RE: Negative Biso / absorption in TOPAS.
Is it possible the data was collected with variable slit aperture? You could try to convert the data to fixed slit aperture intensities and see what happens to the B values. Regards, Jeff Nicolich W.R. Grace & Co. - Conn. 62 Whittemore Ave, Cambridge, MA 02140 +1-617-498-3816 jeffrey.nicol...@grace.com -Original Message- From: Alan Hewat [mailto:he...@ill.fr] Sent: Wednesday, February 10, 2010 5:43 AM To: Jacco van de Streek Cc: rietveld_l@ill.fr; alan.he...@neutronoptics.com Subject: Re: Negative Biso / absorption in TOPAS. > 1. Is absorption indeed the most likely cause of this problem or are > there other possibilities? > > 2. How do we model absorption in TOPAS? Which keyword do we use and > what are reasonable values for the parameters in the expression? 1. You will find discussions of negative B-factors in the Rietveld list archive http://www.mail-archive.com/rietveld_l@ill.fr/info.html Possible causes cited for X-rays include polarisation correction, surface roughness and absorption. You really do need to know how the data were collected - Rietveld refinement should not be treated as a black box. 2. If B-factors are negative, it obviously means that higher order reflections are stronger than they should be. Absorption can cause this, since absorption will be greater for transmission (low angles). For a cylindrical sample it can be shown that moderate absorption has exactly the same effect on intensities as a negative contribution to the overall B factor, and indeed you can calculate the magnitude of the effect if you know the absorption coefficient. Again, use physical information rather than simply refining more parameters. _ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
Re: Find-it
Dear Alan, so it seems I am using an Interface version which is too old. I guess I should urge the responsibles to update it. Thanks for the information. Best wishes, Frank Alan Hewat wrote: Dear Frank, As I wrote, I was referring to my personal opinion and the new ICSD WWW pages on http://icsd.fiz-karlsruhe.de/ which do allow consecutive searches, combination of results and methods of searching for similar structures, structure-types etc. FIZ provides both WWW and FINDIT versions under the exact same conditions, so the choice is yours (previously FINDIT was cheaper). As I wrote, FIZ even allow you to trade in a license for one version against another, and provides demo versions of both to allow you to choose. Certainly each has pros and cons, and I have not tried to list all of them. In particular ICSD is good at drawing structures using Jmol such that it is usually easy to see immediately the chemical interest of the structure. ICSD was the first to use this kind of drawing for inorganic structures, but it has now been adopted by all IUCr journals http://journals.iucr.org/ as well as other databases such as the Cambridge Organic database http://www.ccdc.cam.ac.uk/ the Zeolite database http://www.iza-structure.org/databases/ etc Alan. Frank Girgsdies said: Dear Alan, isn't it a little bit unfair to list just the Pros but skip the Cons? We used to have a FINDIT subscription several years ago. Sadly, our purchasing department canceled the subscription when we got institutional access to the WWW version. If I remember correctly, FINDIT had far more options to customize and tweak a search than the WWW interface offers. You could e.g. search for reduced cells, conduct several consecutive searches and combine the results, select positive hits from the list and skip the rest, export the selected hits etc. All these are features which I miss badly in the WWW version (Or is there a way? Maybe I'm using the wrong engine (http://icsd.fkf.mpg.de)?) So, if you are just looking for a specific crystal structure, the WWW version is nice because it is always up-to-date. But if you try to condcut a systematic and as-complete-as-possible study on a class of compounds, it can be really frustrating to have just the WWW version instead of FINDIT. However, this is just a personal opinion based on experience from several years ago. Maybe the boundary conditions have changed since then... Best wishes, Frank Alan Hewat wrote: I have problem to use FINDIT software: after the starting windows, sometimes it freeze. Is there anyone that know how resolve this problem? FINDIT runs the PC version of the ICSD database. My personal opinion :-) is that you should switch to the WWW version on http://icsd.fiz-karlsruhe.de/ You can get a demo account and you can trade in your FINDIT license for a WWW ICSD license, which will always be up to date and run on any computer. Alan. __ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
Re: Find-it
