Problems with fitting in ATHENA for XANES
Hello Everyone! I am the new user of XAFS method. Recently I obtained some data with XANES method. I have just started using ATHENA for my data analysis. I am using "Using ATHENA" document to learn the software. I have two questions 1.For XANES, Which method is better for knowing unknown data with known standards: Peak fitting analysis or Linear combination analysis. I would like to get some comments from your experiences so far which works better if tried both for same type of data. 2. In the peak fit analysis, in literature related to my work it was mentioned that peak modelled by 50%-50% combination of Gaussian and Lorentzian functions works better. Is it possible to do in ATHENA.? Hope to recieve some useful reply. Regards, Pushan ********************************************* Pushan Shah PhD Candidate CRC for Coal in Sustainable Development Graduate School of The Environment Macquarie University Australia Phone : +61 2 9850 7950 Fax: +61 2 9850 7972 **********************************************
Hi Pushan, Well, it seems that no one bit on your questions and I am now back from my trip to California, so I'll take a stab at this.
I am the new user of XAFS method. Recently I obtained some data with XANES method.
I have just started using ATHENA for my data analysis. I am using "Using ATHENA" document to learn the software.
I have two questions
1.For XANES, Which method is better for knowing unknown data with known standards: Peak fitting analysis or Linear combination analysis. I would like to get some comments from your experiences so far which works better if tried both for same type of data.
I don't see how I can answer this question in an objective way ;-) I rarely find peak fitting to be useful. I often find linear combination analysis to be useful. The caveat about linear combination fitting is that you need to have measured an appropriate set of standards. If you do not have the right standards, you cannot do linear combination analysis effectively. If you do, though, have a good set of standards, LCF is a very powerful tool indeed. The second caveat is that your problem needs to be well-suited for LCF. As an example of what I mean by that, consider the work in a paper that I just submitted with a colleague. In that paper, we report on the reduction of Au(III) chloride to metallic gold via interactions with cyanobacteria. Over the course of tens of hours the gold(III) chloride is converted to gold(I) sulfide then to colloidal gold. We analyzed these data using LCF, fitting those three species to spectra measured over the course of 10s of hours and relating the fractions of the species to a reaction rate constant. In that case, LCF is perfect because the sample itself is a combination of standards that we had measured. The point here is that you need to have a reason to believe that your sample is, in fact, a linear combination for LCF analysis to be useful. Peak fitting is often useful in a situation where the physical structure is known only partially. If some feature of the spectrum changes in response to changes in some extrinsic parameter (polarization, temperature, or some such) then peak fitting can be a good way to quantify that response. My main problem with peak fitting is that it is hard to know how to interpret the physical meaning of the line shapes used to do the fit. That is, it may not be clear how some Gaussian or Lorentzian used in the fit is related in a meaningful way to the electronic or atomic structure of the material.
2. In the peak fit analysis, in literature related to my work it was mentioned that peak modelled by 50%-50% combination of Gaussian and Lorentzian functions works better. Is it possible to do in ATHENA.?
Athena does not currently have a Voight or psuedo-Voight function. As you can see (http://cars9.uchicago.edu/iffwiki/HoraeToDoList) improvements to the peak fitting dialog are on my list of things to do. I would like to add a pseudo-Voight as well as some other lineshapes. One thing that comes to mind is that it would be nice to use the Cromer-Lieberman function rather than an arc-tangent for the step part of the fit. B -- Bruce Ravel ---------------------------------------------- bravel@anl.gov Molecular Environmental Science Group, Building 203, Room E-165 MRCAT, Sector 10, Advanced Photon Source, Building 433, Room B007 Argonne National Laboratory phone and voice mail: (1) 630 252 5033 Argonne IL 60439, USA fax: (1) 630 252 9793 My homepage: http://cars9.uchicago.edu/~ravel EXAFS software: http://cars9.uchicago.edu/~ravel/software/
OK, I'll weigh in:
Peak fitting is often useful in a situation where the physical structure is known only partially. If some feature of the spectrum changes in response to changes in some extrinsic parameter (polarization, temperature, or some such) then peak fitting can be a good way to quantify that response.
I think that PCA is a better way of quantifying the change with respect to an independent variable. For one thing, it tells you whether you can describe the change as a shift between two end-members, as in the reduction example, or whether you would need something more complicated. The obvious 'more complicated' is a combination of more species, for instance an initial form, a final species, and an intermediate for a reaction. However, another possibility, depending on the system, is some continous change in the structure, for instance, a bond angle which moves, so that one always has a single species, but that species changes. That could look enough like a combination to fool you. I don't know how to distinguish this sort of thing from a variable combination without knowing the science behind the system. For polarization, symmetry arguments would tell you that the data should be describable as a combination of two or three contributions, depending on what angle is being varied and what the symmetry of the system is. mam
participants (3)
-
Bruce Ravel -
Matthew Marcus -
Pushan Shah