A white-light interferometry method for 3D measurement of compactly spaced micro-nano structural units.

• We analyze the problem areas and distorted signals that can arise when measuring CSMU using basic white-light interferometry. We specifically focus on addressing the structural areas that are prone to errors, ensuring reliable processing results and improving efficiency. • In the signal processing of the problem areas, we employ characteristic analysis of the distorted signal. By filtering out the disturbed part of the signal and extracting the desired portion containing accurate height information, we are able to reconstruct the 3D topography. • This algorithm can be applied to any traditional optical system utilizing white-light interferometry. It is a simple and accurate approach that only requires a standard experimental setup, making it highly practical. Micro-nanostructures refer to technologies that operate at the nanoscale and provide materials and devices with unique mechanical, physical, and chemical properties. These structures have diverse applications in fields such as microelectronics, biotechnology, aerospace, and metamaterials. As precision machining technology advances, the demand for smaller sizes and higher accuracy of structural units has increased, leading to more compact integration. Consequently, the lossless and high-precision 3D measurement of highly integrated micro-nanostructure components has become increasingly challenging. In this study, we analyzed measurement errors in compactly spaced microstructural units using white-light scanning interferometry and proposed an algorithm to optimize this technology. By combining coherent peak-histogram processing with signal correlation analysis, we were able to improve the measurement accuracy by approximately ten times compared to current mainstream algorithms. To validate our algorithm, we performed experiments using metasurfaces with structural unit spacing at the order of sub-micrometer. The results obtained demonstrated the effectiveness of our algorithm in achieving accurate measurements. [ABSTRACT FROM AUTHOR]