I have come across the term 'Kramer Kronig Residual test' while studying about impedance spectroscopy. For batteries, since the electrode acts as blocking electrode, will it be possible to represent it as the series of RC circuits?. I would like to know how it can be used to check the quality of IS spectra.
Dear Raam,
I am (unfortunately) not familiar with battery tests, but as far as I know KK test is a test of the quality of IS data, independent of the system, so it should work in all cases. (?)
I encourage you to look at papers by prof. Boukamp, who tried how to overcome the limitation of the infinite range of frequency.
You can download his software from:
https://www.utwente.nl/tnw/ims/publication/downloads/
Please look for KK-windows. valuable software. there is also a paper in the zip file that I recommend you to read. KK transform can be useful for correction of inductance at high frequencies and other artifacts, if these are well recognized.
Best regards,
Sebastian
Common KK-Tests have the option to add an additional capacitance to the model that tests for KK compliance.
Follow the link to find a tool to check for KK compliance plus the related paper with the theory behind it.
http://www.iam.kit.edu/wet/english/Lin-KK.php
The Kramers-Kronig (K-K) relations can be used to evaluate data quality. However, the K-K transform requires integration over a range of frequency from 0.0 to infinity. Since the spectral data cannot be measured over that range, evaluation of the K-K relations by integration involves making some assumptions about the behavior of the spectrum outside the frequency over which measurement was taken. Practically however, the K-K analysis can be performed by fitting a generalized model to spectral data.
Hope this helps answer your question.
Professor Yehia Khalil
Yale University
USA
Dear Raam,
I am (unfortunately) not familiar with battery tests, but as far as I know KK test is a test of the quality of IS data, independent of the system, so it should work in all cases. (?)
I encourage you to look at papers by prof. Boukamp, who tried how to overcome the limitation of the infinite range of frequency.
You can download his software from:
https://www.utwente.nl/tnw/ims/publication/downloads/
Please look for KK-windows. valuable software. there is also a paper in the zip file that I recommend you to read. KK transform can be useful for correction of inductance at high frequencies and other artifacts, if these are well recognized.
Best regards,
Sebastian
Are we deciding the quality of spectrum with the magnitude of residuals? Is there any error tolerance upto which the spectrum's quality is acceptable?
To complete all the fine former answers, I just want to add some comment about the magnitude of the residuals and how the quality can be judged.
The KK test is mostly a measure for the stability of the sample during measurement, because the other criteria are usually fulfilled (see paper by Boukamp attached).
That means: Randomly distributed noisy residuals are ok up to a certain level (depending also on the absolute value of the impedance, the measurement device etc.; if that "good" level is exceeded, usually also the impedance itself looks quite noisy). Systematic residuals (not randomly distributed) are a problem. Examples are residuals that are larger than zero for a large frequency range or oscillations with periods of decades or more.
Noisy residuals can often be improved by longer integration times, but if the sample changes over time (temperature, degradation, activation, batteries: SOC,...) we have problems for good measurements that can be detected by systematically distributed residuals. Here, the threshold for good measurements is much lower. General numbers are really difficult to give!
I want to add that the paper by Schönleber (link attached) features some examples especially for batteries. The KK test related to this paper can be downloaded with the link from my previous answer. This test is basically the KK test by Boukamp, only that the number of RC curcuit for the test model are chosen automatically.
Good luck!
http://www.sciencedirect.com/science/article/pii/S0013468614001005
Article A Linear Kronig-Kramers Transform Test for Immittance Data Validation
Amen, to all the fine remarks.
But as far as I understand it, the K-K relations are fundamental to EIS. A K-K validation is highly recommended if you want to fit a model to your data. What is acceptable really depends on the situation. It is not possible to define a limiting value.
Success with the data analysis,
Bernard A. Boukamp
Thank you very much all for your concise explanation and examples. Thank you so much
To study K-K transforms you can use either serial or parallel R-C circuits, as described in papers by Bernard Boukamp. In my book Electrochemical Impedance Spectroscopy and its Applications I have discussesd these methods. You can also find the Boukamp's program together with the excercises (solved) on K-K transforms in the Extra Materials for my book:
http://extras.springer.com/2014/978-1-4614-8932-0
And I like to recommend Andrzej's book to everybody who is involved in Electrochemical Impedance Spectroscopy.
Bernard A. Boukamp
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Electrodes