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

[Ifeffit] What differences are there between Athena and AUTOBK backgrounds?
by Scott Calvin 05 Aug '03

05 Aug '03

Hi all, Today I tried making Athena perform exactly the same background subtraction as AUTOBK 2.61 (all parameters the same, working on the same data, using the same theoretical standard, etc.). I then let FEFFIT take the chi(k) data for each and perform a Fourier transform. When I did this, AUTOBK would always generate a very low r (< 0.2 angstroms) peak of significant amplitude that was nearly absent in the Athena background. There were other subtle differences as well, but that one stood out. Of course, a peak that low in r should have no effect on the fit, but it made me curious--were substantive changes in the background subtraction algorithm made when AUTOBK was incorporated into IFEFFIT? --Scott Calvin Sarah Lawrence College

2 1

Phase Correction using IFEFFIT
by Ian Drake 04 Aug '03

04 Aug '03

Hi Matt and XAFS community I am interested in the details of how IFEFFIT handles phase correction. I asked a similar question in April, but have only now begun to address its use in my work. In your response to my April question, you described how to do a central atom phase shift (without a FEFF path) and a combined central-atom and scattering-atom phase-shift (using a particular FEFF path). In the case of the combined central-atom/scattering-atom phase-shift, is the Chi(k) multiplied by exp(i*Phase(Central-atom)) and exp(i*Phase(scatterer)) before FFT? If this is true, how can I correct for the scattering atom phase shift "only" as has been described by Lee and Beni (Phys. Rev.B, 15, 1977, pg 2862) and Vaarkamp (Cat. Today, 39, (1998), pg 271). At this time, my particular interest is in attempting to identify clearly next-nearest neighbors in the RSF before attempting to do fitting. It is my hope that a positive identification of the next-nearest neighbor by means of IM analysis/ and or k-weighting will add validity to my final fitting results. Thank you for your help, Ian Drake

3 3

phase corrections
by Grant Bunker 04 Aug '03

04 Aug '03

Following Bruce's commentary - it is indeed true that if you have sufficiently heavy backscatterer (think Au or Pt, or even Ag/Pd) the modulations in the backscattering phase and amplitude will mess up your transform peaks, as Ian suggests. Basically you get "sideband modulation" of your transform peaks. To get rid of them you need to not only do a phase correction, but also an amplitude correction (divide out the wiggles). I thinkd oing one without the other doesn't actually make a lot of sense since they are related (because the real and imaginary parts are related by kramers-kronig relation the phase and amplitude are connected). Obviously this is not very difficult to do - I put it into the VAX version of the UW programs in 1985 but that was dropped from the simplified UW/NRL version later. Short term, it would be very easy to do by exporting from ifeffit and using any data massaging program that knows complex numbers. thanks - grant On Mon, 4 Aug 2003 ifeffit-request(a)millenia.cars.aps.anl.gov wrote: > Send Ifeffit mailing list submissions to > ifeffit(a)millenia.cars.aps.anl.gov > > To subscribe or unsubscribe via the World Wide Web, visit > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit > or, via email, send a message with subject or body 'help' to > ifeffit-request(a)millenia.cars.aps.anl.gov > > You can reach the person managing the list at > ifeffit-owner(a)millenia.cars.aps.anl.gov > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Ifeffit digest..." > > > Today's Topics: > > 1. Re: Phase Correction using IFEFFIT (Bruce Ravel) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 4 Aug 2003 09:56:14 -0400 > From: Bruce Ravel <ravel(a)phys.washington.edu> > Subject: Re: [Ifeffit] Phase Correction using IFEFFIT > To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> > Message-ID: <200308040956.14156.ravel(a)phys.washington.edu> > Content-Type: text/plain; charset="iso-8859-1" > > On Saturday 02 August 2003 04:32 pm, Ian Drake wrote: > > At this time, my particular interest is in attempting to identify clearly > > next-nearest neighbors in the RSF before attempting to do fitting. It is > > my hope that a positive identification of the next-nearest neighbor by > > means of IM analysis/ and or k-weighting will add validity to my final > > fitting results. > > Well... I don't know what an "RSF" is or what "IM analysis" is, but I > think I can suggest a way to isolate the scattering atom phase shift > in a way that requires only a small amount of work and can be used > with either ifeffit or artemis as they exist today. > > The second column of the feffNNNN.dat file is the central atom phase > shift. The third and fourth are the scattering atoms magnitude and > phase. Copy your feffNNNN.dat file of interest to another name, say > feff_next_neighbor_no_caps.dat. Edit the new file, replacing the > second column with zeros. Read this into ifeffit or artemis and use > it as the path for phase correction. In artemis, this is done in the > data view using the pull down menu labeled "Path to use for phase > correction". Obviously, you should choose your edited path. In > artemis, you may need to set the amplitude for this path to zero so > that it doesn't really get used in a fit. > > That said, it's not clear to me how phase correction will really do > what you seem to want. It seem unlikely to me that the signal beyond > the first shell will magically become clear simply by doing phase > correction. If the signal is a superposition of various scatterers > before phase correction, it will still be a superposition of various > scatterers afterwards. Removing the scattering atom phase shift does > not affect the terms in the EXAFS equation that make peaks in a > Fourier transform broad, nor does it magically separate signals which > contribute similar Fourier components. Understanding how various > paths contribute similar Fourier components is the reason that Artemis > allows you to easily overplot data with the Fourier transforms of > various paths. Then you can inspect the contributions from each path > before setting up your fitting model. But if playing tricks with > phase correction helps *you*, then great! That is why Artemis allows > you to do this. > > 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/ > > > ------------------------------ > > _______________________________________________ > Ifeffit mailing list > Ifeffit(a)millenia.cars.aps.anl.gov > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit > > > End of Ifeffit Digest, Vol 6, Issue 2 > ************************************* >

