Phase-Controlled Synthesis of Ru Nanocrystals via Template-Directed Growth: Surface Energy versus Bulk Energy

Abstract

Despite the successful control of crystal phase using template-directed growth, much remains unknown about the underlying mechanisms. Here, we demonstrate that the crystal phase taken by the deposited metal depends on the lateral size of face-centered cubic (fcc)-Pd nanoplate templates with 12 nm plates giving fcc-Ru while 18−26 nm plates result in hexagonal closed-packed (hcp)-Ru. Although Ru overlayers with a metastable fcc-(high in bulk energy) or stable hcp-phase (low in bulk energy) can be epitaxially deposited on the basal planes, the lattice mismatch will lead to jagged hcp-(high in surface energy) and smooth fcc-facets (low in surface energy), respectively, on the side faces. As the proportion of basal and side faces on the nanoplates varies with lateral size, the crystal phase will change depending on the relative contributions from the surface and bulk energies.

Phase-Controlled Deposition of Ru on Pd Nanocrystal Templates: Effects of Particle Shape and Size

“…There is a clear trend for the template size in affecting Ru deposition on octahedral Pd nanocrystals, with small ones resulting in the formation of fcc-Ru shell while larger ones give hcp-Ru shell. The observed dependence of crystal phase on both particle shape and size can be rationalized by the symmetry match between the facets of these two systems that allows for the deposition of either hcp - or fcc-Ru on {111} facets, with the outcome determined by the balance between bulk and surface energies as a function of particle size. ,, …”

“…Consequently, it is possible to deposit either fcc - or hcp-Ru overlayers on fcc - {111} facets. The preference for either crystal phase is likely dependent on the template’s size and thus proportions of edges and side faces. , When an fcc-Ru shell is deposited on an fcc-Pd octahedral template, the matching in crystal structure between the core and the shell can give rise to smooth overlayers on both side faces and edges (Figure S1A). ,,, Thus, the resulting particle should have a low surface energy (due to smooth edges), but a high bulk energy (due to the formation of metastable fcc-Ru). ,, This crystal phase is favorable when a high percentage of the total surface is contributed by edges, such as in the case of templates with smaller sizes.…”

“… ,,, Thus, the resulting particle should have a low surface energy (due to smooth edges), but a high bulk energy (due to the formation of metastable fcc-Ru). ,, This crystal phase is favorable when a high percentage of the total surface is contributed by edges, such as in the case of templates with smaller sizes. In contrast, when an hcp-Ru shell is deposited on an fcc-Pd octahedral template, the surface energy of the edges will be increased. , This high surface energy can be ascribed to the inherent misalignment between the ABAB stacking sequence characteristic of hcp packing and the ABCABC stacking of the fcc lattice, causing the edges to be jagged and full of undercoordinated atoms, although hcp-Ru can smoothly cover the fcc-{111} side faces (Figure S1B). , Consequently, the high surface energy (due to jagged edges) and low bulk energy (due to the formation of thermodynamically stable hcp-Ru) would make the deposition of an hcp-Ru shell more favorable on larger templates, where only a small percentage of the total surface area is contributed by the edges.…”

“…For core–shell nanocrystals, we do not expect to observe separate XRD peaks with strong intensities for the core and the shell. ,,, Especially in the case where Pd and Ru adopt the same crystal phase, the observed peaks should be an average of the peaks corresponding to the core and the shell, with relative contributions depending on the shell thickness, particle size, and orientation. , In contrast, a minor separation, expressed by the presence of secondary peaks or small shoulders next to the main peaks, is expected when the core and the shell have different crystal structures. This shoulder peak arising from the core tended to vary in intensity depending on the thickness and crystal phase of the shell, the particle size, and the overall contribution of each phase to the entire sample. ,, As such, we consider the position of the main peak with the strongest intensity as the most important feature to identify the phase of the shell. In the case of Pd@Ru cubic nanocrystals with a relatively thin Ru shell, the main XRD peak around 40.2° could be assigned to an average of the fcc-Pd(111) peak at 40.1° and the fcc-Ru(111) peak at 40.7° because of the similarity in crystal structure between the core and shell.…”

“…Since the faceting of a nanocrystal corresponds to a specific arrangement of atoms on the surface, this parameter could have a major impact on the outcome of a template-directed synthesis . In a recent study involving Pd nanoplates, it was also found that the crystal phase taken by the deposited Ru overlayers depended on the lateral size of the template, with 12-nm plates giving fcc-Ru while 18-, 22-, and 26-nm plates resulted in hcp-Ru . In this case, although Ru can be epitaxially deposited on the basal planes of a Pd nanoplate to take either a stable hcp phase (low in bulk energy) or metastable fcc phase (high in bulk energy), the difference in stacking order will lead to jagged hcp facets (high in surface energy) or smooth fcc facets (low in surface energy) on the side faces, respectively.…”