Equivalent Circuit Modeling in EIS

Introduction above. The behavior of each element is then

described in terms of “classical” electrical
Electrochemical Impedance Spectroscopy (EIS) is a components (resistors, capacitors, inductors) plus a
very powerful tool for the analysis of complex few specialized electrochemical elements (such as
electrochemical systems. See the Basics of EIS Warbug diffusion elements).
Application Note
(http://www.gamry.com/App_Notes/EIS_Primer/EIS_Pr The first step in the process is an educated guess.
imer.htm) on the Gamry Instruments website You predict the system elements that you feel will
(www.gamry.com) for a comprehensive introduction play a part in the cell’s impedance. You then build
to EIS. For the most part, information in that these elements into an Equivalent Circuit Model. The
Application Note will not be repeated here. arrangement of the elements into logical series and
parallel combinations is critical to the success of the
This Note is a practical discussion of the most modeling effort.
common method for EIS data analysis. It should
provide even a novice with the basic approach for Each element in the model has a known impedance
converting an EIS spectrum into meaningful insights behavior. The impedance of the element depends on
into the physical processes within the electrochemical the element type and the value(s) of the parameter(s)
cell. It intentionally avoids most of the complex math that characterize that element. For example, the
associated with EIS data analysis. impedance of a capacitor excited by a sine wave at
frequency f is described by the formula:

EIS Basics ZC =
1
(1)
j 2π f C
In EIS, you measure an electrochemical cell’s complex
impedance over a wide range of AC frequencies. ZC is the complex impedance, j is the Ö-1, f is the
Typically, several cell elements and cell characteristics frequency in Hertz and C is the capacitor’s value in
contribute to the system’s EIS spectrum. A partial list Farads. When you formulate a model of the system,
of possible elements includes: you generally do not know the value of the
• Electrode Double Layer Capacitance parameters for the elements in the system. For
example, you know that a painted metal has a
• Electrode Kinetics coating capacitance but you do not know the value of
• Diffusion Layer that capacitance.

• Solution Resistance
Graphical Model Editor
Unfortunately, the system’s impedance at any given
frequency usually depends on more than one cell The Gamry Instruments EIS300 Electrochemical
element. This greatly complicates the analysis of EIS Impedance Spectroscopy Software includes a
spectra. graphical Model Editor that you can use to build an
equivalent circuit model. Figure 1 shows the Model
The most common method used to analyze EIS
Editor editing the Paint Model supplied with the
spectra is equivalent circuit modeling. You simulate
EIS300. This model represents one possible
the cell incorporating the elements mentioned
assignment of the circuit elements to physical model’s parameters. The optimization program
phenomena in a failing paint film on a metal surface. needs to start with values (often called seed values)
See the EIS Primer mentioned above for a description for all parameters. If the initial values are far from
of this model. the optimal values, the optimization program may be
unable to find the best fit. The first example in this
You can use the graphical Model Editor to make up
Application Note demonstrates this problem.
quite complicated circuits. One advantage of this
approach is that you do not have to deal with the
complex circuit descriptor codes that some older Example 1: Dummy Data Matching the Coated Metal
modeling programs use. Model
This example discusses fitting a model to data when
the form of the model is known. In order to insure a
good fit, the data was recorded on an electronic
dummy cell made up from electronic components.
The components were arranged in the topology of
the paint model seen in Figure 1.
Figure 2 shows the original EIS spectrum in a Bode
format. Gamry recommends that the initial estimate
for fitting parameters be tested in the Bode format –
the lack of frequency information in the Nyquist
format makes it difficult to estimate capacitor values.

Figure 1. Model Editor

The second example in this Applications Note deals

with the problem of fitting EIS data to a model when
the model is not known a priori. Prior to this, we will
discuss the easier case of fitting data to a known
model.

Fitting the Model to Your Data

Once you have a model that you feel represents your
chemical system, you use a non-linear least squares
Figure 2. Original EIS Spectrum of Paint Model
fitting program to fit the model to the experimental
Dummy Cell
data. This program attempts to minimize the
deviation between the spectrum of the model and
If you fit the Paint Model to this data without
the experimental data spectrum.
adjusting the parameter “seed values” (default is all
Similar to other EIS modeling programs, the EIS300 parameters set to 1.00), you get a Math Error
fitting program uses the Levenberg-Marquardt message and a very poor fit (Figure 3). The
technique to adjust the parameter values of the magnitude fit does not appear on the graph at all,
elements in the model. The mathematical details of and the phase fit (the solid line) does not follow the
the fitting process are well beyond the scope of this shape of the experimental data phase.
applications note.
One of the more difficult tasks in equivalent circuit
modeling is determining the initial values for the
 capacitor, ignoring the j in the equation and using f =
10 Hz:

1 1 1
ΖC = = = @ 10 6
jwC 2π f C 6.28 ⋅10 ⋅ Cf

Cf @ 10 −8 Farads

This is the second seed value.

