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

Bkg removal problem
by Alfred Hummer 14 Sep '10

14 Sep '10

Hello, Currently I am working on a Ga XAS spectrum. I'm trying to extract the EXAFS data but I'm not sure if the background function is right, or how to set the right values. I played around with different setting for kmin and Rbkg. So far I know that Rbkg defines the cutoff for the bkg frequencies in the FT optimization and setting it to high leads to loss in information, as more high frequencies are included in the bkg. I also read that it is advisable to exclude strong white lines from the spline range (increasing kmin?). Would this a strategy in my case? I attached the project file. Maybe someone has a clue. The more I tried the bigger my uncertainty. best regards Alfred

3 2

changing valence state
by JeongEunSuk 11 Sep '10

11 Sep '10

Hi all I want to change valence state for XANES calculation in FEFF , from Co2+ to Co3+ Is it possible in FEFF?

1 0

(Reposted) K-space vs R-spce fitting
by Lezama Pacheco, Juan Salvador 11 Sep '10

11 Sep '10

Hi Matt: Thanks so much for your answer. Let's see if I can make it more clear this time. Should dk have an effect on the actual fit, if it is performed in k-space? As I said, I found out that by changing dk (in k space) (keeping (kmin,kmax) constant) values change around 15%, so that tells me the actual fitting range is (kmin-dk/2,kmax+dk/2) is that so? Juan

2 1

Hi ! I am PHD student in Taiwan.
by clo 11 Sep '10

11 Sep '10

Dear joly: I am PHD student at deparment of physics of National Sun Yat-Sen University in Taiwan. I have some problems about DOS simulation by using Feff. 1.I want to compare the simulation result with the LDA+"U" calculation, Could you give some suggestions for presentation the "U" potential part by Feff? thank you for your help Chun Pin Wu

1 0

K-space vs R-spce fitting
by Lezama Pacheco, Juan Salvador 10 Sep '10

10 Sep '10

Dear All: First of all, nicely done you guys! This is indeed a set great of tools the community really appreciates it and I have been putting into heavy use. I have and issue though that has been making me scratch my head from time to time. I've been trying to figure out the feature of k-space and R-space fitting "relationship" for Ifeffit. I usually work out my fits in k-space although I have to admit that R-space fitting in combined approach will definitely be valuable. Now, as I am trying to work some data for Uranium (namely uranyl acetate standard) I am a little bit confused on the real independency of k-space and R-space fitting. This comes from the fact that even if I specified the fitting range in k (fitting in k space), the dk value seems to have an effect in the current fit values (changing as large as 15%). I understand the issue if the actual fitting range is then (kmin-dk/2,kmax+dk/2) but is this the case? Then what is the real relationship-independency between k-space and R-space fitting? For q-space that is of course another story. For the record I am using ixafs 1.2 on Mac OS X 10.5.8. Perl distribution seems to crash "unexpectedly" every time I tried the new 3.0.2, but that is another story in another thread... Thanks so much again Juan S Lezama

2 2

Bkg fit in Artemis
by hemantha aranwela 09 Sep '10

09 Sep '10

Hi All, I like to understand the way how the background fitting is done in Artemis. How the background functions are obtained (or defined)? How do we use the bkg fit in analyzing the data? I have a good idea about how Athena does the bkg reduction. But have no clue in case of Artemis. I would appreciate if somebody can direct me to an article written on this or can give a brief introduction about it. Thanks. Aranwela

