Scanning in coherent Fourier scatterometry using Fresnel zone plate projection on a digital micromirror device

Journal Article (2026)
Author(s)

S. Soman (TU Delft - Applied Sciences)

E.D.M. Özdemir (De Haagse Hogeschool)

S.F. Pereira (TU Delft - Applied Sciences)

Research Group
ImPhys/Pereira group
More Info
expand_more
Publication Year
2026
Language
English
Research Group
ImPhys/Pereira group
Journal title
Measurement Science and Technology
Volume number
37
Downloads counter
19
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We present a digital micro-mirror device (DMD)-enabled scan head for coherent Fourier scatterometry (CFS) that performs lateral scanning without macroscopic moving parts, while maintaining a diffraction-limited probe. A binary Fresnel zone plate (FZP) is displayed and translated on the DMD to steer a single focused spot across the sample, providing an electronically programmable alternative to scanning using piezo-based translation devices. To the best of our knowledge, this is the first published CFS implementation in which a DMD is the primary lateral scanning element. Furthermore, the DMD programmability is used to compensate for the scan-position-dependent aberrations using an iterative optimisation algorithm. Across a 400 × 200 DMD-pixel scan area, the peak-intensity coefficient of variation improves from 39.4% (uncorrected) to 16.1% (after correction) and to 4.47% with additional power normalisation, demonstrating substantially improved probe uniformity. Finally, we demonstrate particle detection on a Si wafer with 1 µm polystyrene latex particles, achieving an signal-to-noise ratio of 16.04 ± 1.11 dB. The results establish DMDFZP scanning with integrated aberration correction as a compact, fast, and scalable CFS architecture, with a clear pathway to higher throughput via multi-spot parallelisation.