Metal additives such as Ag+ and Cu2+ are used in the synthesis of Au nanoparticles to control the particle shape and size. For example, anisotropic Au nanobars can be grown from isotropic seeds Cu, whereas equivalent growth conditions without Cu produce isotropic Au nanocubes [1,2]. The mechanism by which Cu controls shape in Au nanoparticle growth remains a subject of intense debate, in part because the atomic location of Cu, and other additives, are unknown. Direct observations are needed.
Our Energy dispersive x-ray (EDX) mapping reveals the presence of Cu on the {100} surfaces of Au nanobars. However, the results are limited by the poor information resolution (subtly different from image resolution) due to scattering onto adjacent atomic columns [3]. Moreover, the high electron dose required to achieve meaningful statistics can induce structure transformation during acquisition. Other conventional STEM imaging modes, including high angle annular dark field (HAADF), bright field (BF) and annular bright field (ABF), do not show sufficient sensitivity to distinguish the Cu from possible morphological variations at the edges of the Au nanoparticles.
To overcome these limitations, we have identified dynamical diffraction conditions and scattering angles sensitive to the presence of Cu. Using these conditions and a fast electron microscope pixel array detector (EMPAD), we collected 4D-STEM datasets with the probe scanning across the edge of the particle. This allows the collection of full diffraction patterns in a relatively short time with lower dose than EDX. The image formed by integrating signals under these special dynamical diffraction conditions shows a bright monolayer on the surface, which by comparing with simulation can be attributed to the Cu atoms. We show that by integrating segments of the annular detector parallel and normal to the particle surface, we can decouple the contributions from the Cu and surface edge effect [4].