3 2

iXAFS & python scripting
by Brugger, Joel (SAM) 09 Sep '10

09 Sep '10

Hi Matt, Actually we did run . /Applications/iXAFS.app/Contents/Resources/local/bin/iff_init.sh (I was just wondering if it was still required). Hopefully the following output will be clearer than my prose. Line 5 in "iff = Ifeffit.Ifeffit()". The problem seems to be access to "/Users/Newville/Codes/ifeffit/local/share/ifeffit" while running Ifeffit.Ifeffit(). Thanks! Joel + Barb Last login: Fri Sep 10 08:14:26 on ttys004 JoelMacBook-Silver:~ joelb$ echo $PATH /opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin JoelMacBook-Silver:~ joelb$ . /Applications/iXAFS.app/Contents/Resources/local/bin/iff_init.sh JoelMacBook-Silver:~ joelb$ echo $PATH /Applications/iXAFS.app/Contents/Resources/local/bin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin JoelMacBook-Silver:~ joelb$ cd /Users/joelb/data_ifeffit/APS-April08-Pascal/python JoelMacBook-Silver:python joelb$ python aps-12p9mLiCl.py failed to load ifeffit library libifeffit.dylib /Users/Newville/Codes/ifeffit/local/share/ifeffit:/Applications/iXAFS.app/Contents/Resources/local/bin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin Traceback (most recent call last): File "aps-12p9mLiCl.py", line 5, in <module> iff = Ifeffit.Ifeffit() File "/Library/Python/2.6/site-packages/Ifeffit/Ifeffit.py", line 79, in __init__ raise ImportError('Cannot load ifeffit library') ImportError: Cannot load ifeffit library JoelMacBook-Silver:python joelb$ echo $PATH /Applications/iXAFS.app/Contents/Resources/local/bin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin ________________________________________ ------------------------------ Message: 4 Date: Thu, 9 Sep 2010 10:12:09 -0500 From: Matt Newville <newville(a)cars.uchicago.edu> To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> Cc: "barbara.etschmann(a)adelaide.edu.au" <barbara.etschmann(a)adelaide.edu.au> Subject: Re: [Ifeffit] IXAFS3.0.2 and python scripting Message-ID: <AANLkTikYLAwnRcwZWNK+yBL0Byp7NgxNJ=Zrn_v3V56M(a)mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Hi Joel and Barb, Ah, sorry. I believe you need to do . /Applications/iXAFS.app/Contents/Resources/local/bin/iff_init.sh in the shell prior to running python, or add this to your bash .profile file. It would be better if the setup procedure detected that it was on a Mac and set the environment up automatically. I'll work on that (and on the build for10.5). --Matt > Works great on 10.6 for us too. Thanks a million times, and fingers crossed for finding the error in 10.5 ! > > We tried the python scripting, but apparently a /Users/Newville still lurks in there. > > Running the aps-12p9mLiCl.py ?script > > 1. #!/usr/bin/python > 2. > 3. import math > 4. import Ifeffit > 5. iff = Ifeffit.Ifeffit() > > gives the following error: > > bash-3.2$ python aps-12p9mLiCl.py > ?failed to load ifeffit library ?libifeffit.dylib > /Users/Newville/Codes/ifeffit/local/share/ifeffit:/opt/local/bin:/opt/local/sbin:/opt/pgi/osx86/10.4/bin:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Applications/iXAFS.app/Contents/Resources/local/bin:/opt/local/bin:/opt/local/sbin:/sw/bin:/sw/sbin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin:/usr/X11R6/bin:/usr/local/lib:/Users/joelb/Desktop/FitYK/:/usr/local/lib/:/sw/lib/:/sw/include/readline/:/unix/Molecular Dynamics/moldy-2.16e:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Users/joelb/Desktop/fdmnes:/usr/local/lib:/Users/joelb/Desktop/FitYK/:/usr/local/lib/:/sw/lib/:/sw/include/readline/:/unix/Molecular Dynamics/moldy-2.16e:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Users/joelb/Desktop/fdmnes:/Applications/iXAFS.app/Contents/Resources/local/bin > Traceback (most recent call last): > ?File "aps-12p9mLiCl.py", line 5, in <module> > ? ?iff = Ifeffit.Ifeffit() > ?File "/Library/Python/2.6/site-packages/Ifeffit/Ifeffit.py", line 79, in __init__ > ? ?raise ImportError('Cannot load ifeffit library') > ImportError: Cannot load ifeffit library > > > I tried to add /Applications/iXAFS.app/Contents/Resources/local/lib to PATH (there's a libifeffit.dylib there), but that did not help. > > Also notice that my PATH variables have been duplicated. This happens because I still have '. /Applications/iXAFS.app/Contents/Resources/local/bin/iff_init.sh' in my '.profile'. Do I need to keep this? If so, it would be nice to fix iff_init.sh too :-) > > Thanks, > > Joel + Barb >

