A rasterization method for generating exposure pattern images with optical maskless lithography

Abstract

One of the core technologies used for high-density integration of electronic parts is a lithography method that draws a microcircuit by irradiating light onto photoresist. Because maskless lithography technology that employs a digital micromirror device (DMD) uses continuous exposure patterns, a huge amount of image frame data is required, and, moreover, a long calculation time is required to create exposure pattern images. In this paper, we propose a rasterization method that exposes continuous image frames onto a single overlay lithography image using GPU-accelerated path rendering and CPU parallel computing, creating a high-quality exposure pattern image at a high speed. This method enables effective rasterization using multiple CPU and GPU cores to create lithography image data. Additionally, it is possible to reduce the size of the data file by exposing a printed circuit board (PCB) in a single overlay image frame, rather than the conventional consecutive frames.