Dear Frank, As I wrote, I was referring to my personal opinion and the new ICSD WWW pages on http://icsd.fiz-karlsruhe.de/ which do allow consecutive searches, combination of results and methods of searching for similar structures, structure-types etc. FIZ provides both WWW and FINDIT versions under the exact same conditions, so the choice is yours (previously FINDIT was cheaper). As I wrote, FIZ even allow you to trade in a license for one version against another, and provides demo versions of both to allow you to choose. Certainly each has pros and cons, and I have not tried to list all of them. In particular ICSD is good at drawing structures using Jmol such that it is usually easy to see immediately the chemical interest of the structure. ICSD was the first to use this kind of drawing for inorganic structures, but it has now been adopted by all IUCr journals http://journals.iucr.org/ as well as other databases such as the Cambridge Organic database http://www.ccdc.cam.ac.uk/ the Zeolite database http://www.iza-structure.org/databases/ etc Alan. Frank Girgsdies said: > Dear Alan, > > isn't it a little bit unfair to list just the Pros but skip the Cons? > > We used to have a FINDIT subscription several years ago. Sadly, > our purchasing department canceled the subscription when we > got institutional access to the WWW version. > > If I remember correctly, FINDIT had far more options to customize > and tweak a search than the WWW interface offers. You could e.g. search > for reduced cells, conduct several consecutive searches and combine > the results, select positive hits from the list and skip the rest, > export the selected hits etc. All these are features which I miss > badly in the WWW version (Or is there a way? Maybe I'm using the wrong > engine (http://icsd.fkf.mpg.de)?) > > So, if you are just looking for a specific crystal structure, the > WWW version is nice because it is always up-to-date. > But if you try to condcut a systematic and as-complete-as-possible > study on a class of compounds, it can be really frustrating to > have just the WWW version instead of FINDIT. > > However, this is just a personal opinion based on experience from > several years ago. Maybe the boundary conditions have changed since > then... > > Best wishes, > Frank > > Alan Hewat wrote: >>> I have problem to use FINDIT software: after the starting windows, >>> sometimes it freeze. Is there anyone that know how resolve this >>> problem? >> >> FINDIT runs the PC version of the ICSD database. My personal opinion :-) >> is that you should switch to the WWW version on >> http://icsd.fiz-karlsruhe.de/ You can get a demo account and you can >> trade in your FINDIT license for a WWW ICSD license, which will always >> be up to date and run on any computer. >> >> Alan. __ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
Re: Preferred orientation question
Dear Ross, I presume on the flat sample you measured regular theta-2theta diagrams, then only probed those planes parallel to the sample plane. In such a situation, you do not probe correctly the texture, and all the models you could envisage are just giving back parameters that have been refined on non-physic grounds. Harmonics will render a very nice fit, just adding on the number of parameters in the development. Measuring the full spectra in order to resolve structural parameters of a textured sample requires that you acquire data in more than few sample orientations, then the quantitative textural information brings orientational physics in the Rietveld fits (http://www.ecole.ensicaen.fr/~chateign/texture/combined.pdf). For this you need to put the sample in a 4-circle diffractometer. cheers daniel Ross H Colman a écrit : Dear All, Could anybody help me with a question I have related to preferred orientation? I have collected some spectra of a sample, on a D4 using flat plate geometry, that I suspect is showing preferred orientation effects. I was under the impression that a way of testing for this is to re-pack the sample, re-run and compare the spectra, as the level of preferred orientation should change between packings and so the spectra should differ slightly. Despite re-packing and comparing spectra from up to 8 scans the spectra overlay perfectly (within noise). This suggests to me that the refinement residuals are probably not due to preferred orientation. I am using Topas to refine the data and despite the comment above, the residuals seem best fitted using a spherical harmonic-prefered orientation model. I have also tried using other parameters to fit the residuals such as the occupancy of some of the sites, using anisotropic displacement parameters or anisotropic broadening effects, but the SH-PO fits best and seems the most physically reasonable treatment. As a test, I have also run the sample in capillary transmission geometry on another instrument, and the refined value of the SH-PO parameter is reduced. This seems to confirm that PO is the problem, but the question I have is how effective the re-packing comparison method is for confirming PO? If I see no difference in the spectra does it confirm that PO is NOT the problem or can I still have PO effects that are very repeatable upon repeat packing? If PO is not the problem in this case, are there any other structural or instrument effects that could lead to refinement residuals that would be well treated with a SH-PO model? Thank you all very much for any information or light you can shed on the subject. Ross Colman -- -- Daniel Chateigner Professeur, Université de Caen Basse-Normandie Co-editor "Journal of Applied Crystallography", www.iucr.org Editor-in-Chief "Texture, Stress and Microstructure", www.hindawi.com -- address: CRISMAT-ENSICAEN and IUT-Caen, Université de Caen Basse-Normandie, campus 2 6, Bd. M. Juin 14050 Caen, France tel: 33 (0)2 31 45 26 11 fax: 33 (0)2 31 95 16 00 daniel.chateig...@ensicaen.fr http://www.ecole.ensicaen.fr/~chateign/danielc/ -- A Quantitative Texture Analysis Course: http://qta.ecole.ensicaen.fr/ An Open Source for Crystallographic Data: the COD http://sdpd.univ-lemans.fr/cod/ --
