Powder X-ray diffraction is a technique used to solve new crystalline structures, identify phases, analyse phase transformation, and examine micro-structural features. With the recent improvements in X-ray optics and detection, powder microdiffraction experiments can be carried out that provide researchers with images of high lateral resolution in a reasonable scanning time. However, such two-dimensional mappings only provide a global integral of all the diffracted intensity along the X-ray path, thus preventing us from gaining depth resolution information. To study bulk materials, we also require depth resolution information. A local structural probe providing depth resolution images is mandatory in cases where the materials present structural heterogeneities. Three-dimensional diffraction tomography has already been demonstrated for several applications: WAXS was used to study soft tissue achieving millimetre resolution for biological applications, and SAXS was used to study polymers. Furthermore, Laue diffraction and topotomography was used to build 3D mappings of individual grains [1,2].

Here we introduce a new scanning method for synchrotron X-ray diffraction tomography in order to reconstruct cross-section images of unidentified phases in nanomaterials and polycrystalline materials. The technique is a further extension of the X-ray fluorescence micro-tomography technique that has recently been developed at ID22. It involves simultaneous measurement of the absorption, fluorescence and diffraction data while translating (along y) and rotating (around ) a sample illuminated by a focused -or pencil- beam (Figure 126). Thanks to the high flux combined with the use of a fast readout and low noise camera (FRELON) equipped with a taper, the scanning time for a 100 µm3 sample is about 6 hours. This process produces a stack of diffraction images, fluorescence spectra and absorption sinograms. These sinograms are a representation (y,ω) of a particular crystalline phase, elemental distribution, and attenuation coefficient, respectively. Starting from this set of sinograms, a mathematical inversion formula gives rise to cross-section images for each one of the three modalities.

Fig. 126: The ID22 sample environment.

This multi-modal tomographic scheme was first tested on a textbook powder with various grain sizes: a 300 µm capillary was filled with a mixture of chalcedony and iron pigments containing hematite (-Fe2O3). Chalcedony is composed of long quartz micro-fibres, generally less than 100 nm in diameter. A single cross-section image was measured and reconstructed (Figure 127). Even if chalcedony is by far the dominant phase, at least two types of iron grains were observed: hematite and siderite on the one hand, and hematite and phylosilicate phases including greenalite on the other hand. For each grain, a diffraction diagram can be extracted from extremely tiny volumes of powder (as small as the voxel size), allowing subsequent data simulation or structural refinement.

Fig. 127: Reconstructed cross-sections corresponding to: a) the entire diffracted intensity; b) chalcedony; c) hematite and siderite; d) hematite and greenalite; e) fluorescence imaging; f) absorption imaging.

Another carbon-based sample of similar composition and density was also analysed. Five phases were identified and located within the sample (Figure 128). Well-crystallised cubic diamond was located in the central part of the sample, embedded within an amorphous carbon sp3 phase matrix. Furthermore, crystallised ferrite (-Fe) grains located at the sample surface were identified as well as one single impurity grain of calcite (CaCO3), outlining the sensitivity of the method. Ferrite comes from some contamination from the razor blade during sample extraction while calcite is probably a dust particle in contact with the pressure cell. For this weakly-absorbing material, absorption and fluorescence tomography do not provide any contrast, while diffraction tomography reveals the distribution of phases (crystalline and amorphous) inside the sample.

Fig. 128: Cross-section reconstruction of the diamond sample. Glass capillary (gray), sp3 amorphous (red), cubic diamond (red), ferrite (blue) and calcite (white).

Pencil-beam diffraction tomography bridges the gap between the quantitative structural global probe, such as X-ray and neutron diffraction methods, and the local compositional probe, such as X-ray fluorescence and absorption computed-tomography techniques or electron diffraction. Though limited by the number of crystallites in the gauge volume and their size, this method presents a number of advantages. It can obtain contrast where other modalities cannot, provide multi-modal images for a complete sample characterisation and allow the reconstruction of unknown crystalline phases without any a priori information. Furthermore, it is extremely sensitive due to the quantity of diffraction images recorded at different positions and angles of the sample.

 

Principal publication and authors

P. Bleuet (a), E. Welcomme (b), E. Dooryhée (c), J. Susini (a), J-L. Hodeau (c), P. Walter (b), Nature Materials 7, 468 (2008).
(a) ESRF
(b) Centre de Recherche et de Restauration des Musées de France, CNRS-UMR 171, Palais du Louvre, Paris (France)
(c) Institut Néel, CNRS-UPR 2940, Grenoble (France)

References

[1] W. Ludwig, E.M. Lauridsen, S. Schmidt, H.F. Poulsen, J. Baruchel. J. Appl. Cryst., 40, 905 (2007).
[2] B.C. Larson, W. Yang, G.E. Ice, J.D. Budai and J.Z. Tischler. Nature 415, 887 (2002).