1 0

IFEFFIT and PGPLOT on linux
by Kelly, Shelly D. 31 Jul '03

31 Jul '03

Hi Matt, Yesterday, I installed PGPLOT and IFEFFIT from the tar ball on your web page on a clean (no previous installation of any of the XAFS software) Red Hat linux box. I used your script to install PGPLOT and was pleased that it worked just as expected. I had one small issue. When I tried to 'make' IFEFFIT, I got an error message saying that there was nothing for make to do. I realized that the configure step for IFEFFIT was writing out a Makefile but that the current directory had both a makefile and a Makefile. One makefile with a lowercase m and the other with a capital case M. So I removed the makefile and then IFEFFIT installed properly. Then I used the Horae script to install Ravel-ware, without any trouble. Overall the installation went really well...Good going. Shelly Kelly

2 1

Re: [Ifeffit] Self-Absorption Corrections
by Francois Farges 30 Jul '03

30 Jul '03

hi all, Here is a gif of Nb2O5 at Nb K-edge and TiO2 at Ti K-edge. Both collected withb same geometry and detectors and same BL (SSRL 11-2). Fluo was collected at 90° to the beam (while transmission data was collected on the exact same sample). both spectra are normalized. Is there not any remaining self-absorption in the fluo spectrum for Nb2O5, despite data collected at 90° ?? (while TiO2 sounds fine, as Troeger predicted ?). thank for your expertise ! -- Francois FARGES Laboratoire des Géomatériaux Université de Marne la Vallée 5 Bd Descartes-Champs S/Marne 77454 Marne la Vallée cedex 2 TEL: 01 49 32 90 57 from outside France: +33 1 49 32 90 57 FAX: 01 49 32 91 37 from outside France: +33 1 49 32 91 37

3 3

Self-Absorption Corrections, Wavelet Transforms
by Matt Newville 29 Jul '03

29 Jul '03

Hi Everyone, There were several conversations around the XAFS 12 conference on self-absorption corrections for EXAFS. It would be nice to add such corrections (and others, like for deadtime corrections) to Ifeffit. Corwin Booth (at LBL lab, Berkeley) presented a very nice procedure for making these corrections for EXAFS at the conference, improving the work of L. Troger, et al from the mid 1990's. Corwin just sent me the Fortran source code for his program so that it might be included in Ifeffit, and is letting me pass it on. I put the source code (with a Unix Makefile and some instructions) at http://cars.uchicago.edu/ifeffit/contrib/sabcor.tar.gz with a README at http://cars.uchicago.edu/ifeffit/contrib/README.SABCOR I'd be interested in hearing others opinions on this topic, and whether this should be included in Ifeffit. A self-absorption correction that worked better for XANES would be useful too. If you've thought about self-absorption corrections or have some data affected by self-absorption corrections, please check out Corwin's code and let us know what you think of this. Speaking of things to check out, Francois Farges recently created a web page discussingCauchy Wavelet transforms of EXAFS data at http://www.univ-mlv.fr/~farges/wav/ The site includes MATLAB code and references. These transforms are very interesting and I recommend checking out this stuff. There was also a recent email from Harald Funke sent to many people about a different code for Wavelet transforms. If I get permission from Harald, I'll post these codes too. Related to all this, I added a 'Contributions' page to the Ifeffit Web Pages, for "contributions and/or things to consider for XAFS analysis and Ifeffit". Currently, this has links to Corwin's self-absorption correction code, and Francois' Wavelet web page. If anyone has any suggestions for other things to add, let me know. Thanks, --Matt

