Abstract
Optical metrology plays a vital role in a wide range of research and inspection areas in the industry. At present, the market offers a variety of optical metrology instruments, among which the focus variation microscope stands out for its capability of measuring steep surfaces with high slopes. The traditional focus variation (FV) instrument mechanically scans the surface by sweeping the focal plane of the objective lens using linear motion stages and simultaneously capturing images at different scanning positions, forming a stack of images. The mechanical motion stages require regular maintenance and calibration to ensure accuracy over time. Another issue associated with the mechanical scanning methods is their physical size, which creates a limiting factor to compactness for in-situ measurement applications. This work proposes a chromatic focus variation (CFV) method that replaces mechanical scanning with a wavelength scanning mechanism to overcome the above limitations. Unlike traditional focus variation, the CFV system employs a dispersive objective lens (i.e. chromatically aberrated objective lens) to axially shift the focus along the optical axis to provide vertical/depth scanning. This approach brings significant enhancements in measurement speed and reduces the instrument size for on-machine metrology tasks. In this paper, a detailed analysis of the optical performance of the dispersive objective lens is conducted, and then the measurement performance of the proposed CFV system is validated using samples including a step height of 30 μm and a sine wave shape with a peak-to-valley amplitude of 19 μm. The experiment results were compared to those from the state-of-the-art commercial instrument (Alicona G5), which showed a good agreement between the two. Furthermore, a detailed analysis and discussions are provided to investigate the measurement's accuracy.
Original language | English |
---|---|
Pages (from-to) | 35527-35541 |
Number of pages | 15 |
Journal | Optics Express |
Volume | 32 |
Issue number | 20 |
Early online date | 18 Sep 2024 |
DOIs | |
Publication status | Published - 23 Sep 2024 |