- Ifeffit - millenia.cars.aps.anl.gov

RE: [Ifeffit] Debye-Waller factors for metal oxides ?
by Kelly, Shelly D. 07 Feb '05

07 Feb '05

I agree with John Rehr. With a temperature dependent data set, the best model that I have found is to allow each atom pair to have a thermal and static disorder component. Sigma2 = debye(Temp,Td_O1) + stat_O1 (This example is for the first oxygen (O1) shell of the metal oxide) The static disorder component should be the same for all data sets from the same sample at the different temperatures. The thermal component should increase as the temperature increases. If you are lucky a Debye or Einstein model will work well enough. You will have to test both. In addition, the thermal change in sigma2 should be about the same for different samples that are closely related. (We could argue about this last point. The best thing to do is to test this constraint by first not imposing it and determine within the uncertainties in the measurement, if it is true. I have found that it usually is an okay constraint.) In the end, you simultaneously model the temperature dependence and the sample dependence spectra all together. You constrain the static disorder term to be independent of temperature, and you constrain the debye temperature to be independent of sample change. This should impose enough constraints on sigma2 values to accurately look at changes in coordination number. Of course the coordination numbers do not change as a function of temperature but only as a function of sample changes. HTH Shelly Shelly Kelly ---------------------------- skelly(a)anl.gov Bldg 203, RM E113 phone: (1) 630-252-7376 Argonne National Laboratory fax: (1) 630-252-9793 Argonne, IL 60439 www.mesg.anl.gov > -----Original Message----- > From: ifeffit-bounces(a)millenia.cars.aps.anl.gov [mailto:ifeffit- > bounces(a)millenia.cars.aps.anl.gov] On Behalf Of John J. Rehr > Sent: Monday, February 07, 2005 10:07 AM > To: XAFS Analysis using Ifeffit > Cc: John J. Rehr > Subject: Re: [Ifeffit] Debye-Waller factors for metal oxides ? > > > > The lack of accurate Debye-Waller factors in complex materials, like > metal oxides, can be a real problem. In my view, simple phenomenological > methods like correlated Debye and Einstein models can't always be trusted. > > There is an alternative approach to fitting DW factors in such systems > which might be useful. The idea is to fit a few spring constants k_ij > between > the atoms, rather than the DW factors themselves, e.g., for the most > important > sites. This is efficient since the spring constants are temperture > independent. This has been tested by Rossner et al. using their Bayesian > approach [Rossner, H. H. and Krappe, H. J., (2004). Phys. Rev. B 70, > 104102] > The fitting is does via an analytical formula for the Debye-Waller factors > based on a fast recursion method [Poiarkova, A. V. & Rehr, J. J. (1998). > Phys. Rev. B 59, 948]. > > Of course, such an approach still misses contributions to the DW factors > due to structural disorder, which are addititive. > > J. Rehr > > > > On Sun, 6 Feb 2005, Ian James Drake wrote: > > > Hi, > > > > There is little discussion over the significance of fitted Debye-Waller > > factors (ss2) found for metal centers in metal oxides of heterogeneous > > catalysts. I am searching for a better way of determining how > reasonable > > these values are in the fits I get. > > > > 1. Are phenomenological models like the correlated Debye model and > Einstein > > model appropriate? It seems that Debye temperatures are hard to find for > > oxides. > > > > 2. Is there an equivalent value for ss2 in x-ray crystallography data > that I > > can refer to for comparison in these materials? What should I look for > in > > reviewing published x-ray crystallographic data? > > > > 3. Is the temperature dependence reported for the ss2 for some metals > (like > > Cu and Al) similar to their respective oxides? For example, in Debye- > Waller > > factor calculations reported for Al metal by R.C.G Killean (in J.Phys.F: > > Metal Phys., v. 4, pg. 1908, 1974), the ss2 changes by a factor of three > > between 25C and 300C. Would I expect a factor of three increase in > > crystalline zeolites (AlO4 structural units). > > > > 4. Specific to my work: I have studied Al containing oxides at > temperatures > > between 25 and 300 C. I would like to quantify coordination changes > by > > doing EXAFS fitting of the Al K-edge data. The EXAFS was taken at > temperature > > and I observe no change in the broadening of the data. Likewise, the > fitting > > shows no sensitivity in the Debye-Waller factor. It is nearly constant > at > > 0.001 A over the temperatures of interest. Since ss2 and the > coordination > > number (CN) are correlated, I would like a way to bind the error on the > > fitted CN by modeling real physical changes in the Debye-Waller factors. > Do > > you have any suggestions? > > > > Looking forward to your thoughts. > > > > Ian Drake > > Graduate Student > > UC Berkeley > > _______________________________________________ > > Ifeffit mailing list > > Ifeffit(a)millenia.cars.aps.anl.gov > > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit > > > _______________________________________________ > Ifeffit mailing list > Ifeffit(a)millenia.cars.aps.anl.gov > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

