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

A question
by H H 06 Nov '03

06 Nov '03

I have tried to use artemis to fit some data. I faced some problem and I am not sure if I am doing the right thing or not. Here is the problem, after I perform the fit then I can save the data and also the fit in K, R, and Q. However, when I do the phase correction using the option(path to use for phase correction), it did the job, but I did not allow me to save the fit. I am using the latest version posted oonline. I would appreciate if someone tell me if I am doing something wrong or the way to save it. Thanks, Hedy __________________________________ Do you Yahoo!? Protect your identity with Yahoo! Mail AddressGuard http://antispam.yahoo.com/whatsnewfree

2 1

iMol for Mac OSX
by Charles Bouldin 04 Nov '03

04 Nov '03

This may be of some use to the Mac OSX users here: http://www.pirx.com/iMol/index.html

1 0

quick Artemis ?
by k-kupiecki＠northwestern.edu 04 Nov '03

04 Nov '03

Hi, Just a quick Artemis question... I've been pulling in my FEFF8 paths by selecting the feff.inp in the appropriate folder with all the paths.... I've noticed that when I load in my path files this way that all of them are automatically selected to "Include feffNNN.dat in the fit". Is there a way I can get these all turned off at the same time (other than manually to each path)? So that I can add the paths one by one to the fit while they are all listed in Artemis? Thanks, Kristine Northwestern University

4 3

Re: [Ifeffit] quick Artemis ?
by k-kupiecki＠northwestern.edu 04 Nov '03

04 Nov '03

Thanks Mark and Peter for your quick responses.... you just saved me a ton of time today! Thanks! -Kristine

