Diffraction-limit-breaking digital projection lithography via multi-exposure strategies for high-density nanopatterning

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Digital micromirror device-based maskless projection lithography (DMPL) has emerged as a transformative nanopatterning platform, yet faces fundamental challenges in resolving high-density nanostructures due to optical diffraction constraints. The filtering characteristics of the pupil function and the phase modulation caused by defocus lead to pattern distortion and loss of high-frequency information, making it difficult to ensure consistency and distinguishability in dense pattern lithography. Here, we introduce multiple exposures (ME) to DMPL, developing a multiple-exposure without alignment solution to solve the problem of indistinguishable dense pattern exposure. At the same time, we also propose a novel method to maximize the spacing uniformity of the layout decomposition scheme based on the gradient descent algorithm. We further improved the depth of focus by maximizing the uniformity of spacing, achieving decoherence of sub-layouts within the laser field, thereby ensuring the printability of dense patterns. Simulations and experimental results verify that this approach significantly expands the process window and improves lithographic resolution. For one-dimensional dense lines, the minimum resolvable period decreases from 378 nm (half pitch~0.5λ/NA) to 223 nm (half pitch~0.3λ/NA). Furthermore, the metal layer layout of the integrated circuit chip has a minimum gap of 1 DMD pixel (75.6 nm in image plane) is achieved, demonstrating the efficacy of the proposed methods in addressing the challenges posed by dense layouts in DMPL applications. This work provides a paradigm-shifting solution to the persistent "density-fidelity trade-off" in digital nanofabrication, charting new pathways toward next-generation functional meta-devices.

Article activity feed