We have measured the dispersibility of single-walled carbon nanotubes in a range of solvents, observing values as high as 3.5 mg/mL. By plotting the nanotube dispersibility as a function of the Hansen solubility parameters of the solvents, we have confirmed that successful solvents occupy a well-defined range of Hansen parameter space. The level of dispersibility is more sensitive to the dispersive Hansen parameter than the polar or H-bonding Hansen parameter. We estimate the dispersion, polar, and hydrogen bonding Hansen parameter for the nanotubes to be = 17.8 MPa(1/2), = 7.5 MPa(1/2), and = 7.6 MPa(1/2). We find that the nanotube dispersibility in good solvents decays smoothly with the distance in Hansen space from solvent to nanotube solubility parameters. Finally, we propose that neither Hildebrand nor Hansen solubility parameters are fundamental quantities when it comes to nanotube-solvent interactions. We show that the previously calculated dependence of nanotube Hildebrand parameter on nanotube diameter can be reproduced by deriving a simple expression based on the nanotube surface energy. We show that solubility parameters based on surface energy give equivalent results to Hansen solubility parameters. However, we note that, contrary to solubility theory, a number of nonsolvents for nanotubes have both Hansen and surface energy solubility parameters similar to those calculated for nanotubes. The nature of the distinction between solvents and nonsolvents remains to be fully understood.

Spontaneous exfoliation of single‐walled carbon nanotubes on dilution of dispersions in a common solvent, N‐methyl‐pyrrolidone, is demonstrated. The free‐energy of mixing is negative, confirming athermal solubility. Scanning tunneling microscopy measurements show physisorption of the solvent to the nanotube (see figure). Experiments, supported by a simple model, show that successful solvents for nanotubes are those with surface tensions close to that of graphite.

We demonstrate dispersion and exfoliation of nanotubes in two new solvents for nanotubes, cyclohexyl-pyrrolidone (CHP) and 1-benzyl-2-pyrrolidinone (NBenP). Both solvents are structural analogues of the well-known nanotube solvent N-methyl-pyrrolidone. Each solvent can disperse nanotubes at high concentrations, up to 3.5 mg/mL for CHP. The nanotubes in these dispersions are highly exfoliated, even at high concentration. In cyclohexyl-pyrrolidone dispersions, the root-mean-square bundle diameter was ∼3 nm for a concentration of 2 mg/mL. The bundle diameter fell as the concentration was reduced, reaching 1.5 nm at concentrations below 10−3 mg/mL. These dispersions have very large populations of individual nanotubes and small bundles. For CHP the total population of one-dimensional dispersed objects exceeded 100 μm−3 for concentrations >2 mg/mL. Of these ∼10% were individual SWNTs. However, as the concentration was reduced, the fraction of individual SWNTs increased to ∼80% for a nanotube concentration of 10−4 mg/mL. Like other successful nanotube dispersing solvents, both of these new solvents are characterized by surface tensions close to 40 mJ/m2. We believe their ability to disperse and exfoliate nanotubes is due to the low energetic cost of exfoliation in such solvents. Finally, their relative lack of toxicity makes these solvents much more user-friendly than traditional nanotube solvents such as N-methyl-pyrrolidone or dimethyl-formamide.