1 0

EXAFS sample prep
by Kelly, Shelly D. 02 Feb '05

02 Feb '05

Hi Ritika, I hope you do not mind, I'm sending this to the Ifeffit mailing list since others might be interested as well. This email message is about how to prepare an EXAFS sample. We are talking about measuring a Ti-edge from some sample in solution. But any edge and any sample in any matrix will also work just file. You want to set-up your sample so that you have a lot of Ti atoms in the path of the x-ray beam. This is because the EXAFS signal comes from the adsorption from the Ti atoms. Hence more Ti atoms give more adsorption events resulting in a smaller statistical error in the EXAFS spectra. BUT..... In transmission mode the EXAFS signal comes measuring the number of x-rays incident on the sample and transmitted through the sample. If all the x-rays are adsorbed by the Ti atoms in the sample then you do not have any transmitted x-ray signal. Hence you get really noise data because you have no statistically significantly transmitted x-ray signal. So you need your sample to be thin so that a lot of the x-rays pass through the sample such that you have something to measure. In fluorescence mode the EXAFS signal comes from the x-rays emitted by the Ti atoms as the core hole is re-filled. These emitted x-rays need to escape the sample without being re-adsorbed by the sample. Hence the sample needs to be dilute for fluorescence mode measurements. When you make a measurement in fluorescence mode it is always best to collect the transmitted x-ray signal as well. Then you can make a comparison to see if "self adsorption effects" are small. Transmission measurements typically have better signal to noise ratio than fluorescence measurements. So if you can, make your measurements in transmission mode. You can use Tkatoms or Artemis to calculate the best sample thickness and the best concentration to use for the measurements and which EXAFS mode will work best for your samples. >From the components of your sample you can calculate the thickness of the sample that will give 2.5 adsorption lengths. The transmitted x-ray intensity is given by the incident x-ray intensity multiplied by the exponential of the absorption coefficient times the sample thickness. I = I0 * exp(-mu(E) * x) Where I is the transmitted x-ray intensity I0 is the initial x-ray intensity Mu(E) is the absorption coefficient = weighted sum of absorption coefficient from each atom in the beam. X is the thickness of the sample. IF mu(E) * x = 2.5 then the percent of transmitted x-rays is about 10% of the incident x-ray intensity. This is just enough at NSLS where the incident x-ray intensity will be about 10^6 x-rays per second. This means that 10^4 x-rays per second will hit your transmitted Ion chamber. You want to make the transmitted ion chamber adsorb all (99%) of theses x-rays. Then your statistical error in the measurement will be 1/sqrt(N) or 1/sqrt(10^4) or 1%. You need to have at least this many x-rays to measure. At the beam line you measure the total absorption edge. The EXAFS signal, if you are lucky, makes up about 10% of the absorption edge. Hence you need to have a statistical error of no more than 1% to make a measurement. If you can make the sample thinner then you do even better. Now for the other problem: You want to get as many Ti atoms into the sample with a total sample thickness of 2.5 absorption lengths. To do this part you need to calculate the absorption of just the Ti atoms in the sample. I call this delta_mu(E). If you can get the delta_mu(E) * x to be between 0.1 and 1.0 with the total absorption of the sample below 2.5 then you can make the measurement in transmission mode. This means that between 10% and 70% of the absorption in the sample will be due to the Ti atoms. This is enough to get a good transmission signal. If you adsorption from the Ti atoms in the sample is between 1% and 10% then you will hope to make the measurement in fluorescence mode. Now for the practical part: Say you have a solution with 1% Ti-stuff in water by NUMBER of atoms. If you usually work by weight then you need to convert to number for this calculation. Take the feff.inp file for Ti-stuff. Look at the top of the file. You should find total_mu(E) (absorption of the entire sample at the Ti edge), and delta_mu(E) (absorption of Ti atoms at the Ti-edge). Then use Hephaestus to calculate the total_mu(E) for water at the Ti edge energy. To get the total absorption you need to calculate the weighted average for water and Ti-stuff. See how thick you can make your sample and satisfy the criteria for less than a total absorption length of 2.5. (99%_by_number_water_atoms * total_mu(E)_for_water + 1%_by_number_Ti-stuff * total_mu(E)_for_Ti-stuff ) * sample_thickness = 2.5 Once you have that thickness then you can see what the absorption for just the Ti atoms is. 1%_by_number_Ti-Stuff*delta_mu(E)_for_Ti * sample_thickness = absorption_length_for_Ti And then you can figure out if you have a fluorescence or transmission experiment. Absorption_length = 0.1 to 1.0 you have transmission. Absorption_length=0.01 to 0.1 you have fluorescence. Usually you have to go around and around with this sort of calculation to optimize the conditions for the measurements. Hope that Helps Shelly

