“…As a result, if the SiO 2 is <500 nm, a corner-preferred plasma etching will ruin the profile of the mask during a long time etching. This nonuniform electron-attraction model may differ a little with that for photoresist mask, 16,27,31 where the Ar + can penetrate into the photoresist and become a dominant factor. Interestingly, although the profile of SiO 2 mask has deteriorated completely in Fig.…”