Maskless Electron Beam-Induced Etching of Diamond in Air: A Secondary Electron-Driven Mechanism

Abstract

We report a direct, maskless electron beam-induced etching (EBIE) process for diamond in air, enabling high-precision patterning without lithography or plasma processing. Through a comprehensive analysis of electron-gas, electron-diamond, and gas-surface interactions in the SEM environment, we demonstrate that etching is predominantly governed by low-energy secondary electrons, which drive gas dissociation and radical generation. The resulting oxygen- and nitrogen-based radicals chemisorb on the diamond surface, form volatile carbon-containing species, and desorb under continued electron irradiation, enabling controlled material removal. The process exhibits two distinct regimes: a molecule-limited regime governed by gas flux and an electron-limited regime controlled by current density. Etch depths up to 212 nm and lateral resolution down to 200 nm are achieved. Time-dependent anisotropy is observed, with (100) surfaces transitioning to (111)-faceted morphologies, enhancing etch yield. These results establish a general secondary electron-driven mechanism for EBIE in gas environments, providing a maskless, damage-free nanofabrication route for diamond semiconductor and other chemically inert materials.

Figures (10)

• Figure 1: Schematic of the EBIE setup in an SEM chamber. Air is introduced with a total throughput $J_{\mathrm{tot}}$ and directed via a nozzle to produce a local impinging flux $J_i$ toward the diamond surface. An electron beam irradiates the surface under controlled chamber pressure ($10^{-6}$--$10^{-4}$ mbar), maintained by continuous pumping.
• Figure 2: Dependence of etching yield (black circles) and secondary electron SE emission yield (red squares) on primary electron energy. Etching conditions: $2.5 \times 2.5~\mu\text{m}^2$ square scanning area, $10^{-4}$ mbar air, 30 min exposure. Error bars on the etching yield reflect variations associated with the surface roughness of unetched diamond. Secondary electron yield data are reproduced from Ref. ascarelli2001secondary. The observed correlation between SE yield and etching efficiency highlights the influence of secondary electrons in the EBIE process.
• Figure 3: Effect of nozzle-to-beam distance $x_s$ on etching pattern size. (a) Schematic of the EBIE setup with the nozzle positioned at a 30$^\circ$ angle relative to the diamond surface. The origin $x_s = 0$ corresponds to a distance $D = 424\,\mu\mathrm{m}$ from the nozzle projection onto the sample plane. (b) Dependence of pattern size and relative impinging flux on $x_s$. Flux values are obtained from geometric simulation under free molecular flow conditions. Etching conditions: spot scanning, 5 keV electron energy, 0.935 pA current, $10^{-4}$ mbar air, 2 min exposure. The strong correlation between pattern size and impinging flux highlights the spatial dependence of EBIE under these conditions.
• Figure 4: Effect of nozzle-to-beam distance $x_s$ on etching pattern dimensions. SEM images of etched patterns at different $x_s$ positions. Etching conditions: spot scanning, 5 keV electron energy, 0.935 pA current, $10^{-4}$ mbar air, 2 min exposure. Increasing $x_s$ leads to larger, more elongated, and shallower patterns, reflecting the decrease in local gas flux from deeper, localized etching near the nozzle to more diffuse etching at larger distances.
• Figure 5: Etching yield as a function of primary electron current density. Etching conditions: 2.5 $\mu$m $\times$ 2.5 $\mu$m scanned area, 3 keV electron energy, $10^{-4}$ mbar air, 5 min exposure. Error bars reflect surface roughness variations of unetched regions. The etching yield exhibits two plateau regions, indicating a transition between desorption-limited regimes as the current density increases.