Nanoscale dynamics during self-organized ion beam patterning of Si: II. Kr$^+$ Bombardment

TL;DR

This study investigates the nanoscale dynamics of silicon surface patterning under 1 keV Kr$^+$ ion bombardment, revealing similarities and differences with Ar$^+$ patterning, and introduces real-time measurement of ripple motion via speckle analysis.

Contribution

It demonstrates that ion-enhanced viscous flow relaxation is similar for Kr$^+$ and Ar$^+$, and develops a method to measure ripple motion and surface inhomogeneities in real time.

Findings

Surface curvature dependent roughening rate is larger for Kr$^+$ than Ar$^+$.

Fluctuation dynamics show a peak in correlation time at ripple length scale.

Speckle motion analysis enables real-time measurement of ripple velocity and erosion inhomogeneities.

Abstract

Despite extensive study, fundamental understanding of self-organized patterning by broad-beam ion bombardment is still incomplete and controversial. Understanding the nanopatterning of elemental semiconductors, particularly silicon, is both foundational for the broader field and of intrinsic scientific and technological interest itself. This is the second component of a two-part investigation of the kinetics and fluctuation dynamics of self-organized nanoscale ripple development on silicon during 1 keV Ar$^+$ (Part I) and Kr$^+$ bombardment. Here, it's found that the ion-enhanced viscous flow relaxation is essentially equal for Kr$^+$-induced patterning as previously found for Ar$^+$ patterning. The magnitude of the surface curvature dependent roughening rate in the early stage kinetics is larger for Kr$^+$ than for Ar$^+$, qualitatively consistent with expectations for erosive and mass redistributive contributions to the initial surface instability. As with the Ar$^+$ case, fluctuation dynamics in the late stage show a peak in correlation time at the length scale corresponding to the dominant structural feature on the surface -- the ripples. Analogy is made to the phenomenon of de Gennes narrowing in liquids, but significant differences are also pointed out. Finally, it's shown that speckle motion during the surface evolution can be analyzed to determine spatial inhomogeneities in erosion rate and ripple velocity on the surface. This allows the direction and speed of ripple motion to be measured in real time, a unique capability for measuring these fundamental parameters outside the specialized environment of FIB/SEM systems.