1 0

Inconsistency in generaton of feff.inp by atoms
by Carlo U. Segre 04 Nov '03

04 Nov '03

Bruce: I have just run up against an odd error in atoms. I am not sure why it happens but it is consistent among all my computers. I start tkatoms and set up a structural model fo rthe wurtzite structure as follows: ----------------- atoms.inp ---------------- ! This atoms input file was generated by TkAtoms 3.0beta8 ! Atoms written by and copyright (c) Bruce Ravel, 1998-2001 title = ZnS Wurtzite structure space = p 63 m c a = 3.820 b = 3.820 c = 6.260 alpha = 90.0 beta = 90.0 gamma = 120.0 core = Zn edge = K rmax = 7.0 shift 0.0000 0.0000 0.0000 atoms ! elem x y z tag occ. Zn 0.3333 0.6667 0.0000 1.0000 S 0.3333 0.6667 0.3748 1.0000 ----------------- Then I run atoms from within tkatoms and get a feff.inp file: ----------------- first feff.inp (top lines only)---------------- * This feff input file was generated by TkAtoms 3.0beta8 * Atoms written by and copyright (c) Bruce Ravel, 1998-2001 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * * total mu = 761.06 cm^-1, delta_mu = 589.98 cm^-1 * specific gravity = 4.091 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * * Normalization correction: 0.00045 ang^2 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * TITLE ZnS Wurtzite structure HOLE 1 1.0 * Zn K edge (9659.0 eV), second number is S0^2 * mphase,mpath,mfeff,mchi CONTROL 1 1 1 1 PRINT 1 0 0 0 RMAX 7.0 *CRITERIA curved plane *DEBYE temp debye-temp *NLEG 8 POTENTIALS * ipot Z element 0 30 Zn 1 30 Zn 2 16 S ATOMS * this list contains 68 atoms * x y z ipot tag distance 0.00000 0.00000 0.00000 0 Zn 0.00000 2.20546 0.00004 -0.78375 2 S_1 2.34058 -1.10276 -1.90996 -0.78375 2 S_1 2.34058 -1.10276 1.91004 -0.78375 2 S_2 2.34064 0.00000 0.00000 2.34625 2 S_3 2.34625 3.30822 1.91000 0.00000 1 Zn_1 3.82000 -3.30822 1.91000 0.00000 1 Zn_1 3.82000 0.00000 3.82000 0.00000 1 Zn_1 3.82000 3.30822 -1.91000 0.00000 1 Zn_1 3.82000 -3.30822 -1.91000 0.00000 1 Zn_1 3.82000 0.00000 -3.82000 0.00000 1 Zn_1 3.82000 ----------------------------- Then I save the atoms.inp file and when I read it in again and run atoms, I get a completely different result: ------------------- second feff.inp (top lines only) -------------- * This feff input file was generated by Atoms 3.0beta8 * Atoms written by and copyright (c) Bruce Ravel, 1998-2001 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * * total mu = 1522.11 cm^-1, delta_mu = 1179.96 cm^-1 * specific gravity = 8.182 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * * Normalization correction: 0.00045 ang^2 * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * -- * TITLE ZnS Wurtzite structure HOLE 1 1.0 * Zn K edge (9659.0 eV), second number is S0^2 * mphase,mpath,mfeff,mchi CONTROL 1 1 1 1 PRINT 1 0 0 0 RMAX 7.0 *CRITERIA curved plane *DEBYE temp debye-temp *NLEG 8 POTENTIALS * ipot Z element 0 30 Zn 1 30 Zn 2 16 S ATOMS * this list contains 136 atoms * x y z ipot tag distance 0.00000 0.00000 0.00000 0 Zn 0.00000 0.00011 0.00006 0.00000 1 Zn_1 0.00013 -1.10274 1.90987 -0.78375 2 S_1 2.34049 -1.10274 -1.90987 -0.78375 2 S_1 2.34049 2.20548 0.00013 -0.78375 2 S_2 2.34060 2.20548 -0.00013 -0.78375 2 S_2 2.34060 -1.10274 1.91013 -0.78375 2 S_3 2.34070 -1.10274 -1.91013 -0.78375 2 S_3 2.34070 0.00000 0.00000 2.34625 2 S_4 2.34625 0.00011 0.00006 2.34625 2 S_4 2.34625 -3.30811 -1.90994 0.00000 1 Zn_2 3.81987 -3.30811 1.91006 0.00000 1 Zn_3 3.81994 0.00011 -3.81994 0.00000 1 Zn_3 3.81994 3.30822 1.91000 0.00000 1 Zn_4 3.82000 ----------------------------------------------- You can see that the total mu, specific gravity and delta mu are different! I can run atoms from teh command line as well and I get the same incorrect result. I am running atoms-3.0beta8 but I tested beta9 and got the same result. I don't have this problem with the zincblende structure which is cubic so there must be a problem with the hexagonal lattice. Carlo -- Carlo U. Segre -- Professor of Physics Associate Dean for Special Projects, Graduate College Illinois Institute of Technology Voice: 312.567.3498 Fax: 312.567.3494 segre(a)agni.phys.iit.edu http://www.iit.edu/~segre

2 3

Inconsistency in generaton of feff.inp by atoms
by Bruce Ravel 04 Nov '03

04 Nov '03

Carlo, Peter, There is certainly a problem with your example. It seems likely that any hexagonal group will have this problem. Drat! I am very busy this week with some administrative responsibilities, but will look into the problem just as soon as I can and hopefully announce a fixed atoms soon after that. 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 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

Tk804.025_beta
by J. M. An 04 Nov '03

04 Nov '03

Until then, I recommend that you use the work-around I suggested earlier today. As a reminder, I have an older version of perl/Tk on my web site: http://leonardo.phys.washington.edu/~ravel/misc/Tk-800.024.tar.gz <http://leonardo.phys.washington.edu/%7Eravel/misc/Tk-800.024.tar.gz> If you install that, Athena and Artemis should work. If you have other software that requires the very latest perl/Tk -- I apologize for the inconvenience. Thanks, Bruce. Now, my previous problems with Tk804.025_beta disappears, when I install an older version Tk-800.024. I uninstalled the latest version, and intalled the older one. Joon An

