Dear Luca, I'm glad to see your interest to the problem even after such a delay.
>So in the end, your model for faulting as you describe: >"A more general model [J Appl Cryst (2000) 338] is included in DDM:" was to >use a trigonal >cell with hexagonal axis to allow refining the anisotropic shift of the >peaks caused by the planar defects? That is a nice trick, you can try to >justify it by reasoning that one effect of the planar defect is to get >same stacking sequence of the hexagonal (we all know that), but what about >the FCC stacking now? I chose the trigonal setting since the presence of layered faults decreases the symmetry of the material making it intermediate between cubic close-packed and hexagonal close-packed. This is an approximation of the real complex structure, of course, but I find it applicable to most cases of close-packed faulted materials I've dealt with. In some cases one has to choose more sophisticated models in order to account for peculiar diffraction features due to correlations between faults, see, for instance Fig. 2 in [O. Ersen, J. Parmentier, L. A. Solovyov, M. Drillon, C. Pham-Huu, J. Werckmann, P. Schultz, Direct Observation of Stacking Faults and Pore Connections in Ordered Cage-Type Mesoporous Silica FDU-12 by Electron Tomography. J. Am. Chem. Soc. (2008) 16800] >To make the audience aware, just changing the cell was not sufficient, you >have to reproduce the intensities. So in the structure a couple of Cu, Zn >was set in 0,0,0 and another in 0,0,1/3 and the occupancies refined (to >values as 1.10392 for the first position and 0.39608 for the second, but >look like the second is calculated from the first) to adjust the >intensities and the density. Yes, the additional (0,0,1/3) position is included in the model to account for the partial displacement of atoms from ideal CCP sites due to the faulting. It allows accounting for the influence of faults on both the intensities and the peak broadening according to the methodology described in J Appl Cryst (2000) 338. >But actually this didn't work out completely >as the resulting quantitative phase analysis is completely wrong. Why do you think that the QPA is wrong? Just because it differs from what you have from Maud??? >And where are the crystallite sizes? Planar defect densities? So what kind >of results did you get from the material science point of view? I can't determine the crystallite size from this pattern since I don't have the instrumental broadening information. As for the faulting probability, it can be derived form eq. 5 of J Appl Cryst (2000) 338. For the alpha-brass phase, the fraction C of defective cells (atoms displaced by faults) is given by the total occupancy of the pseudo-position at (0,0,1/3) that is refined to 0.132. The reciprocal values (1/t1 - 1/t) are listed in the column hkl of the DDM-output reflections listing. For the faulting direction [003] this value is 64.2A. Thus, the fault probability: p = 2Cd(001)(1/t1 - 1/t) = 2*0.132*2.1464/64.2 = 0.0088 This value, however, may be biased as I don't have the instrumental broadening parameters. I must also note that for such low-quality data one can hardly expect highly-reliable microstructural characteristics. Best regards, Leonid ******************************************************* Leonid A. Solovyov Institute of Chemistry and Chemical Technology 660036, Akademgorodok 50/24, Krasnoyarsk, Russia http://sites.google.com/site/solovyovleonid ******************************************************* ----- Original Message ----- From: Luca Lutterotti <luca.luttero...@ing.unitn.it> To: rietveld_l@ill.fr Cc: Sent: Thursday, April 3, 2014 3:08 AM Subject: Re: Stacking faults and antiphase boundary Dear Leonid, sorry to come back to an old thread, but there is something to be know that you didn't tell completely. So finally I came across this morning to a picture of the fit by ddm of this bbm48bis sample with planar defects (coming from Maud examples, picture and example now also on the ddm web site). I didn't see it at that time as it was not included in the zip of Leonid and I didn't have Windows to install ddm and check it (nor the time). Now that I saw in the picture that Leonid did not use a cubic cell to fit the alpha-brass, I took the time to download ddm, and check the solution. So in the end, your model for faulting as you describe: "A more general model [J Appl Cryst (2000) 338] is included in DDM:" was to use a trigonal cell with hexagonal axis to allow refining the anisotropic shift of the peaks caused by the planar defects? That is a nice trick, you can try to justify it by reasoning that one effect of the planar defect is to get same stacking sequence of the hexagonal (we all know that), but what about the FCC stacking now? To make the audience aware, just changing the cell was not sufficient, you have to reproduce the intensities. So in the structure a couple of Cu, Zn was set in 0,0,0 and another in 0,0,1/3 and the occupancies refined (to values as 1.10392 for the first position and 0.39608 for the second, but look like the second is calculated from the first) to adjust the intensities and the density. But actually this didn't work out completely as the resulting quantitative phase analysis is completely wrong. Not to mention a quite high B factor (2.17 compared to the near 0.6 value of Maud and the other BCC phase) for the atoms in the alpha phase, probably to compensate for the wrong structure and kill the two high peaks at high angle resulting from the trigonal structure. Well this kind of trick may be able in certain cases to fit the pattern (here just because peaks are broad and you don't notice the split on the first peak and others of the alpha), but the results you get speak for themselves. And where are the crystallite sizes? Planar defect densities? So what kind of results did you get from the material science point of view? You could have used just a full profile pattern fitting at this point, and at least get the crystallite sizes. So now I am asking myself why I don't use always a triclinic cell to describe my phases, I will have the flexibility to fit everything without resorting to complex "physical" models. Best regards, Luca On Jun 28, 2013, at 6:34, Leonid Solovyov <l_solov...