Enhancing overlay metrology accuracy via E-beam simulation and SEM target optimization

Accurate metrology is a critical component of effective process control, particularly in addressing the challenges of pattern miniaturization and increasing structural complexity in advanced semiconductor manufacturing. As semiconductor devices continue to scale down, the demand for tighter tolerances in critical dimension (CD) and overlay measurements has never been greater. Therefore, providing metrology results with high accuracy and sensitivity for device-scale structures has become indispensable. Enabling this need by Scanning Electron Microscopy (SEM), as discussed in previous studies, demonstrate the directly influences of image quality (particularly contrast and pattern edge resolution) on measurement precision and accuracy. These factors can be quantified using an imaging metrics, such as edge separation, which represents e-Beam resolution, follows a Gaussian-like distribution, where precision peaks at an optimal separation and diminishes when the edges are too close or too far apart. Similarly increased image contrast enhances precision up to a certain point, beyond which noise artifacts may degrade accuracy. Furthermore, sample size is a critical parameter while increasing sample counts initially reduces random errors and improve statistical reliability. Thus, optimizing resolution, contrast, and sampling count is essential for enhancing SEM overlay measurement accuracy and minimizing metrology errors.