If your material is a pure ionic conductor the simplest way to estimate the ionic conductivity from your spectra is to fit semicircle with "Fit Circle" in zView. High intercept will be the value of the sample resistance.

The better way is to fit your spectra with a equivalent circuit.

Also if the the parameters of the interface electrode/electrolyte is not important for you, you may increase low limit to 10 or may be 100 Hz.

hello Bihag, I think that if you use the warburg element to fit the data at low frequency you can found the ionic conductivity.

Also you can transform the impedance data to conductivity vs frequency data to elucidate the mechanism of the conduction.

Good Luck

Thanks Alexey. What the problem which I face is that when I take the X-intercept or the first point in the high frequency region for calculating conductivity , with respect temperature it is not linearly behaving. Where as the semicircle (Rct) is decreasing while increasing the tempt. So is this can be used to find the conductivity. Any way conductivity is the movements of ion under potential bias which is related to the Rct.

Dear Bihag,

It is rather difficult to give an educated opinion when so little details are given. What type of sample are you measuring? And how is the experimental setup (which FRA?).

An important point is to look at the capacitance of the semicircle. My rough estimation is that it is around a 100 pF. Hardly a value for a double layer capacitance, but more likely to be the geometric capacitance. Hence the R of the semicircle is the ionic resistance of your sample, not an R-ct. That the 1 MHz point yields sometimes a negative R-infinitive is just an artifact of the measuring system.

Beste regards,

Bernard A. Boukamp

you can fit the semi circle to get the X-intercept and can use resistance directly to calculate conductivity at a certain temperature.

Thanks Bernard for the comments. The material is Metal organic frame work and in pellet from with a thickness around 1 mm. I am using Solatron for the EIS. My doubt is that the Resistance of the material is given by the semicircle or the first point at the high frequency region?. In cases of the highly conducting material this problem never arises as the there is a clear X-crossing and hardly semicircle will observe.

This is a very typical Nyquist plot for an eletrode-eletrolyte (ionic conductor)-eletrode sandwich. It is wrong to take the X-intercept or the first point of the semicircle in the high frequency region for calculating conductivity. It is widely accepted to take the X-intercept of the low-frequency straight line by extrapolation. The Arrhenius plot of log(conductivity) vs 1/T will tell you the type of ionic conduction mechanism. It's not necessary to be linear.

Exactly Dr.Chen. e low-frequency straight line by extrapolation will be almost smiler to the semicircle dia (Rct) and both give smiler results. Then how about the first point ? what is the significant of that point..?

The 1 MHz point has no significance. It is corrupted by experimental conditions (parasitic capacitances, etc.) This clear because of the quite varying results from experiment too experiment (sometimes ne.gative). It is quite difficult to measure accurately in the high frequency regime (>100kHz).

Yes I agree Bernad. But this type of data can be fitted by simple Randlers circuit (Rs+(Rct+Wd/CPE)). And it is believed that Rs is bulk resistance and Rct as charge transfer(ionic). In such case ionic conduction nothing to with the Rs..?

Hi. Is there any book or material which is fruitful for these measurements?

Hi. Maybe this book is helpful : Impedance Spectroscopy-

Theory, Experiment, and Applications; Evgenij Barsoukov, J. Ross Macdonald; John Wiley & Sons

Dear Anothumakkool,

You might have a look on this paper:

“An Elementary Picture of the Dielectric Spectroscopy in Solids: Physical Basis”, M. F. García-Sánchez, J.-C. M´Peko, A. R. Ruiz-Salvador, F. Fernández-Gutierrez, G. Rodríguez-Gattorno, A. Delgado, Y. Echevarría, J. Chem Education 80 (2003) 1062.

http://dx.doi.org/10.1021/ed080p1062

https://www.researchgate.net/publication/257132596_An_Elementary_Picture_of_Dielectric_Spectroscopy_in_Solids_Physical_Basis?_iepl%5BviewId%5D=3uD4Ro0bFJeri4G311BO9B0C&_iepl%5BprofilePublicationItemVariant%5D=default&_iepl%5Bcontexts%5D%5B0%5D=prfpi&_iepl%5BtargetEntityId%5D=PB%3A257132596&_iepl%5BinteractionType%5D=publicationTitle

Best regards,

A. Rabdel Ruiz-Salvador

Article An Elementary Picture of Dielectric Spectroscopy in Solids: ...

