Sometimes in EIS measurements, at high frequencies, a noise-like curve shows up that it is a repeatable phenomenon. To what could it be related?

[Mehrzad Soleimany]

Dear Friends,

Sometimes in EIS measurements, at high frequencies, a noise-like curve shows up that it is a repeatable phenomenon. To what could it be related?

I attached some of the EIS results. The tests were taken at different temperatures and in the air atmosphere from a solid oxide proton-conducting electrolyte, and the Ag paste was used as electrodes material. The applied voltage amplitude was 40 mV, and the equipment was a Solartron 1260 in combination with a Solartron1286.

Thank you in advance for your attention

[Eugene Engmann]

  Hi Mehrzad,

Thank you for your email. Here are some suggestions I have. The noise at high frequency could be inductance from the electrodes as a result of passage of a potential. You can model that using a inductor in your circuit. Also, make sure that your electrodes are not touching and there is enough distance between them. This could be a reason for the noise as well. Is there a reason why your amplitude is 40 mV (rms) and not 10 mV (rms)?

Eugene Engmann

--
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-spectro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/9ad56374-2d75-4465-a1e3-eb03442d14c4%40googlegroups.com.

[Mehrzad Soleimany]

Dear Eugene,

I appreciate your kind answer. I completely agree with you. 

Actually, no. 40 mV was the amplitude my professor said that I have to choose. Are you asking it because 40 mV could lead to nonlinearity in comparison with 10 mV?

And do you think there would be any problem if I model only the clean part and remove the noise-like curve at high frequency?

On Tuesday, April 14, 2020 at 8:09:29 PM UTC+4:30, Eugene Engmann wrote:

  Hi Mehrzad,

Thank you for your email. Here are some suggestions I have. The noise at high frequency could be inductance from the electrodes as a result of passage of a potential. You can model that using a inductor in your circuit. Also, make sure that your electrodes are not touching and there is enough distance between them. This could be a reason for the noise as well. Is there a reason why your amplitude is 40 mV (rms) and not 10 mV (rms)?

Eugene Engmann

On Tue, Apr 14, 2020 at 5:18 AM Mehrzad Soleimany <mehrzad....@gmail.com> wrote:

Dear Friends,

Sometimes in EIS measurements, at high frequencies, a noise-like curve shows up that it is a repeatable phenomenon. To what could it be related?

I attached some of the EIS results. The tests were taken at different temperatures and in the air atmosphere from a solid oxide proton-conducting electrolyte, and the Ag paste was used as electrodes material. The applied voltage amplitude was 40 mV, and the equipment was a Solartron 1260 in combination with a Solartron1286.

Thank you in advance for your attention

--
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.

[Eugene Engmann]

Hi Mehrzad,

Most of the literature out there use 10 mV (rms) (it is small enough, preventing any damage to the surface of the electrode. Rather than take out the noise, see if you can fit it using an inductor. The reason is because the noise is part of the scan. Ask you professor to see if you can try a 10 unless there is a reason to stick with 40.

Eugene

To unsubscribe from this group and stop receiving emails from it, send an email to impedance-spectro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/9ad56374-2d75-4465-a1e3-eb03442d14c4%40googlegroups.com.

--

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-spectro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/d588cba8-e5f4-4775-bc38-a2fb0f2a23dd%40googlegroups.com.

[Gavin Reade]

Are you sure it is not related to the bandwidth of the 1286?.  It looks to me you are trying too high a frequency start for your equipment combo.

Greadey

To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/CAE_U5d75E4yeooSdGerQDcyHfw%2BWAiAe%3DCrONji1u5_b9_q3dw%40mail.gmail.com.

[Mehrzad Soleimany]

Thanks a lot, Eugene. I'll try to examine what you kindly suggested. Unfortunately, I lost my samples because of the SEM analysis, and I can't reproduce all of those samples again.

On Wednesday, April 15, 2020 at 8:08:58 PM UTC+4:30, Eugene Engmann wrote:

Hi Mehrzad,

Most of the literature out there use 10 mV (rms) (it is small enough, preventing any damage to the surface of the electrode. Rather than take out the noise, see if you can fit it using an inductor. The reason is because the noise is part of the scan. Ask you professor to see if you can try a 10 unless there is a reason to stick with 40.

Eugene

On Wed, Apr 15, 2020 at 7:46 AM Mehrzad Soleimany <mehrzad....@gmail.com> wrote:

Dear Eugene,

I appreciate your kind answer. I completely agree with you. 

Actually, no. 40 mV was the amplitude my professor said that I have to choose. Are you asking it because 40 mV could lead to nonlinearity in comparison with 10 mV?

And do you think there would be any problem if I model only the clean part and remove the noise-like curve at high frequency?

On Tuesday, April 14, 2020 at 8:09:29 PM UTC+4:30, Eugene Engmann wrote:

  Hi Mehrzad,

Thank you for your email. Here are some suggestions I have. The noise at high frequency could be inductance from the electrodes as a result of passage of a potential. You can model that using a inductor in your circuit. Also, make sure that your electrodes are not touching and there is enough distance between them. This could be a reason for the noise as well. Is there a reason why your amplitude is 40 mV (rms) and not 10 mV (rms)?

