International Union of Crystallography

Science coming out of the Zürich School of Crystallography

The recent paper in Acta Crystallographica B69, 457-464, TiGePt - a study of Friedel differences by Ackerbauer et al. (2013) is a product of the Zürich School of Crystallography held in June 2011. The objective of these two-yearly schools is to teach 20 young participants, mostly PhD students, the basic notions of crystallography by a combination of lectures and practical classes in which each student solves their own crystal structure.

[TiGePt structure]This study arose in a singular way. The crystal of TiGePt used for structure determination by Ackerbauer et al. (2012) (Chem. Eur. J. 18, 6272-6283) was submitted to the Zürich School of Crystallography [Linden & Bürgi (2008) Acta Cryst. A64, C30 and http://www.chem.uzh.ch/linden/zsc] by one of the 20 student-participants (S.-V. Ackerbauer) as her project study. Diffraction measurements (Mo Kα radiation) were made by the school organizers and the student had to solve and refine the project structure, once two example structures provided by the school had been completed. The intermetallic compound TiGePt is atypical in its chemical composition and symmetry compared with most crystals submitted by the other student participants. At an R value of 1.1%, the study of TiGePt was still producing furrowed brows amongst the ten highly experienced tutors and the student. The values of statistics concerning the fit of Friedel opposites looked weird. In particular, it was not entirely clear whether the space group was non-centrosymmetric or not, and in the hustle and bustle of the school, there was no time to pursue these problems further. A lively e-mail discussion was undertaken following the school and its results are presented in the paper.

The X-ray single-crystal diffraction intensities of the intermetallic compound TiGePt were analysed. These showed beyond doubt that the crystal structure is non-centrosymmetric. The analysis revolves around the resonant-scattering contribution to differences in intensity between Friedel opposites hkl and hkl. The following techniques were used: Rmerge factors on the average (A) and difference (D) of Friedel opposites; statistical estimates of the resonant-scattering contribution to Friedel opposites; plots of 2Aobs against 2Amodel and of Dobs against Dmodel; the antisymmetric D-Patterson function. Moreover it was possible to show that a non-standard atomic model was unnecessary to describe TiGePt. Two data sets were compared. That measured with Ag Kα radiation at 295 K to a resolution of 1.25 Å-1 is less conclusive for absolute-structure determination than the one measured with Mo Kα radiation at 100 K to the lower resolution of 0.93 Å-1.

Hans-Beat Bürgi, Howard Flack and Anthony Linden