1 0

Tk804.025_beta
by J. M. An 03 Nov '03

03 Nov '03

On Sunday 02 November 2003 06:43 pm, J. M. An wrote: >> (3) in using artemis, here is the problem >> >> Typing "artemis" brought up a window, but in "import a Feff calculaton" >> window under menu-bar "File", >> I couldn't move to any directory where "feff.inp" is, causing the >> following error message: >> >> Tk::Error: Can't locate Tk/Unselect.pm in @INC (@INC > > >Hi, >Yours is the second email this morning reporting this problem. I >don't know the answer, but I'll look into it and make an announcement >to the mailing list when I figure it out. Dear Bruce, As for the inability to move into/out of directories, the following error message may help to track down its cause: Tk::Error: Can't locate Tk/ListBrowse.pm in @INC (@INC contains: /usr/lib/perl5/5.8.0/i386-linux-thread-multi /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl /usr/lib/perl5/vendor_perl/5.8.0/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.0 /usr/lib/perl5/vendor_perl /usr/lib/perl5/5.8.0/i386-linux-thread-multi /usr/lib/perl5/5.8.0 .) at /usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi/Tk/Widget.pm line 270. I suspect that some ListBrowse.pm and Unselect.pm files are missing. Joon >From LBL

3 2

Regarding IFEFFIT updates and WIN98
by k-kupiecki＠northwestern.edu 03 Nov '03

03 Nov '03

Hi, I've been using IFEFFIT etc. on a Win98 machine. When I went to use Artemis today I received the message regarding an IFEFFIT update being available. I went to install it but it froze... so I figured it might be because of it being Win98..... So I just uninstalled my old version of IFEFFIT and installed the new version - which works great.... But I'm still getting this IFEFFIT prompt for updates even though I have the updated version of everything.... I'm just wondering if there is a way for me to turn that off.... Thanks, Kristine Northwestern University

2 1

Tk804.025_beta
by J. M. An 03 Nov '03

03 Nov '03

Dear helper, Sorry to interrupt to this installation procedure for questions, but it was definitely helpful to install horae-025 package. I'd like to report a problem in using artemis. I use RH 8.0 to install the package, following all your instructions. Log in as a normal user. (1) For Tk804.025_beta installation, xhost + localhost su export LANG=lang_us perl -MCPAN -e shell cpan>install Tk In the installation process, I encountered the following warnings: cc1: warning: changing search order for system directory "/usr/local/include" cc1: warning: as it has already been specified as a non-system directory But the installation was successful after all. (2) For horae-025 intallation, perl Makefile.PL make make intall I also encountered the same warnings as the above, but the installation seems ok. (3) in using artemis, here is the problem Typing "artemis" brought up a window, but in "import a Feff calculaton" window under menu-bar "File", I couldn't move to any directory where "feff.inp" is, causing the following error message: Tk::Error: Can't locate Tk/Unselect.pm in @INC (@INC contains:/usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl /usr/lib/perl5/vendor_perl/5.8.0/i386-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.0 /usr/lib/perl5/vendor_perl /usr/lib/perl5/5.8.0/i386-linux-thread-multi /usr/lib/perl5/5.8.0 .) at /usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi/Tk/Widget.pm line 270. Joon An From LBL.

3 3

[Ifeffit] problems using Athena and Artemis with perl/Tk 804.025 on linux and other unixes
by Bruce Ravel 03 Nov '03

03 Nov '03

Hi folks, The problem I am reporting in this email mostly effects users of linux and other unix systems. This does not effect Windows users of Athena and Artemis. There seem to be some severe problem in how Athena and Artemis interact with the latest version of perl/Tk. At least a couple of people have done the very sensible thing of installing the latest version of perl/Tk and then installing my software. Unfortunately, if you do so, you will run into lots of problems. As soon as possible, I will release new versions of Athena and Artemis that will play well with the latest version of perl/Tk. Until then, the best advice I can give is to "downgrade" your version of perl/Tk. Here is a copy of a version that I believe will allow you to use Athena and Artemis fully: http://leonardo.phys.washington.edu/~ravel/misc/Tk-800.024.tar.gz If this conflicts with other programs or projects for any individual user, I apologize and, as I said, will fix the problems with Athena and Artemis ASAP. So, to summarize, Athena and Artemis currently do not work with the most recent version of perl/Tk. This is my fault not the fault of perl/Tk. The solution is either to wait for me to fix the problems or to install an older version of perl/Tk. Regards, Bruce -- 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 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

