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

a limitation of Athena, Artemis, and Ifeffit
by Bruce Ravel 17 Jun '03

17 Jun '03

Hi folks, An irritating limitation of the codes was just brought to my attention. My colleague here at NRL had his data file in a deeply nested folder and each folder name was long and descriptive. The end result was that the fully resolved file names (i.e. path and name) was many dozens of characters. When he tried to import these data into Athena, Athena failed silently. That is, Athena failed to import the data and failed to report a reason why. The reason is that there is, in ifeffit, a limit to the total number of characters for the fully resolved filename. My colleague, with his long, descriptive folder names, had exceeded this limit. In that situation, the current version of Athena (and Artemis, as well) fails silently. In the next versions of the codes, they will issue a warning if your fully resolved filename is too long. For the moment, however, the best I can do is to make you all aware of the problem and to suggest either short folder names or placing your data high up in the file tree. Regards, B -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b, X24c, U4b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

1 0

athena 0.8.020 and artemis 0.6.005
by Bruce Ravel 16 Jun '03

16 Jun '03

Hi folks, Just in time for the XAFS conference, I am announcing new releases of Athena and Artemis. There are not many major new features this time, but there is a long list of bug fixes and usability enhancements. If you have not yet upgraded to Ifeffit 1.2.1, you should do so, then download the latest versions of my codes. As always, new source code tarballs and windows executables can be found on my downloads page. For those Windows users new to the mailing list, I'll explain how to install the new executables. It's quite simple. Grab the .exe files from my downloads page and drop them in the C:\Program Files\Ifeffit\ folder on top of the old versions. If you installed to some other location, then drop th new executables there. Doing this is enough to leave the desktop and Start Menu icons functional. The one major new feature in Artemis is the ability to set restraints on your parameters. It's as simple as setting a math expression on the "Guess. Def, Set" page and flagging the parameter as a restraint. The restraint defined this way will then be used in the fit. If you are interested in using restraints, you should read the following: http://cars9.uchicago.edu/ifeffit/refman/node55.html http://cars9.uchicago.edu/xafs/Advanced_Methods/BNLworkshop2001/newville.pdf There is an example project file using a restraint in the source code tarball which follows very closely the restraints example from the ifeffit source code tarball. There really are a lot of improvements this time around and it is well worth updating. Check out the change logs on my web site for all the details. The acknowledgment list is a long one this time. Among those with helpful bug reports and good suggestions: Scott Calvin, Vince Harris, Dale Brewe, Adam Webb, Mark Jensen, Paul Fons, and Grazia Proietti. Thanks to all of you. The award for the loopiest suggestion which I had no good reason to refuse goes to Mark Bondin, who wants to use Athena for background removal and Xfit for data analysis! B P.S. There are a few of you who have expressed an interest in using Artemis to analyze your DAFS data. It was pointed out to me that I had forgotten to include the dphase path parameter in Artemis. That has been fixed, although the dphase parameter is hidden along with the k_array, amp_array, and phase_array path parameters. If you want to use dphase or any of the other three "hidden" parameters, go to the Paths menu and click on the button at the bottom of the menu that reads "Extended path parameters". When you click it on, text boxes for dphase and the _array path parameters will be displayed on the paths page in Artemis. P.P.S. I am giving a poster about Athena and Artemis at the XAFS conference. My time is Wednesday during the 16.15 to 17.45 session. If you'll be at the conference, stop by and say hi. I'd enjoy meeting those of you I don't already know and would enjoy catching up with those I do. -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b, X24c, U4b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

1 0

XAFS 12
by Matt Newville 07 Jun '03

07 Jun '03

Hi Everyone, Many people on this list will be at the upcoming XAFS 12 conference, and have probably seen the notice for the microsymposium on XAFS analysis programs. Bruce will talk about the 'Ravelware', I'll talk about Ifeffit etal, and several real luminaries in the field will talk about XAFS and XANES analysis. The conference is a great chance for many informal discussions about the ifeffit programs. At this point, there are many possible directions for development and improvement and I (and I assume Bruce and Sam) would love to hear feedback on what the priorities for new functionality and development should be. It might be good to set aside a lunch time (or beer-time?) to talk about development priorities. I'd tenatively pick Thursday lunch (after the analysis symposium and most posters/talks to be able to learn some things from the conference), but maybe we can finalize the time and place at the Tuesday evening analysis symposium. A few possible topics of discussion are 'Installation on MacOS X' and 'Getting GUIs that run Feff6', but there may be other issues. If you're interested, please feel free to join the discussion, and see you in Malmo! --Matt

