Doping is a widely applied technological process in materials science that involves incorporating atoms or ions of appropriate elements into host lattices to yield hybrid materials with desirable properties and functions. For nanocrystalline materials, doping is of fundamental importance in stabilizing a specific crystallographic phase, modifying electronic properties, modulating magnetism as well as tuning emission properties. Here we describe a material system in which doping influences the growth process to give simultaneous control over the crystallographic phase, size and optical emission properties of the resulting nanocrystals. We show that NaYF(4) nanocrystals can be rationally tuned in size (down to ten nanometres), phase (cubic or hexagonal) and upconversion emission colour (green to blue) through use of trivalent lanthanide dopant ions introduced at precisely defined concentrations. We use first-principles calculations to confirm that the influence of lanthanide doping on crystal phase and size arises from a strong dependence on the size and dipole polarizability of the substitutional dopant ion. Our results suggest that the doping-induced structural and size transition, demonstrated here in NaYF(4) upconversion nanocrystals, could be extended to other lanthanide-doped nanocrystal systems for applications ranging from luminescent biological labels to volumetric three-dimensional displays.

Electrocatalytic reduction of water to molecular hydrogen via the hydrogen evolution reaction may provide a sustainable energy supply for the future, but its commercial application is hampered by the use of precious platinum catalysts. All alternatives to platinum thus far are based on nonprecious metals, and, to our knowledge, there is no report about a catalyst for electrocatalytic hydrogen evolution beyond metals. Here we couple graphitic-carbon nitride with nitrogen-doped graphene to produce a metal-free hybrid catalyst, which shows an unexpected hydrogen evolution reaction activity with comparable overpotential and Tafel slope to some of well-developed metallic catalysts. Experimental observations in combination with density functional theory calculations reveal that its unusual electrocatalytic properties originate from an intrinsic chemical and electronic coupling that synergistically promotes the proton adsorption and reduction kinetics.

Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges.

We constructed highly oriented and ordered macropores within metal-organic framework (MOF) single crystals, opening up the area of three-dimensional–ordered macro-microporous materials (that is, materials containing both macro- and micropores) in single-crystalline form. Our methodology relies on the strong shaping effects of a polystyrene nanosphere monolith template and a double-solvent–induced heterogeneous nucleation approach. This process synergistically enabled the in situ growth of MOFs within ordered voids, rendering a single crystal with oriented and ordered macro-microporous structure. The improved mass diffusion properties of such hierarchical frameworks, together with their robust single-crystalline nature, endow them with superior catalytic activity and recyclability for bulky-molecule reactions, as compared with conventional, polycrystalline hollow, and disordered macroporous ZIF-8.

Organometallic complexes with metal-nitrogen/carbon (M-N/C) coordination are the most important alternatives to precious metal catalysts for oxygen reduction and evolution reactions (ORR and OER) in energy conversion devices. Here, we designed and developed a range of molecule-level graphitic carbon nitride (g-CN) coordinated transition metals (M-CN) as a new generation of M-N/C catalysts for these oxygen electrode reactions. As a proof-of-concept example, we conducted theoretical evaluation and experimental validation on a cobalt-CN catalyst with a desired molecular configuration, which possesses comparable electrocatalytic activity to that of precious metal benchmarks for the ORR and OER in alkaline media. The correlation of experimental and computational results confirms that this high activity originates from the precise M-N coordination in the g-CN matrix. Moreover, the reversible ORR/OER activity trend for a wide variety of M-CN complexes has been constructed to provide guidance for the molecular design of this promising class of catalysts.