a purely capacitive line on the complex plane plots: α = 90 0 (1 − φ ) . perpendicular to the real axis. (177) may represent pure capacitance for φ = 1. similarly to that used by Cole and Cole99. Only when φ =1. T ≡ Cdl and purely capacitive behavior is obtained.5. was proposed. In such cases the double-layer capacitance may be expressed in terms of a so-called constant phase element. which produces a straight line. pits. and using the Cole-Cole formula for the distributed time constants.. In order to describe such behavior a model of distributed time constants. always present on solid surfaces.38 rotated by (1-φ)900 on the complex plane. which has non-zero real and imaginary components. Figure 27. (173) with the distribution function G(τ) described as: G( τ ) = 1 sin[(1 − φ )π ] 2πτ cosh[φ ln( τ / τ 0 )] − cos[(1 − φ )π ] (176) It represents a lognormal distribution that is it is a normal distribution of a function of ln(τ/τ0). on solid electrodes a straight line with angle lower than π/2 is often observed.

It leads to: T = C dl Rs φ − (1− φ ) (180) which allows for the estimation of the average double-layer capacitance in the presence of the CPE element. Lasia and Rami87 studied the hydrogen evolution reaction on polycrystalline Ni in 1 M NaOH. It was attributed slow adsorption and diffusion of these ions and phase transitions (reconstructions). However. instead of a semicircle centered on the Z’ axis. the double-layer capacitances estimated from eqn.107 et al. Similar results were also obtained on rhodium. They obtained rotated semicircles on the complex plane plots and the values of the parameter T and φ were potential dependent. the complex plane plots represent a rotated semicircle. the CPE phenomenon was usually attributed to surface roughness. equal to ~38 µF cm-2.107 Dispersion was observed. However. however.101 Experiments carried out on monocrystalline Au(111) and Au(100) electrodes in the absence of specific adsorption did not show any frequency dispersion. In their analysis they expressed the electrode impedance as: Z = Rs + ( jωC int )−1 where C int is a complex Cint = T ( jω ) . Consequently. C’int) show formation of deformed semicircles.102-104 Similarly. He has found that the Fourier inversion method is not suitable but the maximum entropy deconvolution works relatively well. They found that surface roughness of the order found on polycrystalline metals could lead to the CPE behavior only at much higher frequencies than those observed experimentally. In the case of a simple CPE circuit this parameter equals: .100. in the presence of specific adsorption of halide ions. which is a reasonable value taking into account some surface roughness. Pajkossy 101. assuming that the faradaic impedance can be expressed as a simple equivalent resistance. it was found that the capacitance dispersion increases markedly with the addition of chloride ions. because this is an ill posed problem the results obtained are very sensitive to the experimental errors (noise). VanderNoot106 tried to extract the distribution function G(τ) from the CPE model. in which instead of the CPE there is a double-layer capacitance in parallel with a series φ −1 electrode capacitance. Historically. the double-layer capacitance in the presence of the faradaic reaction may be obtained as:100 − − T = C dl φ Rs 1 + Rct1 [ ] 1− φ (181) An example of the application of eqn.105 Figure 28. (181) to the reduction of protons and tris-oxalato ferric ions was presented by Brug et al. (181) were constant. however. analysis of the ac impedance spectra in the presence of specific adsorption revealed that the complex plane capacitance plots (C’’int vs.39 where τ 0 = Rs C dl . They concluded that an increase in the surface roughness of polycrystalline Pt practically did not change (even slightly increased) the φ parameter. studied recently the origins of the CPE. Figure 28b. However. they proposed an electrical equivalent model shown in Figure 29. In the presence of the faradaic reaction.

and the semiinfinite Warburg impedance connected with the diffusion of adsorbing species. In fact. (177) in Section V. Fractal geometry was introduced to electrochemistry by Le Méhauté et al. He introduced a socalled bounded CPE (BCP) impedance or finite constant phase element: Z BCP = ^ 1 T ( jω ) φ tanh Rs T ( jω ) [ φ ] (182) where Rs is the solution resistance.262. In order to describe their geometry the concept of fractals was introduced. Figure 30. The line enclosing the object in Figure 32c is the so-called von Koch line.6. For φ = 0.115 It was shown by Nyikos. there is no simple relation between the fractal dimension DH and the parameter φ123 of the CPE .e.108 Stoynov109 has extended the CPE model for finite thickness diffusion. Complex plane plots for the BCP element are presented in Figure 31.5. This element has a physical meaning for φ < 0. blocking) interfaces generates a CPE behavior described by eqn. Comparison of the measured and calculated capacitances (using the model in Figure 29) for Au(111) in 0. at high frequencies. Just as the CPE represents infinite diffusion for φ = 0. Similar analysis of complex impedances obtained from the frequency dispersion on a passive stainless steel was carried out by Devaux et al.112. Such reasoning may also be used for surfaces for which the fractal dimensions may be between 2 and 3.2.113 it is continuous. In the example in Figure 32 the object is magnified three times and the line length is magnified 4 times. Pajkossy and coworkers113. Figure 29. (177) and at low frequencies it reduces to Rs. Rad and Cad.40 connection of the adsorption resistance and capacitance. 116-122 that fractal geometry of ideally polarizable (i.15 mM NaBr is shown in Figure 30. This leads to the fractal dimension of the von Koch line DH = (ln 4)/(ln 3) = 1.111 Simple magnification only increases the size of the object while the fractal magnification reveals its self-similarity at different scales. (182) reduces to a simple CPE. eqn. 3. the fractal dimension of the line may be between 1 and 2.19 Figure 31. However. of infinite length and is nowhere differentiable. in general.114 In general. the BCP represents finite length diffusion for the same value of φ.111. The difference between simple and fractal magnification is shown in Figure 32. Such magnification process may continue. Fractal Model Solid surfaces are usually irregular and their detailed geometry is not known.114. It is interesting to note that the observed (measured) length of the von Koch line is scaled in a complex way: its length depends on the yardstick used to measure it.110 This concept is based on selfsimilarity of surfaces implying different scaling.1 M HClO4 in the presence of 0. eqn.5 BCP has a similar form to that of the impedance in the case of the finite length diffusion. it should be stressed that.

129. in the absence of mass transfer control.125. Sierpinski carpets131. With decrease of parameter φ the semicircles become deformed (skewed).118.119.114 The fractal theory was subsequently extended to irregular or quasi-random surfaces lacking well defined selfsimilarity. (184) reduces to eqn.133. fg =1. exp ct Figure 32. Kinetic analysis of the hydrogen evolution reaction on surfaces displaying fractal ac impedance behavior was carried out by Lasia and coworkers135-137 who obtained rate constants expressed as: k i.98 When the surface is flat and homogeneous. etc. i.41 although higher fractal dimensions lead to smaller values of φ. the fractal theory was extended to faradaic processes.130. . in the absence of the dc current which may introduce a local interfacial potential difference. exp = k i / Cdl was used. Cantor bars128.114.125 Levie has shown that the impedance of a fractal electrode. The complex plane impedance plots obtained from eqn. therefore the intrinsic activity of these electrodes is similar and the observed increase in activity of Raney Ni electrodes arises from their very large real surface area. are formally similar to those found by Davidson and Cole134 in their dielectric studies.120 Pajkossy and Nyikos124 carried out simulations of blocking electrodes with a selfsimilar spatial capacitance distribution and found that the calculated impedances exhibited the CPE behavior. (183) may be applied to fractal electrodes in equilibrium. Figure 33. According to de Levie111 eqn.126 De 111. which may be based on von Koch curves127. φ = 1.111. (112). is given as: ⎛ ⎞ ⎜ ⎟ ^ 1 1 ⎟ Z = Rs + ⎜ b ⎜ 1 + jωC ⎟ ⎜ dl ⎟ ⎝ Rct ⎠ where the parameters b is given by: φ (183) b = f g ρ φ −1 (184) ρ is the solution resistivity and fg is a factor depending on the fractal surface geometry110. exp = b1/ φ k i . The fractal theory was also tested experimentally using fractal electrodes prepared by microelectronic techniques. b = 1 and eqn. It was found that this ratio has similar values for Raney Ni and polycrystalline Ni.132. (183). Examples of the complex plane plots obtained for fractal electrodes are presented in Figure 33. exp / Cdl . When the exact fractal structure is unknown the parameter b cannot be obtained and the only parameters accessible are: (185) C = b1/ φ C and R = b1/ φ R exp dl ct .e.117.118 Subsequently. In order to compare the intrinsic activities of such electrodes with that of polycrystalline nickel the ratio of k i.

where Z is the impedance of pore walls per unit of the pore length. This decrease of the current is proportional to the current flowing to the wall: (i) where 2πrdx is the surface area of a pore section dx and j is the current density.118. Figure 34.5 is formally identical with the semi-infinite porous model presented below. Two cases are considered below: (i) porous electrodes in the absence of internal diffusion and (ii) in the presence of axial diffusion. Because of Ohm’s law a potential drop along the pore also arises: (187) dE ⎛ ρdx ⎞ 1 ⎛ ρ ⎞ ( 2πrdx ) j dI =− = −2πrj dx dx (186) dx = − I⎜ 2 ⎟ = − I⎜ 2 ⎟ ⎝ πr ⎠ dx ⎝ πr ⎠ where ρ is the specific solution resistance and (ρdx/πr2) is the resistance of the section dx of the solution in the pore. eqn. Experimental verifications of the fractal model were also carried out for some electrodes. migration can be neglected). In such cases porous electrodes are used. flows towards the walls and its value decreases with the distance x from the pore orifice.140 4. Porous Electrode Model In electrocatalysis there is a great interest in increasing the real surface area of electrodes.118. the current flowing through an element dx of the surface area. Zf.138 It was also stated that the fractal dimension of the surface may be found from dc experiments. It can be shown that Z = Zel/2πr where Zel is the specific impedance of pore walls in Ω cm2. (2πrdx)j. R = ρ/πr2 and is expressed in Ω cm-1. (187) may be called the solution resistance per unit length of the pore. The last term in the parentheses in eqn. self-similar fractal surfaces do not exists in the real world. Porous Electrodes in the Absence of Internal Diffusion In this case it is assumed that the concentration of the electroactive species is independent of the distance along a pore. Figure 34. The fractal models may only approximate random surfaces. in Ω cm.e. The axially flowing dc current.139.121.114. The fractal model in the presence of diffusion was discussed in refs. which enters the pore. Zel consists of the faradaic.42 In general. a simpler model is usually used in which it is assumed that pores have cylindrical shape with a length l and a radius r. . (183) for φ = 0. and the double-layer impedance: Zel = (1/Zf + jωCdl)-1. Similarly. It is assumed that the electrical potential and concentration of electroactive species depend on the distance from the pore orifice only and there is always an excess of the supporting electrolyte (i. I. may be represented as E/Z.141-145 In order to describe the impedance of such electrodes first a dc solution must be found. Because modeling of real electrodes is difficult. 24. In the next Section we shall see when such an assumption is valid. Besides. 111 and 118.

