STXM data showing NEXAFS spectra at the oxygen edge for unexposed and exposed HSQ (each normalized to taken from a blank window) and the difference spectrum.
Left side: plot of of STXM images of latent e-beam exposure of squares in HSQ. Images are obtained at (A), (B), (C), and (D). The highest contrast is observed at . Right side: the corresponding line profiles from the vertically averaged signal across the squares.
Raw STXM images of features as a function of dose (A) and a plot of (B). [For these features: dose , i.e., dose at ]. There is a significant increase in the background optical density as the feature dose increases. The discontinuity in the graph is where STXM data were not taken.
Plot of of e-beam exposed features as a function of dose and feature size. Each feature is normalized to the same . The chemical contrast increases with increasing dose as expected but has an additional dependence on the size of feature exposed. Features sized were exposed at two pitches: (A) 1.25 and (B) .
AFM measurements of latent image feature thicknesses relative to background. The heights change as a function of dose and feature size.
AFM image of latent (A) and developed (B) features on silicon nitride. Dose is . The feature (a) shows that thickness changes occur outside of the exposed area. The feature (b), an “X” is an artifact of how the e-beam is raster scanned. Features (c) and (d) are explained in the text. As the beam spirals from outside to inside, errors in placement result in lower dose as the beam passes the corners and center of the fill pattern.
Plot of for the ( pitch) features and the background between the features. AFM images [insets (A) and (B)] are taken after development. The lines with a lower mOD compared to background areas (areas between the lines exposed at much higher doses) leave thicker developed HSQ features.
Article metrics loading...
Full text loading...