Progressive steps in the development of electron backscatter diffraction and orientation imaging microscopy

Ten years ago electron backscatter diffraction (EBSD) became available to a wider group active in materials research. This paper highlights some of the more significant developments in camera technology and software developments that have arisen since then. The use of slow‐scan charge couple device cameras for phase identification and rapid determination of orientation image micrographs is reviewed. The current limiting spatial resolution of the technique is shown to be less than 10 nm. A procedure for improving lattice spacing measurement by utilizing the full resolution of the camera is described with experimental measurements on silicon and nickel showing relative errors of plus/minus 3%. An investigation of partially recrystallized aluminium shows how the recrystallized fraction can be extracted with confidence but that the mapping of substructure in the highly deformed regions is questionable. Phase identification is described for complex cases in which the phase data tabulated in standard databases do not correspond to what is observed in the EBSD patterns. Phase mapping in a complex mineral in which chemical data and EBSD data are collected simultaneously is shown to be improved by recording both the chemical and the EBSD data into computer memory and proceeding with the phase discrimination and orientation measurement in off‐line analysis.     

Using EBSD and TEM-Kikuchi patterns to study local crystallography at the domain boundaries of lead zirconate titanate

Reliable EBSD mapping of 90° domains in a tetragonal ferroelectric perovskite has been achieved for the first time, together with reliable automated orientation determination from TEM‐Kikuchi patterns. This has been used to determine misorientation angles at 90° domain boundaries and thus local c/a ratios. The sources of orientation noise/error and their effects on the misorientation angle data have been thoroughly analyzed and it is found that this gives a cosine distribution of misorientation angles about the mean with a characteristic width related to the width of the orientation noise distribution. In most cases, a good agreement is found between local c/a ratios and global measurements by X‐ray diffraction, but some clear discrepancies have also been found suggesting that real local variations are present, perhaps as a consequence of compositional inhomogeneities.     

Grain size measurement by EBSD in complex hot deformed metal alloy microstructures

The measurement of grain size by EBSD has been studied to enable representative quantification of the microstructure of hot deformed metal alloys with a wide grain size distributions. Variation in measured grain size as a function of EBSD step size and noise reduction techniques has been assessed. Increasing the EBSD step size from 5% to 20% of the approximate mean grain size results in a change in calculated arithmetic mean grain size of approximately 15% and standard noise reduction techniques can produce a further change in reported size of up to 20%. The distribution of measured grain size is found not to be log‐normal, with a long tail of very small sizes in agreement with a computer simulation of linear intercept and areal grain size measurements through randomly oriented grains. Comparison of EBSD with optical measurements of grain size on the same samples shows that, because of the ability of EBSD to distinguish twins and resolve much smaller grains a difference of up to 50% in measured grain size results.     

'Five-parameter' analysis of grain boundary networks by electron backscatter diffraction

This paper describes state‐of‐the‐art analysis of grain boundary populations by EBSD, with particular emphasis on advanced, nonstandard analysis. Data processing based both on misorientation alone and customised additions which include the boundary planes are reviewed. Although commercial EBSD packages offer comprehensive data processing options for interfaces, it is clear that there is a wealth of more in‐depth data that can be gleaned from further analysis. In particular, determination of all five degrees of freedom of the boundary population provides an exciting opportunity to study grain boundaries by EBSD in a depth that was hitherto impossible. In this presentation we show ‘five‐parameter’ data from 50 000 boundary segments in grain boundary engineered brass. This is the first time that the distribution of boundary planes has been revealed in a grain boundary engineered material.     

A new method for locating Kikuchi bands in electron backscatter diffraction patterns

Electron backscatter diffraction (EBSD) device can provide crystal structure, orientation, and phase content data through analysis of EBSD patterns. The reliability and precision of these data depend on the quality of the band position and zone axes data. This study introduces a new image processing method that can accurately provide the location of Kikuchi bands and poles. In this method, pattern rotation and gray gradient calculation are employed after for the initial detection of Kikuchi lines. Hough transform and Gaussian function are used for the final definition of bands position. Based on the position of Kikuchi bands, the indices of lattice planes and zone axes can be obtained precisely and easily. Angles between zone axes are calculated using locating results. The maximum error for a single‐crystal silicon sample is only 8.07%, illustrating the accuracy of this new method.     