Z por .43 De Levie’s treatment De Levie146 was the first to describe the impedance of porous electrodes. Taking into account the boundary conditions: x=0 E = E0 and x=l dE/dx = 0 where E0 is the potential at x = 0. the dc solution is: ⎤ ⎡ R cosh ⎢ ^ ( l − x )⎥ ⎥ ⎢ ⎦ ⎣ Z E = E0 ⎛ ⎞ ⎜ R l⎟ cosh ^ ⎜ ⎟ ⎝ Z ⎠ and (192) ⎞ ⎛ R ⎛ dE ⎞ ⎜ R l⎟ = − I R = − E0 ^ tanh ^ ⎜ ⎟ 0 ⎟ ⎜ ⎝ dx ⎠ x = 0 ⎝ Z ⎠ Z The total pore impedance. (186) and (187) in the following form: (188) dI E (a) dx =− ^ Z (189) dE = − RI dx Taking the second derivative of eqn. (188) one obtains: d 2E dx 2 dI ⎛ R ⎞ = −R = ⎜ ⎟E dx ⎜ ^ ⎟ ⎝ Z⎠ (190) This equation describes changes of the electrical potential as a function of pore length. is then obtained as: ^ (193) Z por ⎛ ⎞ ^ E0 ⎜ R l⎟ = = R Z coth ^ ⎜ ⎟ I0 ⎝ Z ⎠ (194) This equation may be rearranged into: . He represented eqns. The solution is: (191) ^ ^ E = C1 e − R/ Z x + C2 e R/ Z x where C1 and C2 are the integration constants. (189) and substituting eqn. De Levie assumed that the impedance of pore walls is independent of the pore distance (Z is not a function of distance) which implies that there is no net dc current.

p 1 ^ Z por = where s = 2πrl is the total pore surface area. coth(Λ1/2) → 1. that is the pores behave as semiinfinite channels. then Λ → ¼. of the alternating signal into the pore. p Λ1/ 2 coth Λ1/ 2 ( ) (195) where RΩ. then its impedance may be described by a semicircle on the complex plane plot. Figure 35a. This parameter is defined as: Λ = l/λ. (195) predicts observation of a straight line at 450 at high frequencies. and: (200) 1/ 2 ^ 1/ 2 ^ Λ = 2πrl Z el Z= RΩ. Zel = Rct. Figure 35. Eqn. Another limiting case is obtained when the penetration depth is much smaller than the pore length. when λ >> l. Figure 35b. The faradaic impedance may be obtained assuming that the Butler-Volmer equation describes adequately the electrochemical process. (199) represents simply the impedance of a flat electrode having surface area s.p = ρl/πr2 and Λ = (2ρl2/r)/ Zel . a more adequate representation of impedance would be with respect to the overpotential η: 1 ^ = Zf dj 1 = = j0 nf α e − αnfη + (1 − α ) e(1− α ) nfη Rct dη [ ] (196) Of course. and the equation becomes: ^ (199) RΩ. It is evident that the faradaic impedance of pore walls was assumed potential independent despite the fact that the potential changes with the pore depth.44 ^ Z por = RΩ. (195) has two limiting cases. At low frequencies the impedance becomes real: Z por Z t = Rs + n ^ ^ (197) Z (ω = 0 ) = R p = ⎛ 2ρ ⎞ coth ⎜ l⎟ ⎝ rRct ⎠ 2π 2 r 3 ρRct (198) Behavior of the porous electrode depends on the penetration depth. λ << l. coth(Λ1/2) → Λ-1/2. λ. the total impedance becomes: Eqn. Λ → 0. under dc conditions. For the ensemble of n pores and in the presence of the solution resistance outside the pores. followed by a semicircle. when ω = 0. First. Eqn. or λ = (r Zel /2ρ)1/2. Although original development was carried out using electrode potential E. In this case the ac signal penetrates to the bottom of the pore and the electrode behaves as a flat one. p Λ1/ 2 ⎛ ρ ⎞ = ⎜ 2 3⎟ ⎝ 2π r ⎠ Z el .

1/3 and 2/3. T. al. (186) and (187). ηl = 0. (204) may be rearranged to: (205) dη ρj ⎛ bη ⎞ dx which has a solution: = −4 0 r sinh ⎜ ⎟ ⎝ 2⎠ . (183) for the fractal model with φ = 0. when the potential at the bottom of the pore drops to zero. In such a case. (202) may be written in a simpler form as: d 2η dx 2 = 4ρj0 sinh( bη) r (203) The first integration of eqn. the parameters Rct. Zr = |Z|2cos(2ϕ) gives a perfect semicircle. In this case eqn. plotting Zi = |Z|2sin(2ϕ) vs. In the rigorous treatment. eqns. In further papers the impedance of double-layer was substituted by the CPE: −1 (201) ^ ^ An example of porous behavior was presented by Los et. (187). Using the CNLS fit.144 In a general case. (202) gives.146 Eqn.5 and these two models are indistinguishable. eqn.45 In this case the complex plane plot presents a deformed semicircle.147 for the hydrogen evolution reaction on LaPO4-bonded Ni powder electrodes in 30% NaOH. it may be solved numerically. (204) may be solved analytically only for the case of semiinfinite length of pores and for α = 1/2. are potential dependent. Examples of the complex plane plots are shown in Figure 36. Let us consider now the case of semiinfinite pores. i. written using the Butler-Volmer expression for current and overpotentials is:148. (200) is formally identical with eqn. and in consequence Zel . The second derivative of eqn. (b) Rigorous treatment It is obvious that the de Levie’s treatment is an approximation. ⎡ φ ⎤ Z el = ⎢1 / Z f + ( jω ) T ⎥ ⎣ ⎦ Figure 36. and Cdl were determined. because Z f . taking into account that at x = l.5 and b = 0.5nf. Figure 35c.e.149 2 (202) d η dx 2 = 2ρ 2ρj0 αnfη − (1− α ) nfη 2ρj0 b1η − b2 η e −e = e −e j= r r r [ ] [ ] Assuming that α = 0. dη/dx = 0 and ηl = η(l): dη = −{( 4ρj0 / r )[exp( b1η) / b1 + exp( − b2 η) / b2 dx − exp( b1ηl ) / b1 − exp( − b2 ηl )]}1/ 2 (204) Eqn. should be solved.

46 ⎛ ρj0 b ⎞ ⎛ bη ⎞ ⎛ bη ⎞ tanh ⎜ ⎟ = tanh ⎜ 0 ⎟ exp⎜ − 2 x⎟ ⎝ 4⎠ ⎝ 4 ⎠ r ⎠ ⎝ (206) It may be noticed that the eqn. i + 1 / Zi −1 ⎟ 2 ⎝ πr ⎠ [ ] where Zel . Substitution of eqn. It was assumed that the electrode process could be described by the currentoverpotential equation: (209) ⎛ ⎞ (ii) C C j = j 0 ⎜ O e bη − R e − bη ⎟ * * CR ⎝ CO ⎠ where Ci and Ci* represent surface and bulk concentrations of Ox and Red. eqn.93. The values obtained for of the parameter φ are between 0. starting from the bottom of the pore: (208) ⎛ ρ ⎞ Zi = ⎜ ∆x + 1 / (2πr∆x ) / Zel . Recently. This result indicates that the Tafel slope on porous electrodes is doubled over its normal value.148. Figure 37. The solution obtained may be compared with de Levie’s solution. instead of a perfect semicircle. (209) can be rearranged into: .i is the specific electrode impedance at the distance x. Figure 37.149 In this case the use of the CPE only hides the inadequacy of the model. This problem has been addressed in numerous papers. However. obtained using the correct analysis is twice that found from de Levie’s equation. respectively. Lasia156 solved the problem for a quasi-reversible process and a finite pore length.91 and 0. there being systematic differences between these two curves. the plot of squared impedances produces a deformed ellipsoid. Porous Electrodes in the Presence of Axial Diffusion During electrolysis.118/α V at 25°C. Complex plane plots obtained for the same conditions using two different approaches are displayed in Figure 37. (206) is formally identical with that developed for the GouyChapman theory of the double-layer. eqn. (189) into (205) gives the expression for the steady-state current on porous electrodes: I = 4πr rj0 sinh( bη0 / 2) ρb (207) which gives. (195). concentration changes in the pores. the approximation is good. the pore impedance may be obtained by numerical summation of the impedances of small sections ∆x of the pore walls. when the CPE is used instead of the double-layer capacitance. Besides. Knowing the potential distribution in pores. Assuming that the diffusion coefficients of the two forms are identical. semiinfinite pores or assuming that the concentration gradient in pores is exponential were usually made. Rp.150-155 Simplifications such as assuming totally irreversible reaction kinetics. It has been shown147 that the CNLS fit of the simulated impedances to the de Levie equation (195) is not good. the Tafel slope of 2 ln(10)/αf = 0. at negative potentials. The analysis of the results indicates that the resistance at ω = 0.

The current flowing through the section dx of the pore walls is related to the changes of concentration: (211) nFr dC nF dN [( ) ] (210) j=− 2πrdx dt =− 2 dt (212) or taking into account diffusion: ∂a ∂2a 2 =D 2 − j * ∂t ∂x nFrCO which. reduces into: d 2a dx 2 − 2 * nFrCO j=0 (213) The potential drop in the pore is still described by eqn. ν ~ 10-5 to 10-4 V. the electrode impedance may be calculated. an analytical solution of eqn. with the current given by eqn. For these conditions the porous behavior is determined by the concentration gradient and the potential gradient down pores is negligible. (202) and (213).47 j = j0 a e bη + m e − bη − ( m + 1) e − bη where a = CO/C*O and m = C*O/C*R. (202) and (213) may be combined together as: (214) d 2η d 2a * =ν where v = nFDCO ρ Eqn. The value of the parameter ν determines whether the porous behavior is determined by the potential or concentration drop. In a limiting case when the potential drop in the pore may be neglected. When ν << 1V the system behavior is determined principally by the concentration gradient and when ν >> 1 V it is determined by the potential drop. As usual. eqn. (212). must be linearized: . under steady-state conditions. In a general case the first integration may be carried out analytically and the next numerically. Eqns. For typical conditions: D = 10-5 cm2 s-1. (202). Only for the extreme conditions where the solvent or the supporting electrolyte (at high concentration) undergo the red-ox reaction. may the process be limited by the potential drop in the pores. (213) may be obtained: a = γ + (1 − γ ) where: cosh [ B (l − x ) cosh ( Bl ) ] (216) γ= e 2bη0 + m m +1 and B = * nFrDCO 2 j0 (e η + m e η ) b 0 −b 0 (217) Knowing the dc solution. The derived dependence of η and a as functions of distance may be used to calculate the impedance. the current. ρ = 10 Ω cm and C* = 10-3 to 10-2 M. (214) has the analytical solution: dx 2 dx 2 η − η0 = ν( a − 1) (215) which allows elimination of one variable from eqns. (210).

in fact. The presence of the concentration gradient in the pores produces two potential-dependent semicircles. . Substitution and rearrangement gives: (220) ~ η ~ ⎛ dj ⎞ d 2 ( a / ~ ) ⎛ a ⎞ ⎡ jω 2 2 ⎛ dj ⎞ ⎤ − ⎜ ~⎟⎢ + ⎜ ⎟⎥ − ⎜ ⎟ =0 * * ⎝ η ⎠ ⎢ D nFrDCO ⎝ da ⎠ ⎥ nFrDCO ⎝ dη ⎠ dx 2 ⎦ ⎣ or ~ (221) d2 y ~ where: ~ dx y = a/ η ~ ~ 2 − yK −L = 0 jω 2 ⎛ dj ⎞ K= + ⎜ ⎟ * D nFrDCO ⎝ da ⎠ ⎛ dj ⎞ L= ⎟ * ⎜ nFrDCO ⎝ dη⎠ 2 (222) with the boundary conditions: ~=0 y x=0 and ~ dy / dx = 0 x=l The analytical solution of eqn. a pore size distribution and the exact surface morphology is unknown. (208) the total impedance may be calculated numerically. (219) gives the faradaic impedance and using eqn. Figure 38 andFigure 39. (212) must be solved for ∆a. (221) is: ~ cosh K ( l − x )) L⎧ ⎪ = ⎨− 1 + ~ K cosh K l ⎪ ⎩ η a [ ( ) ]⎫ ⎪ ⎬ ⎪ ⎭ (223) Now. substitution into eqn. eqn. Figure 38. In real cases the problem is much more complicated. the expression for the impedance is obtained as: ~ (218) ⎛ ∂j ⎞ ⎛ ∂j ⎞ a = =⎜ ⎟ +⎜ ⎟ ^ ~ ⎝ ∂η⎠ ⎝ ∂a ⎠ ~ Zf η η 1 j ~ (219) In order to find a solution for the oscillating concentration. because various pores are present and.48 ⎛ ∂j ⎞ ⎛ ∂j ⎞ ∆j = ⎜ ⎟ ∆η + ⎜ ⎟ ∆a ⎝ ∂a ⎠ η ⎝ ∂η⎠ a Then. Figure 39. It should be added that at high frequencies still a part of a straight line at the angle of π/4 may be observed as in Figure 39.

