Semiconducting and Metallic [5,5] Fullertube Nanowires: Characterization of Pristine D5h(1)-C90 and D5d(1)-C100.

Although fullerenes were discovered nearly 35 years ago, scientists still struggle to isolate "single molecule" tubular fullerenes larger than C90. In similar fashion, there is a paucity of reports for pristine single-walled carbon nanotubes (SWNTs). In spite of Herculean efforts, the isolation and properties of pristine members of these carbonaceous classes remain largely unfulfilled. For example, the low abundance of spherical and tubular higher fullerenes in electric-arc extracts (<0.01-0.5%) and multiplicity of structural isomers remain a major challenge. Recently, a new isolation protocol for highly tubular fullerenes, also called f ullertubes, was reported. Herein, we describe spectroscopic characterization including 13C NMR, XPS, and Raman results for purified [5,5] fullertube family members, D5h-C90 and D5d-C100. In addition, DFT computational HOMO-LUMO gaps, polarizability indices, and electron density maps were also obtained. The Raman and 13C NMR results are consistent with semiconducting and metallic properties for D5h-C90 and D5d-C100, respectively. Our report suggests that short [5,5] fullertubes with aspect ratios of only ∼1.5-2 are metallic and could exhibit unique electronic properties.

Fullertubes: Cylindrical Carbon with Half-Fullerene End-Caps and Tubular Graphene Belts, Their Chemical Enrichment, Crystallography of Pristine C90-D5h(1) and C100-D5d(1) Fullertubes, and Isolation of C108, C120, C132, and C156 Cages of Unknown Structures.

We report a chemical separation method to isolate fullertubes: a new and soluble allotrope of carbon whose structure merges nanotube, graphene, and fullerene subunits. Fullertubes possess single-walled carbon nanotube belts resembling a rolled graphene midsection, but with half-fullerene end-caps. Unlike nanotubes, fullertubes are reproducible in structure, possess a defined molecular weight, and are soluble in pristine form. The high reactivity of amines with spheroidal fullerene cages enables their removal and allows a facile isolation of C96-D3d(3), C90-D5h(1), and C100-D5d(1) fullertubes. A nonchromatographic step (Stage 1) uses a selective reaction of carbon cages with aminopropanol to permit a highly enriched sample of fullertubes. Spheroidal fullerenes are reacted and removed by attaching water-soluble groups onto their cage surfaces. With this enriched (100-1000 times) fullertube mixture, Stage 2 becomes a simple HPLC collection with a single column. This two-stage separation approach permits fullertubes in scalable quantities. Characterization of purified C100-D5d(1) fullertubes is done with samples isolated in pristine and unfunctionalized form. Surprisingly, C60 and C100-D5d(1) are both purplish in solution. For X-ray crystallographic analysis, we used decapyrrylcorannulene (DPC). Isomerically purified C90 and C100 fullertubes were mixed with DPC to obtain black cocrystals of 2DPC{C90-D5h(1)}·4(toluene) and 2DPC{C100-D5d(1)}·4(toluene), respectively. A serendipitous outcome of this chemical separation approach is the enrichment and purification of several unreported larger carbon species, e.g., C120, C132, and C156. Isolation of these higher cage species represents a significant advance in the unknown experimental arena of C100-C200 structures. Our findings represent seminal experimental evidence for the existence of two mathematically predicted families of fullertubes: one family with an axial hexagon with the other series based on an axial pentagon ring. Fullertubes have been predicted theoretically, and herein is their experimental evidence, isolation, and initial characterization.