On-surface synthesis of aromatic cyclo[10]carbon and cyclo[14]carbon

All-carbon materials based on sp 2 -hybridized atoms, such as fullerenes 1 , carbon nanotubes 2 and graphene 3 , have been much explored due to their remarkable physicochemical properties and potential for applications. Another unusual all-carbon allotrope family are the cyclo[ n ]carbons (C n ) consisting of two-coordinated sp -hybridized atoms. They have been studied in the gas phase since the twentieth century 4 , 5 , 6 , but their high reactivity has meant that condensed-phase synthesis and real-space characterization have been challenging, leaving their exact molecular structure open to debate 7 , 8 , 9 , 10 , 11 . Only in 2019 was an isolated C 18 generated on a surface and its polyynic structure revealed by bond-resolved atomic force microscopy 12 , 13 , followed by a recent report 14 on C 16 . The C 18 work trigged theoretical studies clarifying the structure of cyclo[ n ]carbons up to C 100 (refs. 15 , 16 , 17 , 18 , 19 , 20 ), although the synthesis and characterization of smaller C n allotropes remains difficult. Here we modify the earlier on-surface synthesis approach to produce cyclo[10]carbon (C 10 ) and cyclo[14]carbon (C 14 ) via tip-induced dehalogenation and retro-Bergman ring opening of fully chlorinated naphthalene (C 10 Cl 8 ) and anthracene (C 14 Cl 10 ) molecules, respectively. We use atomic force microscopy imaging and theoretical calculations to show that, in contrast to C 18 and C 16 , C 10 and C 14 have a cumulenic and cumulene-like structure, respectively. Our results demonstrate an alternative strategy to generate cyclocarbons on the surface, providing an avenue for characterizing annular carbon allotropes for structure and stability.