Sub-10-nm lithography with light-ion beams

Scanning-electron-beam lithography (SEBL) is the workhorse of nanoscale lithography in part because of the high brightness of the Schottky source of electrons, but also benefiting from decades of incremental innovation and engineering of apparatus around the Schottky source. Light ions are an attractive intermediary between electrons and heavy ions in terms of exposure efficiency and resolution by attaining a minimal interaction volume within the resist layer, if only we had bright sources of these light ions and could thus achieve small spot sizes. In this thesis, I present sub-10-nm lithography at high exposure efficiency using the gas field ionization source (GFIS) with helium and neon ions. I also present preliminary results using the magnetooptical trap ion source (MOTIS) with lithium ions. This work has also challenged the understanding of exposure efficiency as directly proportional to the so-called stopping power of incident beam particles — i.e. the average energy loss per unit path length, particularly for thin (less than 20 nm thick) resist. Values of stopping power are readily obtained via the popular Stopping and Range of Ions in Matter (SRIM) software for a variety of beam species and target materials at various landing energies, making this metric particularly convenient for predicting exposure efficiency. However, the exposure efficiency of neon ions for thin hydrogen silsesquioxane (HSQ) resist on bulk silicon is similar to that of gallium ions at 20-30 keV landing energy despite SRIM indicating a much larger stopping power for the gallium ions. Separating stopping power into nuclear and electronic components reveals that both the neon and gallium ions have similar electronic stopping powers. This correspondence points to electronic stopping power as a better indication of exposure efficiency in ion beam lithography. Unfortunately, the use of electronic stopping power alone to predict exposure efficiency has too been challenged by the data. Whereas the exposure efficiencies of neon and gallium ions were much higher than that of helium ions for the landing energies studied, the electronic stopping powers were all similar. One interpretation