Controlled Positioning of Large Interfacial Nanocavities via Stress‐Engineered Void Localization

Abstract

The Kirkendall effect has intensively been exploited at the nanoscale for the fabrication of hollow nanostructures. [1,2] This fabrication strategy generally starts from a core–shell nanostructure where the core material is a faster diffusing species. Preferred outward atomic diffusion across the interface leads to a net inward injection of vacancies, which are likely to accumulate, supersaturate, and finally condense into a single void. If the core material is only partly consumed, a core–void–shell nanostructure can be produced. [3–7] Investigations on the formation of this ‘‘intermediate’’ state are crucial for a deep understanding of the vacancy coalescence and void evolution process induced by the nanoscale Kirkendall effect. Moreover, nanostructures with controlled porosity would enable greater control of the local chemical environment, which is important for molecule probing, drug delivery and catalysis applications. [2–4] Generally the nucleation of Kirkendall voids is favored at a core–shell interface due to the high defect content and stress there. In most cases of spherical core–shell nanoparticles, uniformly distributed interfacial voids are divided by filamentlike bridges, which act as fast transport paths for the delivery of remaining core material into the shell. [2a,2f,4] In materials with one or more macroscopic dimensions such as nanowires and films, the nanoscale Kirkendall effect usually initiates the formation of multiple voids randomly scattered at the interface during the diffusion evolution. [5] Void localization has been noticed in some systems. [5b,6,7] For example, an asymmetric