Structure Matters – Direct In‐situ Observation of Cluster Nucleation at Atomic Scale in a Liquid Phase

Trond R. Henninen,Debora Keller,Rolf Erni

doi:10.1002/cnma.202000503

Abstract

AbstractThe formation of almost all solid crystalline materials starts with a nucleation reaction, where a handful of atoms come together to form the initial crystal seed, which then grows to a larger crystal. Thus, understanding and controlling nucleation is essential for the synthesis and manufacturing of most material systems, especially nanomaterials. Despite this, little is known from direct experimental observations about the initial steps of nucleation, the formation of sub‐nanometer sized clusters. Here, we directly study the atomic nucleation reactions of such sub‐nm clusters of Pt in‐situ in a liquid phase. We do this by inducing nucleation in suspended nanofilms and supported nanodroplets of an ionic liquid (1‐butyl‐3‐methyl imidazolium chloride, 5–50 nm thickness), which is observed at atomic resolution by scanning transmission electron microscopy. We can observe Pt atoms to nucleate into few‐atom clusters, which coalesce and grow into cluster agglomerates or nanoparticles, or redissolve. When comparing nucleation in nanofilms and carbon‐supported nanodroplets, nucleation is rarely observed in nanofilms, while easily observable at high nucleation rate in nanodroplets. This is due to the presence of the pre‐existing liquid‐solid interface, resulting in heterogeneous nucleation in the nanodroplets while there is only homogeneous nucleation in the nanofilms. Ultimately, our results show that the nucleation pathways of nanoparticles are not just determined by the local chemical environment, but are also influenced by size and structure of the initially formed clusters.

Highlights

Our results show that the nucleation pathways of nanoparticles are not just determined by the local chemical environment, but are influenced by size and structure of the initially formed clusters
When metals solidify from a melt and form a grained microstructure, the grain formation starts with a liquid-to-solid nucleation reaction, and the
The reason for this is that theoretical models, such as the Classical Nucleation Theory (CNT), can only qualitatively model nucleation-type phase transitions, but fail at modelling observable quantities such as nucleation rates.[4,7,8,9]

Summary

Science and Technology

CNT predicts that there is a lower energy barrier for heterogeneous nucleation (ΔGhet) on a pre-existing liquid-solid interface, than for homogeneous nucleation without an interface (ΔGhomo).[10] For many materials, such as Pt or Au, the size of the smallest stable nanoparticles is about 1–1.5 nm (or ca 50–100 atoms), but even smaller particles can be stabilized at higher temperatures or by adding molecules interacting with the surface of the clusters.[11,12,13] More recent nucleation modelling considers the dynamics of discrete atoms in cluster formation, by using. Heterogeneous nucleation reactions in nanodroplets: Nanodroplets on the amorphous carbon film with typical diameters below 100 nm can often be seen forming along the edge of larger microdroplets after a short bake-out of the membranes that is needed to remove residual hydrocarbon contaminants These nanodroplets form due to the phenomenon known as the Coffee-ring effect.[40] As the heating causes a significant amount of ionic liquid to evaporate from the edge of Figure 2. These nanodroplets are highly suitable as nanoreactors for inducing nucleation in a controlled manner

In the current data of movies of nanodroplets ranging ca

Conclusions

Conflict of Interest