1 0

RE: [Ifeffit] Two edge problem (in the same scan)
by Kelly, Shelly D. 26 Jan '05

26 Jan '05

> -----Original Message----- > From: ifeffit-bounces(a)millenia.cars.aps.anl.gov [mailto:ifeffit- > bounces(a)millenia.cars.aps.anl.gov] On Behalf Of Bruce Ravel > Sent: Wednesday, January 26, 2005 1:30 PM > To: asg2(a)lehigh.edu; XAFS Analysis using Ifeffit > Subject: Re: [Ifeffit] Two edge problem (in the same scan) > > > Hello everybody, > > I have a question/problem. Last year during the EXAFS course at NSLS, I > > had requested Bruce if it is possible to analyze the data with two > > edges. For example, I have the data for As and Se K-edges, which I > > collect in the same scan. In WinXAS program, I can section the edges and > > analyze them separately. At that time, the program was not able to > > perform these edges or the sectioning. Have we reached this possibility. > > I tried again using the latest updated Athena, but it seems not to be > > working. Can you please tell me whether the sectioning or dealing with > > two edges in the same spectra can be done > > I don't unfderstand why "sectioning" is necessary. What precludes you > from reading the data set a second time and setting e0 to be appropriate > for the second edge? Then your pre- and post-edge lines will be made > using the second e0 and everything else should just follow normally. > > I understand that Athena does not have a way to specify the second edge in > the column selection dialog, but that seems to me like an inconvenience > more than a real limitation. It is an inconvenience because you then have > to specify the second e0 by hand rather than have Athena's magic find it > for you, but is that really so hard? > > B > [Kelly, Shelly D.] Another feature that I have found useful for data with multiple edges is the truncate function. Once you read the data set into Arthena. Make a copy of the data set. Choose your e-zero for each data set. And then truncate the data sets to the "normal" range. This will help with some of the automatic presets, which use the full data range by default, and can give you a very scary initial background function. SK