Generalized Warburg Element Macdonald18 introduced a generalized finite-length Warburg element described as: ^ Rs φ (226) Z= tanh ⎡ A0 Rs ( jω ) ⎤ ⎢ ⎥ φ ⎣ ⎦ A R ( jω ) 0 s to describe non-uniform diffusion under finite-length transmissive conditions. eqn. For φ = 0. However.157 Such surfaces may be prepared. in which the pore is divided into sections and for each section a transmission line model with constant impedances was used. Figure 40. by abrasion. Keiser et al. Eloot et al. (226) represents uniform finite-length diffusion and not the non-uniform diffusion case. Eqn. The impedance of a V-grooved electrode was studied by de Levie. for example.5. Non- ( ) uniform diffusion arises for example when the diffusion coefficient is a function of the distance. observed for cylindrical pores at high frequencies. Recently. (225) reduces to the impedance of a perfectly flat surface for β = 900 and to the impedance of cylindrical porous electrode for β = 00.159 studied the impedance of arbitrary shaped pores. A0 = l / Rs D . I0 and I1 are the modified Bessel functions of zero and first order and: (iii) λ=2 ρl Z sβ (225) l is the groove depth (normal to the surface) and Zs is the double-layer impedance per unit of true surface area. different forms of plateaux or a semicircle were observed. A similar equation but containing the function coth was used by Inzelt and Láng163 to describe the diffusional impedance of conducting polymers under reflective conditions (see . Instead of a straight line at 45°. This equation is formally identical with Stoynov's finite constant phase element BCP. at high frequencies where the impedance is determined by the double-layer charging. Figure 40. They simulated the complex plane plots in the absence of a faradaic process.160 suggested a new general matrix method for calculations involving noncylindrical pores. De Levie described the impedance of a groove per unit groove length as: ^ ρ I 0 (λ ) Z= (224) tan β λI1(λ ) where ρ is the specific solution resistance.162 5. Direct simulations of the impedances for porous electrodes were also carried out using a random walk method. Stoynov109 stated that eqn. (182).49 Other Pore Geometries For pores of geometry different from cylindrical. where Rs is the solution resistance. instead of a straight line some forms of arc may also be obtained. and it represents a finite-length diffusion. β is the angle between groove wall and the normal to the surface.161. Gunning158 obtained an exact solution of the de Levie grooved surface in the form of an infinite series. Comparison with the de Levie equation shows that the deviations arise at higher frequencies.

Physical meaning of these equations is that the system does not generate noise independent of the applied signal. the physical significance of the parameter A0 for φ < 0. linear systems cannot exhibit hysteresis in their response at ω = 0. the response must be zero for t < 0. Moreover. for t < 0. This also implies that the system is described by a system of linear differential equations (see e. The system is casual if it does not have any singularities. • Causality: The response of the system must be entirely determined by the applied perturbation. eqns. Although models including impedance represented by eqn. Finiteness The impedance technique is often applied to various electrochemical systems which were never studied before. (2) and (7)). Before starting the analysis and modeling of the experimental results one should be certain that the impedances are valid. CONDITIONS FOR "GOOD" IMPEDANCES 1. Linearity. (227) and (228) are mathematical forms of causality. Besides.165 further developed by Bode166 and later applied to EIS. or the sum of residues is zero. The impedance derived is valid provided that the following four criteria are met:33. A stable system remains stable unless excited by an external source and it should return to its original state once the perturbation is removed and the system cannot supply power to the output irrespective of the input.6(ii) and eqn. Stability. • Stability: The stability of a system is determined by its response to inputs.g. The causal system cannot predict what its future input will be. For the linear systems the response is independent of the amplitude. (42) and (130)) for small amplitudes. An electrical model containing this element accounted well for the impedance spectra with a minimal number of free parameters. It is easy to verify the linearity of the system: if the obtained impedance is the same when the amplitude of the applied ac signal is halved then the system is linear. It was introduced by Kramers164 and Kronig. In the complex plane the above criterion requires that.18.169 • Linearity: A system is linear when its response to a sum of individual input signals is equal to the sum of the individual responses.5 is not clear . The complex plane and Bode plots obtained often display shapes that had never been encountered previously. Mathematically this means that the function does not have any singularities that cannot be avoided.e. If the system is at rest and a perturbation is applied at t = 0. VI. ω = 0. (228). eqn. eqn. 167-177 During the impedance measurements a small ac perturbation is applied to the system. Electrochemical systems are usually highly nonlinear and the impedance is obtained by linearization of equations (see e. Causality. it has a derivative at each point)178 must be equal to zero: (227) ∫ Z ( s )ds = 0 C If the function Z(s) has singularities then the sum of the residuals of the poles ai must equal zero: Z ( s) 1 (228) ds = f ( a i ) and ∑ f ( a i ) = 0 ∫ 2πj C s − a i i Eqns.50 Section III. (227). eqns. The impedance Z(s) must satisfy the following conditions: Z(s) is real when s is real . (99)). the integral on and inside a closed path C of an analytic function (i.g. Causal systems are also called physically realizable systems. There is a general mathematical procedure which allows for the verification of the impedance data. that is the output depends only on the present and past input values. The system is stable if its response to the impulse excitation approaches zero at long times or when every bounded input produces a bounded output. (226) may well describe some experimental data.

semi-infinite mass transfer. This condition may be easily checked by repetitive recording of the impedance spectra. The discrepancies that arise may be attributed to the errors of integration or to failure to satisfy the four above conditions. However. However. followed by analytical integration of the polynomials. However. Kramers-Kronig Transforms The Kramers-Kronig relations hold provided the four above constraints are satisfied and allow the calculation of the imaginary impedance from the real part: ⎛ 2ω ⎞ ∞ Z '( x ) − Z '( ω ) Z ''(ω ) = −⎜ ⎟ ∫ dx ⎝ π ⎠ 0 x2 − ω2 ∞ (229) the real impedance from the imaginary part. . etc. Such procedures are important when the impedance goes to infinity at low frequencies (blocking electrodes. CPE. if the zero-frequency asymptote of the real part is known: ⎜ ⎟ Z ''( x ) − Z ''(ω ) ⎛ 2ω ⎞ ∞ ⎝ x ⎠ Z '(ω ) = Z '( 0) + ⎜ ⎟ ∫ dx ⎝ π ⎠0 x2 − ω2 the polarization resistance Rp from the imaginary part: ⎛ ω⎞ (231) ⎛ 2 ⎞ ∞ Z ''( x ) R p = Z '( ∞ ) − Z '( 0) = ⎜ ⎟ ∫ dx ⎝ π⎠ 0 x ⎛ 2ω ⎞ ∞ ln| Z ( x )| dx ϕ(ω ) = ⎜ ⎟ ∫ ⎝ π ⎠ 0 x2 − ω2 (232) or the phase-angle from the magnitude (modulus): (233) Similar transformations may also be carried out for admittances. 2. This last condition ensures that there are no negative resistances in the system. They carried out integration between the frequency corresponding to the maximum of the imaginary impedance and infinity. and multiplied the result by two.51 (that is when ω → 0) and Re[Z(s)] ≥ 0 when ν ≥ 0 (s = ν + jω. Kendig and Mansfeld179 used eqn. their method is limited to systems containing one time constant. al.167-169 and Dougherty and Smedley177 used a polynomial approximation of the impedance function. The impedance measurements must also be stationary. (232) and supposed that the imaginary impedance is symmetric. In particular. Macdonald et. • Finiteness: The real and imaginary components of the impedance must be finite-valued over the entire frequency range 0 < ω < ∞. then the obtained Bode plots should be identical. the impedance must tend to a constant real value for ω → 0 and ω → ∞. that is the measured impedance must not be time dependent.). see Section I. the impedance results are known only over a finite frequency range.1(i)). if the high frequency asymptote for the real part is known: Z '(ω) = Z '(∞) + 2 xZ ''( x) − ωZ ''(ω) dx π∫ x 2 − ω2 0 (230) the real impedance form the imaginary part. The major difficulty in applying the Kramers-Kronig relations is that the integration must be performed over the whole frequency range from zero to infinity.

written for immittances. Orazem and coworkers 175. In this method the explicit KramersKronig integration is replaced by the fit to the Voigt model. (234) and (235) become linear in the Rk values and the problem of approximation reduces from iterative nonlinear to a linear single matrix inversion. The method is quite robust with respect to the choice of the distribution and range of τk values. capacitive and inductive elements. ωmin. where gm are m=1 M dimensionless weighting coefficients and τm characteristic time constants to be determined. it means that they are not transformable. Similarly. which are unknown. To take into account the inductive loops they proposed using negative resistances.52 extrapolation of polynomials over a large frequency range may be unreliable. (173). The latter parameter is chosen in such a way that the real impedance goes to a constant value while the imaginary impedance goes to zero at ω0. Haili180 extrapolated Z'' as proportional to ω as ω → 0 and as inversely proportional to ω as ω → ∞ and Z' → R s . as a sum of M discrete delta functions: G ( τ ) = ∑ gmδ( τ − τ m ) . ω0. Esteban and Orazem170. Through inspection of the relative residual plots it is possible to isolate data . Under such conditions eqns. The problem with this approach is the initial selection of Rk and Ck parameters in the complex nonlinear least-squares approximation. (230) and (231) simultaneously to calculate the impedance below the lowest measured frequency. CPE. the entire circuit must also be transformable. Parameters τk were chosen as equal to the inverse of the experimental angular frequencies ωk. If the data cannot be well approximated. Warburg impedance. Such an approximation may be written as: Figure 41.182 used an approximation to the experimental impedance by the Voigt model (Figure 41) followed by a transformation of the model data. Because each parallel R-C circuit is transformable. which are usually logarithmically distributed (5 to 10 per decade). In practice 6 to 7 time constants per frequency decade (τk = 1/ωk) should be selected to get a good approximation. and continue the integration procedure to three or four decades of smaller frequency. etc. This method was applied to approximate the impedance from various circuits containing resistive. that is the time constants were defined and the adjustable parameters were Rk. Boukamp and Macdonald183 proposed an approximation of the experimental impedances using a distribution of relaxation times. Later.181. This method was further modified by Boukamp184 who also used the Voigt circuit but with a fixed distribution of time constants τk. Z '( ω ) = Rs + ∑ and M k =1 1 + (ωRk Ck ) Rk = Rs + ∑ 2 ωRk τ k M k =1 1 + (ωτ k ) Rk 2 (234) Z ''(ω ) = − ∑ M k =1 1 + (ωτ k ) 2 (235) where τk = RkCk and M is the number of R-C elements used in the Voigt circuit to approximate the experimental impedance.171 proposed using eqns. They represented the function G(τ) in eqn.

189 ⎛ S S* ⎞ ⎜ xy xy ⎟ γ xy = ⎜ S S ⎟ ⎝ xx yy ⎠ 1/ 2 (236) where S xx is the average input signal power spectrum. truncated data. etc.187 could be used to validate the data. S yy is the average output signal power spectrum. For example. Non-Stationary Impedances As mentioned above the measured system should be stationary and should not evolve with time. The coherence function. Such measurements should be carried out quickly. Macdonald185 proposed another form of the Kramers-Kronig integrals in which integration to infinity and the poles are avoided. the negative τk could mask some non Kramers-Kronig transformable behavior. In the case of blocking electrodes the impedance goes to infinity as ω → 0. The sign of Rk parameters is not important but the time constants are always positive (by definition). S xy is the average crosspower spectrum. is defined as:188. An example of such analysis is shown in Figure 42. very often the most interesting features are observed at low frequencies and the experiment may take hours. The coherence function takes values between zero and one and it characterizes statistical validity of the frequency response measurements: it is equal to one when perfect coherence exists. a suitable parallel resistance could be added to the system (in such a system the impedance must always be real and equal to this shunt resistance at ω = 0) and then carry out the transformation of the data obtained. For such electrodes the admittance Kramers-Kronig transformation could be used.190 In order to use such a method the initial conditions must always be the .187 3. corrosion process may continue during the experiment and change the measured impedance.). γ. and Sxx(ω) is the power spectrum of parameter x as defined by its Fourier transform X: (237) S (ω ) = X (ω ) X * (ω ) xx The power spectra may be directly obtained using dynamic signal analyzers which measure signals as a function of time and perform the fast Fourier transform. Kramers-Kronig transforms constitute a very sensitive criterion of the validity of the ac impedance data.53 that do not comply with the Kramers-Kronig transformations. such conditions cannot always be met. symbol * denotes the complex conjugate. However.177 Figure 42. It has been shown that instead of Kramers-Kronig transforms another method involving a coherence function28. Alternatively. This function may be used when the Fourier-transformed data exist and the Kramers-Kronig transforms are difficult to use (unbounded impedance.186 In general. In such cases it is possible to follow evolution of impedances with time at one frequency and then repeat the experiment many times at different frequencies. In practice.

