successfully applied to the growth of AGNRs 11-13 and related structures [14][15][16] . Here, we describe the successful bottom-up synthesis of ZGNRs, which are fabricated by the surface-assisted colligation and cyclodehydrogenation of specifically designed precursor monomers including carbon groups that yield atomically precise zigzag edges. Using scanning tunnelling spectroscopy we prove the existence of edge-localized states with large energy splittings. We expect that the availability of ZGNRs will finally allow the characterization of their predicted spin-related properties such as spin confinement 17 and filtering 18,19 , and ultimately add the spin degree of freedom to graphene-based circuitry.To explore the fundamental electronic and magnetic properties related to zigzag edges and to realize specific carbon nanostructures for the controlled manipulation of their spin states,ZGNRs with atomically precise edges are required. For GNRs with armchair edges, it was demonstrated that atomic precision can indeed be achieved by a bottom-up approach based on the surface-assisted polymerization and subsequent cyclodehydrogenation of specifically designed oligophenylene precursor monomers 11 . The on-surface synthesis has been applied by many groups to fabricate a number of different AGNR structures [11][12][13] , N-doped AGNRs 14,15 as well as AGNR heterostructures 15,16 . It is, however, not directly suited forZGNRs since polymerization of monomers via aryl-aryl coupling does not take place along the zigzag but along the armchair direction (Fig. 1a). In addition, dehydrogenative cyclization of phenyl subgroups is not sufficient to form pure zigzag edges, thus calling for a totally new chemical design. Thereby, additional carbon functions must be placed at the edges of the monomers to complete the tiling toolbox needed for the bottom-up fabrication of arbitrary GNR structures.Here, we report a bottom-up fabrication approach to ZGNRs. In our unique protocol, surfaceassisted polymerization and subsequent cyclization of suitably designed molecular precursors carrying the full structural information of the final ZGNR afford atomic precision with respect to ribbon width and edge morphology. The groundbreaking idea depends upon the choice of a unique U-shaped monomer as 1 shown in Fig. 1b. With two halogen functions for thermally induced aryl-aryl-coupling at the R 1 positions, it allows the polymerization toward a snake-like polymer. It is the beauty of this design that additional phenyl groups at the R 2 position fill the holes in the interior of the undulating polymer. The crucial precursor is monomer 1a which carries two additional methyl groups. In such a case, apart from the 3 polymerization and planarization, an oxidative ring closure including the methyl groups is expected which would then establish two new six-membered rings together with the zigzag edge structure. To our delight, this concept could indeed be synthetically realized under reaction monitoring and structure proof by scanning tunneling microscopy (S...

Graphene nanoribbons (GNRs), defined as nanometer-wide strips of graphene, have attracted increasing attention as promising candidates for next-generation semiconductors. Here, we demonstrate a bottom-up strategy toward novel low band gap GNRs (Eg = 1.70 eV) with a well-defined cove-type periphery both in solution and on a solid substrate surface with chrysene as the key monomer. Corresponding cyclized chrysene-based oligomers consisting of the dimer and tetramer are obtained via an Ullmann coupling followed by oxidative intramolecular cyclodehydrogenation in solution, and much higher GNR homologues via on-surface synthesis. These oligomers adopt nonplanar structures due to the steric repulsion between the two C–H bonds at the inner cove position. Characterizations by single crystal X-ray analysis, UV–vis absorption spectroscopy, NMR spectroscopy, and scanning tunneling microscopy (STM) are described. The interpretation is assisted by density functional theory (DFT) calculations.

Graphite vaporization provides an uncontrolled yet efficient means of producing fullerene molecules. However, some fullerene derivatives or unusual fullerene species might only be accessible through rational and controlled synthesis methods. Recently, such an approach has been used to produce isolable amounts of the fullerene C(60) from commercially available starting materials. But the overall process required 11 steps to generate a suitable polycyclic aromatic precursor molecule, which was then dehydrogenated in the gas phase with a yield of only about one per cent. Here we report the formation of C(60) and the triazafullerene C(57)N(3) from aromatic precursors using a highly efficient surface-catalysed cyclodehydrogenation process. We find that after deposition onto a platinum (111) surface and heating to 750 K, the precursors are transformed into the corresponding fullerene and triazafullerene molecules with about 100 per cent yield. We expect that this approach will allow the production of a range of other fullerenes and heterofullerenes, once suitable precursors are available. Also, if the process is carried out in an atmosphere containing guest species, it might even allow the encapsulation of atoms or small molecules to form endohedral fullerenes.

On-surface synthesis is a powerful route toward the fabrication of specific graphene-like nanostructures confined in two dimensions. This strategy has been successfully applied to the growth of graphene nanoribbons of diverse width and edge morphology. Here, we investigate the mechanisms driving the growth of 9-atom wide armchair graphene nanoribbons by using scanning tunneling microscopy, fast X-ray photoelectron spectroscopy, and temperature-programmed desorption techniques. Particular attention is given to the role of halogen functionalization (Br or I) of the molecular precursors. We show that the use of iodine-containing monomers fosters the growth of longer graphene nanoribbons (average length of 45 nm) due to a larger separation of the polymerization and cyclodehydrogenation temperatures. Detailed insight into the growth process is obtained by analysis of kinetic curves extracted from the fast X-ray photoelectron spectroscopy data. Our study provides fundamental details of relevance to the production of future electronic devices and highlights the importance of not only the rational design of molecular precursors but also the most suitable reaction pathways to achieve the desired final structures.

We report on the atomic structure of graphene nanoribbons (GNRs) formed via on-surface synthesis from 10,10'-dibromo-9,9'-bianthryl (DBBA) precursors on Cu(111). By means of ultrahigh vacuum noncontact atomic force microscopy with CO-functionalized tips we unveil the chiral nature of the so-formed GNRs, a structure that has been under considerable debate. Furthermore, we prove that-in this particular case-the coupling selectivity usually introduced by halogen substitution is overruled by the structural and catalytic properties of the substrate. Specifically, we show that identical chiral GNRs are obtained from 9,9'-bianthryl, the unsubstituted sister molecule of DBBA.

The electronic properties of graphene nanoribbons grown on metal substrates are significantly masked by the ones of the supporting metal surface. Here, we introduce a novel approach to access the frontier states of armchair graphene nanoribbons (AGNRs). The in situ intercalation of Si at the AGNR/Au(111) interface through surface alloying suppresses the strong contribution of the Au(111) surface state and allows for an unambiguous determination of the frontier electronic states of both wide and narrow band gap AGNRs. First-principles calculations provide insight into substrate induced screening effects, which result in a width-dependent band gap reduction for substrate-supported AGNRs. The strategy reported here provides a unique opportunity to elucidate the electronic properties of various kinds of graphene nanomaterials supported on metal substrates.

Recent research on deep learning, a set of machine learning techniques able to learn deep architectures, has shown how robotic perception and action greatly benefits from these techniques. In terms of spacecraft navigation and control system, this suggests that deep architectures may be considered now to drive all or part of the onboard decision making system. In this paper this claim is investigated in more detail training deep artificial neural networks to represent the optimal control action during a pinpoint landing, assuming perfect state information. It is found to be possible to train deep networks for this purpose and that the resulting landings, driven by the trained networks, are close to simulated optimal ones. These results allow for the design of an on-board real time optimal control system able to cope with large sets of possible initial states while still producing an optimal response.

No abstract