1 0

Two edge problem (in the same scan)
by Ashtosh Ganjoo 26 Jan '05

26 Jan '05

Hello everybody, I have a question/problem. Last year during the EXAFS course at NSLS, I had requested Bruce if it is possible to analyze the data with two edges. For example, I have the data for As and Se K-edges, which I collect in the same scan. In WinXAS program, I can section the edges and analyze them separately. At that time, the program was not able to perform these edges or the sectioning. Have we reached this possibility. I tried again using the latest updated Athena, but it seems not to be working. Can you please tell me whether the sectioning or dealing with two edges in the same spectra can be done Ashtosh

2 1

[DATA-REQUEST] Platinum standards
by MauRo 24 Jan '05

24 Jan '05

Hi to all, I'm looking forward for some EXAFS (transmission or fluorescence) spectra of Platinum oxide and Platinum carbide at L_III and/or L_I edge. Is there anyone who has taken this data and is so kind to send them to me? Thanks, Mauro

1 0

Increase NOVP error message meaning
by Paul Fons 23 Jan '05

23 Jan '05

I am trying to run feff on some chalcogenide structures that have rather nearest-neighbor long bond lengths, e.g. > Bond search from target site: Te1 > ...found 6 atoms within 0.00 - 3.20 A of this site > > ------------------------ > dist [Å] atoms > ------------------------ > 1. 3.0488 6 > ------------------------ > Bond search complete When I use this structure in feff82, I get the rather mysterious error INCREASE NOVP: > > Calculating potentials ... > free atom potential and density for atom type 0 > free atom potential and density for atom type 1 > free atom potential and density for atom type 2 > free atom potential and density for atom type 3 > initial state energy > overlapped potential and density for unique potential 0 > overlapped potential and density for unique potential 1 > overlapped potential and density for unique potential 2 > overlapped potential and density for unique potential 3 > muffin tin radii and interstitial parameters > tell authors to INCREASE NOVP Being an experimentalist, I did increase novp until the previous error message stopped, however this resulted in a core dump: > Calculating potentials ... > free atom potential and density for atom type 0 > free atom potential and density for atom type 1 > free atom potential and density for atom type 2 > free atom potential and density for atom type 3 > initial state energy > overlapped potential and density for unique potential 0 > overlapped potential and density for unique potential 1 > overlapped potential and density for unique potential 2 > overlapped potential and density for unique potential 3 > muffin tin radii and interstitial parameters > iph, rnrm(iph)*bohr, rmt(iph)*bohr, folp(iph) > 0 1.76126E+00 1.74898E+00 1.01667E+00 > 1 1.71891E+00 1.63874E+00 1.12886E+00 > 2 1.35785E+00 8.69242E-01 1.15000E+00 > 3 1.78418E+00 1.76792E+00 1.02193E+00 > mu_old= nan > Done with module 1: potentials. > Calculating LDOS ... > LDOS calculation for specified grid > Calculating energy and space dependent l-DOS. > It takes time ... > potential type 0 > Segmentation fault (core dumped) When running the equivalent file in feff6, I get the > Calculating potentials and phases... > free atom potential and density for atom type 0 > free atom potential and density for atom type 1 > free atom potential and density for atom type 2 > free atom potential and density for atom type 3 > overlapped potential and density for unique potential 0 > overlapped potential and density for unique potential 1 > overlapped potential and density for unique potential 2 > overlapped potential and density for unique potential 3 > muffin tin radii and interstitial parameters > phase shifts for unique potential 0 > Hard tests failed in fovrg. > Muffin-tin radius may be too large; coordination number too small. > phase shifts for unique potential 1 > Hard tests failed in fovrg. > Muffin-tin radius may be too large; coordination number too small. > phase shifts for unique potential 2 > Hard tests failed in fovrg. > Muffin-tin radius may be too large; coordination number too small. > phase shifts for unique potential 3 > Hard tests failed in fovrg. > Muffin-tin radius may be too large; coordination number too small. > Preparing plane wave scattering amplitudes... > nncrit in prcrit 9 This message seems similar to the troubleshooting section in feff8 for which it states: • Non-physical, widely spaced distributions of atoms. Symptoms of this common problem are very large muffin-tin radii (see the header of any ‘.dat’ file) and possibly a failure of the phase-shift program to converge. This gives error message hard test fails in fovrg. How to I avoid this problem (obviously the crystal structure is a given!). I tried the INTERSTITIAL card, but to no avail. Any suggestions are welcome. Dr. Paul Fons Senior Researcher National Institute for Advanced Industrial Science & Technology METI Center for Applied Near-Field Optics Research (CANFOR) AIST Central 4, Higashi 1-1-1 Tsukuba, Ibaraki JAPAN 305-8568 tel. +81-298-61-5636 fax. +81-298-61-2939 email: paul-fons(a)aist.go.jp The lines below are in a Japanese font 〒３０５−８５６８茨城県つくば市東１−１−1 つくば中央第４近接場光応用工学センターポール・フォンス主任研究官

