AI-Assisted Composite Etch Model for MPT

For advanced semiconductor nodes, the demand for high-precision patterning of complex foundry circuits drives the widespread use of Lithography-Etch-Lithography-Etch (LELE)—a key Multiple Patterning Technology (MPT)—in Deep Ultraviolet (DUV) processes. However, the interaction between LELE’s two Lithography-Etch (LE) cycles make it very hard to build a model for etching contour simulation and hotspot detection. This study presents an AI-assisted composite etch model to capture inter-LE interaction, which directly outputs the final post-LELE etched contour, enabling Etch Rule Check (ERC)-based simulation detection of AEI hotspots. In addition, the etch model in this study can also be used in predicting the etching bias of different types of pattern (especially complex 2D patterns), which enables automatic re-targeting for ADI target generation. In the future, this Composite model’s framework can be adapted to the Self-Aligned Reverse Patterning (SARP) + CUT process to address more complex MPT challenges.