1 0

ifeffit 1.2.1
by Matt Newville 28 May '03

28 May '03

Hi again, Well, it was supposed to be the 'stable' release, but a few large bugs were noticed quickly in ifeffit 1.2.0: - Feff6L didn't make it past the pathfinder before dying. - some of the output arrays for feffit() with multiple k-weights were still messed up. These turned out to be easy to track down (and the Feff problem was all mine -- not native to Feff!!!), so I fixed these to make version 1.2.1 (both source and Window exe), which is now available at http://cars.uchicago.edu/ifeffit/download.html and at Sourceforge: http://sourceforge.net/project/showfiles.php?group_id=80919&release_id=1603… It usually takes a day for Sourceforge's mirrors to get new files. Since this is supposed to be the 'stable' version, I don't intend to be making releases every 5 days. These problems were big and easy to fix, and I'll be away for a week so thought I should post these now. Really, it will calm down. As others have suggested, I like the idea of being able to update the individual files (athena.exe, ifeffit.dll, feff6l.exe) to an already-installed Windows tree. It should be fairly easy to write a program that looks at some webpage for a zip file with the changes since version "1.2.X" and update the needed files. --Matt

5 10

TkAtoms weirdness
by Carlo U. Segre 23 May '03

23 May '03

Bruce: I have recently encountered a weird problem with TkAtoms. Let me detail it setp by step. 1. Start TkAtoms and build a model from scratch. 2. Press the "Run Atoms" button. The output of this is correct and can be saved to disk. 3. Save the "atoms.inp" file associated with the model that has been successfully run and exit TkAtoms. 4. Start TkAtoms again and load the previous "atoms.inp" file. 5. Press the "Run Atoms" button again. This time the output (the FEFF input file) is completely different than in step 2. It is also completely incorrect. Many of the x,y,z coordinates are set to zero. In fact the x coordinate is always zero. If I clear the data boxes and re-enter the original numbers, the results are correct once again. This is extremely confusing to me. I have only tried it with a single structure but it shows this behavior consistently. I have attached the offending atoms.inp file. Carlo -- Carlo U. Segre -- Professor of Physics Associate Dean for Research, Armour College Illinois Institute of Technology Voice: 312.567.3498 Fax: 312.567.3494 Carlo.Segre(a)iit.edu http://www.iit.edu/~segre

3 2

Re: fitting with artemis
by Bruce Ravel 22 May '03

22 May '03

Gleb, I trust you will not mind if I CC my answer to your question to the Ifeffit mailinglist. It is, I think, of sufficiently general interest. > How can I constrain a guess parameter to some limited range of > values during a fit? For example during a fit I want my DW factor > be varied but in the limited range, e.g. from 0.002 to 0.01? In the > ifeffit doc I hav'nt found a math expression for confining a guess > parameter. There are two possible answers. 1. To constrain a value as you wish, this will work: guess x 0.005 set y max(0.002, min(x, 0.01)) That idiom will require the value never to be smaller than 0.002 and never larger than 0.01. A caveat: If, without this constraint, the fit wants that parameter to become too large or too small, than the likely result of this constraint is that the value will be pinned at one end or the other. The min/max idiom often does not behave as you'd like. 2. The other option is to use a restraint as described in the Ifeffit reference manual: http://cars9.uchicago.edu/ifeffit/refman/node55.html Artemis does not yet have an interface to setting and using restraints, but you could use Artemis to write and save an ifeffit script, add the restraint by hand, and then run the ifeffit script from the command line. The ability to add restraints to a fit will soon be handled well by Artemis, though. HTH, B -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b, X24c, U4b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

