EBSD (Electron Backscatter Diffraction) in Metallurgy: Principles and Applications

Electron backscatter diffraction is the primary technique for spatially resolved crystallographic characterization in the SEM, mapping orientation, phase, grain boundary character, and local strain point by point across a polished metallographic surface. This guide develops the Kikuchi pattern formation and indexing process behind EBSD and its major metallurgical applications, complementing the bulk-averaged perspective of XRD and the atomic-scale chemistry of atom probe tomography covered elsewhere on MetallurgyZone.

What Is EBSD?

Electron backscatter diffraction (EBSD) is a scanning electron microscope technique that determines crystallographic orientation, phase identity, and local strain at each point across a polished sample surface. A stationary electron beam strikes a sample tilted to approximately 70 degrees from horizontal, and a fraction of the backscattered electrons diffract as they exit the crystal lattice, forming a pattern of paired parallel bands, the Kikuchi pattern, captured on a phosphor screen positioned near the sample. Rastering the beam across the surface and indexing the pattern at each point builds a spatially resolved crystallographic map, a direct spatial complement to the bulk-averaged phase and stress information obtained from XRD metallurgical analysis.

Kikuchi Pattern Formation and Indexing

Pattern Formation

Each pair of parallel bands in a Kikuchi pattern corresponds to diffraction from a specific crystallographic plane, with band width inversely related to interplanar spacing and the angles between bands directly reflecting the geometric relationships between crystal planes. Because these interplanar angles are invariant under rotation of the crystal, the full pattern of band positions uniquely encodes the crystal’s orientation relative to the detector geometry, and variations in band intensity and width additionally carry information usable for phase discrimination between candidate crystal structures.

Hough Transform Indexing

The conventional and most widely used indexing method applies a Hough (or Radon) transform to the captured pattern, converting the straight-line band traces into points in Hough space where they can be rapidly and automatically detected. The measured interplanar angles between detected bands are compared against a precomputed lookup table for each candidate crystal structure to solve for the best-fit orientation. This approach enables on-the-fly indexing during live acquisition and is implemented in essentially all commercial EBSD systems, typically achieving angular precision on the order of half a degree with absolute accuracy of one to two degrees.

Advanced Indexing Approaches

Where finer angular precision is required, particularly for high-resolution elastic strain mapping, dictionary-based indexing and direct pattern-matching methods compare the full measured pattern, or substantial portions of it, against a library of simulated or experimental reference patterns using cross-correlation techniques. These approaches can achieve precision an order of magnitude finer than conventional Hough indexing, at correspondingly greater computational cost, and are the basis for high-resolution EBSD (HR-EBSD) elastic strain and geometrically necessary dislocation density mapping discussed below.

Orientation Mapping and Texture Analysis

Systematically indexing every point in a rastered scan produces an orientation map, commonly visualized as an inverse pole figure (IPF) map in which each pixel is coloured according to which crystallographic direction is parallel to a chosen sample reference direction. Aggregating orientation data across a mapped area allows direct calculation of pole figures and orientation distribution functions, providing a spatially resolved, statistically representative measurement of crystallographic texture that complements the bulk texture characterization discussed in our crystallographic texture guide, with the added ability to correlate texture directly with grain morphology and neighbour relationships that bulk XRD pole figures cannot resolve.

Grain Boundary Character and Twin Identification

Because EBSD provides the crystal orientation at every measurement point, the misorientation angle and rotation axis between any pair of neighbouring points can be calculated directly. Grain boundaries are reconstructed wherever this misorientation exceeds a chosen threshold, commonly around 15 degrees for a conventional random high-angle grain boundary, distinguishing them from low-angle sub-grain boundaries associated with dislocation substructure.

Special coincidence site lattice (CSL) boundaries, in which a specific fraction of lattice sites coincide across the boundary for a particular rotation angle and axis, are identified by comparing measured misorientation against the known CSL relationships, typically within an allowed deviation of a few degrees. The most metallurgically significant example is the Sigma-3 (Σ3) annealing twin boundary common in face-centred-cubic metals such as austenitic stainless steel, nickel alloys, and copper, corresponding to a 60-degree rotation about the <111> axis. Because Σ3 and other low-Sigma CSL boundaries generally show lower energy and different corrosion, precipitation, and crack-propagation behaviour than random high-angle boundaries, quantifying grain boundary character distribution by EBSD underpins the field of grain boundary engineering, relevant to the intergranular degradation mechanisms discussed in our sensitization in stainless steel guide.