Re: [Ifeffit] ... and lots of new stuff in Athena and Artemis, too!
by SCalvin100＠aol.com 31 Oct '03

31 Oct '03

Yippee!!! Many thanks to Matt and Bruce for the new releases! --Scott Calvin Sarah Lawrence College

1 0

Ifeffit 1.2.4 Debian Linux packages
by Carlo U. Segre 31 Oct '03

31 Oct '03

Hello all: The latest version of Ifeffit is now available as a Debian packge. There is also an update package of horae 024 with a recompiled wrapper for Ifeffit 1.2.4. Carlo -- Carlo U. Segre -- Professor of Physics Associate Dean for Special Projects, Graduate College Illinois Institute of Technology Voice: 312.567.3498 Fax: 312.567.3494 segre(a)agni.phys.iit.edu http://www.iit.edu/~segre

1 0

RE: [Ifeffit] relation between DW factors in uranyls
by Kelly, Shelly D. 30 Oct '03

30 Oct '03

> This is in response to Shelly's post for MS references. For > linear multiple scattering, I like to use this paper: > > Hudson, EA; et. al. (1996) Phys Rev B 54(1) 156-165 > > It uses the uranyl ion specifically in this example and has a > short section in the introduction about the relations between > single and multiple scattering D-W parameters. > > Hope this is vaguely what you were looking for.... > > > Sam Hi Sam and John, I just wanted to say thank-you to the ifeffit mailing list. Sam, the reference was indeed helpful. And thanks to John for his derivation. This is what I learned about relating sigma2 values for the uranyl. For a linear uranyl moiety (UO2), where the atoms are all in a line O1-U-O2, the sigma2 values can be approximated like this: U-O1-U: sigma2-s.s. (single scattering sigma2) U-O1-U-O1-U: sigma2 = 4 * sigma2-s.s. U-O1-U-O2-U: sigma2 ~ 2 * sigma2-s.s. U-O1-O2-U: sigma2 ~ 2 * sigma2-s.s. This information has been published in Hudson et al. PRB V54 n1, pg 156 (1996). Shelly

1 0

Coordination number
by Dave Baker 30 Oct '03

30 Oct '03

I'm interested in extracting coordination numbers from the S02 parameter in Artemis, and, as the beginner, I'm lost as to what to do. I am using two FEFF paths to describe the first shell, and have given each path its own S02 parameter. Any suggestions? -Dave ------------------------------------------------------------------------ -- North Carolina State University Dept. of Physics 2700 Stinton Dr. Box 8202 Raleigh, NC 27695 Ph: 919-515-5017 Fax:919-515-7331 email: dabaker(a)unity.ncsu.edu ------------------------------------------------------------------------ --

2 1

RE: reference
by Webb, Samuel M. 30 Oct '03

30 Oct '03

This is in response to Shelly's post for MS references. For linear multiple scattering, I like to use this paper: Hudson, EA; et. al. (1996) Phys Rev B 54(1) 156-165 It uses the uranyl ion specifically in this example and has a short section in the introduction about the relations between single and multiple scattering D-W parameters. Hope this is vaguely what you were looking for.... Sam

1 0

reference
by Kelly, Shelly D. 30 Oct '03

30 Oct '03

Hello All, Does anyone know of a reference that explicitly shows that sigma2 for a linear backscattering path is approximately twice a single scattering path? My example is for a uranyl moiety (UO2). The structure of this moiety is symmetric and linear with a dumbbell shape with O1=U=O2 all in a row. I need a reference that does the calculation to show that the 4 leg path (U-O1-U-O2-U) has a sigma2 value that is twice the single scattering path (U-O1-U) from one of the oxygen atoms. Thanks for your help Shelly