1 0

background subtractions
by Bruce Ravel 19 May '03

19 May '03

Hey Carlo, Here is my answer to your other two questions, also CCed to the mailing list. > 2. What is the justification for fitting the background during the fitting > process. I understand that it is important to remove background large k > oscillations but it seems to me that this is better done without the bias > of simultaneously fitting FEFF shells. I know that Steve Wasserman and > Jeff Terry have a Mathematica background subtraction algorithm which > allows you to back FT the region below Rbkg and subtract it from chi. > This seems to be roughly the same idea as the background fit but is less > biased by the fitting. I would appreciate your comments. Also, is this > now a part of FEFFIT or is it solely in Artemis? It is roughly the same idea, but I would say that you *want* the background to be biased by the fitting. The issue here needs to be restated in terms of what is meant by "a good background removal". Implicit in the back transform algorithm that you describe is the assumption that a good background removal makes the low-R region "look pretty". That is, one assumes that the low-R regions should tail away in some monotonic fashion. Basically, this is an esthetic argument rather than a statistical or scientific argument. On some level, subtracting away a portion of the spectrum via a back transform algorithm is not horribly different from modifying the plot of the chi(R) spectrum with Photoshop or The Gimp. First off, I should explain how the background co-refinement in Ifeffit works. You define a lower bound of the fitting range in R-space. This Rmin value takes the role of Rbkg in the autobk algorithm. That is, Rmin is the cutoff in R-space above which you will use Fourier components from feffNNNN.dat files to fit the data and below which you will use a spline to fit the data. The spline is determined in exactly the same way that the spline in autobk is determined. Is this cheating in terms of number of `independent' point? No. The splin has exactly the number of knots corresponding to the band whose width in R is from 0 to Rmin. That is, different information is used by the spline than by the rest of the fit. Why is another pass of the autobkl algorithm needed at this stage? Why couldn't autobk just do the job correctly up front? Well, this is sort of a numerical issue. Autobk has a really hard job to perform. If you Fourier transform the entire mu(E) spectrum, you will find that the Fourier contribution from the mu0(E) function is much much huger than the contribution from the chi(K) function. Thus autobk is attempting to isolate something small out of something large. By the time Artemis sees the data, mu0(E) is all gone. Further refining the background is thus a matter of separating something small from something small. The REAL REASON to do the background co-refinement is because Ifeffit (and thus Artemis) will report on the **correlations between** the background parameters and the things you are really interested in, such as delta_R and sigma^2. This is improtant. Even more it is a way of defining "goodness" of a background removal that removes esthetic concerns. A "well-removed" background is one that is not too highly correlated with the fitting parameters. Of course, "not to highly" is a rather subjective, but it atleast is a number to work with. So what is the right value of Rmin? Well, you want to choose an Rmin that is high enough to make a pretty picture (esthetics is not entirely without merit, after all) but low enough so as not to introduce excessive correlation with the fitting parameters. Rmin, therefore, is one of the parameters that you need to fiddle with in your data. For oxides (and carbides, nitrides, etc) Rmin might be quite low. Indeed, there may not be a value which gives a spline that is loosely correlated with the fitting parameters and which also satisfies the esthetic criteria of a pretty low-R region taht tails off monotonically. To avoid harming the data by excessive correlation with the spline, you may be forced to accept low-R peaks even in your published data. Finally, how does this discussion relate to the background removal done in Athena? Well, it should be clear by now that you cannot know the right value of Rmin a priori. You have to do some fits and measure the correlation between the background and the fitting parameters to judge what Rmin should be. Good practice with Athena is to choose a rather low value of Rbkg and not worry that some icky little low-R peaks might remain in the data. If you choose a low Rbkg in Athena, you can always choose a higher Rmin value in Artemis. But you cannot go the other way. You should not choose a value of Rmin that is *lower* than the Rbkg used in Athena because then you will not correctly evaluate the correlations. > 3. When it comes time to report results, is it better to be able to > subtract the background? Is it legitimate to report fitting parameters > without including the "hidden" background parameters? I say yes, it is better to subtract the background found by Artemis. (That's what the "use previous background" button does.) Esthetics, as I confessed above, is still important. And yes, reporting fitting parameters from a fit with a background co-refinement is most certainly legitamate. Indeed, it is more so because you can also report the level of correlation with the background spline. This gives you reader MORE confidence in your results and a better ability to interpret their validity. Sorry I was so long winded, but these are subtle issues. When I give a series of lectures on data analysis, I usually spend about 30 minutes discussing background issues. HTH B PS: Background co-refinement is, like everything else mathematical, something done by Ifeffit. Ifeffit does all the heavy lifting, Artemis just keeps it all organized. -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b, X24c, U4b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