Comparison of quartz crystallographic preferred orientations identified with optical fabric analysis,electron backscatter and neutron diffraction techniques

Three techniques are used to measure crystallographic preferred orientations (CPO) in a naturally deformed quartz mylonite: transmitted light cross‐polarized microscopy using an automated fabric analyser, electron backscatter diffraction (EBSD) and neutron diffraction. Pole figure densities attributable to crystal‐plastic deformation are variably recognizable across the techniques, particularly between fabric analyser and diffraction instruments. Although fabric analyser techniques offer rapid acquisition with minimal sample preparation, difficulties may exist when gathering orientation data parallel with the incident beam. Overall, we have found that EBSD and fabric analyser techniques are best suited for studying CPO distributions at the grain scale, where individual orientations can be linked to their source grain or nearest neighbours. Neutron diffraction serves as the best qualitative and quantitative means of estimating the bulk CPO, due to its three‐dimensional data acquisition, greater sample area coverage, and larger sample size. However, a number of sampling methods can be applied to FA and EBSD data to make similar approximations.     

Assessment of local deformation using EBSD: Quantification of accuracy of measurement and definition of local gradient

Electron backscatter diffraction (EBSD) in conjunction with scanning electron microscopy was used to assess the magnitude of microstructural scale deformation (local deformation) for deformed Type 316 stainless steel. Local misorientation, which is an averaged misorientation between neighboring measurement points, is often used for assessment of local deformation. However, the local misorientation is unsuitable for a quantitative evaluation because it depends not only on the local deformation but also on various factors such as measurement accuracy and the distance between points used in the misorientation calculation (step size). In this study, first, the measurement accuracy was quantified by a parameter called background noise. The factors which affect the measurement accuracy were then discussed from measurements under various conditions. Secondly, in order to reduce the influence of measurement conditions and exclude the dependency of step size in the local deformation assessment, a parameter called the local gradient G_L was proposed. The local gradient clearly showed the spatial distribution of local deformation regardless of the measurement accuracy, and it had hardly any effect from grid pattern and step size. Finally, the local gradient was correlated with the plastic strain from which it was revealed that the strain gradient near a notch root could be estimated by the EBSD measurement.     

EBSD investigation of the microstructure and texture characteristics of hot deformed duplex stainless steel

The microstructure and crystallographic texture characteristics were studied in a 22Cr‐6Ni‐3Mo duplex stainless steel subjected to plastic deformation in torsion at a temperature of 1000 °C using a strain rate of 1 s^?1. High‐resolution EBSD was successfully used for precise phase and substructural characterization of this steel. The austenite/ferrite ratio and phase morphology as well as the crystallographic texture, subgrain size, misorientation angles and misorientation gradients corresponding to each phase were determined over large sample areas. The deformation mechanisms in each phase and the interrelationship between the two are discussed.     

Crystallographic analysis of facets using electron backscatter diffraction

Applications of electron backscatter diffraction (EBSD), also known as backscatter Kikuchi diffraction in the scanning electron microscope (SEM) are first and foremost microtexture and grain boundary misorientation analysis on a single polished section in the specimen. A more subtle and revealing approach to analysis of these data is to use EBSD to probe the orientations of planar surfaces, i.e. facets, which bound crystals. These surfaces include: • grain or phase boundaries • fractures • cracks It is of great interest to know the crystallography of such facets since it provides a key to understanding the physical properties of them.
As far as investigation methodology is concerned, surfaces or facets associated with polycrystals are of two types: exposed or unexposed. Exposed facets, such as a fracture surface, can be viewed directly in the SEM, whereas unexposed facets, such as a grain boundary, are usually revealed as an etched trace on a polished surface. Photogrammetric methods can be used to obtain the positional orientation of an exposed facet, and the crystallographic orientation is obtained either directly from the surface or by indirect sectioning. Calibrated sectioning is required to obtain the equivalent parameters for an internal surface. The present paper compares the methods for obtaining and interpreting the crystallography of facets, with illustrations from several materials.     

EBSD and TEM investigation of the hot deformation substructure characteristics of a type 316L austenitic stainless steel

