Grain analysis of metallic and alloy samples is a key quality control process in the metallographic laboratory. Most metals have a crystalline structure containing grain boundaries. During processing, the atoms in each growing grain in a metal or alloy are arranged in a specific pattern based on the material’s crystal structure.
The growth of grains affects each other, creating an interface with different atomic orientations. As it is a known fact that a smaller grain size typically improves a sample's mechanical properties, it is essential to control alloy composition and processing in order to achieve the desired grain size.
Drawbacks of Traditional Methods
ASTM E112 is the most widely recognized standard for grain analysis in North and South America. The "Chart Comparison" method is the most commonly used in quality-control laboratories for grain analysis, involving visual calculation of the grain size through comparison of a live image under an optical microscope against a micrograph chart.
An eyepiece reticle consisting of images of predefined grain size patterns can also be used by directly inserting it into the optical path of a microscope, allowing direct comparison within the microscope (Figure 1).
However, operator involvement in these methods for grain size calculation leads to erroneous, unrepeatable, and non-reproducible results. In addition, manual entry of results into computer-based report or spreadsheet further increases the chance for errors.
Figure 1. Example of a microscope eyepiece reticle used to compare against a live image.
Turnkey, Fully Automated Grain Analysis Solution
The "Intercept Method", a well-known digital solution for grain size analysis in the digital metallurgical laboratory, involves overlaying of a pattern atop the live or acquired digital image and drawing an intercept on the image whenever the overlaid pattern intercepts with a grain boundary (Picture 2).
After system calibration, the image-analysis software estimates the ASTM "G-Number" and mean intercept length automatically as a function of the pattern length and intercept count.
Figure 2. Grain Analysis via the Intercept method.
The "Planimetric" method is another approach for grain size analysis, determining the grain size on the live or captured image by estimating the grain count per unit area (Figure 3).
The internal calculation of the results within the image-analysis software removes the guesswork of operators, thus providing accurate, repeatable and reproducible results.
In addition, it is possible to configure the metallurgical-specific microscope image-analysis software packages for automatic archiving of the grain results into a spreadsheet or optional integrated database (Figure 4).
Figure 3. Grain analysis via the Planimetric method.
Figure 4. Results of an ASTM E112 Analysis
Typical Equipment Configuration for Grain Analysis
The experimental setup typically used for grain analysis through digital image analysis comprises an inverted metallurgical microscope, microscope-specific high-resolution CCD or CMOS digital camera, 10x metallurgical objective lens, and material science specific image-analysis software (Figure 5).
Since the flat, polished sample is laying flat on the mechanical stage of an inverted microscope, consistent focus can be ensured while controlling the scanning stage.
Figure 5. Typical equipment configuration: inverted metallurigical microscope, 10x metallurigical objective lens, microscope-specific high-resolution digital camera.
Optional add-on modules can be used to extend material science microscope-specific image analysis software packages, enabling users to perform direct grain analysis as outlined by the international standards such as ASTM E112. At the time of procurement, users must know which method is more appropriate, the Intercept or Planimetric method.
The pixel size or corresponding pixel density has more significance than digital resolution in the selection of a digital camera for grain size analysis. According to the Nyquist Theorem, 2 to 3 pixels are needed to resolve the smallest detail.
For example, if using a 10x objective lens for grain analysis, the optical resolution would typically be roughly 1.1µm for a mid-grade objective lens. Consequently, the size of the actual, calibrated pixel must be less than 366nm. Hence, material-science microscopy specific cameras of 3MP or more can be used for grain analysis.
Since gray scale mode enables reliable grain size analysis, it is necessary to have cameras that can take pictures in gray scale mode. Moreover, cameras capable of achieving fast refresh-rate in live mode are useful during sample positioning and focusing. It is recommended to have a coded manual or motorized revolving objective nosepiece.
The image-analysis software selected must be able to read the objective lens magnification automatically at all times, thus ensuring high measurement accuracy through the elimination of manual entry of the erroneous objective lens magnification into the software.
In addition, the use of a manual or motorized XY scanning stage is essential to position the sample at the area of interest for subsequent analysis. Furthermore, the experimental setup includes a computer fulfilling the minimum system specifications of the camera and image-analysis software with a high-resolution monitor.
The 10x objective lens is selected and the XY stage is used to position the sample to view the area of interest being analyzed under reflected-light, brightfield conditions.
The image-analysis software is used to acquire the digital image by applying required filters to get accurate representation of the intercepts on the image. This capability is available as an interactive feature in most of the software packages, allowing the operators to observe the effects of the filter on the resulting intercepts.
The image analysis is performed as outlined in the chosen standard. The results are recorded directly into a spreadsheet within in the image-analysis software. Grain analysis is typically carried out over five random fields.
The aforementioned steps are repeated for five consecutive times. A report is generated automatically depending on the user’s pre-defined template, featuring the analysis results, supporting grains images, and associated data (Figure 6).
Figure 6. An example report
Modern material-science microscope image-analysis software minimizes human intervention in grain size analysis, thus producing accurate and reproducible results. Many software packages are available to perform grain analysis as per international standards such as ASTM E112.
Besides the minimal effort involved in implementation, these software packages can generate reports automatically depending on the analysis data and provide an integrated database for archiving for future reference.
This information has been sourced, reviewed and adapted from materials provided by Olympus Corporation of the Americas Scientific Solutions Group.
For more information on this source, please visit Olympus Corporation of the Americas Scientific Solutions Group.