Neural networks for rapid phase quantification of cultural heritage X-ray powder diffraction data.

Recent developments in synchrotron radiation facilities have increased the amount of data generated during acquisitions considerably, requiring fast and efficient data processing techniques. Here, the application of dense neural networks (DNNs) to data treatment of X-ray diffraction computed tomography (XRD-CT) experiments is presented. Processing involves mapping the phases in a tomographic slice by predicting the phase fraction in each individual pixel. DNNs were trained on sets of calculated XRD patterns generated using a Python algorithm developed in-house. An initial Rietveld refinement of the tomographic slice sum pattern provides additional information (peak widths and integrated intensities for each phase) to improve the generation of simulated patterns and make them closer to real data. A grid search was used to optimize the network architecture and demonstrated that a single fully connected dense layer was sufficient to accurately determine phase proportions. This DNN was used on the XRD-CT acquisition of a mock-up and a historical sample of highly heterogeneous multi-layered decoration of a late medieval statue, called 'applied brocade'. The phase maps predicted by the DNN were in good agreement with other methods, such as non-negative matrix factorization and serial Rietveld refinements performed with TOPAS, and outperformed them in terms of speed and efficiency. The method was evaluated by regenerating experimental patterns from predictions and using the R-weighted profile as the agreement factor. This assessment allowed us to confirm the accuracy of the results.

Laue microdiffraction on polycrystalline samples above 1500†K achieved with the QMAX-µLaue furnace.

X-ray Laue microdiffraction aims to characterize microstructural and mechanical fields in polycrystalline specimens at the sub-micrometre scale with a strain resolution of ∼10-4. Here, a new and unique Laue microdiffraction setup and alignment procedure is presented, allowing measurements at temperatures as high as 1500 K, with the objective to extend the technique for the study of crystalline phase transitions and associated strain-field evolution that occur at high temperatures. A method is provided to measure the real temperature encountered by the specimen, which can be critical for precise phase-transition studies, as well as a strategy to calibrate the setup geometry to account for the sample and furnace dilation using a standard α-alumina single crystal. A first application to phase transitions in a polycrystalline specimen of pure zirconia is provided as an illustrative example.

CNN-Based Laue Spot Morphology Predictor for Reliable Crystallographic Descriptor Estimation.

Laue microdiffraction is an X-ray diffraction technique that allows for the non-destructive acquisition of spatial maps of crystallographic orientation and the strain state of (poly)crystalline specimens. To do so, diffraction patterns, consisting of thousands of Laue spots, are collected and analyzed at each location of the spatial maps. Each spot of these so-called Laue patterns has to be accurately characterized with respect to its position, size and shape for subsequent analyses including indexing and strain analysis. In the present paper, several approaches for estimating these descriptors that have been proposed in the literature, such as methods based on image moments or function fitting, are reviewed. However, with the increasing size and quantity of Laue image data measured at synchrotron sources, some datasets become unfeasible in terms of computational requirements. Moreover, for irregular Laue spots resulting, e.g., from overlaps and extended crystal defects, the exact shape and, more importantly, the position are ill-defined. To tackle these shortcomings, a procedure using convolutional neural networks is presented, allowing for a significant acceleration of the characterization of Laue spots, while simultaneously estimating the quality of a Laue spot for further analyses. When tested on unseen Laue spots, this approach led to an acceleration of 77 times using a GPU while maintaining high levels of accuracy.

LaueNN: neural-network-based hkl recognition of Laue spots and its application to polycrystalline materials.

A feed-forward neural-network-based model is presented to index, in real time, the diffraction spots recorded during synchrotron X-ray Laue microdiffraction experiments. Data dimensionality reduction is applied to extract physical 1D features from the 2D X-ray diffraction Laue images, thereby making it possible to train a neural network on the fly for any crystal system. The capabilities of the LaueNN model are illustrated through three examples: a two-phase nano-structure, a textured high-symmetry specimen deformed in situ and a polycrystalline low-symmetry material. This work provides a novel way to efficiently index Laue spots in simple and complex recorded images in <1 s, thereby opening up avenues for the realization of real-time analysis of synchrotron Laue diffraction data.

Revealing the role of microstructure architecture on strength and ductility of Ni microwires by in-situ synchrotron X-ray diffraction.

Deformation mechanisms of cold drawn and electropolished nickel microwires are studied by performing in-situ monotonous and cyclic tensile tests under synchrotron radiation. X-ray diffraction tests allow probing elastic strains in the different grain families and establishing a link with the deformation mechanisms taking place within the microwires. The measurements were carried out on several microwires with diameters ranging from as-drawn 100 µm down to 40 µm thinned down by electropolishing. The as-drawn wires exhibit a core-shell microstructure with <111> fiber texture dominant in core and heterogeneous dual fiber texture <111> and <100> in the shell. Reduction of specimen size by electropolishing results in a higher yield stress and tensile strength along with reduced ductility. In-situ XRD analysis revealed that these differences are linked to the global variation in microstructure induced by shell removal with electropolishing, which in turn affects the load sharing abilities of grain families. This study thus proposes a new way to increase ductility and retain strength in nickel microwires across different diameters by tuning the microstructure architecture.