The evolution of crystallographic texture and deformation substructure was studied in a type 316L austenitic stainless steel, deformed in rolling at 900 °C to true strain levels of about 0.3 and 0.7. Electron backscatter diffraction (EBSD) and transmission electron microscopy (TEM) were used in the investigation and a comparison of the substructural characteristics obtained by these techniques was made. At the lower strain level, the deformation substructure observed by EBSD appeared to be rather poorly developed. There was considerable evidence of a rotation of the pre‐existing twin boundaries from their original orientation relationship, as well as the formation of highly distorted grain boundary regions. In TEM, at this strain level, the substructure was more clearly revealed, although it appeared rather inhomogeneously developed from grain to grain. The subgrains were frequently elongated and their boundaries often approximated to traces of {111} slip planes. The corresponding misorientations were small and largely displayed a non‐cumulative character. At the larger strain, the substructure within most grains became well developed and the corresponding misorientations increased. This resulted in better detection of sub‐boundaries by EBSD, although the percentage of indexing slightly decreased. TEM revealed splitting of some sub‐boundaries to form fine microbands, as well as the localized formation of microshear bands. The substructural characteristics observed by EBSD, in particular at the larger strain, generally appeared to compare well with those obtained using TEM. With increased strain level, the mean subgrain size became finer, the corresponding mean misorientation angle increased and both these characteristics became less dependent on a particular grain orientation. The statistically representative data obtained will assist in the development of physically based models of microstructural evolution during thermomechanical processing of austenitic stainless steels.     

Multivariate statistical approach to electron backscattered diffraction

This paper assesses the potential of multivariate statistical analysis (MSA) applied to electron backscattered diffraction (EBSD) data. Instead of directly indexing EBSD patterns on an individual basis, this multivariate approach reduces a large (thousands) set of individual EBSD patterns into a core set of statistically derived component EBSD patterns which can be subsequently indexed. The following hypotheses are considered in this paper: (1) experimental EBSD patterns from a microstructure can be analytically treated as linear combinations of spatially simple components, (2) MSA has an angular resolution on par with standard EBSD, (3) MSA can discriminate between similar and dissimilar phases, and (4) the MSA approach can improve the effective spatial resolution of automated EBSD.     

Misorientation distributions in hot deformed NaCl using electron backscattered diffraction

EBSD orientation mapping has been used to derive subgrain boundary misorientation distributions in a series of hot deformed and etched NaCl samples. The main objective of this study has been to examine the influence of data processing, noise caused by angular resolution limits and step size on the subgrain misorientation distributions in hot deformed NaCl. Processing of non‐indexed EBSD patterns increased the average misorientations in etched NaCl. Noise contributed significantly to low angle misorientation peaks for step sizes less than the minimum subgrain size. Orientation data collected using a step size larger than the average subgrain size cumulated misorientations across individual subgrains and effectively measured an orientation gradient between steps. Orientation gradient distributions were not influenced by noise. Average misorientation values calculated from large step data correlated well with average misorientation from small step size data. Average misorientations showed a power law relationship with strain. Three types of substructures were identified using scanning electron microscopy and EBSD mapping, equiaxed subgrains, long subgrain boundaries and a core‐mantle subgrain arrangement.     

Orientation averaging of electron backscattered diffraction data

The use of data averaging to improve the angular precision of electron backscattered diffraction (EBSD) maps is discussed. It is shown that orientations may be conveniently and rapidly averaged using the four Euler-symmetric parameters which are coefficients of a quaternion representation. The processing of EBSD data requires the use of an edge preserving filter and a modified Kuwahara filter has been successfully implemented and tested. Three passes of such a filter have been shown to reduce orientation noise by a factor of ∼10. Application of the method to deformed and recovered aluminium alloys has shown that such data processing enables small subgrain misorientation (< 0.5°) to be detected reliably.     

The analysis of quartz c-axis fabrics using a modified optical microscope

A new fully automated microfabric analyzer (MiFA) is described that can be used for the fast collection of high‐resolution spatial c‐axis orientation data from a set of digital polarized light images. At the onset of an analysis the user is presented with an axial‐distribution diagram (AVA –‘Achsenverteilungsanalyse’) of a thin section. It is then a simple matter to build‐up c‐axis pole figures from selected areas of interest. The c‐axis inclination and colatitudes at any pixel site is immediately available to create bulk fabric diagrams or to select measurements in individual areas. The system supports both the interactive selection of c‐axis measurement sites and grid array selection. A verification process allows the operator to exclude dubious measurements due to impurities, grain boundaries or bubbles. We present a comparison of bulk and individual c‐axis MiFA measurements to pole figures measured with an X‐ray texture goniometer and to data collected from a scanning electron microscope furnished with electron backscatter diffraction (EBSD) facility. A second sample, an experimentally deformed quartzite, illustrates that crystal orientations can be precisely linked to any location within an individual grain.     