3 2

Re: [Ifeffit] Windows Installer for Ifeffit 1.2.0
by Mark Jensen 19 May '03

19 May '03

Bruce, Matt, and the curious: Thanks for clarifying your distribution philosophy. >I think we haven't figured this out yet. Is it OK to download >ifeffit-1.2.0.exe as the basic installer, and then install >whatever updates to athena/artemis may exist on top of that? >>I thought that people were, by and large, happy with the solution of >>dropping new athena & artemis exes on top of the old ones rather than >>downloading the 30MB installer every time I make an update. Since I I very much like the current system of being able to 'plug-in' new versions of Athena and Artemis exe's as they are made available. I just wasn't sure if that approach was being abandoned. >>In any case, either you got confused updating or we got confused >>building the installer. It *was* our intention that the new installer >>have Athena 0.8.019 and Artemis 0.6.004. As far as what happened in the installation goes I'm not sure who was confused. Maybe I was pushing the installer a little too hard: I definitely chose the correct program to install Ifeffit 1.2.0 because: 1) Its filename is 'ifeffit-1.2.0.exe' which makes the choice pretty obvious 2) Ifeffit 1.2.0 was installed 3) When I repeated the procedure on Monday morning, making absolutely certain that I used ifeffit-1.2.0.exe, I was still left with Athena 0.8.015 and Artemis 0.5.008. (Those must have been the most recent versions of Athena and Artemis that I was using on that computer... I thought I was more up to date.) As I have done for Ifeffit installations in the past, I did no uninstallation or file deletion before trying the new install and I installed into the same directory as the existing version. When this Ravelware trouble caused me to look more closely at the directory contents, the only files that had May 2003 dates were the *.bat files and the new Ifeffit files. The dates on Athena and Artemis were still March 2003! However, during the install, I saw the installer say that it was unpacking Artemis and Athena. Apparently it did not install over the old versions (but it did install Ifeffit 1.2.0?!). The Solution: So, I uninstalled Ifeffit and deleted the remaining contents of the Ifeffit directory (Keeping my fingers crossed that the new install would work). The ensuing activation of ifeffit-1.2.0.exe now gives me the correct versions of Ifeffit, Artemis, and Athena! Now let's do some fitting. Thanks for your thoughts, Mark ------------------------------ Mark P. Jensen Chemistry Division Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439-4831 USA 630-252-3670 630-252-7501 FAX mjensen(a)anl.gov

1 0

Windows Installer for Ifeffit 1.2.0
by Mark Jensen 17 May '03

17 May '03

Bruce and Matt and everybody else, I am probably missing something, but the Artemis and Athena splash screens from the new Ifeffit 1.2.0 windows installer that I just used say that it installed Athena 0.8.015 and Artemis 0.5.008 and Ifeffit 1.2.0. Initially, I thought that the windows installer would (as in the past) include the latest (and greatest) versions of Athena and Artemis. But now that I think about it more, Matt seemed to indicate that he wants a stable distribution, which might mean older versions of Athena and Artemis. Bruce's website, on the other hand, seems to say that the windows installer for Ifeffit 1.2.0 would have the most recent windows versions of Artemis and Athena. I politely ask, which of these ideas is winning and when can windows users hope for some new versions of the Ravel-ware. Mark Jensen ------------------------------ Mark P. Jensen Chemistry Division Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439-4831 USA 630-252-3670 630-252-7501 FAX mjensen(a)anl.gov

3 2

Re: Ifeffit Digest, Vol 3, Issue 8
by k-kupiecki＠northwestern.edu 16 May '03

16 May '03

John and Bruce, Thank you for your kind help. I'm still a little perplexed... Basically I'm taking SEXAFS on a Ni(110) single crystal substrate (with a fraction of a monolayer of Au which substitutes into the first layer... SEXAFS spectra were obtained at the AuL3 edge). I've taken data in four different posistions: "grazing incidence" with the E vector of the x-rays parallel to the surface normal 1) with x-rays along the (001) direction 2) with x-rays along the (-110) direction and "normal incidence" with the E vector of the x-rays perpendicular to the surface normal (E parallel to the surface plane) 3) with x-rays along the (001) direction 4) with x-rays along the (-110) direction After receiving both Bruce and John's emails I tried three different feff.inp files, using my original hemisphere cluster, each using one of the following: POLARIZATION 1 1 0 or POLARIZATION 0 0 1 or POLARIZATION -1 1 0 but upon ploting out chi(k) vs. k from the chi.dat's for each of these they looked identical... did I miss something? Thanks, Kristine Witkowski Northwestern University > > Hi Kristine, > I basically agree with what Bruce Ravel wrote. We've occasionally done > FEFF calculations on surfaces (e.g. with K. Baberschke and students), > and the key to the calculation is to use appropriate Polarization > cards (and Ellipticity cards) in the calculation as described in the > doc. Please let me know if you have problems doing this. > Cheers, > John >

1 0