Hi Dear Kavand,

I think the negative part of the real impedance reveals that your coating's resistance is going down and shows more capacitance behavior than resistive properties.

Please see the following paper which has provided good explanations about EIS analysis.

Article In situ synthesis of quasi-needle-like bimetallic organic fr...

Hi Behnam,

Thanks. Does these results seem OK to you? If we look closer at the Nyquist plot, we can see that the curve is segmented. How is that even possible?

Moreover, when we look at the phase vs freq plot, the information in the low frequency range is weird. How is it possible to have data in the positive phase degrees?

I am not an expert and I am having a rough time interpreting these data.

It seems that there occurred a lot of disruption during the test. I think it might be helpful to conduct the test again. Or you should reconsider your coating because it may not has a firm connection to its substrate.

However, the exact analysis is highly complicated you must conduct the EIS test several times in order to get a good evaluation of your work.

It sure is complicated! I have tested the experiment several times and also tried using the Dropsens electrodes. I reconsidered using potassium ferricyanide solution instead of PBS but I am not still satisfied. Well, Pt electrodes (Dropsens) did very good with the ferricyanide solution (Dropsens carbon electrodes did as bad as my gold electrode chips) but when the coating was involved, things didn't turn up as expected.

In case of a bad coating adhesion, I should get lower impedance and the results should be somehow similar to the bare sample.

Your two electrode cell is not ideal for EIS measurement. It would be better to use a proper reference electrode to repeat your experiments. Also, what were the applied bias potential (voltage for your two electrode cell) and the amplitude in your measurements?

Hi Kavand,

The real component of the Nyquist plot represents the resistance of the system and imaginary component gives the value of capacitive reactance (resistance to the flow of current imposed by a capacitor). Any point on the Nyquist plot gives the resistance and capacitive reactance of the system. I would suggest you read Electrochemical Impedance Spectroscopy and its Applications by Andrzej Lasia. This book can be very helpful for the non-expert to understand EIS.

As far as you data is concerned, it seems that the your coatings are not stable at low frequencies which means that the properties of coating are changing and it no longer satisfies time-invariance condition for the reliable EIS measurement. The bare electrode in green color shows comparatively better results than you coating which again indicates the instability of coatings at low frequencies.

Also, as suggested by Professor Chen, you better use a reference electrode to reliably measure the properties of your coatings. Also, you can use multi-sine feature in the software to measure the impedance, multi-sine function makes the measurement faster and you can probably overcome the time-variance of your system at low frequencies (long measurement time).

Thanks for the inputs! Please take a look at the new data I am attaching here.

George Zheng Chen do you mean any 2 electrode configuration in general or the electrodes I am using? Because in articles I can see the they simply used a 2 electrode system for quantifying the encapsulation quality. Anyhow, I also used 2 different Dropsens electrodes (DRP 110 (Carbon (working and counter) and silver reference) and DRP 550 (Platinum (working and counter) and silver reference)), which has a 3 electrode configuration. Here, I simply had a PDMS coating just to check if things are working OK with these electrodes before going for a more time consuming Parylene coating.

As I was not sure if these results are due to the electrolyte (PBS) that I am using, I also tested with a ferricyanide solution.

Roby Soni can you please explain the relation between frequency and the 'instability' of the coating? The coating should be very stable in terms of moisture barrier.

Hanie Kavand If you wish to study a specific electrode material, it is better to use the three-electrode cell. In some cases the reference electrode may be combined with the counter electrode, such as a lithium metal foil a Li+ ion containing electrolyte, making the cell look like a two-electrode one, but it is still in principle a three-electrode cell. An asymmetrical cell with only two different electrodes is not suitable for suitable for studying a specific electrode material because the EIS would reflect the overall performance of both electrodes.

George Zheng Chen thank you. I am only using EIS to see if my 'coating' is doing well enough. To quantify the moisture barrier of the coating, I am using EIS. As impedance values are very sensitive to any changes, people usually use it to quantify their encapsulation properties. Upon measurements, I was very upset when I did not see the typical Nyquist curves in my results. I initially thought I had connection issues with my 2 electrode configuration but when I also used the commercial 3 electrode Dropsens electrodes, I got weird results which are a bit hard to interpret.

Hanie Kavand If you used a three-electrode cell (I am not sure how the Dropsense cell is constructed), can you tell me what the bias potential and signal amplitude were?

George Zheng Chen this the what I used:

http://www.dropsens.com/en/pdfs_productos/new_brochures/550_c550.pdf

The bias potential was 0.2 V and the sinus amplitude was set at 50 mV. I have attached the results in my previous comment.

Hanie Kavand For a two-electrode cell, a bias voltage of 0.2 V is fine, but your 50 mV amplitude is too large, and should be reduced to 10 mV or smaller.

Thanks George Zheng Chen . I had tried with lower sin amplitudes, even 10 mV, but the results from the coating was always the same; I see weird signals in the lower frequency (lower than 100 Hz to be specific) range for the samples that have a coating. I was told that I might be having 'noise' issue in that frequency. However, the bare sample (without the coating) seems OK. That's why I am not convinced that the whole issue is arising from the noise. I was also told that impedance measurements are super sensitive and maybe I should have a pretty shielded and motion-free setup for that. In the lab, people mostly run impedance measurements with batteries and maybe they don't need such fancy setup.

Hanie Kavand: I looked at your log lZl vs log (frequency) plot again, and noticed that the impedance of your sample at 100 Hz or lower was larger than 100 Mohm. This could be the reason because the input impedance of a frequency analyzer (impedance analyzer) is usually 100 Mohm. In other words, your equipment is coupling its internal impedance with the external impedance, and hence unable to give a stable value of the measured impedance.

You may resolve the problem by using a sample of significantly larger area or much thinner coating so that the sample impedance is reduced by a few orders of magnitude (i.e. below 1 Mohm).