The area in the middle of the spectrum where the
magnitude plot approaches a horizontal line looks
like another resistor. This must be Rp and we can
assign a seed value of about 100 kW.
Figure 3. Fitting With All Seed Values at 1.00
The high frequency region appears to be capacitive.
This error is caused by poor initial values for the
The impedance at 10 kHz looks to be about 10 kW.
model’s parameters. The Levenberg-Marquardt
Following the procedure above, but with 104 ohms at
algorithm in the EIS300 analysis requires initial values
104 Hz, you get a capacitance value of about 10-9
that are within a decade or two of their final values
Farads. This is the seed value for Cc.
before it can fit properly. The inability to fit when
the seed values are far from the optimal values is not The high frequency data never becomes resistive,
unique to the EIS300. Other EIS model fitting indicated by a constant value of impedance. Ru must
programs use the same fitting algorithm and require be smaller than the highest frequency impedance
the same accuracy in the initial values. magnitude. Therefore, use 1W for the seed value.

So how do you estimate initial values? It takes a Now that we have a set of seed values, try the fit
calculator and a little understanding of the behavior again. Type the seed values into the parameter
of the elements in the circuit. Basically, we try to find fields. You can enter the capacitors in E format, e.g.,
areas in the EIS curve where the model’s impedance enter 10-9 as 1E-9. Press the Preview button. You
is dominated by one element and calculate the see a plot similar to that in Figure 4.
approximate value for that element at that frequency.
Notice that the shape of the magnitude and phase
Examine the model in Figure 1. Remember that the plots are similar for the seed curves and the
impedance of a capacitor approaches zero at high experimental data. In general, whenever these
frequency and approaches infinity at low frequency. shapes are similar and the seed values are within a
At low frequency, the impedance of the capacitors in factor of 100 of the optimal value, the model will fit.
the model is very high. Both Cc and Cf are in parallel
When you press Calculate, you see that the fit is
with a resistor. When the capacitors impedance is
indeed excellent (Figure 5). Table 1 shows the
high, the resistor’s impedance dominates. At the
parameter values calculated by the fitter, versus the
lowest frequency in Figure 1, the impedance
values of the components used to make the dummy
magnitude is about 107 ohms and the phase is near
cell.
0° (indicative of a resistor) . This is the sum of Ru,
Rp, and Rf. Assuming that Rf > Rp > Ru, you can
estimate that Rf is 107 ohms. You have your first Element Fit Value Component
seed value! Value
Rf 2.012E+07 ohms 2.0E+07 ohms
The drop in magnitude below 1 Hz is due to Cf. At 10
Cf 2.155E-08 F 2.2E-08 F
Hz, the impedance of the system is about 106 ohms.
Use this in the equation for the impedance of a Rp 1.004E+05 ohms 1.00E+05 ohms
Cc 9.962E-10 F 1E-09 F
Ru 4.185E+02 ohms 4.02E+02 ohms
Table 1. Fit Values and Component Values In this example, the model is unknown prior to
starting EIS data analysis. The sample is a commercial
AA rechargeable alkaline battery. Spectra were
Notice the large uncertainty in the fit value for Ru and
recorded using Gamry’s Hybrid EIS mode after each
the poor agreement between the real Ru value and
cycle of a repeated charge/discharge cycle. The EIS
the fit value. This is easily explained. Remember that
spectra from the charged state were more interesting
the impedance curve never became resistive at high
than the discharged state spectra, so one of the
frequency. In fact, a slight phase shift at high
charged state spectra was chosen for this example.
frequency is the only indication that Ru is present. As
a rule of thumb, if an element’s impedance is not a The spectrum recorded following the first charge
significant factor in the cell impedance in the fit’s cycle is seen in Figure 6 (Bode format) and Figure 7
frequency range, the fit’s parameter value for that (Nyquist format). The Bode plot shows behavior very
element will be poor and the uncertainty in its value different from the previous example, so it has been
will be large. transformed to show linear impedance magnitudes
instead of the more common log magnitudes. The
impedance magnitude changes by less than 30% as
the frequency is varied from by more than 3
decades! A variation this small displayed on a log
scale is very confusing, so a linear scale was chosen.
The variation in phase is also very small – less than 5°
of phase change occurred.

Figure 4. Seed Values – Paint Model

Figure 6. Bode Plot of Charged Battery

The Nyquist plot (Figure 7) shows a depressed semi-

circle, typical of a Randles element. At low frequency,
the plot shows a well-developed diagonal line at an
angle of 45°, indicative of a Warburg impedance.

