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

Re: [Ifeffit] Novice FEFF user question
by Joshua Kas 21 Dec '12

21 Dec '12

Hi Melissa, You will need to set the FMS radius, which is blank at this point as seen in your screenshot. You should also set the SCF radius which is also blank. I would start with FMS 7.0 and SCF 5.0 and you can make them larger to check convergence. Another thing that is a bit strange in your calculation are the options after the XANES card. I would start by leaving these blank, so that the default energy grid is used. Josh Kas Message: 2 > Date: Fri, 21 Dec 2012 16:56:05 -0500 > From: Melissa Menard <mmenard81(a)gmail.com> > To: ifeffit(a)millenia.cars.aps.anl.gov > Subject: [Ifeffit] Novice FEFF user question > Message-ID: > <CA+Rj7Q9UmQ0O0FH210GSoAp+FfRSjtvWK9BjmFshrmkq= > 7foBw(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > I am new to FEFF and I have a small problem. I'm trying to produce a XANES > spectra of Cu (from the examples). I cannot seem to obtain the pre-edge > region in my result. I have attached a screen shot. Could you tell me how > to calculate the area before E0, i.e. from -50 to 50 eV. > > Thanks > Sincerely > > -- > M Menard, Ph.D. >

1 0

Novice FEFF user question
by Melissa Menard 21 Dec '12

21 Dec '12

I am new to FEFF and I have a small problem. I'm trying to produce a XANES spectra of Cu (from the examples). I cannot seem to obtain the pre-edge region in my result. I have attached a screen shot. Could you tell me how to calculate the area before E0, i.e. from -50 to 50 eV. Thanks Sincerely -- M Menard, Ph.D.

1 0

Per-data set R-factor
by Lyle Gordon 21 Dec '12

21 Dec '12

Dear iffefit users, I am using the latest version of Demeter/Artemis to fit some EXAFS data. I'm running Win 7 64. I found in Horae that when I fit multiple datasets it would output the R-factor for each dataset in the fit. I would really like these values in the newest version. I found on the todo list (https://github.com/bruceravel/demeter/blob/master/todo.org) [ ] per-data set R-factor reporting in log file is turned off. Is there a simple way to turn this "on"? If not I guess I can calculate it by hand from the Re/Im terms of the R space data and fit. To that end, if I have a few datasets it is tedious to export the data/fit files for each sample. Is there an automated way to export this data? Thanks very much, Lyle -- Lyle Gordon Department of Materials Science and Engineering Northwestern University 2220 Campus Drive Cook Hall 2036 Evanston, IL 60208 Tel: (847) 491-3584 Mobile: (847) 400-4071 http://lylegordon.ca