2 1

one more thing...
by Bruce Ravel 30 Oct '03

30 Oct '03

Hi folks, The one very obvious change in how Athena behaves is how it plots normalized mu(E) spectra. If you dislike this new feature and would prefer to always see the normalized data as they were previously plotted, it is easy to disable by going to the preferences dialog, selecting the variable Bkg -> flatten, and setting it to false. Below is the description of this new feature from the NEWS page on my web site. B 1. FLATTENED DATA. I added a new features for plotting normalized spectra which I call flattening. In short, this works by subtracting the linear and quadratic portions of the polynomial regressed to the post-edge region of the data from the normalized data after the edge. This has the effect of shifting the oscillatory portion of the spectrum such that it oscillates about 1. That is, the flattened spectrum is presented with its baseline at 0 and it's post-edge at 1. This is purely a plotting device and has no impact on the calculation of chi(k). It can be turned on and off for a data set using a toggle button in the background removal section of the main window. Whether flattening is on or off by default is configurable in preferences dialog. Flattening is sensitive to how the post-edge line is chosen, particularly in XANES data which have a limited energy range. The idea for this feature was blatantly swiped from SixPack. -- 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 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

... and lots of new stuff in Athena and Artemis, too!
by Bruce Ravel 30 Oct '03

30 Oct '03

Hey! As a followup to Matt's announcement, I wanted to mention that the brand new Ifeffit release contains the latest versions of Athena and Artemis, both of which contain lots of new features. Here is an explanation of the new goodies: http://leonardo.phys.washington.edu/~ravel/software/exafs/NEWS Windows users can just run the updater or the installer and you will get the latest codes. Linux and other unix users can grabthe latest tarball from my web site. 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 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

Ifeffit 1.2.4
by Matt Newville 30 Oct '03

30 Oct '03

Hi Everyone, I've uploaded Ifeffit version 1.2.4 to the Ifefft web pages, and to Sourceforge.net. This version has a few long-promised fixes, and improvements. == Bug fixes: the bug of 'n_varys' not being correct is now fixed. Several of the program limits were increased (lengths of strings, number of Feff paths, etc) and better reporting of these limits. Long label lines should be handled better now in read_data() and athena. Feff6 now works on Windows. A few more bug fixes are mentioned in the ChangeLog file. == Feff6, Feffit, Autobk: these 'input-file' based programs can now all take the name of an input file with a name other than 'feff.inp', 'feffit.inp', and 'autobk.inp' as a command line argument (on Unix, Mac OS X, and Windows). On Windows, the batch files now pop up a file browser so that you can browse for a feff.inp file, and Feff6 will be run with that input file, in that directory (and similarly for feffit and autobk). == Diffkk: This has (finally!) been updated to allow f1->f2 transforms. This still needs testing, but works for a test case. Sorry for taking so long, Michel! == Mac OS X: There is now a disk image with a crude installer that will install the latest versions of ifeffit, feff6, feffit, autobk, athena, artemis, tkatoms, and a few other programs for Mac OS X (but not gifeffit or sixpack -- yet). Fink is NOT required to run these programs, but there are a few important caveats: This installer will work only with OS 10.2, requires X11 to run (the X11 beta3 from Apple works OK for me), and assumes the Apple-supplied perl 5.6 (it won't work with perl 5.8). I realize that OS 10.3 just came out (which ships with X11 and perl 5.8) -- I don't yet have it installed yet, but expect to support it relatively soon. If anyone gives this installer a try, please let me know how it works. If you want to get these programs to work with 10.3 and/or 10.2 with perl5.8, let us know: we can make it go much smoother now than in the past. == Windows: There is a full installer (30Mb) or an update (8Mb) from Ifeffit 1.2.3 that can be automatically installed. These include latest versions of athena and artemis (and ifeffit, feff6l, feffit, and autobk) are included. If you install from the update, make sure you double-click on make_bats.exe afterwards!!! == Other Announcements: As some of you may know, , my wife Julie had a baby two weeks ago. We're very happy, and everybody's doing well. But as a result I'm spending a lot less "extra time" at work these days, and it will probably stay that way. This will cut down on my ability to support Ifeffit, especially to build and fix the Windows versions which I have to do at work (though I do count it as real work, it's not always the highest priority). On the other hand, I can work on the source and Mac versions from home between diaper changes and other real work. The Windows installer and updater for 1.2.4 are much less tested than they should be: please report problems and be patient, as it will take time to fix any problems!!! I'd also like to repeat Bruce's plea for help earlier this week. It's likely that the level of support and ability to respond to problems will drop, and we really do need help. If you're using these programs as an integral part of your research, please consider what they're worth to you and what you would have to do if these programs didn't exist. If you can help, either with making releases for Windows or Macs, with program documentation or web pages, or in any other way, please contact us. As always, let us know if there are any questions or problems, --Matt

