Exploiting preferred orientation to resolve overlapping reflections

Lynne B. McCusker

Laboratorium für Kristallographie, ETH Hönggerberg, CH-8093 Zürich, Switzerland

If it weren't for the fact that reflections with similar scattering angles (2q) overlap in a powder diffraction pattern, structure solution for polycrystalline materials would be as straightforward as it is for single crystals. It is the ambiguity in the relative intensities of these overlapping reflections that hinders the determination of the structures of many industrially important materials. In recent years, a number of clever methods have been developed to circumvent this problem, both by adapting existing methods to cope with the intensity ambiguity and by introducing chemical information into the structure determination process [1], but if the ambiguity could be resolved in some way, the powerful techniques that have been developed over the years for single-crystal data could be applied directly.

One way of addressing this problem is to adopt a more elaborate data collection strategy in which several different, but related, data sets are collected on the same polycrystalline sample. For example, a sample with a preferred orientation of the crystallites will yield diffraction patterns whose intensities are dependent upon the orientation of the sample in the X-ray beam. By collecting data with the sample in several different orientations, more information about the relative intensities of reflections that overlap in 2q, but not in orientation space, can be gleaned.

The concept, which involves a full texture analysis followed by a deconvolution procedure, was described by Hedel et al. in 1997 [2], and its practical viability was demonstrated by Wessels et al. in 1999 with the determination of the structure of the high-silica zeolite UTD-1F with 117 non-H atoms in the asymmetric unit [3, 4]. However, the reflection mode geometry used for that structure determination required a relatively large, uniformly textured sample and three days of synchrotron beamtime. In an attempt to reduce the amount of beamtime required and to eliminate the need for a large homogeneous sample, the method has since been adapted to work with data collected in transmission mode using a small sample and a 2-dimensional detector.

The transmission geometry, with an area detector, requires much less synchrotron beamtime, is insensitive to sample inhomogeneities, and yields a dataset that is more complete. However, the reflection geometry, with a pre-detector analyzer crystal, produces data with a much higher resolution in 2q and no limitation (beyond the wavelength) on d_min. Examples of the application of both methods and a discussion of their advantages and disadvantages and of possible improvements will be presented.

1. David, W.I.F.; Shankland, K.; McCusker, L.B.; Baerlocher, Ch. (Eds): Structure Determination from Powder Diffraction Data, Oxford University Press, 2002.

2. Hedel, R.; Bunge, H.J.; Reck, G.: Textures Microstruct. 29 (1997) 103-126.

3. Wessels, T.; Baerlocher, Ch.; McCusker, L.B.: Science 284 (1999) 477-479.

4. Wessels, T.; Baerlocher, Ch.; McCusker, L.B.; Creyghton, E.J.: J. Am. Chem. Soc. 121 (1999) 6242-6247.