Re: Find-it
Dear Alan, isn't it a little bit unfair to list just the Pros but skip the Cons? We used to have a FINDIT subscription several years ago. Sadly, our purchasing department canceled the subscription when we got institutional access to the WWW version. If I remember correctly, FINDIT had far more options to customize and tweak a search than the WWW interface offers. You could e.g. search for reduced cells, conduct several consecutive searches and combine the results, select positive hits from the list and skip the rest, export the selected hits etc. All these are features which I miss badly in the WWW version (Or is there a way? Maybe I'm using the wrong engine (http://icsd.fkf.mpg.de)?) So, if you are just looking for a specific crystal structure, the WWW version is nice because it is always up-to-date. But if you try to condcut a systematic and as-complete-as-possible study on a class of compounds, it can be really frustrating to have just the WWW version instead of FINDIT. However, this is just a personal opinion based on experience from several years ago. Maybe the boundary conditions have changed since then... Best wishes, Frank Alan Hewat wrote: I have problem to use FINDIT software: after the starting windows, sometimes it freeze. Is there anyone that know how resolve this problem? FINDIT runs the PC version of the ICSD database. My personal opinion :-) is that you should switch to the WWW version on http://icsd.fiz-karlsruhe.de/ You can get a demo account and you can trade in your FINDIT license for a WWW ICSD license, which will always be up to date and run on any computer. Alan. __ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
Re: Find-it
> I have problem to use FINDIT software: after the starting windows, > sometimes it freeze. Is there anyone that know how resolve this problem? FINDIT runs the PC version of the ICSD database. My personal opinion :-) is that you should switch to the WWW version on http://icsd.fiz-karlsruhe.de/ You can get a demo account and you can trade in your FINDIT license for a WWW ICSD license, which will always be up to date and run on any computer. Alan. __ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
The 5th Reynolds Cup
Dear people interested in the Rietveld method, you may be interested in the Reynolds Cup 2010 will see the staging of the 5th Reynolds Cup competition in quantitative mineralogy. The biennial Reynolds Cup was started in 2000 by Dougal McCarty, Jan Środoń, and Dennis Eberl in honour of Bob Reynolds and his pioneering work on quantitative clay mineralogy. Participation in the Reynolds Cup has increased steadily over the past 10 years with 43 labs taking part in the 2008 cup and it is now established as the foremost round robin in quantitative mineralogical analysis. The primary purpose of the Reynolds Cup is a blind round robin tool to aid in the improvement of quantitative methods by participation. By taking part you can identify the weaknesses in your methods and then seek to improve them. The results of the 2010 Reynolds Cup will be announced at the CMS Tripartite meeting in Seville. Only the names of the top three contestants will be published. The names of the other participants will remain confidential. The competition is open to anyone interested in quantitative mineral analysis and any method or combination of methods is permissible. Send an e-mail to reynolds...@macaulay.ac.uk to register your interest in participating in the 2010 Reynolds Cup. Your e-mail will then be used to inform you when the samples become available and how to obtain them. For more information about the competition go to http://www.clays.org/SOCIETY%20AWARDS/RCintro.html steve hillier -- Please note that the views expressed in this e-mail are those of the sender and do not necessarily represent the views of the Macaulay Institute. This email and any attachments are confidential and are intended solely for the use of the recipient(s) to whom they are addressed. If you are not the intended recipient, you should not read, copy, disclose or rely on any information contained in this e-mail, and we would ask you to contact the sender immediately and delete the email from your system. Thank you. The Macaulay Land Use Research Institute is a company limited by guarantee, registered in Scotland under company number 16190 and a registered Scottish charity, number SC011922. Registered office Macaulay Drive, Craigiebuckler, Aberdeen, AB15 8QH.
Find-it
Hallo, I have problem to use FINDIT software: after the starting windows, sometimes it freeze. Is there anyone that know how resolve this problem? Thank you Davide Prof. Davide Levy DSMP Università degli Studi di Torino tel. +390116705122 fax +390112365122 skype davidelevy <>
Re: Negative Biso / absorption in TOPAS.
