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

RE: [Ifeffit] testing ravelware
by Kelly, Shelly D. 07 Nov '03

07 Nov '03

> Bruce says: > P.S. Once again, on the topic of how people can contribute to the > ifeffit project without writing code, volunteering to conceive of > a battery of tests and then actually running them would be an > excellent contribution. Any takers? I would think that someone would want to write a perl script to execute a bunch of Athena/Artimis commands and then check the results. Then have the script write out a report. (Kind of like what tkperl does when you do install the software?) Can Athena/Artimis commands be executed from a script? If someone was willing to figure out how to do this for one or two of the commands, I would be willing to extend it to the rest. Shelly

2 1

Carlo's problem with wurtzite
by Bruce Ravel 06 Nov '03

06 Nov '03

Hi folks, Last week Carlo posted a problem he was having with a hexagonal space group. He was getting different results when entering data directly into TkAtoms than when he saved the input data and reread it into TkAtoms. I finally found the source of the problem. All output files in TkAtoms are formatted according to "template files". The reason these template files exist is to make it easy for the user to tweak the appearence and content of the files Atoms writes out. Unfortunately there is a bug in the template that is used to save the input data to an atoms.inp file. In Carlo's example, one of the atoms was at the position (1/3,2/3,0). When Atoms save the data, it saved that position as (0.3333,0.3333,0), i.e. with four digits of precision. It turns out that Atoms considers *five* digits of precision, so it was not recognizing that 0.3333 was meant to represent 1/3. This lead directly to the problem that Carlo reported. Eventually Carlo was able to explain this to me in a way that made it into my very dense skull! ;-) This problem will be fixed in the next release of Atoms. Anyone who wants to fix this problem beforehand, here are the simple instructions: unix people: 1. Copy the atoms.atp file to your ~/.atoms/ directory. (Create it if it does not already exist.) The atoms.atp file will be found in someplace like /usr/lib/perl5/site_perl/5.8.0/Xray/atp/ 2. Edit ~/.atoms/atoms.atp and change the precision in the first non-comment line from 8.4 to 9.5. 3. Save the file and fire up TkAtoms. Should work now. windows people: 1. Using WordPad or NotePad, edit the file C:\Program Files\Ifeffit\perl\Xray\atp\atoms.atp 2. Change the precision in the first non-comment line from 8.4 to 9.5. 3. Save the file and fire up TkAtoms. Should work now. There is a moral to this story. Atoms handles numerical precision in a quirky manner. If it is possible to use a factional representation of a number, do so. (I.e. always use "1/3" rather than 0.33, 0.333, 0.3333, or 0.33333 -- even though the last one will, in fact, work correctly.) Thanks to Carlo for finding this problem and to Peter for a valient if unsuccessful stab at solving the problem. Thanks guys! 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/

2 1

Carlo's problem with wurtzite
by John J. Rehr 06 Nov '03

06 Nov '03

Hi everyone, We encountered a similar problem with the default atoms precision of 5 digits for the Ti-omega phase structure. For that structure atoms gave different results for the Wyckoff coordinates: 1/3 2/3 1/2 and ..3333 .6666 .5 This (apparently) confused the degeneracy checker in atoms causing atoms to print out pairs atoms at nearly identical points. Although there may be cases where a difference between say 0.67 and 2/3 is a significant distortion, I think it's very rare that crystallographic parameters are known to better to 4 digits. Thus I'm wondering whether reducing the precision that atoms demands of a user might be advantageous. J. Rehr

1 0

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 03 Nov '03

03 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