4 7

compiling ifeffit with pgplot on unix
by Bruce Ravel 28 Jul '03

28 Jul '03

Hi folks, The other day a user of ifeffit on linux asked a question that I have seen several times now and so thought it would be prudent to explain the situation on the mailing list. It is fairly easy to get confused about the sequence of how things must be installed when installing ifeffit on a unix machine for the first time. The explanation below only applies to unix users (including, but not limited to, linux users) and does not apply to MS Windows users. (The situation is different for OSX users, as well, as indicated by last week's flurry of email.) When installing this stuff for the first time on a unix machine, the correct sequence is: 1. install pgplot 2. install ifeffit 3. install Atoms, Athena, Artemis, GIfeffit, and/or SixPack The easiest way to install pgplot is to use the PGPLOT_install script that Matt supplies with the ifeffit source code tarball. Just unpack the ifeffit tarball, cd into the build directory, and do ~> sh PGPLOT_install before doing the "./configure; make; make install" incantation. This script will try very hard to fetch the pgplot tarball from the web. Once downloaded, it will unpack it, compile it, and install it automagically. Once this is done, go ahead and install ifeffit. In the future, when you upgrade ifeffit, it should not be necessary to reinstall pgplot (unless, of course, you have done something to change things on your computer in the interim). When pgplot is installed correctly, the "./configure" step on the installation will end with a message that says something like this: === === ifeffit 1.2.1 Configuration Results: === linking to PGPLOT with: -L/usr/local/pgplot -lpgplot -lpng -lz -L/usr/X11R6/lib -lX11 === === ready for next step: type 'make' then 'make install' Note that in the second line it has correctly identified the location of the pgplot installation and its library. It has also identified several other libraries necessary for drawing stuff to the screen under X. If pgplot is not found on your computer, "./configure" will end with a message like this: === === ifeffit 1.2.1 Configuration Results: === linking to PGPLOT with: /home/bruce/codes/ifeffit-1.2.1/src/pgstub/libnopgplot.a === === ready for next step: type 'make' then 'make install' Note that, in this case, the ifeffit library will be linked to something called `libnopgplot.a'. This is a stub that allows ifeffit to continue working as a data analysis tool even in the absence of pgplot. That is, you will still be able to import and export data, do background removals, and so, but you will not be able to plot stuff. This seems like a good decision that Matt made -- that is, it seems appropriate that ifeffit should still be functional even in the absence of pgplot. Athena and Artemis, however, will not function in the absence of pgplot. If pgplot was not found and you did not notice that ifeffit was linking to libnopgplot.a, then you will not realize that there is a problem until *after* you have installed one of the GUIs and tried to run them. At that point, the program will fail with a variety of cryptic error messages like so: Can't load '/usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi/auto/Ifeffit/Ifeffit.so' for module Ifeffit: /usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi/auto/Ifeffit/Ifeffit.so: undefined symbol: pgqndt_ at /usr/lib/perl5/5.8.0/i386-linux-thread-multi/DynaLoader.pm line 229. at /usr/bin/athena line 55 Compilation failed in require at /usr/bin/athena line 55. BEGIN failed--compilation aborted at /usr/bin/athena line 59. The common feature of these error messages is the failure to find symbols that begin with the letters "pg", as in the 5th line of the error message I replicated above. If you run into this situation, the thing to do is to go back to the beginning. Install pgplot and make sure that it got installed correctly. Then build ifeffit *from scratch*. It is not enough just to install pgplot. You MUST rebuild ifeffit after installing pgplot. Once ifeffit it rebuilt, go ahead and install one or more of the GUIs. Again, it is insufficient just to rebuild ifeffit and not to rebuild the GUIs. If you are careful to follow the procedure of installing pgplot, then ifeffit, then GUIs, you should find that everything will work as advertised. And finally, if you have any questions or suggestions regarding installation, don't hesitate to ask them here on the mailing list. If you ask questions on the mailing list, then everyone can benefit by the answer. HTH, Bruce P.S. Carlo Segre is kindly packaging ifeffit into Debian package files. This is a *great* way for people to help out with the ifeffit project without needing to write code. If any of you are users of Red Hat, SuSE, Mandrake, Connectiva, or any other linux distribution (such as the über-cool Gentoo) and wanted to contribute packages guaranteed to work on those systems, that would be great! Just let me or Matt know and we will help you get started. The same would be true for anyone using any of the commercials unixes (Solaris, SGI, etc.) which have their own packaging systems. -- 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/