2 1

How to refine the float
by Zhaomo Tian 21 Dec '12

21 Dec '12

2 1

How to refine the float
by Zhaomo Tian 20 Dec '12

20 Dec '12

2 1

Re: [Ifeffit] Demeter installation (under Ubuntu)
by Denis Testemale 19 Dec '12

19 Dec '12

Hello Bruce. There are no special characters only letters, numbers, and underscores. To be sure, I tried by placing the athena projects in ~/temp with the same result. Actually, I played a little bit more, and the behaviour is really strange: - consider the home folder /home/denis (I'll write ~ in the following) and a subfolder ~/A - I place an old style athena project in ~: named athena.prj - start dathena - open ~/athena.prj in dathena. It works. - cp athena.prj A/athena.prj - open ~/A/athena.prj in dathena. It works. - rm ~/athena.prj - reopen ~/A/athena.prj in dathena. It says: "/home/denis/athena.prj is not readable" - cp ~/A/athena.prj ~/. - reopen ~/A/athena.prj in dathena. It works. - and finally, I replace ~/A/athena.prj by another athena project (different), i.e., cp ~/some_random_athena_project ~/A/athena.prj - reopen ~/A/athena.prj in dathena. No error message BUT the project it opens is the old one, not the new random one I replaced the original one with... An additional info: the result of this sequence is the same if I let dathena run between each step ((but close the projects CTRL+W) or if I close (CTRL+Q) dathena at every step. What do you think of my Xmas gift Bruce? :-) Cheers denis Le 18/12/2012 19:54, ifeffit-request(a)millenia.cars.aps.anl.gov a écrit : > ------------------------------ > > Message: 2 > Date: Tue, 18 Dec 2012 13:47:32 -0500 > From: Bruce Ravel <bravel(a)bnl.gov> > To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> > Subject: Re: [Ifeffit] Demeter installation (under Ubuntu) > Message-ID: <3657129.uFgO795gez(a)lachesis.bnl.gov> > Content-Type: text/plain; charset="us-ascii" > > > Hi Denis, > > As always, thanks a million for your feedback. This kind of thing has > a profoundly positive impact on the quality of the package. I am > really grateful. > > The gnuplot-x11 dependency is a good catch. Thanks. > > Unsurprisingly, I have no problems opening files on my computers. So > we need to figure out what is happening on yours that is not happening > on mine. My first suspicion involves character sets. > >> - if I copy athena.prj from /home/denis/pathtothefolder/ to /home/denis/ >> then it works. > What is "pathtofolder"? Specifically does it contains any accented > characters, ess-zets, eastern characters or anything else involving > unicode? > > B > -- Denis Testemale Institut Néel FAME beamline at ESRF +33 476 881 045