1 0

Re: [Ifeffit] This might be a silly question....
by k-kupiecki＠northwestern.edu 29 Oct '03

29 Oct '03

Thank you so much! That's very helpful! BTW... so far I've found the Artemis manual very helpful in getting started (especially the explanation of IFEFFIT)... thanks!!!! -Kristine On Wed, 29 Oct 2003 18:44:03 -0400 Bruce Ravel wrote: > On Wednesday 29 October 2003 04:55 pm, k-kupiecki(a)northwestern.edu wrote: > > Hi, > > > > I'm also a novice with the Athena/Artemis Ifeffit software. > > > > I've read through the online documentation for both and I'm now > trying to > > go through the Cu example in Chapter 2 of the Artemis manual. > > > > Can someone please tell me where I input the following information into > > Artemis (I've tried in the fields on the Path parameter screen, but it > > gives me an error when I go to fit)? (from page 19 of the manual): > > > > We will then define extremely simple math exressions for the > > corresponding path parameters of the first shell. > > > > s02 0:1 amp > > e0 0:1 e0 > > delr 0:1 delr > > sigma2 0:1 ss > > > Kristine, > > Hmmm.... this is a little embarassing. I have to apologize for the > state of the Artemis document. It suffers rather alot right now from > bad stream-of-consciousness writing. > > My goal in the chapter on the copper fit was to walk through the steps > of an increasingly complicated fit. Start with single shell, then two > shells, then three. Then do four shells + all the necessary multiple > scattering. Then introduce the concept of multiple-data-set fitting. > As I was writing it a few months ago, I didn't want to interupt my > (ever-so-brilliant) train of thought to make a bunch of screenshots > demonstrating what I was talking about, so I tried to develop a > text-only shorthand to explain what I meant. > > The "0:1" was my shorthand for indicating that this was path number 1 > (i.e. the first one) from data set number 0 (also intended to be the > first one). Obviously confusing. Sigh. > > In an attempt to clarify the situation, I made four screenshots from > the single shell project file that gets distributed with the program. > > Point your browser at this directory: > http://feff.phys.washington.edu/~ravel/misc/ > > You will find the four screenshots I just made: > > artemis_cu1.png the view of the data > artemis_cu2.png the view of the guess and set parameters > artemis_cu3.png the view of the description of the feff > calculation > artemis_cu4.png the view of the first path > > I think this should make it clear what goes where. Note that these > shots were taken of Artemis getting ready to fit just the fist shell > of the copper data. Thus paths contained in feff0002.dat through > feff0005.dat are not included in the fit. The ones not included in > the fit are written in brown text in the list on the right side of the > window. In subsequent fits, when those included in the fit, you can > edit their parameters by clicking on them. That will display > something that looks very much like the artemis_cu4.png image, but > with the information relevant to that path. > > Taking language from those images on my web site, the bit in the > document that you are asking about would have been much better > written as: > > FEFF0: feff0001.dat > s02 = amp > e0 = e0 > delr = delr > sigma2 = ss > > That would have, perhaps, been clearer. Certainly more consistent > with the appearence of the program. And still more concise than an > entire screenshot at that point in the doc. > > > Hope that helps, sorry for the confusion, and don't hesitate to keep > asking questions on the mailing list. And, contrary to the sentiment > in your subject line, there are no silly questions, just silly > programmers. > > 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 > 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

This might be a silly question....
by k-kupiecki＠northwestern.edu 29 Oct '03

29 Oct '03

Hi, I'm also a novice with the Athena/Artemis Ifeffit software. I've read through the online documentation for both and I'm now trying to go through the Cu example in Chapter 2 of the Artemis manual. Can someone please tell me where I input the following information into Artemis (I've tried in the fields on the Path parameter screen, but it gives me an error when I go to fit)? (from page 19 of the manual): We will then define extremely simple math exressions for the corresponding path parameters of the first shell. s02 0:1 amp e0 0:1 e0 delr 0:1 delr sigma2 0:1 ss Thanks, Kristine Witkowski Northwestern University

