Modeling and Optimization of Remote Plasma-Assisted Photoresist Stripping in High-Aspect-Ratio Structures

As advanced semiconductor packaging continues to evolve toward finer and denser integration, the removal of photoresist from structures with high-aspect-ratio (HAR) features such as deep trenches and blind vias remains a critical challenge. Conventional wet chemical cleaning and plasma cleaning processes often suffer from the limitations in insufficient penetration and inadequate reactivity at the feature bottom. To address these limitations, this study develops a two-dimensional axisymmetric simulation model in COMSOL Multiphysics to evaluate the combined effects of structural optimization and bias voltage on downstream plasma behavior. Results indicate that reducing the aspect ratio from 1.5 to 1.15 leads to significant improvement of plasma uniformity, an increase in the bottom electron density to over 1.4×10¹⁶ m⁻³ and stabilized the electron temperature around 1.8 eV. With an applied bias voltage of 10 V, the temperature gradient across the trench is reduced from 1.4 eV to less than 0.2 eV, enhancing vertical ion transport and minimizing thermal imbalance. The results also suggest that the combination of structural adjustment and bias control may offer synergistic benefits in improving plasma distribution and enhancing activity at the trench bottom. These findings provide a valuable reference for design of the plasma system and optimizing remote plasma photoresist removal processes in high-aspect-ratio structures.