2 1

Beryllium XAFS in proteins
by grant bunker 18 Dec '12

18 Dec '12

Gina - measuring Be in proteins is extremely difficult and probably unprecedented, but it may be possible to substitute another element that is chemically similar and probe that instead. For example, often metalloenzymes remain functional under such substitutions, e.g. if you substitute a Mn or a Co in a Ca site. The covalent radius of Be is very small, so substituting Ca might not be feasible without excessively distorting the site, but Aluminum isn't far off, and might work. Of course Aluminum itself isn't easy because of its low k-edge, but it may be feasible. If a heavier element (Ca or heavier) could be made to substitute and retain the functionality/properties you are studying, it might be a suitable probe. good luck! Grant Bunker On Dec 18, 2012, at 3:19 PM, 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: Beryllium and EXAFS (Erik Farquhar) > 2. Larch 0.9.17 available for download and testing (Matt Newville) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 18 Dec 2012 12:50:30 -0800 (PST) > From: Erik Farquhar <efarq(a)yahoo.com> > To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> > Subject: Re: [Ifeffit] Beryllium and EXAFS > Message-ID: > <1355863830.4845.YahooMailNeo(a)web141002.mail.bf1.yahoo.com> > Content-Type: text/plain; charset="iso-8859-1" > > A bit unrelated to the question at hand, but regarding #5, yes, people do neutron PX.? The focus?is almost entirely on locating hydrogen atoms (e.g. for protonation state), so I wonder what?the neutron PX guys?would say about looking for a Be atom. The downside is that you usually need?quite large crystals compared to X-ray PX and long data collection times. > ? > Erik > > > ________________________________ > From: Matthew Marcus <mamarcus(a)lbl.gov> > To: Erik Farquhar <efarq(a)yahoo.com>; XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> > Sent: Tuesday, December 18, 2012 1:39 PM > Subject: Re: [Ifeffit] Beryllium and EXAFS > > > Just to stir the pot a little more: > 1.??? ALS has some beamlines which can go down as far as Li.? Work has been done on Li, but only on concentrated stuff. > 2.??? The sample is probably really dilute in Be.? The signal will be pathetic. > 3.??? Radiation damage will be a major issue, especially with X-ray Raman, which is a highly inefficient technique.? Even LHe cooling doesn't solve that problem. > 4.??? Are there model compounds available for proposed local structures of the Be center?? You'd need that to have any hope of making sense out of Be XANES. > 5.??? Is there such a thing as neutron PX?? That might see the Be. > ??? mam > ? > ----- Original Message ----- >> From: Erik Farquhar >> To: XAFS Analysis using Ifeffit >> Sent: Tuesday, December 18, 2012 5:32 AM >> Subject: Re: [Ifeffit] Beryllium and EXAFS >> >> >> To add to Matt's last paragraph, you might also consider beamline 6-2 at SSRL (http://www-ssrl.slac.stanford.edu/beamlines/bl6-2/) as they do a fair amount of work with proteins and have an X-ray Raman setup. I think the beamline scientist occasionally?posts to this list. >> >> >> Good luck, >> Erik >> >> >> -- >> Erik Farquhar, Ph.D. >> Case Center for Synchrotron Biosciences >> Brookhaven National Lab >> NSLS, Building 725A-X3 >> Upton, NY? 11973 >> efarquhar(a)bnl.gov >> +1-631-344-8174 >> >> >> >> ________________________________ >> From: Matt Newville <newville(a)cars.uchicago.edu> >> To: XAFS Analysis using Ifeffit <ifeffit(a)millenia.cars.aps.anl.gov> >> Sent: Tuesday, December 18, 2012 7:52 AM >> Subject: Re: [Ifeffit] Beryllium and EXAFS >> >> Hi Gina, >> >> On Mon, Dec 17, 2012 at 5:23 PM, Clayton, Gina <ClaytonG(a)njhealth.org> wrote: >>> Hi there >>> >>> I am trying to find out more about EXAFS. I am a protein crystallographer. >>> >>> >>> I want to identify the environment of a beryllium ion in the protein, I work with,? of which I have a crystal structure.? EXAFS was recommended as a good technique for identification of the beryllium site and species. The beryllium is an acidic pocket of the protein surrounded by oxygens from the protein, likely covalently bound. However I have