@yahoo.com> wrote: Dear Luca, My doubts regarding the applicability of the faulting model in Moud are due to the absence of its description in the program and the non-availability of options required for crystal systems where the faulted atomic displacements and domain planes are not obviously pre-defined as in the close-packed metals. The model included in DDM is, indeed, simplified and phenomenological to some extent (as well any other model) but it is equally applicable to any structure with any crystallographic symmetry and any faulting type that may be described by a lattice of unit cell domains differing in the atomic arrangement. For your bbm48bis example it gives a similar profile fit as that in Maud in both Rietveld and DDM mode with even fewer number of variable parameters - 14 for DDM and 17 for Rietveld compared to 22 in Maud: http://sites.google.com/site/ddmsuite/tutorials/bbm48bis.zip The profile R-factor output by DDM in the Rietveld mode is high (20.77%) since it is calculated for the background-subtracted data. The equivalent “standard” Rwp is 7.88% (and 7.59% in Maud). Some microstructural variable parameters in the Maud model bbm48bis.par are, apparently, excessive as their estimated errors exceed the refined values. Concerning the FeAl.ddm example, it does contain antiphase domains whose influence on the selective broadening is allowed for by the connection between the pseudo-position occupancy and the size-ellipsoid (regulated by the NDPAR field). Best regards, Leonid ******************************************************* Leonid A. Solovyov Institute of Chemistry and Chemical Technology 660049, K. Marx 42, Krasnoyarsk, Russia http://sites.google.com/site/solovyovleonid ******************************************************* From: Luca Lutterotti <luca.luttero...@ing.unitn.it> To: "rietveld_l@ill.fr" <rietveld_l@ill.fr> Sent: Wednesday, June 26, 2013 5:19 PM Subject: Re: Stacking faults and antiphase boundary I made a quick analysis on Leonid FeAl example. Very simple, no stacking/twin faults, no antiphase domains, only stoichiometry is 15% in favor of Fe. So you only fit the relative occupancy of Al and the different sites accordingly. So if you want to compare the two programs on that simple example, you can download: http://www.ing.unitn.it/~maud/FeAl.zip there are the same files of the example of Leonid, plus I added: FeAl_m.dat (reformatted data file to comply with Maud dat format) FeAl_Maud.par (the analysis file you can load with Maud to see how the fitting was done and what parameter I refined and occupancies setting/binding/refinement) FeAl_Maud_fit.pdf (figure with the final fitting, plotted in sqrt root of the intensities). The step I used: - loaded the datafile - imported FeAl structure from Maud database - automatic quantitative analysis - you recognize the first peak wrong intensity, so set the binding of the occupancies and refined both Al1, Fe1 occupancies. - you can get lower occupancy for Al1, so fixed the Fe1 to 1 and only refined the Al1 occupancy - refined also the instrument asymmetry as I started with the default instrument resolution in Maud (that has a higher asymmetry) Now I wait for the Leonid refinement of bbm48bis (beta-brass milled 48 hours sample bis) example of Maud. Best regards, Luca On Jun 26, 2013, at 11:43 AM, Luca Lutterotti <luca.luttero...@ing.unitn.it> wrote: Well at this point I need to post a clarification, what Leonid call a classical treatment (Warren) is still the best and more comprehensive theory on antiphase domain and stacking/twin faults in close packed alloys. I took the time to read the Leonid paper of 2000 and it doesn't offer anything more than a phenomenological approach to anisotropic broadening, even taking my old way of doing it in 1990 that I am not using anymore after the much better Popa treatment. But the Warren theory on ordered/disordered alloys is a "little" bit more than that: - antiphase domains cause additional broadening only on so-called "superstructure" peaks. This is not well accounted for with just some anisotropic broadening. In fact having worked a lot with such alloys I had to implement the Warren theory to really have a good fit on them (and get the interesting quantities). - stacking and twin faults, depending on the symmetry (BCC, FCC, HCP, etc.) cause not only anisotropic broadening but also: - anisotropic peak shift and even splitting - peak asymmetry and this can be in one side or the other and alternating - the Warren theory gives you directly the antiphase, stacking faults (intrinsic and extrinsic) and twin faults probabilities that are what a metallurgist or a material scientist is interest in, not simply phenomenological parameters to do the fit. So if you really are interested in trying a pattern with some of these effects (not very good from statistical point of view, but sufficient to appreciate the "problems" you encounter with this kind of samples), take the example bbm48bis.par of Maud (the datafile is bbm48bis.dat) and try to do the fitting. You need to download Maud, run it at least once so it extracts all the examples and needed files in a directory you choose and there you will find that example. With