The classical theory of Cabrera–Mott describes passivation film formation on metals, where they predicted that this film grows as a uniform layer due to a field-enhanced ionic transport mechanism. Here we present experimental evidence, based on in situ transmission electron microscopy of copper oxidation, that the passivation film nucleates and grows as oxide islands, not as a uniform layer. We propose an alternative phenomenological theory to describe passivation film formation, based on island growth followed by coalescence.

We report the in situ microscopy observation of an unnatural phase of Ni, a highly strained hexagonal close-packed ͑hcp͒ form which we believe is stabilized by heteroepitaxial growth on the ͑001͒ face of MgO. We find that the nanosized hcp nickel islands transform into the normal face-centered cubic structure when the size of the islands exceeds a critical value ͑about 2.5 nm thick with a lateral size of ϳ5 nm͒. The structural transition proceeds via a martensitic change in the stacking sequence of the close-packed planes. The formation of hcp Ni nanostructures with an unusually large crystallographic c / a ratio ͑ϳ6% larger than ideal hcp͒ is very interesting for spintronic and recording applications where large uniaxial anisotropies are desirable.

We have observed the formation of heteroepitaxial interfacial layers between silver nanoparticles and a single crystal copper surface by a phenomenon we term “contact epitaxy.” Upon depositing Ag nanoparticles (5–20 nm diameter) onto clean (001) Cu in an ultrahigh vacuum in situ transmission electron microscope, a thin (111)-oriented layer of Ag was detected at the interface between the substrate and particles. Molecular dynamics simulations reveal that the epitaxial layers form within picoseconds of impact, with rapid alignment arising from mechanical relaxation of the highly stressed interface formed upon initial contact. The simulations also show that multiple grains form in the nanoparticle as a consequence of this relaxation process. The unique structure of the nanoparticles, induced by contact epitaxy, is expected to significantly influence physical properties such as interfacial bonding, diffusion, chemical activity, and electrical transport, as well as forming a nucleus for grain growth and epitaxy which we also observe. Due to its simple origin, the phenomenon should also apply to materials systems beyond the field of nanoparticles with implications for cluster deposition, adhesion, rheology, and catalysis.

Metal nanoparticles can display a unique behavior when deposited on substrates with a significantly lower surface energy. Co nanoparticles in the 10 nm size regime burrow into clean Cu(100) and Ag(100) substrates when deposited at 600 K and also assume the substrate orientation. Deposition at room temperature fails to show either burrowing or reorientation. Crucial in understanding these results are the capillary forces and surface tension associated with a nanoparticle: they must be high enough to drive atoms away from underneath the cluster.