Ideally. Selection of the Model The aim of the analysis of the EIS data is to elucidate the electrode process and to derive its characteristic parameters. process175. transients) should also be carried out. Section IV. etc. etc.181. However. In such cases a priori model predictions are difficult or impossible.g. hydrodynamic conditions. diffusion coefficients. many different conditions arise from the numerical calculations. cylinder. C. together with good physical knowledge of the system (solution and surface composition.) or nonuniform diffusion. first. the process model constructed in consistency with the parameters found and the physico-chemical properties of the process. This method was applied recently to follow the electrochemical impedance of a guillotined aluminum electrode. and is often misunderstood and wrongly interpreted. may produce two semicircles in the complex plane plots. In such a case approximation to a full model introduces too many free parameters and a simpler model containing one time constant should be used. etc.196 or structural19. etc.191-194 introduced a new mathematical method for nonstationary impedances involving four-dimensional analysis and so-called rotating Fourier transforms. and L. but experimentally only one semicircle is often identified. Other electrochemical experiments (dc. This is a quite difficult task which must be carried out very carefully. Interpretation of impedance data requires use of an appropriate model. finite length or spherical diffusion. for porous electrodes (correct solution). presence of various layers. the number and nature of parameters should be determined and.). nonlinear diffusion (to disk. etc. VII. It should be stressed here that the EIS is a very sensitive technique but it does not provide a direct measure of the physical phenomena. e. the number and the nature of the circuit elements should be identified and then the process modeling should be carried out. Process modeling links measured impedances with physicochemical parameters of the process (kinetic parameters. the system containing one capacitive .199 modeling and (ii) measurement.196 formal19 or mathematical200. Another problem of data modeling is connected with the fact that the same data may be represented by different equivalent circuits. a system with one adsorbed species. MODELING OF EXPERIMENTAL DATA 1. For example. It may also be carried out for the circuits containing distributed elements that can be described by a closed form equation: CPE. semiinfinite. sample geometry. Usually. Measurement modeling explains the experimental impedances in terms of mathematical functions in order to obtain good fit between the calculated and experimental impedances. It should be stressed that proper modeling is the most difficult part of the analysis. Such a procedure is relatively elementary for a circuit containing simple elements: R. porosity. However.201 modeling. In the latter case the parameters obtained do not necessarily have a clear physico-chemical significance. then. an equivalent circuit is chosen and the fit to the experimental data is performed using the complex nonlinear least-squares technique (CNLS). concentrations. thickness.197.190 Stoynov et al.195. hydrodynamic conditions. The modeling may be classified as: (i) physico-chemical.200 For example. Although the simulations were carried out they were not applied in experimental studies of non-stationary systems.2. nonhomogeneous materials (conducting polymers).198.195. first the measurement modeling should be carried out.).54 same. Therefore. the model deduced from the reaction mechanism may have too many adjustable parameters while the experimental impedance spectrum is simple.

Their admittance may be written as: a ( jω ) 2 + a1 jω + a 0 ( 240) Y= 2 ( jω ) 2 + b1 jω + b0 And the behaviors of these three circuits are also indistinguishable. In fact. These authors also studied a more complex case. substituting circuit resistances and capacitances by the CPE elements. For example. (25)). having two time constants. Figure 23. Zoltowski200 proposed using ladder circuits for measurement modeling. They stated that the ambiguity demonstrated above is common to model identification for all electrochemical measurements and presents the greatest challenge for the analysis of impedance data.e. Development of a proper model requires knowledge of the chemistry and physics of . Figure 43. The complex plane plot represented a depressed semicircle The authors found that the input data could be quite well approximated by a circuit consisting of a solution resistance in series with the parallel connection of a resistance and a CPE element (see Figure 28) and three other circuits containing two time constants (ladder. that the admittance of these two circuits may be written in the general form: a jω + a 0 a −a b (238) = a1 + 0 1 0 Y= 1 jω + b0 jω + b0 where a1 = R1-1. may be exactly described by either of the two equivalent circuits shown in Figure 43. Figure 4. Very often modeling depends on the errors of the experimental data. (195). i. Circuits the most often used in measurement modeling are: the Voigt. that is for a proper choice of the parameters they display the same impedance spectrum over at all frequencies. Voigt and mixed). a0 = 1/(C2R1R2) and b0 = (R1-1 + R2-1)/C2 for circuit (a) (239) or a1 = R1-1 + R2-1. Similarly. Equation (238) indicates that there is only one time constant of the system (see eqn.196 studied approximation of synthetic data corresponding to the impedance response of a single electrochemical reaction on a rotating disk electrode under the conditions of nonuniform current and potential distribution. as presented in Figure 45. impedance of a porous electrode may be described by the Voigt circuit with a sufficient number of R-C elements or by eqn.55 loop. Figure 44. Orazem et al. Figure 45. may be adequately described by the three circuits in Figure 44. a system displaying two capacitive loops. a0 = 1/( C2R1R2) and b0 = 1/(R1C2) for circuit (b) and these two cases are indistinguishable. ladder and Maxwell circuits.

If the process converges to another minimum. may be described by the ladder circuit shown in Figure 21. Such decrease must be statistically important. 202. In general. Usually. a simpler model is first used. (139).202-204 This is a nonlinear least-squares fit of the real and imaginary parts. To do this a complex nonlinear least-squares (CNLS) program is used. or magnitude and phase-angle of the experimental impedance/admittance to a given model.calc and Z"i. disk rotation rates. the sum of squares: N ⎧ 2 2⎫ (241) S = ∑ ⎨wi ' Z i' − Z i' . Usually. 2. finally. (135) leads to simpler data analysis.203. Stoynov and collaborators19. Sometimes vary flat minima are obtained leading to large values of the relative standard deviations of the measured parameters. They also used spectral analysis which could identify the number and nature of time constants existing in the system. CNLS Approximations CNLS Method After validated data are obtained one can proceed with modeling. and the summation runs over all N experimentally used frequencies.56 the system.calc are the values calculated from a given model. parameters A. some prior information about it and good understanding of the characteristics of the measured values. The number of adjustable parameters should be kept to a minimum. In such cases it is advisable to repeat the approximation starting with different initial parameter values. calc ⎬ ⎭ i =1⎩ must be minimized. B. eqn. The F-test of addition of k parameters to the approximating function is described as: (i) [ ] [ ] .197-199 developed mathematical methods for the structural and parameter identification from impedance data. It may be tested using the F-test for the additional parameter. otherwise the CNLS method may become divergent. where Z'i and Zi'' are the real and imaginary parts of the experimental impedances at the frequencies ωi. i. Z'i. Another problem is connected with the goodness of fit and the number of free parameters used in the approximating function. (141). and C are simpler functions of the kinetic parameters than the electric parameters Ra and Ca. the approximation starts with the smallest possible number of parameters. then an additional parameter is added and the decrease in the sum of squares must be compared. For example. are complex functions of the parameters A. The minimization is most often carried out using the iterative Marquard-Levenberg algorithm. all the parameters are used as adjustable. w'i and wi" are the statistical weights of the data. Such a procedure may be tricky and in some cases local minima are found. The identification procedure may give the number and nature of the elements in the circuit. B. Direct use of equivalent circuits may lead to more complex data analysis.18. The parameters Ra and Ca describing the faradaic impedance.205 Because of the iterative nature of the algorithm the initial choice of the parameters is very important: they must lie relatively close to the real values. and C while direct use of eqn. the relative values of the weighted sum of squares or χ2 should be compared. temperatures.e. for a system containing one adsorbed species. several parameters determined. Such a model identification procedure should be supported by a series of measurements at different potentials. Addition of some elements in the circuit may be connected with addition of more than one parameter. then they are kept constant as new parameters are added and. calc + wi '' Z i'' − Z i''.206. eqn. concentrations. etc.

They found that the standard deviations of the real and imaginary impedances are identical and may be described by: 2 ' + β Z " + γ Zi σ i = α Zi i Rm (ii) ( 243) . Another alternative proposed by Macdonald18 was to use proportional weighting.N-p).57 Fexp = [ S ( N − p ) − S ( N − p − k )] / k S ( N − p ) /( N − p ) (242) where S(N-p) is the sum of squares.N-p. If the model is valid the residuals should behave randomly.196. This test should be applied with prudence and always the smallest possible number of added parameters k should be used. and a level of confidence α.α) for k and N-p degrees of freedom.208 Orazem and coworkers195.210 studied the error structure of the impedances measured using a Solartron FRA. (241). Statistical Weights The choice of statistical weights in the CNLS fit is very important. only the largest impedances contribute to the sum of squares S. several repetitions of the experiment allows for the determination of the standard deviation of each point (σi' and σi") and the statistical weights may be obtained as: wi' = (σi')-2 and wi" = (σi")-2. Using this technique it was possible to determine how the impedance errors are transferred to the determined parameters.N-p. these parameters are often measured using the same sensitivity for both components. Such weighting methods mean that the real and imaginary parts of the impedance may be independently determined and that their precisions are independent. Because the measured impedances may vary at different frequencies over several orders of magnitude when unitary weights (wi = 1) are used. in practice. The phase-angle Bode plot is particularly sensitive in detection of time constants.N-p.α).203. depending on their relative values. a better weighting procedure may be use of modulus weighting: wi = 1/(Z' 2 + Z" 2) . However.209. that is taking weights inversely proportional to the measured or estimated impedances: wi = 1/Zi2 or wi = 1/Z2i. eqn. Besides comparing the sum of squares the comparison of the experimental and simulated data should be carried out using complex plane and Bode plots. This method allows sensitivity of the parameters to the random noise to be determined. However. This parameter should be compared with function F(k. If Fexp is lower than F(k. Macdonald207 studied precision of the parameters determined by the EIS. from the statistical tables.α). He added a noise to the simulated impedance data and used the CNLS technique to determine the parameters together with their standard deviation. the hypothesis that the function has p+k parameters must be rejected. although for k parameters Fexp is also > F(k. therefore.calc. In such a case low time constants may be overlooked. In general. If they display regular tendencies it may mean that the model is not correct and further elements should be added. Boukamp203 proposed to study the residual sum of squares after subtracting the assumed model values from the total impedance data. for p parameters and N-p degrees of freedom. S(N-p-k) is the sum of squares for p+k free parameters and N is the number of points. the variations of the residuals should be statistically important. Although this is the best approach such a procedure is time consuming and rarely used in practice. It may happen that when k parameters are added only the first one increases the value of Fexp > F(1.

Macdonalds's program18. and allowed for better determination of system parameters. it may introduce a phase shift. FORTRAN is a language which contains intrinsically complex number calculations which facilitates the programming process. Eco Chemie BV: Autolab. and allows for easy modifications of the subroutines for the model impedance calculations. Tacussel: Z-Computer. S . BAS-Zahner (Thales). etc. although they may range to MHz. is very popular. e.217 but the experiments were carried out without a potentiostat on ultramicroelectrodes. Inc.g. Such an error structure was verified for solid state and electrochemical systems under a wide variety of experimental conditions. Sirotech Ltd. Scribner Associates. Much higher frequencies up to 10 MHz were used by Bara½ski et al. together with the complex sensibility.212-215 VIII.218 If the response of the electronic system is linear the parasitic impedances. Some other programs have been developed in the literature without being widely commercialized. cell. There is no possibility of changing the subroutines or introducing new equations.204.58 where α. Equivalent circuits are constructed from several predefined elements. and for errors ranging form mΩ to MΩ. etc. many weighting possibilities.). from Gamry Instruments. It is based on Macdonald's algorithm and the data analysis was simplified. Boukamp' program203 written in Pascal.211 developed a software which performs a similar task and is able to choose the best equivalent circuit. Zin and Zoc . Its disadvantage is a special formatted data input (however. because at high frequencies. It contains various models already predefined. Inc. Figure 46. (iii) • • • • • • AC Modeling Programs Several programs for the EIS modeling are available: J.216.R. and on the stray capacitance and inductance of the experimental setup (cables. With increase of sensitivity the potentiostats tend to slow down and the response on a 10 mA current scale is much faster than that on 10 µA scale. Several software packages (often with very limited modeling capabilities) come with the hardware. INSTRUMENTAL LIMITATIONS EIS measurements should be carried out over a wide frequency range in order to identify all time constants in the circuit (usually 10 frequencies per decade). It uses a number of predefined circuits without the possibility of modification. distributed with EG&G software. A typical range in modern systems is 20 to 50 kHz. β and γ are constant parameters determined for a given instrumental system and Rm is the value of the current-measuring resistor.202. may be obtained from: S Z m = Z in + ^ ^ ^ 1 1 ^ + 1 ^ (244) Z el Z oc . it can be easily corrected). developed a software (Zplot) for data acquisition and analysis. Corrections for slow response of the potentiostat may be carried out and increases the effective bandwidth by about one order of magnitude.: EIS900. The highest frequency depends on the potentiostat used. source code available. written in FORTRAN.