had? zero result finding a beam line that can tune their setups to the Be edge at 111 eV. Also some EXAFS users say that EXAFS technique would be good for working out the environment of the Be but some say it would not . In particular? I have been advised that? Be EXAFS would not be useful since? the Be? EXAFS signal would have unacceptable interference from the carbon K-edge at 285 eV and the oxygen K edge signal at 535 eV. >>> >>> Would EXAFS work or something in a similar vein? Or, for instance,? could the acidic pocket be identified using EXAFS, by measuring the oxygens (instead of the Beryllium) using the protein with and without beryllium? >>> >>> >>> Thank you so much for any advice. >>> >>> Gina >> >> As you and as the person who advised you suspect, you face a number of >> challenges: >> >> 1.? The extremely low energy (111 eV).? You might be able to find a UV >> source or x-ray beamline that can reach this energy , but often such >> work is done in vacuo, and so may not be suitable for proteins.? Using >> electrons (from a TEM, say) might be possible too -- these can >> sometimes do C and N edges either directly or looking at Electron >> Energy Loss (send in an electron of high energy E1, and look for >> electrons of energy E1 - E, and scan the analyzer for the loss energy >> E through the absorption edge).? Again, electron probes may not be >> what you want for your proteins. >> >> 2. The C K-edge will limit your signal to "only" 175 eV.? I would say >> to not worry so much about this limitation.? You'll really be >> measuring NEXAFS (near-edge X-ray absorption ....)? but that's OK. >> People measure the B, C, N, and O K edges, all facing the same >> problem, especially for the O K edge: our planet is more or less >> covered with the stuff ;).? These spectra tend to be very rich and >> sensitive to ligands. >> >> 3. Not much (zero?? I can almost believe that!) literature on Be XAFS. >> This is unfortunate because most NEXAFS is measure and compare to >> known spectra.? I think that might mean you have to spend effort to >> building up a spectral library for Be in organic phases. >> >> I think trying to measure the O K edge with and without Be is nearly >> hopeless.? All the X-ray spectroscopies average equally over all the >> atoms of the absorbing species.? In addition all the spectra are >> grossly the same (a step function from 0 below the edge to 1 above the >> edge).? These make it difficult to see changes below about 1%.? ? So I >> think you would just have too high of an O::Be ratio to see anything >> useful. >> >> One option to consider is using X-ray Raman scattering >> (http://en.wikipedia.org/wiki/X-ray_Raman_scattering).? Here, one hits >> the sample with high-energy X-rays (say, 10 keV), which are highly >> penetrating for reasonably sized protein sample.? One analyzes an >> energy loss (again, E1 - E, where here E1 is around 10 keV) to get the >> edge of interest.? I'm familiar with (been involved with) studies on >> B, C, and O using this technique (all in solids, mostly at high >> pressure).? See, for example,? S. K. Lee, et al "Probing of bonding >> changes in B2O3 glasses at high pressure with inelastic X-ray >> scattering",? Nature Materials 4, pp 851-854 (2005), and also the work >> of Simo Huotari.? ? I'm not aware of anyone doing Be K-edge this way, >> but it should definitely be possible. For the work I've been involved >> with, Be would be challenging because there are other excitations in >> solids that appear around 100 eV.? I believe that effect would be far >> suppressed in a protein, though I'm not sure. >> >> That's somewhat more exotic than straightfoward XAFS,, but it might be >> the best chance for measuring Be in a protein.? ? I'm not sure I know >> any beamlines that do X-ray Raman on proteins, but I'd suggest APS >> beamline 20-ID or ESRF beamline as places to start.? ? The next >> questions would be "how many Be atoms are in your protein?", and "how >> much material do you have?".... >> >> Hope that helps, >> >> --Matt Newville >> >> _______________________________________________ >> 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 1