Eugene Engmann

On Tue, Apr 14, 2020 at 5:18 AM Mehrzad Soleimany <mehrzad....@gmail.com> wrote:

Dear Friends,

Sometimes in EIS measurements, at high frequencies, a noise-like curve shows up that it is a repeatable phenomenon. To what could it be related?

I attached some of the EIS results. The tests were taken at different temperatures and in the air atmosphere from a solid oxide proton-conducting electrolyte, and the Ag paste was used as electrodes material. The applied voltage amplitude was 40 mV, and the equipment was a Solartron 1260 in combination with a Solartron1286.

Thank you in advance for your attention

--
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.

To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/9ad56374-2d75-4465-a1e3-eb03442d14c4%40googlegroups.com.

--
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.

[Mehrzad Soleimany]

Dear Greadey,

Thank you for your kind answer. Actually I'm not very familiar with the equipment. Could you please explain it a little more? I chose the frequency so that the curve intercepts the Z' axis. The frequency I used was around 1E7. And do you think it is ok if I omit the noise-like curve and fit the clean parts?

On Wednesday, April 15, 2020 at 9:13:48 PM UTC+4:30, greadey wrote:

Are you sure it is not related to the bandwidth of the 1286?.  It looks to me you are trying too high a frequency start for your equipment combo.

Greadey

On Wed, 15 Apr 2020, 16:38 Eugene Engmann, <englis...@gmail.com> wrote:

Hi Mehrzad,

Most of the literature out there use 10 mV (rms) (it is small enough, preventing any damage to the surface of the electrode. Rather than take out the noise, see if you can fit it using an inductor. The reason is because the noise is part of the scan. Ask you professor to see if you can try a 10 unless there is a reason to stick with 40.

Eugene

On Wed, Apr 15, 2020 at 7:46 AM Mehrzad Soleimany <mehrzad....@gmail.com> wrote:

Dear Eugene,

I appreciate your kind answer. I completely agree with you. 

Actually, no. 40 mV was the amplitude my professor said that I have to choose. Are you asking it because 40 mV could lead to nonlinearity in comparison with 10 mV?

And do you think there would be any problem if I model only the clean part and remove the noise-like curve at high frequency?

On Tuesday, April 14, 2020 at 8:09:29 PM UTC+4:30, Eugene Engmann wrote:

  Hi Mehrzad,

Thank you for your email. Here are some suggestions I have. The noise at high frequency could be inductance from the electrodes as a result of passage of a potential. You can model that using a inductor in your circuit. Also, make sure that your electrodes are not touching and there is enough distance between them. This could be a reason for the noise as well. Is there a reason why your amplitude is 40 mV (rms) and not 10 mV (rms)?

Eugene Engmann

On Tue, Apr 14, 2020 at 5:18 AM Mehrzad Soleimany <mehrzad....@gmail.com> wrote:

Dear Friends,

Sometimes in EIS measurements, at high frequencies, a noise-like curve shows up that it is a repeatable phenomenon. To what could it be related?

I attached some of the EIS results. The tests were taken at different temperatures and in the air atmosphere from a solid oxide proton-conducting electrolyte, and the Ag paste was used as electrodes material. The applied voltage amplitude was 40 mV, and the equipment was a Solartron 1260 in combination with a Solartron1286.

Thank you in advance for your attention

--
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.

To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/9ad56374-2d75-4465-a1e3-eb03442d14c4%40googlegroups.com.

--
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.

To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/d588cba8-e5f4-4775-bc38-a2fb0f2a23dd%40googlegroups.com.

--
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.

[gyirga]

Hi,
From my experience i found such type of curves at high frequency region. Especially when the charge transfer resistance is very small. What i did is I decreased the high frequencies region. For instance your sample charge transfer resistance looks very small. I think no need to apply such higher frequency, you can run the experiment from 50 kHz to 1Hz, with 5 or 10 mV amplitude which probably solve your problem. One additional information is the number of scanned points are very few, for example for the first graph you have only one data point in the high frequency region. It is impossible to associate one data point with inductor property or with any property. I am sure you can minimize the error By lowering the high frequency values.
Best,
Getachew

[Gavin Reade]

I agree with gyirga about running from 50 kHz.  From memory the 1286 has a maximum bandwidth of 100 kHz.  In zplot (if you are using it - there is an option somewhere to change it but max. is 100 kHz).  Are you measuring 1 cell and if so what is the voltage that is going to the 1286?  It could be that you ditch the 1286 and connect the 1260 direct.

greadey

--
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-spectro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/impedance-spectroscopy/a3c3a521-62e0-4ad9-8283-25f7e1d19805%40googlegroups.com.

[Gavin Reade]

it's also worth remembering that to get to f > 1 MHz really does require attention to wiring etc, etc.