EBSD.com is an educational site by Oxford Instruments

A grain is a three dimensional crystalline volume within a specimen that differs in crystallographic orientation from its surroundings but internally has little variation. Grain size is an important characteristic used in understanding the development, engineering and potential failure in materials. The mechanical and physical properties of metallic materials are often related to grain size e.g. via the Hall-Petch relationship where strength is inversely dependent on the square root of grain size. Electron backscatter diffraction (EBSD) is an ideal technique for this determining grain size, it offers microstructural characterisation including grain size, grain boundary characterisation and texture quantification.

Grain size and grain parameters

To accurately measure grain size, it is imperative that all of the grain boundaries are detected. Therefore the technique used must produce the highest degree of grain boundary delineation. Traditionally grain size was measured using light optical microscopy (LOM) and some of the grain size standards still reference this method. This optical technique often requires a chemical etching of the surface in order to highlight the grain boundaries. However, this etching can be influenced by the existing microstructure in the sample which can be problem for fine structured materials. In addition, as the trend is towards nano scale materials, there is a limit to the grain size which can be detected by LOM. Therefore, EBSD becomes the only viable alternative to measuring grain size. In addition EBSD can provide additional information about the microstructure in excess of that achievable with optical techniques.

To identify the grains based on EBSD requires the definition of a critical misorientation angle, so that all boundary segments with an angle higher than this defined critical angle are considered grain boundaries. By measuring the misorientation between all pixel pairs it is possible to identify the boundaries enclosing the individual grains. If this information is used with the phase information then it is possible to determine the grain size distribution for the individual phases within the sample.

Grain size is a key parameter effecting a materials properties.  However EBSD data provides much more information, so it is possible to extract grain specific parameters on both the morphology and orientation changes within the grains.

Reporting grain size information is described in ASTM standard (E2627) as a grain number. If the grain size information is to be determined accurately then even at the data acquisition stage it is important to have an idea about the grain size so the   data can be collected at a suitable resolution, such that the grains are well defined within the map. It is recommended to have at least 100 pixels within each grain, and to sample at least 500 grains for the grain size information is to be statistically meaningful.

Grain statistics, relating to entire data set or selected phases are generated from the grain measurement.

Figure 1 EBSD data from a single phase steel sample.

figure 1a

(a) Grain map showing grains in random colour. Grains were detected using grain detection angle of 10 degree and minimum 100 pixels within a grain. 1378 grains are detected with mean grain diameter of 25.5μm.

figure 1b

(b) Grain details and statistic summary.

In addition to morphological measurements the grain detection also offers quantitative data of orientation variation within each grain. This can also offer insight when investigating the impact of materials processing, as shown in the example of a shot peened Al.

Figure 2 Grain Orientation Spread (GOS) map of a shot peened aluminium in cross section. This valuable primary strain analysis tool reveals grains that show the most deformation - it illustrates their spatial distribution and numerical prevalence. It is a whole grain classification tool: for each grain in the mapped area, GOS measures the degree of orientation change between every pixel in the grain and the grain's average orientation. The grain is then coloured by the average measurement for all of its constituent pixels. Grains with a higher level of strain, measured by the internal degree of lattice rotation, are coloured at the yellow - red end of the rainbow colour scale shown in the histogram. In this example, the grains exhibiting higher levels of strain are concentrated near the surface, within a damage zone extending to approximately 150μm below the surface.

figure 2a

Grain Orientation Spread (GOS) map of a shot peened aluminium in cross section.

figure 2b

The corresponding misorietation distribution histogram.

Grain boundary

In grain boundary engineering it can be important to enhance or reduce the relative abundance of certain grain boundary types in order to optimise the properties of the final material. EBSD is well suited to extract this type of information as it gives both statistical and spatial information about the grain boundaries.

Grain boundaries can be identified through a misorientation distribution plot. Here the plot will have a distinct peak if there are many grain boundaries with the same misorientation angle. This method is typically used to get a quick overview of the boundary occurrences in the sample.

Similarly, generating a map which shows the spatial distribution of the grain boundaries can provide additional microstructural information. A typical example, shown below, includes a map showing low angle boundaries (highlighting the substructure of individual grains) combined with high angle boundaries (defining the actual grain structure).

Figure 3 Grain boundary from a steel sample determined by use of EBSD. The sample contains Ferrite (white) and Austenite (red) phases.

figure 3a

(a) Grain boundary positions of ferrite phase superimposed on the phase map. Grain boundaries between 2-10 degrees are in green (low angle boundaries), higher than 10 degrees are in black (high angle boundaries). It illustrates the grain structure and substructure of individual grains in ferrite.

figure 3b

(b) The corresponding misorientation distribution plot shows the frequency of grain boundaries in Ferrite.

When the misorientation angle is calculated it is also possible to calculate the misorientation axis. This means that EBSD data can be used to identify specific boundaries defined not just by a misorientation angle but by a combination of misorientation angle and misorientation axis. The most common example is the identification of twin boundaries, which are a subset of the coincident site lattice (CSL) boundaries.

CSL boundaries

CSL boundaries are special boundaries which fulfil the coincident site lattice criterias whereby the lattices are sharing some lattice sites. CSLs are characterized by Σ where Σ is the ratio of the CSL unit cell compared to the standard unit cell. Two examples of CSL relationships are shown below.

Figure 4 (a) The sigma 3 boundary (twin boundary) is a 60° rotation about the [111] direction.
(b) The sigma 5 boundary is a 36.9° rotation about the [100] direction.

figure 4a

CSL boundaries typically have a significant impact on the material properties, which means that from a materials engineering viewpoint it is important to determine both the ratio of CSL boundaries and their distribution within the material. An example from twinned steel is shown below.

Figure 5 CSL boundary data measured by EBSD for a steel sample.

figure 5a

(a) Pattern quality map for the steel sample;

figure 5b

(b) Coincident site lattice (CSL) boundary positions superimposed on the previous pattern quality image;

figure 5c

(c) The boundaries are colour coded by CSL type as shown in the histogram of CSL found.

Go To Top