Experimental and numerical investigation of nanoparticle removal using acoustic streaming and the effect of time

Abstract

The removal of nanoparticles is becoming increasingly challenging as the minimum linewidth continues to decrease in semiconductor manufacturing. In this paper, the removal of nanoparticles from flat substrates using acoustic streaming is investigated. Bare silicon wafers and masks with a 4 nm silicon cap layer are cleaned. The silicon-cap films are used in extreme ultraviolet masks to protect Mo-Si reflective multilayers. The removal of 63 nm polystyrene latex (PSL) particles from these substrates is conducted using single-wafer megasonic cleaning. The results show higher than 99% removal of PSL nanoparticles. The results also show that dilute SC1 provides faster removal of particles, which is also verified by the analytical analysis. Particle removal from the 4 nm Si-cap substrate is slightly more difficult as compared to bare silicon wafers. The experimental results show that the removal of nanoparticles takes a relatively long removal time. Numerical simulations showed that the long time is due to particle oscillatory motion and redeposition, and that this phenomenon is not observed in the removal of sub-mu m or larger size particles.