Argentum Solutions, Inc.

    Sterling guidance on corrosion and materials degradation


 

Potential-pH Diagrams
THERMEXPERT - Potential-pH diagram generator

Intelligent Tools

POLEXPERT - Polarization Scan Artificial Neural Network Expert System

SEQEXPERT - Sequential Immersion Test Artificial Neural Network Expert System

CYLEXPERT - Rotating Cylinder Electrode Intelligent Rotation Rate Calculator

Corrosion Calculator

Corrosion Rate Calculator


Corrosion Economics Estimator

FINCALCULATOR - Corrosion Economic Calculator


TUTORIAL ON ELECTROCHEMICAL IMPEDANCE SPECTROSCOPY

David C. Silverman


Table of Contents

Overview of Tutorial
Overview of EIS
Analysis Using Simple Circuits
Constant Phase Element
Corrosion Rate Estimation
        Two Capacitive Relaxation Time Constants
Critical Criteria for Proper Spectra
Diffusion Impedance
Pseudoinductance

Constant Phase Element

The constant phase element arose during early studies in the 1940’s of dielectrics and was attributed to a combination of dispersion and absorption. The complex dielectric constant was found to be a function of the frequency raised to a fractional power. This analysis is directly applicable to electrochemical impedance spectroscopy. It offers a convenient method of accounting for the non-idealities alluded to in Analysis Using Simple Circuits. The impedance is affected by the interaction between frequency and all of the physical and chemical processes that respond to that frequency change within the electrochemical cell and across the corroding interface. The result is that one process may have multiple closely spaced dependencies on changing frequency. Or multiple processes may have similar but not the same responses to changing frequency. The result is often modeled as a phenomenological "RC" time constant the value of which may or may not be associated with one actual physical process. Some causes are:
  • Surface roughness or surface heterogeneities
  • Slow reactions affected by frequency
  • Diffusion processes affected by frequency (Warburg impedance)
  • Improper electrode spacing
  • Transmission line
An excellent discussion of the constant phase element can be found in J. R. Macdonald, Impedance Spectroscopy, John Wiley & Sons, NY, 1987. These interactions occur often enough that the constant phase element should always be considered in place of capacitance in any attempts to model practical impedance spectra. Use of constant phase element may be the only viable approach to extracting needed values from the impedance spectra when the complex chemistry prevents a kinetic model from being developed or when time and money preclude an extensive study.

The dispersion of relaxation across frequencies results in an alteration of the impedance spectrum. For example, the equation for the simple circuit presented in the section on Analysis Using Simple Circuits is:

                                                              (14)

This equation results in a semi-circle whose diameter is Rp. The curve would intersect the real axis at Rs as the frequency approaches infinity and at Rp+Rs when the frequency approaches zero. The angle of intersection with the abscissa would be 90o. When the time constant (RpC) is dispersed across a range of frequencies, the semicircle becomes shifted and appears as in this figure . Note that the entire semicircle would still be in the first quadrant but the distance between the intersections at high and low frequency would be less than the value of Rp. The quantity ß is the constant phase element. It defines the amount of depression below the axis in Nyquist format. It defines the amount of dispersion for the relaxation process governing it. The equation that describes the impedance as a function of frequency for one relaxation time constant is

                                                                                            (17)

where (jωτ)β accounts for depression below the real axis (e.g. dispersion) and τ is the phenomenological relaxation time constant loosely related to an RC time constant. The mathematical equations are generated by remembering that the term is expandable in terms of sines and cosines of the constant βπ/2.

This figure shows the effect of a relatively small amount of dispersion on the impedance response to changing frequency in Nyquist format . and in Bode format . The polarization resistance is 10000 ohm-cm2, the solution resistance is 10 ohm-cm2, and the time constant is 1 second. Note that the presence of the constant phase element even with such a relatively high exponent as 0.9 has a significant effect on the impedance spectrum. Attempting to model these spectra without accounting for dispersion would be wrong. Neglecting the effect of the constant phase element during curve fitting could affect the estimate of the polarization resistance and could result in an erroneous estimate of the corrosion rate. One additional point needs to be made. The CPE has been written in a number of different ways, for example (jωτ)β and (jAω)β. When the latter term is used, "A" is not equal to a capacitance.

The following figures show how typical circuits might be written to take account of the constant phase element . The first two show the models for one and two distinct constant phase elements. The latter might be found for the frequency response to a coated electrode or one in which the surface has imperfect inhibition as in the case of imperfect inhibitor adsorption. The second two show a diffusion process with one constant phase element and inductance with one constant phase element. More is said about inductance in the section entitled Pseudoinductance One of the observations has been that most spectra can be modeled by one of these four circuits with with the majority being modeled by the first two. This observation is not unreasonable considering that the first circuit contains four adjustable parameters and the second contains seven.

Some typical impedance spectra that at first glance might appear to be modeled by a simple RC time constants but which could only be fit by taking such dispersion into account are shown in these figures, and . The first shows the impedance spectrum for steel in polyaspartic acid at a pH of 10 (D. C. Silverman, Corrosion, 51, 818 (1995) 1    (500k). The exponent is about 0.9. The second shows the impedance spectrum for steel in a complex waste stream (D. C. Silverman, Electrochimica Acta, 38, 2075 (1993)). In the latter case, the exponent was about 0.6 which lead to the large decrease in the phase angle maximum. Both of these spectra were generated under turbulent flow conditions in the rotating cylinder electrode to minimize mass transfer effects. Inclusion of one constant phase element led to an excellent fit of the impedance spectrum and reasonable agreement with corrosion rate estimates from mass loss of the cylinder.




Next Page: Corrosion Rate Estimation

Previous Page: Analysis Using Simple Circuits

Return to Table of Contents



1 © NACE International publication and year shown in citation above. All rights reserved. Displayed with permission from NACE International, Houston, TX (http://www.nace.org). Published in Corrosion, in the month and year shown in the citation above.






David C. Silverman, Ph.D. - Primary Consultant
E-Mail:     dcsilverman@argentumsolutions.com
Phone:     314-576-3586
Fax:         314-754-9825
Address:   The Argentum House
                14314 Strawbridge Ct.
                Chesterfield, MO 63017