Gap-matching algorithm with the impCEEMDAN in scanning white-light interference microscopy.

Area of Science:

• Optical Metrology
• Surface Characterization
• Signal Processing

Background:

• Coherence scanning interferometry (CSI) is widely used for 3D surface profiling.
• Batwing artifacts and ghost steps are significant challenges in CSI, limiting measurement accuracy.
• Existing methods often require prior knowledge of sample geometry, which is not always feasible.

Purpose of the Study:

• To develop a novel algorithm for addressing batwing artifacts and ghost steps in CSI.
• To enable accurate 3D surface profile measurements without requiring prior knowledge of the sample's geometry.
• To improve the reliability and applicability of CSI for micro-component analysis.

Main Methods:

• A gap-matching algorithm was developed and integrated with improved complete ensemble empirical mode decomposition with adaptive noise (impCEEMDAN).
• The impCEEMDAN technique was employed for advanced signal processing to decompose and filter interferometric data.
• The proposed method was evaluated using micro-components with known step heights (500 nm, 1200 nm) and a 10 µm standard step.

Main Results:

• The proposed gap-matching algorithm effectively suppressed the batwing effect in CSI measurements.
• Ghost steps, a common artifact in CSI, were successfully eliminated by the developed approach.
• Simulations and experimental results demonstrated good accuracy and repeatability for the method.

Conclusions:

• The novel gap-matching algorithm combined with impCEEMDAN provides an effective solution for common obstacles in CSI.
• This approach enhances the accuracy and reliability of 3D surface profiling, particularly for micro-components.
• The method's ability to perform measurements without prior geometric knowledge broadens its practical applications in metrology.