Introduction to Orientation Imaging Microscopy
When the beam of a Scanning Electron Microscope (SEM) strikes a crystalline material mounted at an incline around 70º, the electrons disperse beneath the surface, subsequently diffracting among the crystallographic planes. The diffracted beam produces a pattern composed of intersecting bands, termed electron backscatter patterns, or EBSPs. The patterns can be imaged by placing a suitable film or phosphor screen in close proximity to the sample in the SEM sample chamber.
An example of such a pattern is shown here. The bands in the pattern are referred to as Kikuchi bands and are directly related to the crystal lattice structure in the sampled region. As such, analyzing the pattern and bands can provide key information about the crystal structure for the measured phase:
- The symmetry of the crystal lattice is reflected in the pattern.
- The width and intensity of the bands are directly related to the spacing of the atoms in the crystal planes.
- The angles between the bands are directly related to the angles between planes in the crystal lattice.
Electron Backscatter Diffraction Patters (EBSD patterns or EBSPs) can also be used to determine the orientation of the crystal lattice with respect to some laboratory reference frame in a material of known crystal structure.
The core to most everything that goes on in EBSD is "indexing" the pattern. If the sample produces good diffraction patterns, getting the proper indexing is a process of:
- Locating the bands
- Determining the angles between the bands
- Comparing the angles to theoretical values
- Determining the phase
This technique allows microstructural phase and crystal orientation information to be determined at very specific points in a sample. The spatial resolution varies with the accelerating voltage, beam current, and spot size of the SEM along with the atomic number of the sample material. Indexable patterns can be obtained from about 0.05 microns with a field emission source.
Historically, one of the most difficult parts in the process of indexing was the first: properly identifying the Kikuchi bands. Their locations were usually determined by an operator tediously locating and drawing the lines on an image. Variations in image intensity, background, pattern quality, etc., frustrated most attempts to automate band identification with image-analysis techniques until the Hough transform was applied to the process.
Basically, the Hough transform converts bands in an image to points in Hough space, which are subsequently easier to identify and localize in an image using software. In the images below, the color-coded Kikuchi bands in the right image have been identified from the same colored peaks in the Hough space image shown to the left.
Once the bands have been identified, the next step, determining the angles between the bands, is pretty much straight math. The subsequent process, which determines the actual indexing, involves comparing the information derived from the Kikuchi bands to the theoretical values for reflectors in known phase reference tables.
For modern, automated EBSD mapping and phase identification applications, the majority of commercial systems place a specially coated phosphor screen inside the specimen chamber, in close proximity to the sample. YAG crystals have also been used in place of the phosphor, but their cost precludes more general usage.
A camera is mounted on the SEM and images the phosphor screen. The electron beam is focused on a particular point of interest in the sample. The interaction of the beam and the microstructure results in an EBSD image forming on the phosphor screen, which is captured by the camera and then further processed. Depending on the system, the image is typically adjusted for background effects using either a dedicated signal processor or PC software. Ultimately, a digital image of the Kikuchi bands is present in the computer for indexing.
In OIM mapping, a variety of data and parameters are calculated and recorded, including the orientation of the crystal, a quality factor defining the sharpness of the diffraction pattern (IQ), the "confidence index" (CI – patented by TSL) indicating the degree of confidence that the orientation calculation is correct, Hough data, the phase of the material, and the location (in x,y coordinates) where the data was obtained on the specimen, etc.
Large amounts of crystallographic data from a specimen can be collected by positioning the electron beam on the specimen under manual or automatic control and repeating the indexing procedure at each beam position. When under automatic control, the beam can be positioned sequentially on the points of a grid to scan an area of interest on the specimen. This allows data to be collected without any operator intervention after the initial setup. Alternatively, a scan can be performed using automatic stage control.
The stored data (location, orientation, image quality, confidence index, and phase) can be processed to create Orientation Imaging Micrographs, enabling a visual representation of the crystallographic microstructure. Each point can be assigned a color or gray scale value based on a variety of parameters such as orientation, image quality, confidence index, phase, etc.
For example, an orientation map is generated by shading each point in the OIM scan according to some parameter reflecting the crystallographic rotation. The map below and left is an Inverse Pole Figure (IPF) map in which the colors correspond to the crystal orientations as shown in the projection. Crystals with their 111 axis normal to the surface of the sample will be blue, and so on.
A grain boundary map can be generated by comparing the orientation between each pair of neighboring points in an OIM scan. A line is drawn separating a pair of points if the difference in orientation between the points exceeds a given tolerance angle. The map below shows grain boundaries color coded according to the ranges given in the key to the right.
These figures illustrate a side-by-side comparison of optical (left) and OIM micrographs of an intergranular stress corrosion crack in a nickel-based super alloy.
The OIM micrograph is based upon 30,000 local orientation measurements. The underlying determinant of resolution of the crack in the OIM micrograph is the spatial scale of the OIM data. In this example, 5 micron steps separate each point in the grid of orientation measurements. In this OIM micrograph, the gray scale at each point in the image is reflective of the quality of the corresponding EBSP. Dark points are associated with poor patterns while light points are associated with good patterns. The crack is readily apparent in the OIM micrograph due to the dark shaded points resulting from the absence of EBSPs from the cracked region. Grain boundaries with misorientations greater than 5º are overlaid on the gray scale image.
The OIM micrograph accurately reproduces the features visible in the optical micrograph, but contains inherently greater crystallographic detail. The OIM micrograph is based upon quantitative and spatially specific crystallographic data, which can be manipulated and displayed according to the research interests of the scientist.