1 0

atoms input file
by mohamed sobhy 09 Sep '10

09 Sep '10

Dear I would like to fit the signal from (U0.8Ce0.2)O2, so I tried to put Ce in the same position of U but all atoms generated files give no Ce in the scattering paths and some fitting give negative amplitude values. Could you help me with the Atoms input file? Mohamed

3 3

IXAFS3.0.2 and python scripting
by Brugger, Joel (SAM) 09 Sep '10

09 Sep '10

Dear Matt, Works great on 10.6 for us too. Thanks a million times, and fingers crossed for finding the error in 10.5 ! We tried the python scripting, but apparently a /Users/Newville still lurks in there. Running the aps-12p9mLiCl.py script 1. #!/usr/bin/python 2. 3. import math 4. import Ifeffit 5. iff = Ifeffit.Ifeffit() gives the following error: bash-3.2$ python aps-12p9mLiCl.py failed to load ifeffit library libifeffit.dylib /Users/Newville/Codes/ifeffit/local/share/ifeffit:/opt/local/bin:/opt/local/sbin:/opt/pgi/osx86/10.4/bin:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Applications/iXAFS.app/Contents/Resources/local/bin:/opt/local/bin:/opt/local/sbin:/sw/bin:/sw/sbin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin:/usr/X11R6/bin:/usr/local/lib:/Users/joelb/Desktop/FitYK/:/usr/local/lib/:/sw/lib/:/sw/include/readline/:/unix/Molecular Dynamics/moldy-2.16e:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Users/joelb/Desktop/fdmnes:/usr/local/lib:/Users/joelb/Desktop/FitYK/:/usr/local/lib/:/sw/lib/:/sw/include/readline/:/unix/Molecular Dynamics/moldy-2.16e:/Users/joelb/Public/XDS/XDS-OSX_Intel:/Users/joelb/Desktop/fdmnes:/Applications/iXAFS.app/Contents/Resources/local/bin Traceback (most recent call last): File "aps-12p9mLiCl.py", line 5, in <module> iff = Ifeffit.Ifeffit() File "/Library/Python/2.6/site-packages/Ifeffit/Ifeffit.py", line 79, in __init__ raise ImportError('Cannot load ifeffit library') ImportError: Cannot load ifeffit library I tried to add /Applications/iXAFS.app/Contents/Resources/local/lib to PATH (there's a libifeffit.dylib there), but that did not help. Also notice that my PATH variables have been duplicated. This happens because I still have '. /Applications/iXAFS.app/Contents/Resources/local/bin/iff_init.sh' in my '.profile'. Do I need to keep this? If so, it would be nice to fix iff_init.sh too :-) Thanks, Joel + Barb ________________________________________ Today's Topics: 1. Re: iXAFS3.0.2 (Matt Newville) ---------------------------------------------------------------------- Message: 1 Date: Wed, 8 Sep 2010 11:36:06 -0500 From: Matt Newville <newville(a)cars.uchicago.edu> To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> Subject: Re: [Ifeffit] iXAFS3.0.2 Message-ID: <AANLkTinmG5RGjBwj5eX5yui2JQo_nEP+1JJuMMdWPndh(a)mail.gmail.com> Content-Type: text/plain; charset=windows-1252 Hi Brandon, Thanks for the report. I've seen the same errors on the only system I have access to running 10.5.something: All the executable programs (perl, ifeffit) report the same dyld error. I thought these messages were perhaps because I had built a "universal binary" that included x86_64 and i386, so I rebuilt for i386 but not x86_64, hoping that would solve the problem. But it didn't. I've looked into a little bit more (if anyone knows more, please let me know!!), and believe the issue may be solved by adding compiler flags of -mmacosx-version-min=10.5 -isysroot /Developer/SDKs/MacOSX10.5u.sdk I have not tried this yet, but will try to get this rebuilt in the next day or two. Sorry for the trouble. FWIW, I still get results I don't understand on PPC, where ifeffit and artemis run but hephaestus are artemis do not.... isn't that amazing? --Matt

