Hi Yang:

If this data is not due to a poorly prepared sample that can be improved or mismatched responses in the detectors then you will simply have to live with it. The data between 8-10 can certainly be included but with a bit more car in background subtraction. It looks like there might be a long wavelength oscillation in k-space. I would also use a window function with a dk of at least 2 on these data. The noise level is not unreasonable in the range below 10 A-1 but I would like to see the FTs. As for your question about how the range of k-space included affects the FT, the wider the range of k-space included, the sharped the features in the FT. The information is still there but you will not have as many variable parameters available for your fits.

Carlo

On Mon, Aug 30, 2021 at 3:33 PM Hu, Yang (HIU) <yang.hu@kit.edu> wrote:

Dear Ifeffit members,

I have some fundamental questions regarding noisy data, and I am wondering how to tell whether the data quality worth doing Fourier transformation/EXAFS fitting or not.

For example, the attached MnO chi(k) data becomes noisy from 7 Å-1 when it was measured up to 12 Å-1. The deglitching (left: red -> blue) mitigated the strong “dips” but the high-k end still has too much noise. I practiced the FT and fitting the 1^st and 2^nd shells by using the lower part. But My questions here are:

1)    How noisy would be “too noisy”? Like the data between 8 and 10 Å-1 in the attached file, can they still be included for the FT?
2)    We can choose the high-k end based on the signal-to-noise ratio, but to what extend? With data being noisy from even 5 or 6 Å-1, can they still be used?
3)    For fitting the 1^st and 2^nd shells, I still lack of clear understanding how the high-k portion can influence. If I measure a set of samples, and one of them has much noisy data so a shorter k-range is picked up for background subtraction and FT. In this case, can I still consider the change or evolution in the fitted parameters “systematic”?

Thank you in advance,
Best regards,
Yang

_______________________________________________
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit