Robert Krakow

Scanning precession electron tomography for three-dimensional nanoscale orientation imaging and crystallographic analysis

Three-dimensional (3D) reconstructions from electron tomography provide important morphological, compositional, optical and electro-magnetic information across a wide range of materials and devices. Precession electron diffraction, in combination with scanning transmission electron microscopy, can be used to elucidate the local orientation of crystalline materials. Here we show, using the example of a Ni-base superalloy, that combining these techniques and extending them to three dimensions, to produce scanning precession electron tomography, enables the 3D orientation of nanoscale sub-volumes to be determined and provides a one-to-one correspondence between 3D real space and 3D reciprocal space for almost any polycrystalline or multi-phase material.

On three-dimensional misorientation spaces

Determining the local orientation of crystals in engineering and geological materials has become routine with the advent of modern crystallographic mapping techniques. These techniques enable many thousands of orientation measurements to be made, directing attention towards how such orientation data are best studied. Here, we provide a guide to the visualization of misorientation data in three-dimensional vector spaces, reduced by crystal symmetry, to reveal crystallographic orientation relationships. Domains for all point group symmetries are presented and an analysis methodology is developed and applied to identify crystallographic relationships, indicated by clusters in the misorientation space, in examples from materials science and geology. This analysis aids the determination of active deformation mechanisms and evaluation of cluster centres and spread enables more accurate description of transformation processes supporting arguments regarding provenance.

Inter-phase Relationships Revealed in 3-Dimensional Orientation Spaces

Spatially resolved orientation mapping is increasingly performed using electron microscopy techniques, including: electron backscatter diffraction, transmission Kikuchi diffraction and scanning precession electron diffraction. The resulting orientation maps contain a wealth of information with crystal phase and orientation specified at each pixel. However, the depth of this data is often underutilized owing to challenges posed by the analysis of such large quantities of data. In the context of understanding complex and multi-phase materials it is important to characterize inter-phase relationships between nanoscale precipitates and the surrounding matrix, which both affect properties and are indicative of formation pathways. Revealing inter-phase relationships requires statistical assessment of the orientation relationship across the phase boundary, the spatial occurrence of particular boundaries and the surfaces of contact at interfaces. Analysis procedures that highlight relationships in both spatial and orientation dimensions are therefore required. Here, we present an approach to revealing inter-phase relationships, based on considering orientation data in 3-dimensional vector spaces constrained to fundamental zones defined by the crystal symmetry of both crystals.

Multi-Dimensional Machine Learning Aided Analysis of a Nickel-Based Superalloy

In the wake of improved X-ray detector efficiencies [1] and advanced data processing techniques [2-3] emerges the practicality of multi-dimensional electron microscopy, an analytical approach to materials characterization that combines spatial and spectral information [4]. In this work we combine electron tomography with energy dispersive X-ray (EDX) spectroscopy, a technique some term 4D microscopy, to obtain three-dimensional (3D) chemical information at the nano-scale. Such information is essential to fully understand structure-property relationships in materials. The approach is used to analyze a nextgeneration nickel-based superalloy, a critically important high strength material used in the aerospace and power industries. Mapping the 3D distribution of the alloying elements, both common and exotic, is essential to better understand the structural origin of its exceptional strength and ultimately aid the design of new superalloys capable of withstanding increased service temperatures.

Decomposing electron diffraction signals from multi-component microstructures

The question of overlapping signals in measurements recorded in the transmission electron microscope is not a new one. The very name, transmission indicates that the recorded image will be a projection through a given thickness of the sample. In this process the ability to distinguish detail along the beam direction is usually lost. One common approach to separating the combined signals in TEM is through tomography, where multiple projections of the sample from different angles (typically through a systematic tilt-series) are combined to create a volume representation of the sample [1]. In this way some unique signal can be localized within the volume.