Beryllium and EXAFS
by Clayton, Gina 18 Dec '12

18 Dec '12

Hi there I am trying to find out more about EXAFS. I am a protein crystallographer. I want to identify the environment of a beryllium ion in the protein, I work with, of which I have a crystal structure. EXAFS was recommended as a good technique for identification of the beryllium site and species. The beryllium is an acidic pocket of the protein surrounded by oxygens from the protein, likely covalently bound. However I have had zero result finding a beam line that can tune their setups to the Be edge at 111 eV. Also some EXAFS users say that EXAFS technique would be good for working out the environment of the Be but some say it would not . In particular I have been advised that Be EXAFS would not be useful since the Be EXAFS signal would have unacceptable interference from the carbon K-edge at 285 eV and the oxygen K edge signal at 535 eV. Would EXAFS work or something in a similar vein? Or, for instance, could the acidic pocket be identified using EXAFS, by measuring the oxygens (instead of the Beryllium) using the protein with and without beryllium? Thank you so much for any advice. Gina NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

6 8

plotting backends in Demeter
by Bruce Ravel 18 Dec '12

18 Dec '12

Hi, This afternoon I noticed that someone who goes by the handle "khuonggit" posted a question at David Hughes' gist about installing Demeter on Ubuntu. https://gist.github.com/3959252 I hope he is reading this mailing list. khuonggit asked: I am beginner in EXAFS. I tried to install Demeter in Ubuntu 12.04 64 bit following the guidance here and the installation was successful. However, when I tried to open (d)athena and import some data, the default plotting backend is PGPLOT, not GNUPLOT. Could you please show me how to change it to GNUPLOT ? Good question. In the time since khuonggit posted that question, I have modified the build script to attempt to detect the presence of gnuplot and modify Demeter's configuration accordingly. Hopefully, this will become a non-issue in the future. The plotting backend can be set by hand at any time in any of the GUI programs by editing the preferences. In Athena, choose "Preferences" from the main menu. In Artemis, choose "Edit preferences" from the file menu. In Hephaestus, click on the Configure icon at the bottom left. Once at the preferences tool, select "Plot" then "plotwith". Select "gnuplot" then click "Apply and save". B -- Bruce Ravel ------------------------------------ bravel(a)bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel

1 0

Re: [Ifeffit] How to do with diffraction peak
by Matt Newville 15 Dec '12

15 Dec '12

Hi Matthew, On Wed, 12 Dec 2012, Matthew Marcus wrote: > > On 12/12/2012 8:03 PM, Matt Newville wrote: >> Hi Matthew, Bruce, All, >> >> Sorry for not being able to join this discussion earlier. I agree >> that having glitch-free data is preferred. But I also think it's OK to >> simply remove diffraction peaks from XAFS data -- you're just >> asserting that you know those values are not good measurements of >> mu(E). I don't see it as all that different from removing obvious >> outliers from fluorescence channels from a multi-element detector -- >> though that has the appeal of keeping good measurements at a >> particular energy. > > That's kind of my point. The standard methods, which require uniform > tabulation, cause you to fill gaps in data by interpolation. > Effectively, you're claiming knowledge of data for which you don't have > knowledge. Better to use a method which preserves > agnosticism about the bad data. The method I described for deglitching with > reference scalers tries to preserve what information > you do have by allowing you to ignore some channels. That only works if > there are 'good' channels to serve as references. I would say that identifying a point as an invalid measure of mu(E) and simply not excluding it in the analysis is OK: it is claiming knowledge that we didn't measure that point well. I just don't see it as much different from claiming that outliers in the set of fluorescence intensities are not good measures of fluorescence intensity. You might argue (I would agree) that with multiple measures of fluorescence you can determine outliers more rigorously. But I would also say that then the best thing to do then is to not simply remove outliers but to take the average AND standard deviation of the samples and include both in the analysis. A glitch would then not be removed, but would be reflected by having a less certain value for that energy point. >> I'm not sure that "slow FT" versus "discrete FT" is that important >> here, though perhaps I'm missing your meaning. My vies is that The >> EXAFS signal is band-limited (finite k range due to F(k) and 1/k, and >> finite R range due to F(k), lambda(k), and 1/R^2), so that sampling >> on reasonably fine grids is going to preserve all the information that >> is really there. > > Again, FFT requires a uniform tabulation of data in k-space, which is > generally secured by interpolation. Traditionally, this has > been histogram interpolation, which tries to take into account the > possibility of having multiple data points within one of the new > intervals in k. That method, however, has to 'make up' data when there isn't > any in one or more intervals. > > Actually, there are three possible tiers of speed. There's FFT, there's > discrete FT done the slow way by computing the sums as > they appear in the textbook (possibly good if you want a funny grid in R > space), and then there's what I propose. I agree with your larger point that we can definitely afford the CPU cycles to do better analysis than use something simply because it is Fast. I'm not sure that a slow discrete FT would be significantly different than an FFT, though. I think (but am listening...) that the continuous v discrete FT is like classic audiophile snobbery about the superiority of analog over digital music. With sufficient sampling and resolution and dynamic range in the signal, the difference has to be trivial. I'm willing to believe that a digital recording has less dynamic range, but I'm also willing to believe that this is swamped by my stereo and ears (and I mostly listen to mp3s through cheap headphones these days, so I clearly am not buying a tube amplifier anytime soon...). But for EXAFS, we know the limits of the signals in k and R (say, 50 Ang^-1 and 25 Ang for extrema), and so can choose samplings that definitely over-sample enough to not cause significant artifacts. There can be dynamic range issues (especially with standard pre-amplifiers and VtoF converters), but I think this is a slightly separate issue from "the glitch problem". Interpolation methods are definitely worth discussing and studying. I'm not sure that the energy values that happened to have been where the monochromator was stopped for a measurement of mu(E) are more sacred or a better sampling of the first shell of an EXAFS spectra than a set of energies interpolated from those measurements to be on uniform k grid, provided the the original data isn't woefully undersampled. With a typical step size of 0.05Ang^1 you can easily miss several points and still accurately measure distances up to 5 or probably even 10 Angstroms. >> I do think that having richer window functions and spectral weightings >> would be very useful. You could view chi(k) data with glitches ias >> wanting a window function that had a very small value (possibly zero: >> remove this point) exactly at the glitch, but had a large value (~1) >> at nearby k-values. >> >> Another (and possibly best) approach is to assign an uncertainty to >> each k value of chi. At the glitch, the uncertainty needs to go way >> up, so that not matching that point does not harm the overall fit. > > A window function which went to 0 in the glitch region is tantamount to an > assertion that chi(k)=0 there, which is generally not so > and is probably a lot worse than the error you make by even the lamest > interpolation through the glitch. The uncertainty makes more > sense, except that if you filter data, the noise spreads out in a correlated > manner over adjacent regions. Also, the tacit assumption > of an uncertainty-based method is that the noise is uncorrelated within the > high-uncertainty region. That said, I can imagine that > this is one of those methods that works better than it has any right to. I think of the window as weighting of measurements to consider, and that a weight of zero is applied to data with infinite uncertainty. I don't think of chi(k) as zero outside the window range (say, below kmin or above kmax), just that we don't want to consider that in the analysis. Again, I think the better thing would be to have an uncertainty for each value of chi(k), and propagate that through the analysis. > Now, in my proposed system, how would you fit filtered (q-space) data? The > first step is to do the very slow FT (VSFT) on the input > data with no interpolation. This leads to a set of sines and cosines which > replicate the data. I would then apply the window and > do the back-summation of all those sines and cosines at the given points. > This will undoubtedly introduce some artifacts near > the edges of the deleted regions simply due to our ignorance of what the data > really do in there. Next, in each fit iteration, I would > evaluate the model function at each point, then do the VSFT and filtering > exactly as had been done on the data, which causes the fit to > have the same artifacts. Since the VSFT and filter operations are linear, it > should be possible to pre-compute a kernel which relates the > values of unfiltered data to those of filtered data. This would be a matrix > of size Np*Np with Np=# of points. Using the kernel instead > of evaluating gazillions of trig functions would be reasonably fast; I > suspect that ifeffit does a similar trick. Similarly, fitting in > R-space would be accomplished by computing the VSFT on data and model > function. > > Yes, I know I talk a good game. Unfortunately, I'm too swamped to attempt > implementation. Also, I'd have to reinvent a number of > very well-engineered wheels to do it. I hear you! I'm interested in pursuing "more advanced weightings and transforms", but don't have much time to spend on such things these days, so development is slow.... >> Larch has a separate Transform class used for each dataset in a fit - >> this includes the standard FT parameters and fit ranges. It is >> intended to be able to do this sort of advanced windowing (and in >> principle, advanced weighting too). Currently (I'd be ready for a >> release but am commissioning our new beamline and am hoping to post a >> windows installer by the end of the month) this has "the standard >> XAFS window functions", but one can create their own window functions >> (for k- and/or R-space) with suppressed points/regions, etc, and use >> them as well. I have to admit I haven't tried this, but it was >> definitely part of the intention, and should work. My hope is to >> extend this to include some wavelet-like weightings as well. >> > What is Larch? Is it a replacement for ifeffit? By 'class' do you mean > object as in OOP? > "And now for something completely different. The larch." - Monty Python > mam Yes, Mostly yes, and, Yes, you get the reference exactly ("How to recognize different types of trees from quite a long way away, Number 1: The Larch"). http://xraypy.github.com/xraylarch/ Another wide-open appeal: I'd love help in implementing improvements like these (better windows, comparing continuous v discrete transforms, using wavelets, etc), and think work like this would be publishable. There are several other improvements I'd like to see, such as generating Feff Paths on-the-fly, and so forth. Essentially all of Larch is in python so that it's very easy to modify and add features like this. The available numerical libraries for python are plentiful and well-maintained. It's also easy to add wrappers for C and Fortran code if that is needed..... --Matt

2 4