Three measurements: one in the open circuit and two with two different resistances instead of Zel allow the determination of three unknown complex parameters and further correction of the measured impedance. However. Such resistance should be at least 100 times larger than the measured impedance.223 Similar artifacts. This error becomes negligible at higher frequencies (> 10 Hz for the Solartron) because of the internal delay of the measurements at each frequency.219 Figure 46.222 This inductance arises form that of the leads and the current measuring resistor. It depends on the initial phase of the sinusoidal excitation. It is relatively easy to get good precision of measurements for impedances between 1 and 105 Ω at frequencies below 5×104 Hz. Such corrections should be repeated at each frequency and the cable configuration used for calibration should be the same as for the electrochemical measurements. a calibration procedure is necessary. The errors for high impedance measurements originate from the finite potentiostat input impedance.223 In principle. The use of FRAs may lead to erroneous results when the frequency is swept too fast. Another distortion is observed at very low impedances. are also observed in highly . Figure 47 presents examples of the complex plane plots obtained at several distances in acetic acid. error is introduced into the results. Such effects may be often minimized by shortening the cables and shielding the reference electrode. corresponding to an inductance in series with the electrode impedance. Such behavior arises from the contribution of the reference electrode to the measured impedances and it was explained introducing the resistance and capacitance of the reference electrode and a capacitive coupling between the reference electrode and counter and working electrodes. The distance between the working electrode and the Luggin capillary may also affect the impedance measurements and lead to artifacts.221. Another method involving two working electrodes for high frequency (≤ 5 MHz) impedance measurements was proposed by Schöne and Wiesbeck.220 The change of frequency may lead to a transient regime. This limit is connected with the possible changes in the state of the electrode during long-period measurements. in measurements of protective coatings on metallic surfaces and very low impedances are found in molten salts. increase of the distance between the Luggin capillary tip and working electrode should only increase the solution resistance. Very high impedances are found. which shifts the complex plane plot along the real axis. distortions may be observed. It is observed at high frequencies and leads to large positive imaginary impedances. for example. for lower and higher impedances. If the measurements are performed during this transient. The lowest frequency typically used is 10-3 Hz. observed at high frequencies. It can be neglected when 10 cycles of signal integration and at least five steps per decade are used in measurements. if not. Figure 47. At this frequency measurements averaged over five wave periods take 1 hour 23 min.59 where Zm is the measured impedance and Zel is the impedance of the electrochemical cell.30. The measurement at all frequencies takes a much longer time.

. batteries and fuel cells. It is now possible to acquire. the most difficult problem in EIS is modeling of the electrode processes.) and the main effort is now applied to understand and analyze these processes. They can be minimized by inserting a thin platinum wire into the Luggin capillary and the salt bridge. CONCLUSION EIS has become a mature and well understood technique. These applications will be subject of a second review by the author in a forthcoming volume in this series. that is where most of the problems and errors arise. There is almost an infinite variety of different reactions and interfaces that can be studied (corrosion. conducting polymers.60 conducting solutions224 when the Luggin capillary is located too close to the electrode surface. etc. semiconductors. The above review was dedicated to the understanding of the fundamental processes of diffusion and faradaic reaction at electrodes. validate and quantitatively interpret the experimental impedances. electrocatalytic reactions.224-226 IX. However. chemical reactions coupled with faradaic processes. coatings.

............ Rs = 10 Ω.............. The solid line is a sum of all three contributions... (d) 0..... R = 100 Ω..................... Dependence of real and imaginary parts of the faradaic impedance of a diffusionkinetic process on ω-1/2 at constant potential....21 Figure 14...12 Figure 8............... ...... Solid lines show response without noise.......................... Cdl = 20 µF.. (c) plots for semi-infinite spherical diffusion..............20 Figure 12. C = 20 µF.......linear semi-infinite........... (c) and Bode (d).... Faradaic (dashed line) and total (continuous line) impedance for a reversible reaction in the conditions of semi-infinite cylindrical diffusion.........no optimization........................... (b) 0.........................1. Complex plane (b) and Bode (c)............. Rct = 100 Ω................................... Dependence of cot ϕ as a function of potential for TiCl4 reduction in H2C2O4 at various frequencies... Cdl = 20 µF.. FFT analysis of the sum of sine waves perturbation.......................... ... (e) 0...finite transmissive............ (b) voltage and current phasors in frequency space.... Rct = 100 Ω....... (e).... Solid lines show response without noise............................................. sphericity parameter 2 Di / r0 : (a) ∞ .. (e).......... . presented in frequency domain.................. dashed line: faradaic impedance...................... Faradaic (a) and total (b)-(d) impedances for: a) .... right optimization of amplitudes. plots for a series connection of a resistance and a capacitance.........................................................14 Figure 9............. FFT amplitude......... b) perturbation voltage in the frequency domain................................... in (b) continuous line: total impedance..........no optimization.......... |H(jω)|.... .. σ = 10 Ω s-1/2.......... ... plots for the circuit (a)... left .9 Figure 4..61 List of Figures Figure 1......... Complex plane (a) and Bode (b)... ........ (c) and Bode (d).......23 Figure 15.......10 Figure 6.................4....20 Figure 13........ Rct = 100 Ω.......................................... (c) and Bode (d)...... (38).............................77 .................... c) current response with 10% noise added....51 ............ a) perturbation voltage in the time domain................2 s-1/2.......... (d)............ c) complex plane plots of simulated impedance spectra with 5% noise added to the current response....51 ............................................... (c) 0........... plots for a parallel connection of R and C (a)................................ eqn.. Rct = 100 Ω...23 Figure 16......25 Figure 17...................12 Figure 7......................... (26)................. and c) ....... Dependence of logarithms of Rct and |ZW| on potential for α = 0. Frequency response of a FRA averaging filter for different numbers of integration cycles44... Complex plane (b).................8 Figure 3........ Rs = 10 Ω......... of duration T0 = 1 s.........28 ...... FFT analysis of the sum of sine waves perturbation................... R = 100 Ω................... .... d) complex plane plots of simulated impedance spectra with 10% noise added to the current response.......................... Dependence of the real and imaginary admittances for a diffusion-kinetic process on the electrode potential.......15 Figure 10. Rs = 10 Ω............... Faradaic (dashed line) and total (continuous line) impedances in the conditions of the diffusion to a disk and a reversible charge transfer... b) ....................26 Figure 18..................... ........................... of the pulse function......05...................... Complex plane (b)............ Rs = 10 Ω.......... plots for the semi-infinite linear diffusion model (a).............. Rs = 10 Ω............... Schematic operation of a lock-in amplifier. Construction of the Bode magnitude plot for the circuit in Figure 4a using eqn..................7 Figure 2................ Complex plane (b)........ Representation of ac signals: (a) rotating voltage and current vectors in time space........................finite reflective boundaries..................................... (a) ..... b) perturbation voltage in the frequency domain...27 Figure 19.....................02.. (E)................. C = 20 µF.... left ..9 Figure 5............ right optimization of phases...linear diffusion.....15 Figure 11.......................................................... a) perturbation voltage in the time domain.

...... Rs = 10 Ω... j0 = 10-6 A cm-2..................... T = 0....... l = 0.........1. η0: a) 0....Rct |B| < 0........ ..Rct |B| = 0.................faradaic impedance....62 Figure 20..................107 ......... r = 10-4 cm...total impedance.........94 . r = 10-4 cm.........01 F cm-2 sφ-1... Complex plane plots for LaPO4-bonded Ni powder electrodes during hydrogen evolution in 30% NaOH at 700C.................. eqn....48 Figure 40........ Some examples of the complex plane plots obtained for the case of two adsorbed species.......... D = 10-5 cm2 s--1........... (195): (a) general case................. continuous line .. Cdl = 2×10-5 F............... Complex plane plots on fractal electrodes for different values of parameter φ.........32 Figure 22.......... (144)(146)............. dashed lines ............ Influence of the depolarizer concentration on the complex plane plots for a porous electrode.. Cdl = 2×10-5 F...........40 Figure 30....... eqn....... ....................total impedance................... concentrations in mol cm-3 are indicated in the figure... parameters used: Rct = 100 Ω..025. Complex plane plots for a porous electrode in the presence of a concentration gradient at different overpotentials................ ..........48 Figure 39.........111 ... Complex plane plots for the bounded thickness impedance (BCP) for Rs = 100 Ω........concentrations given in the figure.. pore parameters: l = 0.......33 Figure 25...5 V.46 Figure 38.............41 Figure 34............ dashed line ............................ The distribution function τG(τ) versus ln(τ/τ0) according to eqn..... Complex plane plot for the case of one adsorbed species....... Complex plane plots for a porous electrode at different overpotentials ..........total impedance..159.. lower three curves at E = 0.............. parameters used: Rct = 100 Ω...42 Figure 35......... eqn... continuous line ............................ Cdl = 2×10-5 F.............4. Rs = 10 Ω....... Comparison of a simple (b) and fractal (c) magnifications of the image (a). ρ = 10 Ω cm.......faradaic impedance......... (139)(140).39 Figure 29... Ca = 2×10-3 F........05 cm........ .... CO = 0.....38 Figure 28... Complex plane plots for a porous electrode according to de Levie’s model....107 .... η = 0......... d) 0... Br..................2.......... upper three curves at E = 0...................... Equivalent-circuit model of the electrode in the presence of adsorption. continuous line .32 Figure 23......... L = 0.........................2 H. and f) 0... Ra = 100 Ω......28 Figure 21............149 ......... Calculated impedances for various shapes of a single pore blocking electrode................ (b) limiting case for shallow pores..... ω1/2 for TiCl4 reduction in H2C2O4......3 V............................... and different values of parameter φ......................... j0 = 10-6 A cm-2...01 M........... Complex plane plots for the case of one adsorbed species and B < 0.. Nyquist plot of complex capacitances for Au(111) in 0........... ...40 Figure 31.. ................total impedance.......... eqns..1 V vs...... Ra = -200 Ω..faradaic impedance..... other parameters as in Figure 38.... eqns. Ca = 2×10-3 F............... Continuous lines ....... ρ = 10 Ω cm................... eqn..... Rs = 10 Ω..... parameters used: Rct = 100 Ω........41 Figure 33..... continuous line ...................................(195)....... Complex plane plots in the presence of a CPE element: a) ideally polarizable electrode........49 Figure 41. SCE..........calculated using de Levie's model..... dashed line ......... dashed line ..(195).... Equivalent circuit for the case of one adsorbed species: a) for B < 0........44 Figure 36........... (176).......... (c) limiting case for very deep pores.....52 .........faradaic impedance........................ e) 0. Complex plane plot for the case of one adsorbed species.........2 V................................. b) 0..... dashed line .......147 ................simulated............... Ro = 40 Ω.............. b) in the presence of faradaic reaction.. b) B > 0........ Cdl = 2×10-5 F.....05 cm......45 Figure 37......................... .... j0 = * 10-7 A cm-2.... Ca = 2×10-3 F... B < 0 and C ..........................................1 M HClO4 in the presence of bromide ions....40 Figure 32.... (199). eqn........ m = 1.....33 Figure 26. parameters used: Rct = 100 Ω.. Rs = 10 Ω.. ...... B < 0 and C . Complex plane plots for the case of one adsorbed species and B > 0................................. 77....... (200)................. Data from ref.................. Voigt (a) and ladder (b) models.. ..33 Figure 24...........36 Figure 27.. Pore model...................... Dependence of cot φ vs....3........ c) 0........

............ (b) and (c)....55 Figure 45.........63 Figure 42.................................53 Figure 43.............. (c) ladder.......................................223 ..................... (b) Voigt................18................55 Figure 44................... (b) 100% acetic acid..................................59 Figure 47... (c) mixed.....................177 ................................ Typical circuits used in ac modeling....... Impedance spectrum for an Al electrode in water (a) and its Kramers-Kronig transforms..................................................................... they are experimentally indistinguishable: (a) Voigt.................................................................. .... Equivalent circuit for the case of a slow response of the potentiostat...................59 ............. Three circuits describing a system displaying two capacitive loops: (a) ladder.............. Alternative circuits for the impedance behavior of a system containing one capacitive loop.............55 Figure 46.................................................. (b) Maxwell....... Impedance diagrams for different distances x between the working electrode and the Luggin capillary tip in: (a) 80% acetic acid.......