1 0

Windows update for athena/artemis
by Matt Newville 18 Jan '05

18 Jan '05

Hi, I've built a Windows update file to upgrade to the latest versions of Athena and Artemis. Since Bruce is off-line for a while and since there is no easy way for us to make these executables, there is not much we'll be able to do if there is a problem with this version. Because of this, I am NOT releasing this as a normal update yet, but would like the braver of you to download and install this update by hand and give the new versions of Athena and Artemis a test drive. The update file is at http://cars9.uchicago.edu/ifeffit/updates/ifup_013.zip which is just a simple zip file that can be unpacked in your Ifeffit folder. (Probably C:/Program Files/Ifeffit). If you're willing to try these versions, please let me know how it goes. I'll wait until I hear some positive responses before I make a general update release that would automatically update everyone to these versions. Thanks, --Matt

1 0

RE: [Ifeffit] Some questions about ATOMS and FEFF
by Barton, David (DG) 16 Jan '05

16 Jan '05

Patrick, Welcome to the mailing list! This list is an excellent forum for newcomers and veterans alike. > 2) Running the FEFF6 software directly or through the Artemis > bundle, I never get any "chi.dat" output file as a weighted sum of the > individual This feature is built into Artemis. After you run the FEFF calculation with Artemis and choose which paths to import, just highlight the paths in the DATA & PATHS window that you would like to sum and then choose DATA-->SUM OF PATHS-->SELECTED DATA PATHS from the menu. This will produce the chi(k) sum which can be plotted along with your data set. A shortcut to this menu item can also be found by right-clicking on the selected paths and choosing DATA-->SUM OF PATHS-->SELECTED DATA PATHS. I often use this feature to get a rough feeling for the relative contributions of each path in the calculation by plotting the SUM along with all of the individual paths. Have Fun, Dave _________________________________ David Barton, PhD. The Dow Chemical Company Catalysis R&D 1776 Building Midland, MI 48674 USA Ph: (989) 636-4753 Fax: (989) 638-6225 Email: dgbarton_remove_me(a)dow.com > -----Original Message----- > From: ifeffit-bounces(a)millenia.cars.aps.anl.gov > [mailto:ifeffit-bounces@millenia.cars.aps.anl.gov] On Behalf Of Patoo > Sent: Sunday, January 16, 2005 9:07 AM > To: ifeffit(a)millenia.cars.aps.anl.gov > Subject: [Ifeffit] Some questions about ATOMS and FEFF > > > Hi, > > I am just starting to work with the ARTEMIS/ATHENA software suite for > EXAFS data processing and analysis and a newcomer in this > mailing list. This project is really very nice and useful for > the EXAFS community and > I really enjoy to learn how to use it. > I read most of the documents which are available on the website and > around, but I have already two questions : > 1) in the "feff.inp" command file, what are the options of the cards > CONTROL and PRINTS > 2) Running the FEFF6 software directly or through the Artemis > bundle, I > never get any "chi.dat" output file as a weighted sum of the > individual > "feffNNNN.dat" individual path contribution. Is there any forgotten > option to get it ? > > Thanks in advance for your help > > Patrick > > -- > *********************************************************** > Dr. Patrick Judeinstein > R.M.N. en Milieu Orienté, (UMR 8074), Bâtiment 410 > I C M M O (Institut de Chimie Moléculaire et des Matériaux d'Orsay) > Université Paris-Sud, 91405 Orsay Cedex - France > Tel : +33 (0)1 69 15 70 17 ou ++ 33 (0)1 69 15 47 68 ; Fax + > 33 (0)1 69 15 81 05 > > e-mail : pjudeins(a)icmo.u-psud.fr > web site : http://patoo.hope.free.fr > > *********************************************************** > > > > _______________________________________________ > Ifeffit mailing list > Ifeffit(a)millenia.cars.aps.anl.gov > http://millenia.cars.aps.anl.gov/mailman/listi> nfo/ifeffit >

