Problems with Rietveld analysis

kimenking

Dear all,

I'm brand new to the TOPAS-Academic software and am having trouble getting it to execute on my data files. I'm using a Bruker D2 Phaser with a Lynxeye detector, which creates files marked .raw. I am trying to follow the same order of operations as are given in John Evans' Tutorial #13 at http://community.dur.ac.uk/john.evans/topas_workshop/pcg_workshop_menu.htm in order to carry out a multi-phase Rietveld refinement. My XRD traces are of mixtures of calcite, gypsum, and quartz, so I downloaded .cif files for these minerals at http://rruff.geo.arizona.edu/AMS/amcsd.php. I followed the instructions given in Tutorial #13 exactly (using jEdit to create the input file for TOPAS), but substituting my raw file and the .cif files I downloaded for the tutorial files Dr. Evans provides. When I tried to execute TOPAS, I was told that there was an Abnormal program termination and the log file told me "Structure has no atoms or sites."

In looking carefully at Dr. Evans' tutorial again, I discovered that TOPAS was not able to load .sg files for the spacegroups of the minerals in my mixtures. In investigating further, I found that the sg folder in TOPAS did not appear to contain any of the correct .sg files, so I thought that I had uncovered the source of my problem. On a whim, however, I decided to look for montmorillonite in my mixtures, whose space group did appear to have a .sg file in the sg folder. I deleted the str information for calcite, gypsum, and quartz from the input file and put in the information for montmorillonite and re-ran TOPAS. This time the software said it was able to load the xyz's for the space group from the file but then gave a new error message that says "Cannot find in file isotopes.txt." I also discovered, contrary to my previous belief, that the .sg file for the montmorillonite space group information was freshly created in the sg folder, suggesting that TOPAS should have been able to create the .sg files for the space groups associated with calcite, gypsum, and quartz in the first place.

At this point, I'm at a loss as to how to proceed and would welcome any suggestions anyone might have to get past this impasse.

Thanks for any help you can provide. Kirsten Menking

mfisch

Hi Kirsten

(1) I never had any problem with a Topas installation (neither Academic, nor Bruker). To my knowledge, Topas does not create the .sg files. You should probably reinstall and then leave the Topas folder untouched.

(2) Depending on the syntax in the CIF, jEdit may not be able to get the correct space group from the CIF file. You may need to enter it manually, e.g. directly above the sites as:
space_group "P21/c" or whatever spacegroup you need. You can enter the HM symbol or the Intl. Tables number.

johnsoevans

Just to confirm Martin's response. I think you have two different problems. The first is likely to be the space group label being wrong in your .inp file. Once it's correct topas will create the .sg file if it doesn't already exist. The secod problem is that the atom type you have after "occ" in the site line is probably wrong. The errors both probably arise as the cif isn't being read in properly or contains slightly unusual information. One common cause is if the cif has atom types as e.g. Ca2+ or O2- and those labels aren't in topas' scattering factor list.