Accurate process window determination for EUV lithography using large field of view eBeam metrology and stochastic-aware modeling

Extreme ultraviolet (EUV) lithography offers an alternative to deep ultraviolet (DUV) methods by enabling singlepatterning processes, simplifying the complexities of multiple-patterning techniques such as Self-Aligned Quadruple Patterning (SAQP) and Litho-Etch-Litho-Etch (LELE). However, EUV's increased sensitivity to stochastic variations necessitates more precise process window characterization than traditional critical dimension (CD) or defect-based methods can provide. This paper presents an advanced approach for EUV process window (PW) analysis, combining large field of view (FOV) scanning electron microscopy (SEM) with a process window metrology (PWM) analysis software that incorporates stochastic-aware modeling for feature failure predictions. We demonstrate the effectiveness of this integrated solution by evaluating the process windows of three Optical Proximity Correction (OPC) designs for an EUV-processed DRAM device featuring both 2D island patterns in the device array and 1D line patterns in the periphery simultaneously. Our results show that while traditional CD-based methods fail to differentiate between OPC designs, PWM identifies significant differences, highlighting the most effective OPC designs with the largest process window. This approach offers a reliable, feature-level process window analysis, providing actionable insights for optimizing EUV lithography performance.