Using solution thermodynamics to describe the dispersion of rod-like solutes: application to dispersions of carbon nanotubes in organic solvents.

We report a simple model describing the solubility of rods in solvents, expressing the final result explicitly in terms of the surface entropy and the enthalpy of mixing. This model can be combined with any expression for the mixing enthalpy depending on the requirements. For example, in one instance it predicts the dispersed concentration of rods to decrease exponentially with the Flory-Huggins parameter of the dispersion. Using a different enthalpy function, it predicts a Gaussian peak when concentration is plotted versus solvent surface energy. The model also suggests specific solvent-rod interactions to be important and shows the dispersed concentration to be very sensitive to ordering at the solvent-rod interface. We have used this model to describe experimental results for the concentration of dispersed nanotubes in various solvents. Qualitative agreement with these predictions is observed experimentally. However, we suggest that the fact that quantitative agreement is not found may be explained by solvent ordering at the nanotube surface.

High quality dispersions of hexabenzocoronene in organic solvents.

We have studied the exfoliation and dispersion of hexabenzocoronene (HBC) in 28 different solvents. We see a wide range of dispersed concentrations and aggregation states, all of which can be related to the solvent properties. To a first approximation, the dispersed concentration is maximized for solvents with Hildebrand solubility parameter close to 21 MPa(1/2), similar to graphitic materials such as nanotubes and graphene. We have also studied the concentration dependence of the absorbance and photoluminescence of HBC for both a good solvent, cyclohexyl pyrrolidone (CHP), and a poor solvent, tetrahydrofuran (THF). In both cases, we observe features that can be associated with either individual molecules or aggregates, allowing us to establish metrics both for aggregate and individual molecule content. While the aggregate content always increases with concentration, good solvents disperse individual molecules at relatively high concentrations while poor solvents display aggregation even at low concentrations. Using these metrics, we determine that large populations of individual molecules are present at low concentrations in certain solvents with Hildebrand solubility parameters close to 21 MPa(1/2). However, the aggregation state of HBC is considerably more sensitive to solvent Hildebrand parameter for halogenated solvents than for amide solvents. We find a combination of high overall concentrations and large populations of individual molecules in four solvents: cyclohexyl pyrrolidone, 1-chloronaphthalene, 1-bromonaphthalene, and 1,2,4-trichlorobenzene. Scanning tunnelling microscopy (STM) measurements show the formation of self-assembled monolayers at the interface between a HBC-solvent dispersion and a highly oriented pyrolytic graphite (HOPG) substrate. Similar structures were observed on ultrathin supports by aberration-corrected transmission electron microscopy (TEM). Also observed were graphitic objects of size ~1 nm consistent with monomers or aggregated stacks of very few monomers. We believe this is strong evidence of the presence of individual molecules in dispersions prepared with appropriate solvents.

Versatile solution phase triangular silver nanoplates for highly sensitive plasmon resonance sensing.

Solution phase triangular silver nanoplates (TSNP) with versatile tunability throughout the visible-NIR wavelengths are presented as highly sensitive localized surface plasmon refractive index sensors. A range of 20 TSNP solutions with edge lengths ranging from 11 to 200 nm and aspect ratios from 2 to 13 have been studied comprehensively using AFM, TEM, and UV-vis-NIR spectroscopy. Studies of the localized surface plasmon resonance (LSPR) peak's sensitivity to refractive index changes are performed using a simple sucrose concentration method whereby the surrounding refractive index can solely be changed without variation in any other parameter. The dependence of the TSNP localized surface plasmon resonance (LSPR) peak wavelength lambda(max) and its bulk refractive index sensitivity on the nanoplate's structure is determined. LSPR sensitivities are observed to increase linearly with lambda(max) up to 800 nm, with the values lying within the upper limit theoretically predicted for optimal sensitivity, notwithstanding any diminution due to ensemble averaging. A nonlinear increase in sensitivity is apparent at wavelengths within the NIR region with values reaching 1096 nm.RIU(-1) at lambda(max) 1093 nm. Theoretical studies performed using a simple aspect ratio dependent approximation method and discrete dipole approximation methods confirm the dependence of the LSPR bulk refractive index sensitivity upon the TSNP aspect ratio measured experimentally. These studies highlight the importance of this key parameter in acquiring such high sensitivities and promote these TSNP sols for sensing applications at appropriate wavelengths for biological samples.

Etching-resistant silver nanoprisms by epitaxial deposition of a protecting layer of gold at the edges.

The protection of silver nanoprisms against etching by the epitaxial deposition of a thin layer of gold in solution has been investigated. It has been found that at low Au/Ag ratios (approximately 0.08 to 0.17) a thin layer of gold is deposited on the edges of the nanoprisms as expected, but without the structural damage typically associated with galvanic replacement. Furthermore, this layer of gold provides robust protection against etching of the nanoprisms by chloride and is strong evidence that etching by chloride is face-selective and does not take place at the flat {111} faces of the nanoprisms. Additionally, the deposition of a protecting layer of gold results in only a small red shift in the position of the main plasmon resonance. We have investigated the sensitivity of the localized surface plasmon resonance (LSPR) to changes in the bulk refractive index of the solution and find that the gold-protected silver nanoprisms are promising candidates for the development of new refractive index-based biosensors.

Spontaneous exfoliation of single-walled carbon nanotubes dispersed using a designed amphiphilic peptide.

We have observed concentration dependent exfoliation of single-walled carbon nanotubes dispersed in solutions of the synthetic peptide nano-1. As the nanotube concentration is reduced, the bundle diameters tend to decrease before saturating at <2.0 nm for concentrations below 6 x 10(-3) mg/mL. The fraction of individual nanotubes increases with decreasing concentration, saturating at approximately 95% at low concentration. This concentration dependent exfoliation happens even if the dispersions are not sonicated on dilution, albeit over a longer time scale. The populations both of individual nanotubes and of bundles are much higher than expected at high concentrations, indicating the presence of repulsive internanotube interactions stabilizing the dispersions.