Damaskin.. Varma and J. J. 1991. . Fundamentals of Electrochemical Science.D. Burlington House. Bond.J. Vol. 19 Z. J. Stoynov. Ed. New York. 241. p. Oldham and J. Wiley. Dekker. and Material Scientists. VCH. 41 (1947) 441. A Specialist Periodical Report. London. A. Rev. Silverman and M. Armstrong. J. Bell. Macdonald. 9 K. 8 P. I. Nauka. Electrochemical Impedance. New York. p. Electrochemistry.D. P. 1978. Principles.W. 1966. Plenum Press. London. The Chemical Society. Galus. Interscience. New York. Rubinstein. 1. New Instrumental Methods in Electrochemistry. New York. 1991. 1977. 6 D.R. New York. Wiley & Sons. and B.R. Methods. and A. Vol. Brett and A. Vetter.B. V. R. Electrochemical Kinetics. p. Moscow. 1954. Vol. 16 H. 12 Instrumental Methods in Electrochemistry. Smith in Electroanalytical Chemistry. Bauer. Macdonald in Techniques for Characterization of Electrodes and Electrochemical Processes. B. p. 1980. San Diego. 1993. Mohilner in Electroanalytical Chemistry. 1966. 1965.I.. 581. Delahay. B.M. Gileadi. Ellis Horwood. New York. 14 C. 17 Physical Electrochemistry. Metcalfe in Electrochemistry. Philadelphia.M. Oliveira Brett. Wiley-Interscience.C. Eds. 7. 24 S. Savova-Stoynova.R. 5 B. 18 J. 6. 98.. Selman.J. Burlington House.J. Archer and R. 1980. Modern Aspects of Electrochemistry. Wiley. Academic Press.F. H. Transient Techniques in Electrochemistry. Lorenz in Techniques for Characterization of Electrodes and Electrochemical Processes. 25 Electrochemical Impedance: Analysis and Interpretation. New York. 7 A. and Applications.J. Thirsk. 1991 (in Russian). Engineers.A. Macdonald. p. and Applications. Thirsk.C. Ed.R. Armstrong. New York. p. New York. 1994. Alternating Current Polarography and Tensammetry. Electrode Kinetics for Chemists. The Chemical Society. by J. Principles. Wiley-Interscience. 103. 1994.V.B. Faulkner. Delahay. 3 P.M.C. 22 D. 1. Fundamentals of Electrochemical Analysis. 21 W. 11 A. Oxford University Press. Methods. Parsons. Ed. New York. ASTM. Chemical Analysis Series. Scully. in Electrochemistry. 515. Selman.B. Bard. 1993. Wiley & Sons. 13 E. Ed. Kendig. A. 20 R. 1993. Chem. 23 F. Kolthoff. 10 D. 15 Z. Myland. New York. Electrochemical Methods. Southampton Electrochemistry Group. Graham. Academic Press. 4 D. by R.M..R. 157.S.R. New York. Chichester. Ed.M. 1991. Mansfeld and W. Szpak in Techniques for Characterization of Electrodes and Electrochemical Processes. 1954.. Wiley & Sons.R.. Breyer and H. 1963. Bard.A.D. Ed.H. Ed. Impedance Spectroscopy Emphasizing Solid Materials and Systems.R. New York. Elving and I.D.64 References 1 2 D. 1987. Ed. Vol.. Varma and J. H.M. Dekker. Dekker. Ed. 677. p. M. Dekker. 1967. by R. Ellis Horwood. Grafov. A Specialist Periodical Report. New York. 1995. Selman. 1. D.J. Bard and L. Varma and J. 1980. 1985.J. Elkin.E. New York. by R. New York. Double Layer and Electrode Kinetics. Modern Polarographic Techniques in Analytical Chemistry. p.

J. Instrum. 1984. J. 139. Chem. Sluyters. Calculations. Doblhofer and A.. Frequency Response Analysis. 1981.H. C.J. Thirsk.. and H. Larsen. Wilson. P. J. Tribolet. S. and D. 139. R. Hampshire. Mattson. R... Motilla. Chem. 13 (1968) 215. R. by A.G. J.M. and D. G. Jr. M.. Sluyters-Rehbach and J. Pilla in Electrochemistry.. ASTM.C. Electroanal.. H. Electroanal.J. Mark. and H. Analytical Chemistry. Keddam. Takenouti in Electrochemical Methods in Corrosion Research. Identification of Electrochemical Processes by Frequency Response Analysis.J. Goldman. Materials Science Forum. Birgham. Simulation and Instrumentation. J.B. Chem. New York. 177. p.E.E. Pilla. O’M. Schwall. and D. New York. Commun. G.W. Husovsky. Solartron Instruments. Acta. Bond. J.. J. Electroanal. R. 64 (1993) 3111. Transformation Calculus and Electrical Transients.S.D. J. Evans.M. and Instrumentation. 14.. N. 1953. Anal. . Gabrielli. The Fast Fourier Transform. R. Technical Note No 10.. 1986. Keddam and H.H. Ed. Electroanal. 85 (1977) 231. W. Technical Report No 004. 39 (1972) 91. 47 (1973) 9. Ed. Bulgarian Chem. Princeton Applied Research. p. 727. Keddam. Eds. Bond. B. Doetsch. MacDonald. Bard. Diard.E. D. Smith. M. Marcel Dekker. Gabrielli.E.H. B. Conway and R. M.. Gabrielli. Chem. Marcel Dekker. J. A.S. Loyd. by J. Pilla in Electrochemistry. New York. 318 (1991) 145.S. Williams. C. April 1996. White. Chem. p. Marcel Dekker. 61.E.. Anal. Hayes.. Chem. Electrochem. Electrochim. Technical Report No 24.J. 1970. 377 (1994) 61. p. 27 (1994) 459. 1990. Prentice-Hall. 1974. Electroanal. Electrochemistry Newsletter. 35 (1990) 1553. Vol.. 48 (1976) 221A. III.E.A.J. Chem. 1950. 1982. 20 (1969) 181. Sluyters. Takenouti. 2nd. E. Use and Applications of Electrochemical Impedance Techniques. 1990. S. McKurbe in Electrochemical Corrosion Testing. J. Electroanal.N. Calculations. 1. Ed. Mark. C.. D.P. Wellstead. 252 (1988) 151. H. Pilla.C. Gabrielli. de Levie and A. Popkirov and R. A. Chem. C.D.C.J. J.C. Smith. Muszalska and J.B. Jr. Dover.F.. A Lock-In Primer. Acta.J. and H.65 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 M. A. Prentice-Hall. and E. R. and O.O. Le Gorrec. Chem. Solartron. Vol. Comprehensive Treatise of Electrochemistry. Chandrasekara Pillai.A. K. Acta. Schwall. ed.N.H.J. A. 49 (1977) 1797. D. Vol. M. Laplace Transformation. 141 (1994) 957. MacDonald. 1972. Waghorne. Theory and Problems of Feedback and Control Systems.R. McKurbe in Modern Aspects of Electrochemistry. Armstrong.D.E. Sluyters-Rehbach and J. Electroanal. Electrochem. Stubberud and I. Jr. J. Lizee. Duprat. Ed. Soc. Schindler. and J... Electroanal. Schaum’s Outline Series. DiStefano. Race. Montella. Bockris. Plenum Press. p. Jarzebska. Rev. by J. 34 (1989) 831. New York. EG&G. C. C. 9. McGraw-Hill. Diez. Chem. 205 (1986) 59. p. and H. 8. Vol. 1972. Electrochim. Plenum Press. A. p. Gabrielli and B. A. Mattson. J. Macdonald and M. Simulation. Smith. Solartron. and C. W. J. M. Sci. M. Princeton. Soc.E. Schindler. 11. Hampshire. 4. Creason. Philadelphia. 117 (1970) 467.A. Popkirov and R.. Macdonald and M. K.R.A.A. N. A. J. ASTM Special Publ. 417. Gabrielli. G.S. Smith. 1984. Gomez.J.-P.C.. Electrochim. 1986.. New York.

Thirsk.D. Firman. Firman. Chim. 22 (1969) 29.-P. J. 294 (1990) 123. Grahame. 11 (1996). Montella. Discuss.E. J. and E. Corrosion. Chem. 29. Montella. White. 1053 (1997). Electroanal. 317 (1991) 1. D. Chem.-N. 272 (1989) 11. Moreno. Scantlebury. Rec. S. L... K. 30.R. 4 (1970) 1.. J. 39 (1983) 61. A. Soc.. Fleischmann.. T. Chem. Pronzato. Warburg. 35 (1990) 831. Chem.A.B. 42.-P.S. 326 (1992) 13.P. J. Bockris. Chem. T. 90 (1993) 2069.-P. Chap. Walter. Diard. F. Phys.D. Elman. S. Electrochim. Berthier. Electroanal. J.E. J. M. Diard. J. D.. Electroanal. Philadelphia. K. Electroanal.. E. Berthier. 35 (1990) 837. and C. and E. Pons. and E. J.. Electroanal. J. R. Armstrong and R. p. 51 (1995) 664. 327 (1994) 33. Chem. 300 (1991) 199. Chem. J. Arvia. 410 (1996) 247. Conway. J. 38 (1993) 1803. Darowicki. Taylor and E. S. Acta.. Acta. by M. Acta. Salvarezza and A. Chem.-P. 137 (1990) 3737.. Soc. and J. Sci. Jacobsen and K. Yeager. Soc. Landaud.. Transactions of the Symposium on Electrode Processes. 209. Montella and C. J. J. J. J. Acta. E. 40 (1995) 767. 30 (1995) 255. Chim. Acta. Sluyters. 56 (1973) 244. Chem. Electroanal. Corr. Bai and B. Le Gorrec.-P. J. 45 (1973) 3. Montella.A. NC. Walter. Darowicki. Farad. 2. 1 (1947) 11. Le Gorrec. D.E. Randles. Electrochim. 403.H.C. Chem. F. Electrochim..-P. Walter.B. Cao. Montella. Ultramicroelectrodes.E. C. R. Electroanal. Electroanal. Diard and C. Electroanal. L. F. J.. Pons and J.. Morganton. Sagüés. Diard.D. Daschbach. A. 42. Cao. B.E. Electrochim. Chem.J. J. J. 4-6 July 1994. Schmidt.-P. 1959. Pronzato. Electrochem. X. Electrochim. Callow. Inc.C. New York. Electrochim. and C. 352 (1993) 1. Berthier.A. 32 (1996) 973.J.D. Electroanal. . Electrochem. D. Acta. M.. Harrington. Eds. B. Farad. J. Bai and B.C. Montella. De Levie and A. VanderNoot. Wu and W. L.. Ed.. Chem. Chem. Soc. Fleischmann. Electroanal. 383. Chem. C. O'M. Chem. Montero-Ocampo. R.H. Acta. Diard.-N. Electrochem. Armstrong and M. Conway.. Wiley. Neumannspallart and M. Phys. Electroanal. F. S. Berthier. Sluyters. 99 (1952) 370C.. 362 (1993) 13. J.. J. C.E. Armstrong... Diard. Zhang. P. Le Gorrec. R. Edt. Electrochim. J. M. B. Ye and K.R. Chem. (1995) 1. Rami. Electroanal. L.. J. 1987. Chem.66 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 J. Henderson. and H.. and J. Sluyters-Rehbach and J. J. Automatica. Diard. J. Smith. 1073 (1997). Trav. Electroanal. J. Lasia and A. L. West. Randles. Krans. B. R. 39 (1972) 81. 79 (1960) 1092. Vol. Soc. 403 (1996) 11. Doblhofer. 67 (1899) 493. L. Fleischmann and S. J. and E. Chem. Disc.M.R. Copenhagen. Electroanal.. J. Pons. Conway. Husovsky. J. Ann. F.. Electroanal. J. Anal. Electrochim. 250 (1988) 277. Berthier.-P. Harrington. Modern Aspects of Electrochemistry.E. J.E. 41 (1996) 1239. Chem. Pronzato. Proceedings of the 10th IFAC Symposium on System Identification.I. Gill. D. C. 2661.. Datatech Systems.A. R. Acta. R. 35 (1963) 610. Gileadi..E. and C. M. Plenum.. Diard. Rolison and P.