2 1

iXAFS3.0.2
by Matt Newville 08 Sep '10

08 Sep '10

Hi All, OK, I put a next attempt for a Mac distribution at http://cars9.uchicago.edu/~ifeffit/src/iXAFS3.0.2.dmg which should have some improvements. I believe this should work on 10.6 without issue -- I tested it on one other system at work, and everything launched without trouble. The Python module should now install well. With earlier attempts, I still had problems with a 10.5 machine, and suspected something wrong with i386 v x86_64. So I rebuilt everything (including perl) to be for i386 and ppc architectures, but not x86_64. In fact, I noticed some Aquaterm-related issues that also seemed x86_64 v i386, and went back to using (and including in the distribution) the standard issue Aquaterm1.0.1 that has i386 and ppc binaries, but not x86_64. Thus, iXAFS3.0.2 has no x86_64 binaries. For one thing, this makes the image "only" 60Mb. I could not find suitable 10.5 or PowerPC machines at work, but will test later today. For PPC, the iXAFS launcher application is now universal, and runs on PPC (G4) machine. With an earlier incarnation (when x86_64 was still included) I could run the Ifeffit shell and Athena. Unfortunately, both Artemis and Hephaestus failed to load with dyld: lazy symbol binding failed: Symbol not found: ___gcc_gmul Referenced from: ....local/bin/perl Expected in: /usr/lib/libSystem.B.dylib Hephaestus showed a splash screen, Artemis did not. I grepped out all the 'use' statements from Hephaestus, and these can all be loaded. In addition, all '*.bundle" files (Perl's binaries) claim to be universal binaries (including ppc). I don't really know what to make of this, but will try again with the latest build. Any feedback on what works and what doesn't would be greatly appreciated. Cheers, --Matt On Mon, Sep 6, 2010 at 11:06 AM, Jeff Terry <terryj(a)iit.edu> wrote: > Hi All, > > Looks like it works now on intel machines.. > I can run ifeffit from the iXAFS shell. > > The iXAFS app is not universal though, it is Intel only. > > Nice job Matt. > > Jeff > > > On Sep 6, 2010, at 10:17 AM, Matt Newville wrote: > >> Hi Joel, All, >> >> Yes, sorry for that. I'm still learning how to do this. Here's the >> latest news (some progress, incomplete): >> >> With Jeff's earlier message I looked into and solved at least one of >> the problems: that /Users/Newville/Codes/ifeffit/local/bin/perl >> was all over the place and expected. Of course, this worked when I >> tested it on my own machine!! I also found an issue with png >> support. >> >> I put a 3.0.1 dmg with these changes at >> http://cars9.uchicago.edu/~ifeffit/src/iXAFS3.0.1.dmg >> >> I believe this may work on 10.6 systems. >> >> I tried this on a 10.5.7, i386 system (Julie's). The main iXAFS app >> runs, but I unfortunately get 'dyld' loading errors for all >> executables (including perl and ifeffit). I do not understand this, >> as thhe executables are definitely listed as "universal", ut I suspect >> it is a x86_64 v i386 issue. If anyone does understand this, please >> let me know. FWIW, the machine I'm building on is less than 3 months >> old, so I think it's x86_64, but 'arch' and 'uname -p' report i386. >> On the build machine, running >> arch -x86_64 ifeffit >> runs fine. >> arch -i386 ifeffit >> runs, but Aquaterm does not generate a plot (using PGPLOT_DEV=/XS does >> generate a plot with both settings). So my sense is that my box is >> really x86_64, but reports i386 and this may be causing trouble on >> real i386 machines. >> >> What's stranger is that I also tried this on a very old G4 powerpc, >> running 10.5. Here the iXAFS.app does not run, and indeed >> the iXAFS executable (iXAFS.app/Contents/MacOS/iXAFS) is indeed i386 >> only. But, the ifeffit application DOES RUN(!!!) on this machine. >> Perl