Extracting twins from orientation imaging microscopy scan data

Automated electron backscatter diffraction or orientation imaging microscopy (OIM) provides spatially specific measurements of crystallographic orientation. These measurements are typically collected on regular grids. By inspecting the misorientation between neighbouring measurements on the grid, potential twin boundaries can be identified. If the misorientation is within some given tolerance of a specified twin misorientation, the boundary separating the two measurements may be identified as a potential twin boundary. In addition, for a coherent twin, the twinning planes must be coincident with the grain boundary plane. As OIM scans are inherently two‐dimensional, the scan data provide only limited information on the boundary plane. Thus, it is not possible to ascertain definitively whether the twinning planes are coincident with the boundary plane. Nonetheless, the alignment of the surface traces of the twinning planes with the trace of the boundary provides a partial indication of coincidence. An automated approach has been developed that allows data concerning both twin criterion to be extracted from OIM scans. Application of the methodology to deformed zirconium suggests that the twinning planes remain coherent during deformation. The methodology was also used to improve grain size distributions measured by OIM. These results more closely match those obtained by conventional metallography.     

A statistical analysis of the variation in measured crystal orientations obtained through electron backscatter diffraction

Electron backscatter diffraction (EBSD) techniques are used to determine the crystallography of individual metal grains. This paper examines the variability in the orientation of measurements obtained by EBSD. Although precision and statistics of orientation have been explored in the literature, little attention has been paid to formal statistical inference for quantifying variation in orientation measurements. Our intention is to study precision by developing statistical analyses for quantifying multiple sources of orientation variation, given repeat scans of a metal sample. Three sources of variability are simultaneously explored: variation in repeat measurements at a fixed location, variation among locations within a grain, and grain-to-grain variation. Bayes statistical methods will be applied to a hierarchical model with the uniform-axis-random-spin (UARS) components of Bingham et al. [1] to quantify these sources of variation. Repeat scans of a Inconel 600 specimen will be used to provide an illustrating example of how the statistical methods can be used to arrive at precision estimates.     

Determination of pattern centre in EBSD using the moving-screen technique

The ‘moving‐screen’ or ‘pattern magnification’ method of calibration for electron backscatter diffraction (EBSD) was reformulated to develop a high‐precision technique requiring no crystallographic knowledge of the specimen and no initial estimates of the calibration parameters. The technique depends upon the accurate displacement of the screen and camera assembly. Corresponding points are selected, interactively, from EBSD patterns. It is suggested that, as an alternative, the selection of points from the Hough transform could lead to a completely automated routine.     

Factors affecting the accuracy of high resolution electron backscatter diffraction when using simulated patterns

High resolution EBSD directly compares electron backscattering patterns (EBSPs), generated in a scanning electron microscope, to measure relative strain and rotation to a precision of ～ 10(-4) in strain and 10(-4)rad (0.006 °) in rotation. However the measurement of absolute strain and rotation requires reference EBSPs of known strain and orientation (or a far field region of known strain). Recent suggestions of using simulated EBSPs with known strain show much promise. However precise measurement of the experimental geometry (pattern centre) is required. Common uncertainties of 0.5% in pattern centre result in uncertainty of ～ 10(-3) in strain state. Aberrations in the compact lenses used for EBSP capture can also result in image shifts that correspond to strains/rotations of ± 10(-3) between experimental and simulated EBSPs. Simulated EBSPs can be generated using dynamical or kinematic models (or a combination of the two). The choice in simulation model has a significant effect on the measured shifts, particularly at zone axis and high structure factor bands, due to large intensity variations, and for simple kinematic simulations can result in the measurement of rogue shifts and thus erroneous strain measurements. Calibrant samples of known strain provide a method of measuring the experimental geometry but imprecise stage movement combined with the high depth of field in the SEM could also result in uncertainties in strain of ～ 10(-3).     

Improving EBSD precision by orientation refinement with full pattern matching

We present a comparison of the precision of different approaches for orientation imaging using electron backscatter diffraction (EBSD) in the scanning electron microscope. We have used EBSD to image the internal structure of WC grains, which contain features due to dislocations and subgrains. We compare the conventional, Hough-transform based orientation results from the EBSD system software with results of a high-precision orientation refinement using simulated pattern matching at the full available detector resolution of 640 × 480 pixels. Electron channelling contrast imaging (ECCI) is used to verify the correspondence of qualitative ECCI features with the quantitative orientation data from pattern matching. For the investigated sample, this leads to an estimated pattern matching sensitivity of about 0.5 mrad (0.03°) and a spatial feature resolution of about 100 nm. In order to investigate the alternative approach of postprocessing noisy orientation data, we analyse the effects of two different types of orientation filters. Using reference features in the high-precision pattern matching results for comparison, we find that denoising of orientation data can reduce the spatial resolution, and can lead to the creation of orientation artefacts for crystallographic features near the spatial and orientational resolution limits of EBSD.