Graphene nanoribbons-narrow and straight-edged stripes of graphene, or single-layer graphite-are predicted to exhibit electronic properties that make them attractive for the fabrication of nanoscale electronic devices. In particular, although the two-dimensional parent material graphene exhibits semimetallic behaviour, quantum confinement and edge effects should render all graphene nanoribbons with widths smaller than 10 nm semiconducting. But exploring the potential of graphene nanoribbons is hampered by their limited availability: although they have been made using chemical, sonochemical and lithographic methods as well as through the unzipping of carbon nanotubes, the reliable production of graphene nanoribbons smaller than 10 nm with chemical precision remains a significant challenge. Here we report a simple method for the production of atomically precise graphene nanoribbons of different topologies and widths, which uses surface-assisted coupling of molecular precursors into linear polyphenylenes and their subsequent cyclodehydrogenation. The topology, width and edge periphery of the graphene nanoribbon products are defined by the structure of the precursor monomers, which can be designed to give access to a wide range of different graphene nanoribbons. We expect that our bottom-up approach to the atomically precise fabrication of graphene nanoribbons will finally enable detailed experimental investigations of the properties of this exciting class of materials. It should even provide a route to graphene nanoribbon structures with engineered chemical and electronic properties, including the theoretically predicted intraribbon quantum dots, superlattice structures and magnetic devices based on specific graphene nanoribbon edge states.

The fabrication of nanoscale membranes exhibiting high selectivity is an emerging field of research. The possibility to use bottom-up approaches to fabricate a filter with porous graphene and analyze its functionality with first principle calculations is investigated. Here, the porous network is produced by self-assembly of the hexaiodo-substituted macrocycle cyclohexa-m-phenylene (CHP). The resulting porous network exhibits an extremely high selectivity in favor of H(2) and He among other atmospheric gases, such as Ne, O(2), N(2), CO, CO(2), NH(3), and Ar. The presented membrane is superior to traditional filters using polymers or silica and could have great potential for further technological applications such as gas sensors or fuel cells.

Graphene nanoribbons-semiconducting quasi-one-dimensional graphene structures-have great potential for the realization of novel electronic devices. Recently, graphene nanoribbon heterojunctions-interfaces between nanoribbons with unequal band gaps-have been realized with lithographic etching techniques and via chemical routes to exploit quantum transport phenomena. However, standard fabrication techniques are not suitable for ribbons narrower than ~5 nm and do not allow to control the width and edge structure of a specific device with atomic precision. Here, we report the realization of graphene nanoribbon heterojunctions with lateral dimensions below 2 nm via controllable dehydrogenation of polyanthrylene oligomers self-assembled on a Au(111) surface from molecular precursors. Atomistic simulations reveal the microscopic mechanisms responsible for intraribbon heterojunction formation. We demonstrate the capability to selectively modify the heterojunctions by activating the dehydrogenation reaction on single units of the nanoribbons by electron injection from the tip of a scanning tunneling microscope.

Atomically precise graphene nanoribbons (GNRs) can be obtained via thermally induced polymerization of suitable precursor molecules on a metal surface. This communication discusses the atomic structure found at the termini of armchair GNRs obtained via this bottom-up approach. The short zigzag edge at the termini of the GNRs under study gives rise to a localized midgap state with a characteristic signature in scanning tunneling microscopy (STM). By combining STM experiments with large-scale density functional theory calculations, we demonstrate that the termini are passivated by hydrogen. Our results suggest that the length of nanoribbons grown by this protocol may be limited by hydrogen passivation during the polymerization step.