Figure 5. Final Fit – Paint Model

Example 2: Rechargeable Alkaline Battery

 Examining the Bode plot, you see that the impedance
at high frequency is about 0.7 W. Use this as an
initial value for Resr.
The low frequency impedance (at 10 Hz) is about 0.85
W. Still ignoring the Warburg impedance, this is the
sum of Re1, Re2 and Resr. Since our estimate for Ru
is 0.7 W, our estimate for Re1 plus Re2is 0.15 W.We
have no good way of dividing up the contributions of
the two resistors, so we’ll assign each a seed value of
0.075 W.
We also cannot visually separate the contributions
from Ce1 and Ce2. Between about 6000 Hz and 100
Hz, the cell impedance changes by about 0.15 W.
Plugging the “mid-point” of 1000 Hz into the
Figure 7. Nyquist Plot of Charged Battery capacitive impedance equation given above, Ce1 and
Ce2 in series have a value of about 1 mF. Capacitors
What is a good starting point for a model of this in series combine similarly to resistors in parallel so
system? we can use a seed value of 2 mF each for Ce1 and
Ce2.
We know that the sample is a battery measured in a
two-electrode mode, so it has two When you preview the seed curve using these values
electrode/electrolyte interfaces. plus a value of 1.0 for the Warburg coefficient, you
get a plot that looks like Figure 9.

Figure 8. First Cut Model – Battery Data

We can assume that each interface has a double layer

capacitance and a charge transfer resistance. Let’s
assign the Warburg impedance to only one interface.
We also know that there is a resistance in the solution Figure 9. Preview of Seed Curve – Rough
path between the electrodes. This is the battery’s Estimates
equivalent series resistance (esr). Putting these facts
and assumptions into a model, we get the diagram The shape of both the magnitude and phase curves
shown in Figure 8. looks similar to the shape of the data curves, so we’re
safe if we select the Calculate button.
Once again, we need seed values before the fit will
converge (a mathematical term meaning a good fit If we want to improve the initial values, we can note
will be obtained!). Looking at the model (and that the increase in impedance at low frequency is
temporarily ignoring the Warburg impedance), the caused by the Warbug impedance. Lowering the
limiting impedance at high frequency is Resr. Warburg coefficient will make the increase smaller at
any given frequency. Figure 10 shows the seed value
curve with the Warburg coefficient lowered to 0.35.

Figure 12. Battery Model with CPEs

We can use the values found in the previous fit to

calculate seed values for the new CPE components.
The previous fit values can be seen in a table in
Figure11. Remember that the first CPE parameter is
equivalent to the inverse of the capacitance.
Therefore, the seed value for CPEe2 is 1/0.075 or 13.3
and the seed value for CPEe1 is 1/0.0027 or 370. Use
1.0 as the seed value for both CPE’s alpha
Figure 10. Seed Curve with Warburg Coefficient of
parameters.
0.35
When we enter these new seed values into the new
With either set of seed values, selecting Calculate model and press Calculate Best Fit, we see the plot in
Best Fit results in the plot shown in Figure 11. Figure 13.

Figure 11. Fit with Initial Model Figure 13. Fit with a Second Cut Model

The fit is not bad, but there are significant errors, A careful comparison of Figure 11 and Figure 13
especially in the phase. Let’s try replacing both shows that the CPE elements have slightly improved
capacitors with constant phase elements (CPE). The the phase fit in the region between 10 Hz and 100
new model is shown in Figure 12. Hz. The improvement may not be large enough to
justify the inclusion of the CPEs and their arbitrary
alpha parameter.
Note that one cannot distinguish between the anode
and cathode interfaces of a battery when only a two
terminal EIS measurement is available. If a reference provide a pretty fit, but offer no practical information
electrode can be placed in the cell, the impedance of about the cell’s behavior.
a single interface can be measured.
Once you have established a model, first estimate the
resistor values. Look for horizontal regions on a Bode
Summary plot and assign them to individual resistors or series
combinations of resistors. Estimate the value of
Fitting an EIS model to experimental data can be a capacitors in the model as demonstrated above.
fairly straightforward task. It requires a little Don’t try to be too accurate when you make these
knowledge of the cell being studied and its estimates. Any value within a factor of ten of the true
mechanisms and a basic understanding of the value is a good starting point.
behavior of cell elements.
Use the Preview feature to check your estimates.
If you do not have a model, start by examining the This is a good time to fine-tune the initial values for a
data. Look for valleys in the phase that indicate Warburg impedance in the model. In general, the fit
inflection points in the data. Use your knowledge of algorithm will converge whenever the shape of the
the cell and the experimental data to propose a seed curve is similar to the shape of the data curve. If
model. In many cases, the standard models provided the two curves have a different shape, the fitting
with the EIS300 will provide either a final model or a routine is likely to fail.
good starting point for development of your own
model. Finally, perform the fit and examine the result. If the
fit does not look good, you may have to adjust the
Gamry does not recommend that you add elements model and repeat the procedure.
to a model until all visible errors in the fit are
eliminated. Models that include elements with no
basis in the chemical processes in the cell may Application Note Rev. 1.0 11/9/2016 © Copyright 1990-2016
Gamry Instruments, Inc.

734 Louis Drive • Warminster PA 18974 • Tel. 215 682 9330 Fax 215 682 9331 • www.gamry.com •[email protected]