runs too, but it appears I forgot to rebuild Ifeffit.pm for perl >> after some change for png support because >> perl -e 'use Ifeffit' still complains about png support. This is >> fixable. I'll look into making the iXAFS launcher application be >> universal. >> >> The python problem is definitely fixable too -- sorry for such poor testing! >> >> If you or Jeff or someone else wants to try this v3.0.1 out, that >> would be fine, but there are definitely "known problems". I'll make >> a v3.0.2 with the perl and python fixes today. I'll look into the >> i386 v x86_64 issues too, but that will take some research... >> >> Cheers, >> >> --Matt >> >> On Mon, Sep 6, 2010 at 7:34 AM, Brugger, Joel (SAM) >> <Joel.Brugger(a)samuseum.sa.gov.au> wrote: >>> Hi Matt, >>> >>> Not much luck with iXAFS3.0 so far. >>> >>> Under 10.5.8 (two macbooks), iXAFS crashes at startup. >>> >>> Under 10.6.4 (macbook pro), IXAFS opens, but clicking on athena/Artemis/Hephaestus only opens an X11 terminal. >>> >>> I think that somehow the installer expects perl to live in /Users/Newville/: >>> >>> $ pwd >>> /Applications/iXAFS.app/Contents/Resources/local/bin >>> b$ artemis >>> -bash: /Applications/iXAFS.app/Contents/Resources/local/bin/artemis: /Users/Newville/Codes/ifeffit/local//bin/perl: bad interpreter: No such file or directory >>> $ >>> >>> >>> >>> I also tried to install python, but something must be wrong with setup.py since files are copied to /Users/Newville. I now have such an inhabitant on my HD thanks to sudo, but even then the install failed. See details below. >>> >>> Hope it’s nothing traumatic! >>> >>> Thanks, JOEL >>> >>> >>> >>> JoelMacBook-Silver:IfeffitPython-2.0.0 joelb$ python setup.py install >>> running install >>> running build >>> running build_py >>> creating build >>> creating build/lib >>> creating build/lib/Ifeffit >>> copying lib/__init__.py -> build/lib/Ifeffit >>> copying lib/Ifeffit.py -> build/lib/Ifeffit >>> copying lib/ifeffit_config.py -> build/lib/Ifeffit >>> copying lib/shell.py -> build/lib/Ifeffit >>> running install_lib >>> creating /Library/Python/2.6/site-packages/Ifeffit >>> copying build/lib/Ifeffit/__init__.py -> /Library/Python/2.6/site-packages/Ifeffit >>> copying build/lib/Ifeffit/Ifeffit.py -> /Library/Python/2.6/site-packages/Ifeffit >>> copying build/lib/Ifeffit/ifeffit_config.py -> /Library/Python/2.6/site-packages/Ifeffit >>> copying build/lib/Ifeffit/shell.py -> /Library/Python/2.6/site-packages/Ifeffit >>> byte-compiling /Library/Python/2.6/site-packages/Ifeffit/__init__.py to __init__.pyc >>> byte-compiling /Library/Python/2.6/site-packages/Ifeffit/Ifeffit.py to Ifeffit.pyc >>> byte-compiling /Library/Python/2.6/site-packages/Ifeffit/ifeffit_config.py to ifeffit_config.pyc >>> byte-compiling /Library/Python/2.6/site-packages/Ifeffit/shell.py to shell.pyc >>> running install_data >>> creating /Users/Newville >>> error: could not create '/Users/Newville': Permission denied >>> >>> >>> Running as SUDO: >>> >>> JoelMacBook-Silver:IfeffitPython-2.0.0 joelb$ sudo python setup.py install >>> Password: >>> running install >>> running build >>> running build_py >>> running install_lib >>> running install_data >>> creating /Users/Newville >>> creating /Users/Newville/Codes >>> creating /Users/Newville/Codes/ifeffit >>> creating /Users/Newville/Codes/ifeffit/local >>> creating /Users/Newville/Codes/ifeffit/local/bin >>> error: can't copy 'ifeffit_shell.py': doesn't exist or not a regular file >>> >>> >>> >>> >>> >>> _______________________________________________ >>> 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 > > > _______________________________________________ > Ifeffit mailing list > Ifeffit(a)millenia.cars.aps.anl.gov > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit > >

2 2