We present systematic molecular dynamics simulation studies of hydrogen storage in single walled carbon nanotubes of various diameters and chiralities using a recently developed curvature-dependent force field. Our main objective is to address the following fundamental issues: 1. For a given H 2 loading and nanotube type, what is the H 2 distribution in the nanotube bundle? 2. For a given nanotube type, what is the maximal loading (H 2 coverage)? 3. What is the diameter range and chirality for which H 2 adsorption is most energetically favorable? Our simulation results suggest strong dependence of H 2 adsorption energies on the nanotube diameter but less dependence on the chirality. Substantial lattice expansion upon H 2 adsorption was found. The average adsorption energy increases with the lowering of nanotube diameter (higher curvature) and decreases with higher H 2 loading. The calculated H 2 vibrational power spectra and radial distribution functions indicate a strong attractive interaction between H 2 and nanotube walls. The calculated diffusion coefficients are much higher than what has been reported for H 2 in microporous materials such as zeolites, indicating that diffusivity does not present a problem for hydrogen storage in carbon nanotubes.

Complementary atomic force microscopy (AFM) measurements and molecular dynamics (MD) simulations were conducted to determine the work of adhesion for diamond (C)(111)(1 × 1) and C(001)(2 × 1) surfaces paired with carbon-based materials. While the works of adhesion from experiments and simulations are in reasonable agreement, some differences were identified. Experimentally, the work of adhesion between an amorphous carbon tip and individual C(001)(2 × 1)-H and C(111)(1 × 1)-H surfaces yielded adhesion values that were larger on the C(001)(2 × 1)-H surface. The simulations revealed that the average adhesion between self-mated C(001)(2×1) surfaces was smaller than for self-mated C(111)(1×1) contacts. Adhesion was reduced when amorphous carbon counterfaces were paired with both types of diamond surfaces. Pairing model diamond nanocomposite surfaces with the C(111)(1 × 1)-H sample resulted in even larger reductions in adhesion. These results point to the importance of atomic-scale roughness for adhesion. The simulated adhesion also shows a modest dependence on hydrogen coverage. Density functional theory calculations revealed small, C-H bond dipoles on both diamond samples, with the C(001)(2 × 1)-H surface having the larger dipole, but having a smaller dipole moment per unit area. Thus, charge separation at the surface is another possible source of the difference between the measured and calculated works of adhesion. of hydrogen into DLC films was shown to reduce friction in self-mated DLC-DLC contacts [18].By definition, the work of adhesion (W ) is the energy per unit area required to separate two semi-infinite surfaces from their equilibrium separation to infinity. For two dissimilar materials, this quantity is equivalent to a sum of the surface energies, γ 1 and γ 2 , of the two surfaces (because, effectively, the separation process creates the two surfaces), minus the interfacial energy, γ 12 . Hence, W = γ 1 + γ 2 − γ 12 , where W is also known as the Dupré energy of adhesion. To control adhesion forces between two materials, the work of adhesion can be varied by changing the surface termination. However, this assumes that the two surfaces are perfectly flat. If one or both of the surfaces is rough, then the actual area of contact will be less than the apparent contact area [19], even at the atomic scale [20][21][22]. Taller asperities (or protruding atoms) at the interface will prevent some of the smaller features (or other atoms) from closely approaching each other, increasing the separation between different regions of the surface and, therefore, reducing the energy of interaction. While the extent to which nano-or even atomic-scale roughness affects adhesion measurements is not fully understood [23], recent MD simulations indicate that the effects could be dramatic [20,21].At present, both modeling and experiment have been used to investigate adhesion at the nanometer scale. In an AFM experiment, adhesion is measured by recording pull-off forces (L C ) between an AFM tip and a sample surface in a controlled...

A computational approach has been developed for performing efficient and reasonably accurate toxicity evaluation and prediction. The approach is based on computational neural networks linked to modern computational chemistry and wavelet methods. In this paper, we present details of this approach and results demonstrating its accuracy and flexibility for predicting diverse biological endpoints including metabolic processes, mode of action, and hepato- and neurotoxicity. The approach also can be used for automatic processing of microarray data to predict modes of action.