Atomic-scale effects behind structural instabilities in Si lamellae during ion beam thinning

TL;DR

This study combines atomistic simulations and experiments to reveal atomic-scale effects that limit the thinning of silicon lamellae below 26 nm using focused ion beam techniques, explaining the success of alternative methods.

Contribution

It identifies the atomic-scale mechanisms causing thickness limitations in FIB thinning and validates an alternative approach that overcomes these physical constraints.

Findings

Conventional FIB thinning cannot produce lamellae thinner than 26 nm at 30 keV due to atomic effects.

Atomic-scale effects at the lamella edges cause shrinkage during ion beam milling.

An alternative FIB method avoids these limitations by eliminating the physical processes responsible.

Abstract

The rise of nanotechnology has created an ever-increasing need to probe structures on the atomic scale, to which transmission electron microscopy has largely been the answer. Currently, the only way to efficiently thin arbitrary bulk samples into thin lamellae in preparation for this technique is to use a focused ion beam (FIB). Unfortunately, the established FIB thinning method is limited to producing samples of thickness above ~20 nm. Using atomistic simulations alongside experiments, we show that this is due to effects from finite ion beam sharpness at low milling energies combined with atomic-scale effects at high energies which lead to shrinkage of the lamella. Specifically, we show that attaining thickness below 26 nm using a milling energy of 30 keV is fundamentally prevented by atomistic effects at the top edge of the lamella. Our results also explain the success of a recently proposed alternative FIB thinning method, which is free of the limitations of the conventional approach due to the absence of these physical processes.