116 L. Electrochim. Sci. Pajkossy. 281 (1990) 23. Cole. Gray and S. G.. A. Electrochem. 281 (1990) 1. Nyikos and T. 23 (1993) 135. Devaux. Stoynov. 22 (1992) 376. Liu. Cole and R. Peyrière. J.. Electrochim. Wang.H. Electrochim. Solid State Ionics. 33 (1988) 445. Electrochim. 279 (1990) 67.S. Reinhard. Chazalviel and J. Pajkossy. Sakharova. Electrochem. Electroanal. Chem. Chem. 38 (1988) 5867. Electrochim. J. Acta. San Francisco. Sluyters-Rehbach. Chem. 126 W. 306 (1988) 117. Mat. 28/30 (1988) 1441. Kaplan S. 107 T. J. 121 A. 108 R. Chem. J. T. 9/10 (1983) 17.H. Lett. 102 Z. Daccord and R. Heterogen. Acta. Rotenberg and N. Electroanal. Nyikos and T. Pajkossy. Acta. Nyikos. Solid State Ionics.. Rev. 109 Z.W. Electroanal. Wrona. Phys. Nyikos and T.M. 110 B. Acta. Chem. Appl. 132 B..-N. 300 (1991) 1. 136 L. 36 (1991) 163. 138 E. Rev. Appl. Reinhard. 122 T. Pajkossy and L. Fys. 35 (1990) 1493. 134 D. Acta. G. 31 (1986) 1347. Electrochem. Acta. Liu. 35 (1990) 1045. 125 R. de Levie.. Chazalviel and J.-N. 139 T. Rami and A. 23 (1987) 253. 128 S. Gray. Le Méhauté. Electroanal. Phys. J. Electrochim.. Pajkossy and L. Phys. Vogt. and J. Electrochim.M. Pajkossy.. Electrochim. Chem. 261 (1989) 1.. Keddam. 414 (1996) 209. 25 (1989) 651. Peyrière. VanderNoot. Acta. and A. 115 A. Mulder. A. Chem. Rammlet and G. Acta. Electrochim. J. and M. G. Nyikos and T. Acta. Chem. Pajkossy. Electrochem. 1 (1904) 681. 9 (1941) 341.C. 35 (1990) 1567. Electrochim. Lasia.. C. 117 L. Solid State Ionics. Chem. and D. J. J. 113 L. Los. 326 (1992) 231. 35 (1987) 5379. Phys. 34 (1986) 4870. J..A. de Levie. A. 140 R. 33 (1987) 707. J. Pajkossy. Chassaing. 130 T. L. Electroanal.V.J.. J. J. Pajkossy. Crépy. Nyikos. Liu and L. Sapoval. Corr. Nyikos. Soc. Acta. 176 (1984) 275. Pajkossy. Lasia. Phys. Ménard. 105 P.H. 38 (1993) 1615.G. Nyikos and T. Keddam and H. Kaplan. Electroanal. 133 B.. 27 (1987) 373. Nekrasove. de Becdelievre.. Le Méhauté and G. 2 (1995) 143. Borossy. Rev. C. 123 M. J. L.B. Sapoval. Rev. 129 T. Electroanal.P. 104 U. Rev.. Pajkossy.K. Sapoval.H. Electrochim. J. Acta. 137 P. Sapoval. 127 J. 37 (1992) 1283. 386 (1995) 57. Brug. R. 112 H. Chen and A.. Freeman. 135 A.H. 139 (1992) 3458. de Levie and A. Phys. Crépy and A. Soc. Lenormand. 332 (1992) 55. Nyikos and T.J. A. 106 T... The Fractal Geometry of the Nature. Rami. 1982. Acta. 19 (1951) 1484. J. Cole. Astron. 124 T. Electrochem.. Chem. von Koch. J. Chem. 37 (1992) 973. Wandlowski. Takenouti. 103 U. M.. 131 B. Paris. Duret-Thual. Kolb. 118 T.. Hurd. 34 (1989) 181. Sov. J. Lasia. 30 (1985) 1533.. 100 . Lasia. Ark. L. Pajkossy. J. Rammlet and G. 133 (1986) 2061. 120 L. Electroanal. B. Electrochim. M. 111 R. Chem. Lessard. A.L.J. Electroanal. van den Eeden. 278 (1990) 25.. Sci. 364 (1994) 111. B. Electroanal. Acad. 119 A. J.J. Chem.. Electrochim. 34 (1989) 171. 101 T. Pajkossy and L. 55 (1985) 529.67 99 K. and H. A. Davidson and R. 114 T.H. Chem. Mandenbrot. Acta. Sluyters. Electroanal.

Electroanal. 176 C. 397 (1995) 27.W.. de Levie. Ksenzhek. Moors.. Orazem. Scully. Jüttner. Halsey and M. S. in Electrochemical Impedance: Analysis and Interpretation. and L. Macdonald. 157 R.D. J.S. ASTM.. 177 B.C. 164 H. 160 K. Interscience. 145 J. Keddam.R. Los. Chem. 168 M. 154 C. Real and D.E.P. J. p..C. Kending. Leibig. Eds. Scott. Takenouti. Smedley in Electrochemical Impedance: Analysis and Interpretation. 137 (1990) 3306. Acta.W. Van Nostrand. 1993.J. J. Chem. and H. Real and D. J. 163 G. 151 L. 152 S. 159 H. 173 M.D. J. 39 (1994) 1297. Chem. 13 (1983) 709. Appl. and M. Electroanal. 166 H. and M. 35 (1990) 1559. Electrochem. F. ASTM. 140.D. Raistrick. J. 175 P. 392 (1995) 1. van Peteghem.E. Vol.. 360 (1993) 101.D.M.. 378 (1994) 39. Electrochim. 155 I. Kending. O. 138 (1991) 67. Opt. 21 (1976) 539. Electroanal.K.C.A. Posey. Am. Agarwal. Lasia. Urquidi-Macdonald. J. Soc. 10 (1965) 113. 329. Chem. 1945. and M. J. 155 (1997). in P. 165 R. 162 M. J. 170 J.C. Eloot. J.. 171 M.. Rangarajan. 172 D. Eds. J. J. Electrochim. Electroanal. Urquidi-Macdonald. Beccu.A. Electrochim. New York. Network Analysis and Feedback Amplifier Design. Orazem. Silverman. Philadelphia.D. and A. 334. 154. Electrochim. Inzelt and G. 1967. 133 (1896) 2018. de L. Russ. Zeit. Bisang. Garcia-Rubio in Electrochemical Impedance: Analysis and Interpretation. 36 (1962) 331. Macdonald. Leibig and T. J.A. 169 M. Z. and L.M. Electrochem.. 144 F. Electrochem.N. Soc. Electrochem. 1993.R. Silverman. Silverman.. Acta. S. K. A. Electrochem. Electrochem. 22 (1969) 89. D. 23 (1949) 1477. Esteban and M. 150 J. Rakomotavo.. Philadelphia.. p.. 147 P. 47 (1991) 248. 137 (1990) 515. Keddam. New York. 36 (1991) 1699. Soc. Cachet and R. J. Electrochim.S... ASTM STP 1188. Austin and H. 132 (1985) 2316. M. Lasia.E. Electroanal. 25 (1995) 326. Macdonald. Macdonald. M.. ASTM STP 1188. Kronig. Kreysa. Soc. Dougherty and S. Kramers. Acta. 143 A. J. Soc. M. Zh. Lasia. Takenouti...W.. 35 (1990) 1579. Keiser.. Tobias.. J. Soc. Gutjahr.C. Urquidi-Macdonald and D. Eds. 6. Scully. Soc. Chem. Wiart. Phys.D. 156 A. K. J. Corrosion. 161 T. Brossard. and M. 115. 9 (1964) 1469. J. 148 K. . p. J. Frumkin. Electroanal. J. Kending. Scully. Esteban and O. 111 (1962) 1173. Ménard.. J.R. Acta. 428.. Halsey. and G. Soc. Orazem. Phys. and H. Acta. 12 (1926) 547. Electrochem. Appl. Electroanal. ASTM. Acta.G. Advances in Electrochemistry and Electrochemical Engineering. Gunning. Chem. Electroanal. Acta. Appl.H. Chem. Urquidi-Macdonald.68 141 142 A.. J. ASTM STP 1188. Bode. 109 (1962) 1183. Gabrielli. Townley.M.D.. D. Fiz. Lerner. Electrochem. 30 (1929) 52. p. 149 A. 174 D. 146 R.. Láng. 66 (1962) 287. H. Macdonald and M. 153 M. 1993.. Electrochim. Elektrochem. 158 J. Moghissi. Chem. Uurquidi-Macdonald. Soc. J. D. Chem.I. J. C. Electrochem.). 137 (1990) 3305. de Levie. 358 (1993) 77. 195 (1985) 21. J. Khim. Winsel. M. 14 (1984) 437. Electrochim. Electrochem. Debuyck. Delahay (Ed. Philadelphia.W.W. Electrochem. 167 D. Newman and C.

Brown. Electroanal.. Numerical Recipes. Savova-Stoynov and Z. 4149 (1995). Kendig and F. 186 G.A.D. Macdonald. Savova-Stoynova. Vetterling. Acta. 188 J. Electrochim. Electrochim. New York. 187 J. Stoynov.R. 74 (1994) 85.B..N. B. 216 A... and C. Tsai and D. 4th edition. P. Acta. Solid State Ionics. Chem.W.D. J. 37 (1992) 2357. Max. Garcia-Rubio. McGraw-Hill. Orazem.. Corr. Cambridge. Macdonald.V. Agarwal. Agarwal. Landaud. Agarwal. 375 (1994) 45. 34 (1989) 1187. 41 (1996) 985. Agarwal. Burstein and C. Wiley. 214 J.. Chem. 203 B.E.W. J. Agarwal and M. and M. Electroanal. Stoynov and B. Schoonman.T. 183 (1985) 133. Solid State Ionics. Electrochim. J.B.S. 4159 (1995). Acta.R. Piersol. J.. L. 210 P. Chem. McGraw-Hill. 213 M.H.-T. P. J. 38 (1993) 1919. 1969. J. Thesis. Méthodes et techniques de traitement du signal. Szulborska. Mansfeld. Acta. Liu. Zoltowski. Diard. 39 (1994) 2585. Chem. Chem.. Barański and A. 195 M. Nguyen. 206 P. Electrochim. Flannery. Chem. Orazem. 202 J. Paris (1985).V. Kocsis. Electrochem.W. 35 (1990) 1493. 201 P. 38 (1993) 1883. Láng.Sc. B. Whitmore. 180 C. Acta. Teukolsky. Electroanal. Zoltowski. J. 41 (1996) 1017. Z. Soc. Chem. Berkeley. 183 B. Montella.A. Garcia-Rubio. Macdonald.R.. Walter. Acta. 215 G. Stoynov. 37 (1992) 2353. Churchill and J.-P. 38 (1993) 1047. Bendat and A. Electrochim. Garcia-Rubio. Mansfeld. 139 (1992) 1917. Boukamp.G. Soc. New York. A. 1987.S.M. W. 204 R. Electrochem. Stoynov and B. Wojcik. Jansen. 184 B. 200 P. Sirotech Ltd. M. Le Gorrec. J. G. 1966. 196 M.A. Boukamp. Milocco. Orazem.E. Acta.E. Soc. J. Israel.R. 197 Z. Kendig. Montella. and L. J. Diard. Ness-Ziona. 142. Solid State Ionics. P. Electrochem. Acta. Electrochim. 178 (1984) 11. 378 (1994) 51. New York. 193 Z. Electroanal. Press.H. Electrochim.. and F. Madurasinghe. Electrochem.B. E.N. Data Reduction and Error Analysis for the Physical Sciences. Orazem.E. S. J. 181 P. Le Canut. 185 J. L.S. 205 (1986) 77. Corrosion. and L. Le Gorrec and C. and A. Orazem.A. 192 B. Acta.E. 209 (1986) 11. Haili. Cambridge University Press. Sci. 68 (1994) 1171. J. O.T. The Art of Scientific Computing. Boukamp. 209 P. and B. 212 Y. Zoltowski. 211 Equivalent Circuit Analysis Technology. 307 (1991) 1. Orazem. Electroanal. Acta. 20 (1986) 31.B. 1994.H. M..H. 191 Z. and L. Complex Variables and Applications.. Masson. Electrochim. 38 (1993) 1903. Electrochim.E. Electrochim. J. and L. . 199 Z. 7 (1982) 129. 205 W. Garcia-Rubio. 131 (1982) 77. Polish J. Savova-Stoynov.P. M.P. 182 P. 179 M. University of California. Crisalle. Bevington. P. Measurements and Analysis of Random Data. 142. Lehnen.H. 273 (1989) 243. Acta. Acta. 190 G. 189 J. 207 J. 1990. 39 (1994) 873. Meyer. J..-M.. Inzelt. M. Stoynov. 1992. Electroanal. Garcia-Rubio. and G. 23 (1983) 1007.A. 39 (1983) 466.H.-P. Electrochim.R. Electroanal. Soc. Electroanal. Stoynov. 37 (1992) 245. 208 P. J. Electroanal. Stoynov. Chem. Acta. Electrochim. D. Lindberg. Electrochim.W.69 178 R. J. Chem. 142 (1995) 1885. Chem. 198 Z.. Macdonald.H. Agarwal..