Typical EBSD grain boundary classification thresholds:

Misorientation < ~2-5 deg    Sub-grain / low-angle boundary (LAGB)
Misorientation ≥ ~15 deg    Random high-angle grain boundary (HAGB)

Sigma-3 CSL boundary (FCC twin): 60 deg rotation about <111>
  (accepted within an allowed deviation, commonly a few degrees,
   per the Brandon criterion or similar tolerance rule)

Phase Mapping

Because Kikuchi band geometry and intensity depend on the specific crystal structure being probed, EBSD can distinguish between crystallographically distinct phases present within the same field of view, provided a candidate structure file is available for each phase expected. This makes EBSD phase mapping particularly valuable for quantifying and spatially resolving multi-phase microstructures, such as ferrite-austenite distribution in duplex stainless steel, retained austenite morphology in advanced high-strength steel, or carbide and intermetallic phase identification, complementing the bulk phase quantification obtained from XRD with direct spatial and morphological context.

Grain Size Measurement: ASTM E2627

ASTM E2627 provides a standard practice for determining average grain size from EBSD orientation maps in fully recrystallized polycrystalline materials, defining grains as contiguous regions of measurement points bounded by a specified minimum misorientation angle, and applying minimum grain size and pattern confidence index criteria to exclude indexing noise from the calculation. Compared to traditional linear-intercept or comparison-chart grain size methods performed on chemically etched optical micrographs, the EBSD-based approach uses an objective, quantitative misorientation criterion rather than relying on the visibility of etched grain boundaries, and simultaneously provides orientation, phase, and texture information from the same underlying dataset, connecting directly to the general grain size concepts in our ASTM E112 grain size calculator and guide.

Local Strain Analysis: KAM and GND Density

Beyond orientation and phase, EBSD provides spatially resolved information about local plastic deformation through two related metrics.

Kernel Average Misorientation (KAM)

KAM is calculated by averaging the crystallographic misorientation between each measurement point and its immediate neighbours within a defined kernel (typically the nearest one or two rings of neighbouring points), producing a map in which higher values indicate greater local lattice curvature. Because plastic deformation introduces geometrically necessary dislocations that bend the lattice, elevated KAM correlates with regions of higher local plastic strain, making it a widely used, computationally inexpensive strain proxy for visualizing deformation heterogeneity around indentations, inclusions, and grain boundaries.

Geometrically Necessary Dislocation (GND) Density

GND density is derived more rigorously from the spatial gradient of orientation across an EBSD map, using the mathematical relationship between lattice curvature and the minimum dislocation content required to produce that curvature. While still an indirect estimate rather than a direct dislocation count, GND density mapping provides a more quantitatively grounded measure than KAM alone and is widely used in deformation and fatigue studies to correlate local dislocation accumulation with microstructural features, connecting to the dislocation mechanics discussed in our strain hardening and cold working guide.

Sample Preparation

EBSD Compared with XRD and Atom Probe Tomography

These techniques form a natural characterization hierarchy from bulk statistical averages down to atomic-scale chemistry, and a complete metallurgical investigation frequently combines all three: bulk XRD to establish overall phase fractions and stress state, EBSD to visualize and quantify the spatial grain, phase, and strain distribution responsible for that bulk behaviour, and atom probe tomography to resolve fine-scale segregation or precipitate chemistry at specific features identified by EBSD.

Industrial and Research Significance

EBSD is central to modern alloy development and failure analysis wherever microstructural orientation, phase distribution, or local deformation state governs performance: recrystallization and texture control in rolled sheet product, grain boundary engineering for corrosion and creep resistance, dual-phase and advanced high-strength steel microstructure quantification, weld HAZ characterization, and strain localization studies around fatigue crack initiation sites. Its ability to link crystallographic orientation directly to spatial microstructural features, at a resolution and statistical scale unavailable from either bulk XRD or point-analysis techniques alone, has made it a standard tool across both academic materials research and industrial quality control laboratories.

Frequently Asked Questions

Recommended Reference Reading

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.

Further Reading