Maud you can load as an analysis file the bbm48bis.par file, and press the hammer button on the toolbar to do the final fitting (the example is not at convergence). See if you can fit it as well with another model/program. Editing either phases, under the microstructure tabpanel you will find the microstructural model used and the probabilities. In this case, being a brass alloy, by X-ray we cannot fit the occupancies of the atoms as the contrast is only one electron, otherwise in Maud you just duplicate the sites changing the atom type and use the "equal to" binding of the parameters to set the rules based on the alloy compositions and site exchange. Nothing difficult. As an additional note, if you select the Warren model with the wrong symmetry, Maud will disable it; so it enforces to use it only when it is applicable. Again, if you don't work with close packed alloys and you need to model serious faultings (like turbostratic faultings) that the anisotropic broadening model of Popa cannot fit (heavy asymmetry etc.), in Maud is implemented the single layer model of Ufer (the same implemented in bgmn) that works very well in such cases. Here are the two papers detailing the Maud/Warren models for antiphase domains and stacking/twin faults and Ufer model (so there you find the original Ufer and bgmn references): L. Lutterotti and S. Gialanella, "X-ray diffraction characterization of heavily deformed metallic specimens", Acta Mater., 46[1], 101-110, (1998) L. Lutterotti, M. Voltolini, H.-R. Wenk, K. Bandyopadhyay and T. Vanorio, "Texture analysis of a turbostratically disordered Ca-montmorillonite", Amer. Mineral. 95, 98-103, (2010) Finally (in response to the original poster) the Maud site is not out of date. There is a new version coming very soon. It is at the moment still in beta but downloadable from the Maud site (updated nearly every week or two). But the old stable version works quite well anyway. And sorry if I don't have all that time to maintain a fully featured site always updated to the last news. Best regards, Luca Lutterotti On Jun 26, 2013, at 10:07 AM, Leonid Solovyov <l_solov...@yahoo.com> wrote: Dear David, As I may guess from the microstructural options in Maud, the Warren model of faults and antiphase domains there seems to be applicable only to “simple” classical cases of cubic close-packed metals and alloys such as AuCu etc. A more general model [J Appl Cryst (2000) 338] is included in DDM: http://sites.google.com/site/ddmsuite An example DDM-template for a hypothetical semi-ordered FeAl with antiphase domains separated by (001) planes may be downloaded here: http://sites.google.com/site/ddmsuite/tutorials/AlFe.zip In this example, the variable (Fe,Al) site occupancy parameter number is entered in the NDPAR field to be connected with the selective (superstructure) peak broadening due to the antiphase domains. The domain shape is approximated by the size-ellipsoid. The domains are considered to divide the crystal by layers along [001] direction. The ellipsoid parameters hh and kk are set to zero since the domain sizes along [100] and [010] directions are equal to the crystal size. Note that for the anisotropic domain size modeling in cubic system the ‘Expanded hkl’ generation option is chosen instead of the ‘Symmetry-filtered’ one. For particular sample and faulting model the respective parameters may be adjusted and refined. Other examples for faulted structures may be found in the \EXAMPLES\Defects\ folder of DDMsuite. Best regards, Leonid ******************************************************* Leonid A. Solovyov Institute of Chemistry and Chemical Technology 660049, K. Marx 42, Krasnoyarsk, Russia http://sites.google.com/site/solovyovleonid ******************************************************* From: David Martínez Blanco <martinezbda...@uniovi.es> To: rietveld_l@ill.fr Sent: Tuesday, June 25, 2013 4:50 PM Subject: Stacking faults and antiphase boundary Dear Rietvelders, I am currently analyzing X-ray powder pattern diffraction for the FeAl binary system obtained with Cu anode in Bragg-Bretano geometry, especially focus on the study of order-disorder transition. I usually do Rietveld refinement by means FullProf Suite package but I do not know the specific formulation to deal with stacking faults and antiphase boundaries, in particular for body centred cubic structures. Surfing in the web I discover the program Maud that seems to be capable for analyze these kinds of defects but I can not find related information in the web page of the Maud program (it looks like is out of date since 2011) to the following topics that are specific to my samples: Which are the expressions of the different models using to model the line broadening due to antiphase boundaries and planar defects? Has anyone checked that they function correctly? Maud program includes and works properly with corrections for the micro-absorption for Fe rich phases? Any case, Is it possible to deal the line broadening ascribe to these defects also in the FullProf program for these structures? Best regards and thanks in advance, David ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Please do NOT attach files to the whole list <alan.he...@neutronoptics.com> Send commands to <lists...@ill.fr> eg: HELP as the subject with no body text The Rietveld_L list archive is on http://www.mail-archive.com/rietveld_l@ill.fr/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Please do NOT attach files to the whole list <alan.he...@neutronoptics.com> Send commands to <lists...@ill.fr> eg: HELP as the subject with no body text The Rietveld_L list archive is on http://www.mail-archive.com/rietveld_l@ill.fr/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++