2 1

status: potential surface scanning
by Norbert Weiher 28 Jul '03

28 Jul '03

Dear friends, just a small update for the potential scanning stuff: It works fine for one shell. Really straightforward thanks to the ifeffit() calls. Some corrections and I can provide you the program ;) But there is one little thing I am thinking about, which is fairly general, I think: If I just have one shell, everything is fine. The program iterates through all possible variable combinations. But... if I have two shells I just manage to let him vary the shells independently. I know there must be a way to tell the algorithm that if there are 2 shells it needs to vary each of the eight parameters against each other and not two times the four from each shell. I know there are some good PERL hackers around. So do you have an idea how to do this? Recursion shurely helps - I experimented a lot with it but it still does not do what I need it to do. Just for your convenience: I stored the parameters in an array of hashes, so that I can have something like $shell[0]{E0} = $something $shell[0]{dr} = $somethingelse... So the loop for one shell is just looping through the hashkeys of element 0 of this array... Thanks for your suggestions. Cheers, Norbert -- Dr. rer. nat. Norbert Weiher (norbertweiher(a)yahoo.de) Laboratory for Technical Chemistry - ETH Hönggerberg HCI E 117 - 8093 Zürich - Phone: +41 1 63 3 48 32

3 2

Re: [Ifeffit] Getting started in OS X...
by dean_hesterberg＠ncsu.edu 23 Jul '03

23 Jul '03

Hello, I just wanted to echo Scott's sentiments. If someone has the time and energy to create for the Mac an executable version of the Artemis/Athena/IFEFFIT programs (and keep them updated), I'd also be very grateful. Being unfamiliar with Unix, I experienced the same problems as Scott in trying to get the programs up and running on my Mac. So, I bought a PC laptop specifically for doing data analysis with these very nice programs. Dean Hesterberg ************************************************** Hi again, > >OK, I've reached a conclusion. For someone not currently using the >open source UNIX tools and the like on the Mac, the project of >getting Artemis, Athena, and IFEFFIT up and running is immense, >because it means getting a large amount of additional infrastructure >in place. Consider: Fink is required. Once Fink is installed, the >x-free87-base package is required. That package, in turn, requires >downloading 12 separate files (one-by-one, as far as I can tell). And >that's just one part of a part of what is required (g77, pgplot, >perl...). My estimate is that, with my lack of experience, it would >take me a week full-time to get all this stuff working right. That's >not a problem, of course, for people who already do development on >the Mac and have most of this stuff set up anyway, or at least find >uses for it other than getting the EXAFS packages to run. But it's >too much of a time investment for me, and I suspect that the same is >true for a lot of others in the field. > >So I'm going to scrounge a pc from somewhere, transfer my data files >>from the Mac, and download the executable there. > >If at some point an executable appears for the Mac, I will thank the >authors profusely...I realize that no one is getting paid to work on >this! > >--Scott Calvin >Naval Research Lab >Code 6344 >_______________________________________________ >Ifeffit mailing list >Ifeffit(a)millenia.cars.aps.anl.gov >http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ------------------------------------------------------ DEAN HESTERBERG Associate Professor Department of Soil Science College of Agriculture and Life Sciences North Carolina State University Box 7619 (3235 Williams Hall) Raleigh, NC 27695-7619 E-mail: dean_hesterberg(a)ncsu.edu FAX: (919) 515-2167 tel: (919) 513-3035

7 14

Problems with Reference Paths
by Norbert Weiher 14 Jul '03

14 Jul '03

Cheers friends, now a short question concerning reference paths calculated by FEFF (I use FEFF8.1): I have a series of spectra of catalysts containing V,P and O. Data was taken at the V K edge. I know from literature, that the first coordination shell should contain some 5-6 O atoms. The problem (and this is really the first time I ever had problems with this) is that ifeffit (V 1.2.1) is not capable of fitting even the first shell. As it works fine for model systems, e.g. Au foil, I wonder if I have problems with the calculation of the references... As first try, I re-calculated the feffxxxx.dat files with the SCF flag turned on (which I never used before for the EXAFS) - the problem persisted. Has one of you experienced similar problems and knows how to come over? Norbert -- Dr. rer. nat. Norbert Weiher (norbertweiher(a)yahoo.de) Laboratory for Technical Chemistry - ETH Hönggerberg HCI E 117 - 8093 Zürich - Phone: +41 1 63 3 48 32

3 2