a b s t r a c tThe dynamics of oxidation of cobalt nanoparticles were directly revealed by in situ environmental transmission electron microscopy. Firstly, cobalt nanoparticles were oxidized to polycrystalline cobalt monoxide, then to polycrystalline tricobalt tetroxide, in the presence of oxygen with a low partial pressure. Numerous cavities (or voids) were formed during the oxidation, owing to the Kirkendall effect. Analysis of the oxides growth suggested that the oxidation of cobalt nanoparticles followed a parabolic rate law, which was consistent with diffusion-limited kinetics. In situ transmission electron microscopy allowed potential atomic oxidation pathways to be considered. The outward diffusion of cobalt atoms inside the oxide layer controlled the oxidation, and formed the hollow structure. Irradiation by the electron beam, which destroyed the sealing effect of graphite layer coated on the cobalt surface and resulted in fast oxidation rate, played an important role in activating and promoting the oxidations. These findings further our understanding on the microscopic kinetics of metal nanocrystal oxidation and knowledge of energetic electrons promoting oxidation reaction.

SummaryIn this study, size‐selected platinum (Pt) nanoclusters were imaged with aberration‐corrected scanning transmission electron microscopy in high‐angle annular dark field (HAADF) mode. For image analysis, a relatively simple macro program was developed by making the use of existing ImageJ plug‐ins. The macro allows effectively for assessing criterions chosen for intensity threshold and filter blurring factors. It can extract the integrated HAADF intensity, peak intensity and projected area of the clusters. Here, the effects of magnification and objective lens defocus on nanocluster measurement were investigated. It was found that the integrated HAADF intensity of Pt clusters is a more robust sample descriptor than the peak intensity and the projected area. The macro program developed is freely available.Lay DescriptionMeasuring precisely the size of nanoclusters plays an important role in the investigation of nanocluster‐based material systems. Aberration‐corrected scanning transmission electron microscopy (STEM) is one of the most powerful tools to extract the size of clusters directly from their images. In this study, we developed a macro program based on existing ImageJ plug‐ins, allowing easy‐assessment of criterions chosen for image intensity threshold and filter blurring factors. It can be used to extract the integrated intensity, peak intensity, and projected area of the clusters for size determination. Using the program, we investigated the effects of magnification and objective lens defocus on measurements performed on size‐selected platinum (Pt) nanoclusters, and found that the integrated intensity of Pt clusters is a more robust sample descriptor than the peak intensity and the projected area. The macro developed allows a rapid assessment of factors affecting the accuracy with which size information can be obtained from clusters.