Formation of bimetallic clusters in superfluid helium nanodroplets analysed by atomic resolution electron tomography

Metallic nanoparticles consisting of a few thousand atoms are of large interest for potential applications in different fields such as optics, catalysis or magnetism. Structure, shape and composition are the basic parameters responsible for their properties. To reveal these parameters in three dimensions at the nanoscale, electron tomography is a powerful tool. Advancing electron tomography to atomic resolution in an aberration-corrected transmission electron microscope is challenging though [1–3], and the ultimate goal of resolving position and type of each single atom inside a material remains elusive. Here we demonstrate atomic resolution electron tomography on silver/gold core/shell nanoclusters grown in superfluid helium nanodroplets [4,5] (Figure 1a). Superfluid helium droplets represent a versatile, novel tool for designing such nanoparticles, allowing fine-tuned synthesis of pure or composite clusters for a wide range of materials. Using ultra-high vacuum conditions and getting on without solvents or additives compared with chemical synthesis, the method delivers high purity materials, which can be well controlled in terms of size and composition. Analytical TEM investigations reveal smaller clusters mainly consisting of a single silver core, surrounded by a gold shell, whereas larger clusters contain two or more silver grains embedded in a gold matrix [6] (Figure 1b&c). The measured transition between single- and double-core growth appears at a cluster size of about 5000 atoms and similar values are found when cluster agglomeration inside the He-droplet is simulated for the used process parameters [7] (Figure 1d). One cluster with two silver cores was analysed three-dimensionally for its atomic structure, shape and composition [6] (Figure 2). We identify gold- and silver-rich regions in three dimensions and we are able to estimate atomic positions inside the nanocluster. Two silver cores are visible as darker regions, separated by gold with a minimal thickness of 2–3 atomic layers. The cluster appears in a highly symmetric multiply twinned structure, shaped roughly as an icosahedron, which is structurally modified due to binding of the cluster to the surface. This work demonstrates estimation of atomic positions within nanoparticles in 3D without any prior information, while at the same time information about the local elemental composition is retrieved at near-atomic resolution. Our results give insight into the growth and deposition process of composite nanoclusters created in superfluid helium droplets. This understanding will allow fine-tuning of process parameters for optimizing nanoparticle properties.