1 0

Some questions about ATOMS and FEFF
by Patoo 16 Jan '05

16 Jan '05

Hi, I am just starting to work with the ARTEMIS/ATHENA software suite for EXAFS data processing and analysis and a newcomer in this mailing list. This project is really very nice and useful for the EXAFS community and I really enjoy to learn how to use it. I read most of the documents which are available on the website and around, but I have already two questions : 1) in the "feff.inp" command file, what are the options of the cards CONTROL and PRINTS 2) Running the FEFF6 software directly or through the Artemis bundle, I never get any "chi.dat" output file as a weighted sum of the individual "feffNNNN.dat" individual path contribution. Is there any forgotten option to get it ? Thanks in advance for your help Patrick -- *********************************************************** Dr. Patrick Judeinstein R.M.N. en Milieu Orienté, (UMR 8074), Bâtiment 410 I C M M O (Institut de Chimie Moléculaire et des Matériaux d'Orsay) Université Paris-Sud, 91405 Orsay Cedex - France Tel : +33 (0)1 69 15 70 17 ou ++ 33 (0)1 69 15 47 68 ; Fax + 33 (0)1 69 15 81 05 e-mail : pjudeins(a)icmo.u-psud.fr web site : http://patoo.hope.free.fr ***********************************************************

2 1

re:fits
by k-kupiecki＠northwestern.edu 15 Jan '05

15 Jan '05

Thank you Bruce and Shelley! I just wanted to thank Matt, Bruce and everyone involved with IFEFFIT/Artemis/Athena... it has been a Godsend.... I couldn't have stumbled on it at a better time (I'm frantically trying to analyze about 4 beamtimes worth of SEXAFS data for papers and my thesis and WINXAS wasn't really working out for me)!!!!! Does anyone have any suggestions for additional literature in regards to better understanding the quality of fits and the parameters involved?.... or perhaps know when we might be able to expect a updated version of the Artemis literature? (What is currently there has helped me sooo much that I've been checking for an update every couple of weeks hoping to get see sections 2.2-3.2..... ) Also.... I guess I've realized that only knowing IFEFFIT with Artemis I kinda take things for granted (like I didn't know that IFEFFIT was having problems, in certain cases since I had only been checking the "Results" and the error messages on the bottom of the window....) which meant I missed errors like this in the IFEFFIT (since I didn't notice anything unusual in the "Results"): ............ # Do the fit! ## fitting difference of Cov_0_06ML_bl1435_111703.chi and one or more paths ... ## === data set #1 of 1 feffit(1-8, group=data0_fit, chi=data0.chi, k=data0.k, kweight=2.00, rmin=1, rmax=4, kmin=2.5, kmax=15, dk=2, kwindow=kaiser-bessel, fit_space=R, do_bkg=no, data_set=1, data_total=1) feffit fitting 1 data sets fitting ... fitting ... fitting ... fitting ... fitting ... fitting ... fitting ... fitting ... no uncertainties estimated! fit gave a warning message: too many fit iterations. try again with better guesses or a simpler problem. ............................................... I guess my question is.... is there anything else I should be looking for in the IFEFFIT printout that I might not otherwise notice by just using Artemis and looking at the error messages and "Results"? Thanks for the input! Sincerely, Kristine Witkowski Northwestern University

6 8