Hello, finally what was the solution? , please I also have the same doubt because I understand that the conductivity is a function of the geometry and the inverse of the Z * so it depends on the frequency and has both real and imaginary parts. Therefore, what is the standard criterion to take a single value of the conductivity ?. The conductivity formula based in EIS is mentioned in many papers and results are plotted as a function of frequency showing conductivity behavour, but in others papers they give a value conductivity directly without clarifying it.

Dear Veliz Noboa,

I suggest to extract DC conductivity from the AC impedance or dielectric spectroscopy data for each electroactive specie. Then, you can relate this value to the physical/chemical features of the system. Note that the shape of the AC curve provides information on the correlation of interactions of the electroactive species.

Best regards,

A. Rabdel Ruiz-Salvador

Hi All,

Great discussion here. I am in a similar need to measure ionic conductivity of synthesized polymer thin film using potentiostat. I copper taped two ends of the rectangular polymer film. I connected one end of the film to working electrode and another to the ground. I used Ag/AgCl as a reference electrode. Using the Impedance Spectroscopy, I have the Nyquist plot. What is best approach to estimate ionic conductivity from here? I appreciate your insights. Thank you

Dear Rashupati,

the set-up of your cell is not clear to me. How is the reference electrode connected? Is one copper tape the working electrode (ground) and the other the counter electrode?

Best regards,

Bernard A. Boukamp

Dear Rahupati,

Your answer disappeared suddenly from the list. Can you resend it? By the way, I know that you use Ag/AgCl reference electrode, but how is it place in the cell? Is there a liquid electrolyte?

Best regards,

Bernard A. Boukamp

Dr. Boukamp,

Thank you for the response. You are exactly right that one end of the tape is working and another is the counter These are marked with different color alligator clips. Silver/Silver Chloride electrode is the reference electrode.

1 Working electrode - Green

2 Reference electrode - White

3 Counter electrode - Red

From the data that consists: frequency, phase angles, and Impedance. I calculated the real part of the impedance as resistance (R) = |Z|*Cos(Phase). I am more interested to measure ionic conductivity through this polymer thin film. Thank You

Sincerely,

Pashupati Adhikari

For using gamry system, just sandwitch your electrolyte between the two SS plates of same thickness. Make sure there is no air trape between the ss plate and electrolyte. Connect counter and reference terminals to one ss plate while the working and wrking sense to another. Connect the couter sense to ground.

Start your EIS nyquist plot per your required parameters. The initial point on the x intercept at the high frequency region of the plot would be your resistance. From resistance you can calulate your ionic conductivity by using k=l/RA where R is resistance, A is contacted surface area of the electrolyte and l is the thickness of the electrolyte and k is ionic conductivity. Hope this will help

Dear Pashupati, tell us please:

1. Is your sandwich placed in a liquid electrolyte in the cell?

2. Or maybe you didn't use a liquid electrolyte?

3. What place of cell/sandwich is reference electrode placed in?

It's important, cause you sandwich setup does not provide 3rd (reference) electrode. It looks like a symmetrical two-electrode cell without any additional electrodes and external electrolyte.

It will be possible to answer your question more accurately, after receiving your exact answers.

(I support Bernard A. Boukamp clarification)

What is the way to calculate ionic conductivity without using extrapolating linear portion or determining diameter of the semi circle? Because in my case they give me different values. You can see the attached figure.

Dear Orhan,

first of all, you have to describe your system. The best method is to use CNLS-fitting. It will take in account all contributions to the dispersion. But a good CNLS-fit depends also on the quality of the data. Data validation should be done with a Kramers-Kronig test.

You may send me your data (Excel or just text, but with freq, Zreal and Zimag. I can then give a more realistic answer.

Best regards,

Bernard Boukamp

[email protected]