fitting sulphur using standards by Athena
Dear Sir, I face one quite confusing question. Hope to get your help about using Athena to do the fitting. What I am concerning is sulphur species in my char sample. However, I cannot get the good fitting results due to the standards i used. Please see the following figure. S1 in the figure means the dibenzothiophene standard, c26 is my char sample. I intend to fit the char sample by using the standards such as dibenzothiophene, dibenzyl disulfide, sulphate and so on, to get the sulphur species. However, because of the self-absorption, the pure standard spectra intensity is even lower than our char sample as shown in fig. 1. If I do like this, the fraction is higher than 1 in my sample, I cannot get the reasonable fitting data. So how to do the fitting? As I known, the spectra for standard diluted with BN and non-diluted are different. The peak of the non-diluted is low due to self-absorption, as shown in the figure below. S1 is non-diluted and s1_10 is diluted sulphur to about 3%. For dilution, the position of the peak is also changed. Therefore the non-diluted s1 spectra isnot the real spectra of dibenzothiophene. The question is what percentage should I dilute the standard to get the credible spectra for the standards?and how to explain the peak position change? Can I use Athena to counteract the self-absorption and get the credible spectra for sdandards? Waiting for your reply. Thank you very much. Best regards Juan Chen
On Tuesday, February 14, 2012 07:55:54 am 陈娟 wrote:
As I known, the spectra for standard diluted with BN and non-diluted are different. The peak of the non-diluted is low due to self-absorption, as shown in the figure below. S1 is non-diluted and s1_10 is diluted sulphur to about 3%. For dilution, the position of the peak is also changed. Therefore the non-diluted s1 spectra isnot the real spectra of dibenzothiophene. The question is what percentage should I dilute the standard to get the credible spectra for the standards?and how to explain the peak position change?
Although this sort of thing can be approximated from tables of absorption coefficients, the advice I usually give to my users is to prepare a series of samples of diminishing concetrations and measure then. Having a good intuition for sample preparation and the effects of poor sample preparation is always important and even more so for low energy edges. The best way to develop that intuition is by doing the experiment yourself.
Can I use Athena to counteract the self-absorption and get the credible spectra for sdandards?
Yes it can -- in principle. Doing a good job of self absorption correction requires a lot of knowledge about your sample and its matrix -- often more than one has. The use of the self absorption tool is explained in the Athena document. I have also written a lecture on this topic. Go to https://github.com/bruceravel/XAS-Education/downloads and download the "selfabs.pdf" file. B -- Bruce Ravel ------------------------------------ bravel@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 My homepage: http://xafs.org/BruceRavel EXAFS software: http://cars9.uchicago.edu/ifeffit/Demeter
To reiterate a point which is often not understood - dilution won't fix overabsorption if the particles of the concentrated material
which are mixed in with the diluent are too big. For concentrated S compounds, "too big" is in the low microns.
When I run a S standard, I do it in total electron yield to avoid this problem. When I fit a S spectrum, I use a crude
oveabsorption model with a single parameter which is treated as an unknown fitting parameter, with the assumption
that the reference spectra were collected in a way which avoids overabsorption.
mam
----- Original Message -----
From: "Bruce Ravel"
On Tuesday, February 14, 2012 07:55:54 am 陈娟 wrote:
As I known, the spectra for standard diluted with BN and non-diluted are different. The peak of the non-diluted is low due to self-absorption, as shown in the figure below. S1 is non-diluted and s1_10 is diluted sulphur to about 3%. For dilution, the position of the peak is also changed. Therefore the non-diluted s1 spectra isnot the real spectra of dibenzothiophene. The question is what percentage should I dilute the standard to get the credible spectra for the standards?and how to explain the peak position change?
Although this sort of thing can be approximated from tables of absorption coefficients, the advice I usually give to my users is to prepare a series of samples of diminishing concetrations and measure then.
Having a good intuition for sample preparation and the effects of poor sample preparation is always important and even more so for low energy edges. The best way to develop that intuition is by doing the experiment yourself.
Can I use Athena to counteract the self-absorption and get the credible spectra for sdandards?
Yes it can -- in principle. Doing a good job of self absorption correction requires a lot of knowledge about your sample and its matrix -- often more than one has.
The use of the self absorption tool is explained in the Athena document. I have also written a lecture on this topic. Go to
https://github.com/bruceravel/XAS-Education/downloads
and download the "selfabs.pdf" file.
B
--
Bruce Ravel ------------------------------------ bravel@bnl.gov
National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973
My homepage: http://xafs.org/BruceRavel EXAFS software: http://cars9.uchicago.edu/ifeffit/Demeter
_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
participants (3)
-
Bruce Ravel -
Matthew Marcus -
陈娟