> 1. Is absorption indeed the most likely cause of this problem or are > there other possibilities? > > 2. How do we model absorption in TOPAS? Which keyword do we use and what > are reasonable values for the parameters in the expression? 1. You will find discussions of negative B-factors in the Rietveld list archive http://www.mail-archive.com/rietveld_l@ill.fr/info.html Possible causes cited for X-rays include polarisation correction, surface roughness and absorption. You really do need to know how the data were collected - Rietveld refinement should not be treated as a black box. 2. If B-factors are negative, it obviously means that higher order reflections are stronger than they should be. Absorption can cause this, since absorption will be greater for transmission (low angles). For a cylindrical sample it can be shown that moderate absorption has exactly the same effect on intensities as a negative contribution to the overall B factor, and indeed you can calculate the magnitude of the effect if you know the absorption coefficient. Again, use physical information rather than simply refining more parameters. _ Dr Alan Hewat, NeutronOptics, Grenoble, FRANCE +33.476.98.41.68 http://www.NeutronOptics.com/hewat __
Preferred orientation question
Dear All, Could anybody help me with a question I have related to preferred orientation? I have collected some spectra of a sample, on a D4 using flat plate geometry, that I suspect is showing preferred orientation effects. I was under the impression that a way of testing for this is to re-pack the sample, re-run and compare the spectra, as the level of preferred orientation should change between packings and so the spectra should differ slightly. Despite re-packing and comparing spectra from up to 8 scans the spectra overlay perfectly (within noise). This suggests to me that the refinement residuals are probably not due to preferred orientation. I am using Topas to refine the data and despite the comment above, the residuals seem best fitted using a spherical harmonic-prefered orientation model. I have also tried using other parameters to fit the residuals such as the occupancy of some of the sites, using anisotropic displacement parameters or anisotropic broadening effects, but the SH-PO fits best and seems the most physically reasonable treatment. As a test, I have also run the sample in capillary transmission geometry on another instrument, and the refined value of the SH-PO parameter is reduced. This seems to confirm that PO is the problem, but the question I have is how effective the re-packing comparison method is for confirming PO? If I see no difference in the spectra does it confirm that PO is NOT the problem or can I still have PO effects that are very repeatable upon repeat packing? If PO is not the problem in this case, are there any other structural or instrument effects that could lead to refinement residuals that would be well treated with a SH-PO model? Thank you all very much for any information or light you can shed on the subject. Ross Colman
Re: Negative Biso / absorption in TOPAS.
Lubomir Smrcok wrote: > could you kindly tell us the composition of the sample and the technique > used to collect the data ? The technique to collect the data is unfortunately not known to me, I'd have to ask. The composition is not a secret: it is an organic pigment, Pigment Yellow 110, and its formula is C22 H6 Cl8 N4 O2. Space group P-1, Z'=1/2. Best wishes, -- Dr Jacco van de Streek Senior Scientist Avant-garde Materials Simulation Freiburg im Breisgau, Germany
Re: Negative Biso / absorption in TOPAS.
Jacco, could you kindly tell us the composition of the sample and the technique used to collect the data ? Lubo On Wed, 10 Feb 2010, Jacco van de Streek wrote: Dear colleagues, I am helping some people from the University of Frankfurt with their Rietveld refinements, and usually their powder patterns are measured on their own instrument and the refinements are excellent. This time, however, they have to work with a powder pattern that was sent to them. It is laboratory X-ray data, copper radiation, visible peaks out to 60 degrees 2theta, reasonable counting statistics, almost certainly flat plate (reflection). The crystal structure is known. For the refinement, they have TOPAS available. The Rietveld refinement progresses very smoothly, but when the global Biso is refined, it becomes very large and negative (-4). If we allow the Biso to become negative, the difference curve is a flat line, in other words, the fit is perfect. If we fix the global Biso at, say, 3.0, the fit is still very good, but not as good as we are used to and the difference curve starts to show some small residuals. We have no experience with this, but some Google searches suggest "absorption" as the problem. We have already modelled anisotropic peak broadening and in our experience the background is modelled very well in TOPAS and does not correlate with other parameters. So our main two questions are: 1. Is absorption indeed the most likely cause of this problem or are there other possibilities? 2. How do we model absorption in TOPAS? Which keyword do we use and what are reasonable values for the parameters in the expression? Best wishes, -- Dr Jacco van de Streek Senior Scientist Avant-garde Materials Simulation Freiburg im Breisgau, Germany
Negative Biso / absorption in TOPAS.
Dear colleagues, I am helping some people from the University of Frankfurt with their Rietveld refinements, and usually their powder patterns are measured on their own instrument and the refinements are excellent. This time, however, they have to work with a powder pattern that was sent to them. It is laboratory X-ray data, copper radiation, visible peaks out to 60 degrees 2theta, reasonable counting statistics, almost certainly flat plate (reflection). The crystal structure is known. For the refinement, they have TOPAS available. The Rietveld refinement progresses very smoothly, but when the global Biso is refined, it becomes very large and negative (-4). If we allow the Biso to become negative, the difference curve is a flat line, in other words, the fit is perfect. If we fix the global Biso at, say, 3.0, the fit is still very good, but not as good as we are used to and the difference curve starts to show some small residuals. We have no experience with this, but some Google searches suggest "absorption" as the problem. We have already modelled anisotropic peak broadening and in our experience the background is modelled very well in TOPAS and does not correlate with other parameters. So our main two questions are: 1. Is absorption indeed the most likely cause of this problem or are there other possibilities? 2. How do we model absorption in TOPAS? Which keyword do we use and what are reasonable values for the parameters in the expression? Best wishes, -- Dr Jacco van de Streek Senior Scientist Avant-garde Materials Simulation Freiburg im Breisgau, Germany