J. Shiller. Electrochem. Electroanal. 297 (1991) 297. R. 229 (1987) 407.D. Chem.S.. J. Lizee. and H. 267 (1989) 279. Barański and W. P. Chem. A. 226 S. 180 (1984) 273. Acta. Teschke. 12 (1996) 3295... Chem. Hughes. Takenouti. and C. Chem. 220 C. Barton. Electroanal. Soares. H. J. M. Chem. M. 1991. Appl. Karunathilaka.. Wiesbeck. Göhr. 225 W. Schöne and W. Lasia and P. Jr. 163 (1984) 419. Electrochim.. Electroanal. Lu. Langmuir. Electroanal. and N. unpublished data. 219 G. D. Keddam and J... Mirnik. Fletcher and M. Los. 224 A. Eichner.F. Chechirlian.A..M. M. Hampson. Chem.G. Moyana. 15 (1985) 251. Botter. 221 H.S. 222 S. M.70 217 218 A. Horne. Keddam. Gabrielli. Mazille. J. 223 S. J. and O.A. . 35 (1990) 1125. 260 (1989) 1. J. Electroanal. Electroanal.R. J.A. Barański and A.

. (b) voltage and current phasors in frequency space. Representation of ac signals: (a) rotating voltage and current vectors in time space.(a) (b) i ~ i E f ~ f E Figure 1.

. Complex plane (b).a) C R b) c) d) e) Figure 2. C = 20 mF. plots for a series connection of a resistance and a capacitance. (a) . (c) and Bode (d). R = 100 W. (e).

(E). C = 20 mF. Complex plane (b). (c) and Bode (d). .a) C R b) c) d) e) Figure 3. plots for a parallel connection of R and C (a). R = 100 W.

Rct = 100 W.a) Cdl Rs Rct b) c) d) e) Figure 4. (c) and Bode (d). Rs = 10 W. Complex plane (b). Cdl = 20 mF. . plots for the circuit (a). (e).

(26).Figure 5. Construction of the Bode magnitude plot for the circuit in Figure 4a using eqn. The solid line is a sum of all three contributions. .

Eout Input signal Mixer Reference signal Phase shifter Low-pass filter Output signal Figure 6. . Schematic operation of a lock-in amplifier.

Figure 7.44 . Frequency response of a FRA averaging filter for different numbers of integration cycles.

eqn. .Figure 8. |H(jw)|. FFT amplitude. (38). of the pulse function. of duration T0 = 1 s.

c) complex plane plots of simulated impedance spectra with 5% noise added to the current response. right optimization of phases.51 . left .no optimization.Figure 9. b) perturbation voltage in the frequency domain. a) perturbation voltage in the time domain. Solid lines show response without noise. FFT analysis of the sum of sine waves perturbation.

d) complex plane plots of simulated impedance spectra with 10% noise added to the current response. left . presented in frequency domain. Solid lines show response without noise. a) perturbation voltage in the time domain. FFT analysis of the sum of sine waves perturbation. c) current response with 10% noise added.51 .Figure 10. b) perturbation voltage in the frequency domain.no optimization. right optimization of amplitudes.

in (b) continuous line: total impedance. Rct = 100 W. Complex plane (b) and Bode (c). . Rs = 10 W.(a) (b) Cdl Rs Rct (c) RW (d) Figure 11. Cdl = 20 mF. s = 10 W s-1/2. dashed line: faradaic impedance. plots for the semi-infinite linear diffusion model (a). (d).

Dependence of logarithms of Rct and |ZW| on potential for a = 0. .4.Figure 12.

. sphericity parameter 2 Di / r0 : (a) ¥ .02. Rs = 10 W.linear diffusion. Complex plane (a) and Bode (b). Rct = 100 W.1. (b) 0.2 s-1/2.05. (c) plots for semi-infinite spherical diffusion. (e) 0. (d) 0.(a) (b) (c) Figure 13. (c) 0.

Figure 14. Faradaic (dashed line) and total (continuous line) impedance for a reversible reaction in the conditions of semi-infinite cylindrical diffusion, Rs = 10 W.

Figure 15. Faradaic (dashed line) and total (continuous line) impedances in the conditions of the diffusion to a disk and a reversible charge transfer.

(a)

(b)

(c)

(d)

Figure 16. Faradaic (a) and total (b)-(d) impedances for: a) - linear semi-infinite, b) - finite transmissive, and c) - finite reflective boundaries, Rs = 10 W, Rct = 100 W.

Figure 17. Dependence of the real and imaginary admittances for a diffusion-kinetic process on the electrode potential. .

Figure 18. . Dependence of real and imaginary parts of the faradaic impedance of a diffusionkinetic process on w-1/2 at constant potential.

Figure 19. Dependence of cot j as a function of potential for TiCl4 reduction in H2C2O4 at various frequencies.77 .

77. w1/2 for TiCl4 reduction in H2C2O4. . Data from ref. Dependence of cot f vs. .Figure 20.

(a) (b) Figure 21. . b) B > 0. Equivalent circuit for the case of one adsorbed species: a) for B < 0.

dashed line . Rs = 10 W. Cdl = 2´10-5 F. parameters used: Rct = 100 W. Complex plane plots for the case of one adsorbed species and B < 0. continuous line . (139)-(140). Ca = 2´10-3 F.total impedance. eqns. Ra = 100 W.Figure 22. .faradaic impedance.

parameters used: Rct = 100 W.faradaic impedance. .Rct |B| = 0. Cdl = 2´10-5 F. Complex plane plot for the case of one adsorbed species.total impedance. Rs = 10 W. Ca = 2´10-3 F. B < 0 and C . dashed line .Figure 23. continuous line .

parameters used: Rct = 100 W. Rs = 10 W.Rct |B| < 0. continuous line . Cdl = 2´10-5 F. Ra = -200 W. dashed line .total impedance.Figure 24. .faradaic impedance. Complex plane plot for the case of one adsorbed species. B < 0 and C . Ca = 2´10-3 F.

(144)-(146). continuous line . Rs = 10 W.total impedance. eqns. dashed line .Figure 25.2 H.faradaic impedance. Ro = 40 W. . Complex plane plots for the case of one adsorbed species and B > 0. parameters used: Rct = 100 W. Cdl = 2´10-5 F. L = 0.

94 . Some examples of the complex plane plots obtained for the case of two adsorbed species.Figure 26.

. The distribution function tG(t) versus ln(t/t0) according to eqn. (176).Figure 27.

Complex plane plots in the presence of a CPE element: a) ideally polarizable electrode.a) b) Figure 28. b) in the presence of faradaic reaction. .

Equivalent-circuit model of the electrode in the presence of adsorption.107 .Figure 29.

Br.1 V vs.3 V. Nyquist plot of complex capacitances for Au(111) in 0. lower three curves at E = 0.concentrations given in the figure.107 . SCE.1 M HClO4 in the presence of bromide ions.Figure 30. upper three curves at E = 0.

01 F cm-2 sf-1. . T = 0.Figure 31. and different values of parameter f. Complex plane plots for the bounded thickness impedance (BCP) for Rs = 100 W.

Comparison of a simple (b) and fractal (c) magnifications of the image (a).Figure 32.111 .

. Complex plane plots on fractal electrodes for different values of parameter f.Figure 33.

I r l Figure 34. . Pore model.

eqn. (c) limiting case for very deep pores. (b) limiting case for shallow pores. (199). . (200).Figure 35. eqn.(195). eqn. eqn. . Complex plane plots for a porous electrode according to de Levie’s model. (195): (a) general case.

Complex plane plots for LaPO4-bonded Ni powder electrodes during hydrogen evolution in 30% NaOH at 700C.Figure 36.147 .

3. r = 10 W cm. pore parameters: l = 0. j0 = 10-6 A cm-2. eqn.025.4.simulated.1. Complex plane plots for a porous electrode at different overpotentials .calculated using de Levie's model. h0: a) 0. d) 0. r = 10-4 cm.05 cm.a) b) c) e) d) f) Figure 37. b) 0.(195). e) 0. dashed lines .2.5 V. and f) 0.149 . c) 0. Continuous lines .

j0 = . l = 0.01 M. D = 10-5 cm2 s--1. r = 10-4 cm.* 10-7 A cm-2. CO = 0. Figure 38. r = 10 W cm. m = 1. Complex plane plots for a porous electrode in the presence of a concentration gradient at different overpotentials.05 cm.

other parameters as in Figure 38.2 V. Influence of the depolarizer concentration on the complex plane plots for a porous electrode. j0 = 10-6 A cm-2.Figure 39. concentrations in mol cm-3 are indicated in the figure. . h = 0.

159 . Calculated impedances for various shapes of a single pore blocking electrode.Figure 40.

. Voigt (a) and ladder (b) models.(a) C1 R0 R1 C2 R2 C3 R3 C4 R4 Cn Rn (b) R1 R2 C1 R3 C2 R4 C3 Figure 41.

177 .Figure 42. (b) and (c). Impedance spectrum for an Al electrode in water (a) and its Kramers-Kronig transforms.

Alternative circuits for the impedance behavior of a system containing one capacitive loop.2 1 2 2 1 2 Figure 43. .

(a)

C2 R1 R2 R3

(b)

C3

C2 R1

(c)

C3 R3

R2

R3 R1

C3 C2 R2

Figure 44. Three circuits describing a system displaying two capacitive loops: (a) ladder, (b) Voigt, (c) mixed.

Figure 45. Typical circuits used in ac modeling; they are experimentally indistinguishable: (a) Voigt, (b) Maxwell, (c) ladder.18

Zin Zel Zoc

Figure 46. . Equivalent circuit for the case of a slow response of the potentiostat.

223 . Impedance diagrams for different distances x between the working electrode and the Luggin capillary tip in: (a) 80% acetic acid. (b) 100% acetic acid.Figure 47.

Sign up to vote on this title

UsefulNot useful- Electrochemical Impedance Spectroscopy Manual
- Electrochemical Impedance Spectroscopy
- Electrochemical Impedance Spectroscopy Library
- Thesis Kraner Impedance Spectroscopy Nyquist
- Handbook of electrochemical impedance spectroscopy.pdf
- Eis
- Piezoceramics
- Basics of EIS
- Basics of Electrochemical Impedance Spectroscopy
- Fundamentals of Electrochemistry (Bagotsky, Wiley 2006, ISBN 0471700584)
- Electrochemical Impedance and Noise
- Cadex Electronics - Batteries in a Portable World
- Fundamentals of Crystallography_Giacovazzo 2000
- Absorption Characteristics of Fibrous Material Covered
- 01451223
- r-c combination.pdf
- 04205953
- Single Phase Tap Fault Calculation
- 261_CourseNotes_Part3c
- NEW CHAPTER 4 SHORT CIRCUIT CALCULATIONS (1).pdf
- The Fifty Ohm Enigma2
- 27847840 Distance Protection Utility Main Protection for Transmision Lines
- 2nd Sem DIP Electrical Circuits - May 2015.pdf
- Vibration Monitoring System-3500
- r05310205 Linear and Discrete Systems Analysis
- PSAO Problems
- lecture3 (1)
- The Art of Protecting Electrical Systems 11
- Unit-4
- A Method of a Lightning Surge Analysis Recommended in Japan Using EMTP
- Impedance