2 1

Re: [Ifeffit] Novice seeks advice
by SCalvin100＠aol.com 29 Oct '03

29 Oct '03

Oops. Scott meant to send this to the list, of course. --Scott In a message dated 10/29/03 10:21:28 AM Eastern Standard Time, ravel(a)phys.washington.edu writes: On Wednesday 29 October 2003 10:12 am, SCalvin100(a)aol.com wrote: > I would like to add to Bruces' post one other quick "reality check." > Look at Athena and see where E0 was picked there. It's possible that > for some reason you have an E0 in Athena way above the actual rising > portion of the energy spectrum, in which case Artemis is actually > doing the right thing by moving it back to the correct energy > origin. You really don't want to depend on Artemis to fix a poor E0 > choice in Athena, though, so if that is the case then you should > choose a more reasonable E0 in Athena and feed the resulting chi > file back in to Artemis. Scott, Did you send this to the list or just to me? The headers seem to suggest the latter -- in which case it is weird the Scott is referring to Bruce in the third person ;-) B

1 0

Another stab at the data
by Dave Baker 29 Oct '03

29 Oct '03

Bruce, Thanks for the advice. I've run through the list, redone processing and started the analysis with said list in mind, and the most plausible solution is that I shouldn't be using a crystalline material for the FEFF calculation. One interesting thing (at least to my untrained eye) to note is that the shapes of the first path and the peak that I'm trying to fit (this is before any attempt at fitting) are nearly identical, but the FEFF path is shifted about 2~3 angstroms higher than the peak. The only improvement that I could muster is that the e0 variable has gone down from 80 to 50. -Dave Baker ------------------------------------------------------------------------ -- North Carolina State University Dept. of Physics 2700 Stinton Dr. Box 8202 Raleigh, NC 27695 Ph: 919-515-5017 Fax:919-515-7331 email: dabaker(a)unity.ncsu.edu ------------------------------------------------------------------------ -- On Tuesday 28 October 2003 11:00 pm, Dave Baker wrote: > I am a novice student in the EXAFS world, and I am starting to work my > way through Athena and Artemis. Currently, I'm working with amorphous > GeTe and the Atoms/FEFF runs are based on the crystalline data. I have > performed a fit in Artemis of the Ge K edge, and my e0 for three > contributing paths is exceedingly large, (-70 ~ -80 volts). Two > questions: 1.) What symptom is this an indicator of, and 2.) What are > some possible solutions? Hi Dave, Glad to see you were able to get started using the software after that very short conversation we had at X11A a couple weeks ago. I have seen this effect on occassion when, for some silly reason, I have made the Feff calculation using the wrong edge (i.e. K instead of L3 or vice versa). Another possible indication of an absurdly large e0 shift might be if the distances between atoms in the feff calculation are very different than the distances between atoms in the real material. That is, if the material used in the feff calculation is a very poor approximation of the real material. Another possibility that occurs to me from the way you phrased your question is that you are allowing e0 to float independently for each path. That's almost certainly not physically supportable. At least as a first stab at theproblem, you should use the same e0 for each path. One thing you might try is to not allow everything to vary in the fit. Try fixing e0 to be something sensible -- like 0.0. Vary all the rest of the parameters to get a sense of how the parameters behave and how well they describe the data. This sort of manual examination of the parameter space might help give you a clue of why you are getting such an odd result. HTH, B

2 1

Novice seeks advice
by Dave Baker 29 Oct '03

29 Oct '03

I am a novice student in the EXAFS world, and I am starting to work my way through Athena and Artemis. Currently, I'm working with amorphous GeTe and the Atoms/FEFF runs are based on the crystalline data. I have performed a fit in Artemis of the Ge K edge, and my e0 for three contributing paths is exceedingly large, (-70 ~ -80 volts). Two questions: 1.) What symptom is this an indicator of, and 2.) What are some possible solutions? -Dave Baker

2 1