Scanning Electron Microscopy versus Transmission Electron Microscopy for Material Characterization: A Comparative...

Area of Science:

• Materials Science and Engineering
• Metallurgy
• Microscopy and Imaging Techniques

Background:

• Traditional microstructure characterization of quenched and tempered steels relies on Transmission Electron Microscopy (TEM) due to its high resolution.
• Scanning Electron Microscopy (SEM) offers advantages in cost, availability, and area coverage, with significant advancements in resolution.
• A need exists for cost-effective, high-statistical significance characterization methods in materials science.

Purpose of the Study:

• To compare the microstructure characterization capabilities of TEM and SEM electron channeling contrast techniques in high-strength steels.
• To evaluate the effectiveness of SEM, complemented by electron backscatter diffraction and deep learning, for quantitative microstructure analysis.
• To demonstrate a lower-cost, higher-statistical significance alternative for materials characterization.

Main Methods:

• Comparative analysis of quenched and tempered high-strength steel microstructures using TEM and SEM electron channeling contrast.
• Utilized electron backscatter diffraction for martensite size distribution and retained austenite fraction assessment.
• Employed SEM secondary imaging and a deep learning method for quantitative characterization of carbide precipitation (size, shape, distribution).

Main Results:

• Both TEM and SEM provided similar conclusions regarding the distribution of martensite laths/plates and nanoscale features like nanotwins and dislocations.
• Quantitative measurements of carbide precipitation down to nanometer scale using SEM and deep learning showed good agreement with TEM-based results.
• Electron backscatter diffraction effectively assessed martensite size and retained austenite fraction.

Conclusions:

• SEM electron channeling contrast techniques are a viable and effective alternative to TEM for characterizing quenched and tempered steel microstructures.
• The integration of SEM with electron backscatter diffraction and deep learning enables cost-effective, statistically significant quantitative analysis of nanoscale features.
• This study encourages the adoption of advanced SEM-based methods to reduce characterization costs and improve statistical significance in materials science.