The moving-beam diffraction geometry: the DIAD application of a diffraction scanning probe

Understanding the interactions between microstructure, strain, phase and material behavior is crucial in scientific fields such as energy storage, carbon sequestration and biomedical engineering. However, quantifying these correlations is challenging, as it requires the use of multiple instruments and techniques, often separated by space and time. The Dual Imaging and Diffraction (DIAD) beamline at Diamond Light Source is designed to address this challenge. DIAD allows its users to visualize internal structures (in two and three dimensions), identify compositional/phase changes and measure strain. It enables in situ and operando experiments that require spatially correlated information. DIAD provides two independent beams combined at one sample position, allowing 'quasi-simultaneous' X-ray computed tomography and X-ray powder diffraction. A unique functionality of the DIAD configuration is the ability to perform 'image-guided diffraction', where the micrometre-sized diffraction beam is scanned over the complete area of the imaging field of view without moving the specimen. This moving-beam diffraction geometry enables the study of fast-evolving and motion-susceptible processes and samples. Here, we discuss the novel moving-beam diffraction geometry, presenting the latest findings on the reliability of both the geometry calibration and the data-reduction routines used. We provide a comprehensive quantitative assessment of the moving-beam diffraction geometry implemented at the DIAD beamline, which will serve as a reference for beamline users. Our measurements confirm that diffraction is most sensitive to the moving-beam geometry for the conventional transmission geometry of the detector. The observed data confirm that the motion of the Kirkpatrick-Baez mirror coupled with a fixed-aperture slit results in a rigid translation of the beam probe, without affecting the angle of the incident-beam path to the sample. Our measurements demonstrate that a nearest-neighbor calibration can achieve the same accuracy as a self-calibrated geometry when the distance between the calibrated and probed sample regions is smaller than or equal to the beam spot size. The absolute error of the moving-beam diffraction geometry at DIAD with typical calibration setup remains below 0.01%, which is the accuracy we observe for the beamline with stable beam operation.