Interpretation of EIS time constants of an oxide layer?

[Gokul OS]

Hello all.

     I have collected few electrochemical impedance spectra of Cr2O3-based oxide layer formed during high temperature oxidation of alloys in a non-aggressive electrolyte. The qualitative Bode phase angle plot exhibits around 3 time constants (CPE peaks at high-, mid- and/or low- frequencies) indicating various dielectric features within the oxide layer. Generally, I attribute the time constant in the high-frequency region (above 10³ Hz) with that of the bulk oxide because of its lower capacitance (from high thickness). Meanwhile, the other time constants located in the mid- and/or low-frequency time constants look complicated, though I suppose it must be the response from nano-porous fractured paths observed in the oxide that may have electrolyte accommodation till the metal-oxide interface.

     The general oxide models (porous and sandwich structured)1 seems irrelevant when much electrical components needed to be added for fitting spectra with much dielectric components. Also it seems that the transmission line model cannot be applicable in this case because of the multiple time constants observed here. In this regard, how can I associate these time constants with their respective dielectric features?

     Thank you.

    

Reference:
1. Characterization of High-Temperature Oxide Films on Stainless Steels by Electrochemical-Impedance Spectroscopy (1998).

[Yevgen Barsukov]

Your assignments of time constants make sense. It also looks like all semicircles are depressed.

This could be due to porosity of the material, which causes all surface impedances to be "distributed".

This situation can be described with  transmission line model with conductive boundary where surface impedance is the inside "Z-box" in the

transmission line ladder, while ionic conductance is its distributed conductivity.

I copied this info for similar element difshort from MEISP help. You would need to modify the surface impedance that is distributed to add more RC elements for different

processes that you have, so instead of "Cd" you would have Z(s) representing 2 or more Rc elements.

I am not sure if you also have a corrosion process present because your semicircles never reach vertical line behaviour at low frequency in Nuqyist.

Maybe you need to  measure to lower frequency range to fully resolve all available processes.

Regards,
Yevgen

--
You received this message because you are subscribed to the Google Groups "Impedance Spectroscopy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to impedance-spectroscopy+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Gokul OS]

Dear Yevgen Barsukov
     Thanks for your reply. The objective of my EIS analysis was to study the structure of oxide layer after corrosion in a non-aggressive buffer solution. In this case no corrosion or charge transfer process would be expected when EIS was measured at OCP. Though no vertical line behavior at low frequencies observed as you said, aren't the corrosion processes (fast electron transfer process) are expected in the high frequency region?
     In my situation, I am trying to determine the reasons for those multiple relaxation features observed, though no corrosion processes were involved. Also I could not find relevant literatures with multiple time constants (R-C) fitted with transmission line model. It would be great if you suggest me few references for determining the physical meaning of those relaxation features.
Thank you.

Regards
Gokul.