In contrast to crystalline solids--for which a precise framework exists for describing structure--quantifying structural order in liquids and glasses has proved more difficult because even though such systems possess short-range order, they lack long-range crystalline order. Some progress has been made using model systems of hard spheres, but it remains difficult to describe accurately liquids such as water, where directional attractions (hydrogen bonds) combine with short-range repulsions to determine the relative orientation of neighbouring molecules as well as their instantaneous separation. This difficulty is particularly relevant when discussing the anomalous kinetic and thermodynamic properties of water, which have long been interpreted qualitatively in terms of underlying structural causes. Here we attempt to gain a quantitative understanding of these structure-property relationships through the study of translational and orientational order in a models of water. Using molecular dynamics simulations, we identify a structurally anomalous region--bounded by loci of maximum orientational order (at low densities) and minimum translational order (at high densities)--in which order decreases on compression, and where orientational and translational order are strongly coupled. This region encloses the entire range of temperatures and densities for which the anomalous diffusivity and thermal expansion coefficient of water are observed, and enables us to quantify the degree of structural order needed for these anomalies to occur. We also find that these structural, kinetic and thermodynamic anomalies constitute a cascade: they occur consecutively as the degree of order is increased.

An approach for directly determining the liquid–vapor phase equilibrium of a model system at any temperature along the coexistence line is described. The method relies on transition matrix Monte Carlo ideas developed by Fitzgerald, Picard, and Silver [Europhys. Lett. 46, 282 (1999)]. During a Monte Carlo simulation attempted transitions between states along the Markov chain are monitored as opposed to tracking the number of times the chain visits a given state as is done in conventional simulations. Data collection is highly efficient and very precise results are obtained. The method is implemented in both the grand canonical and isothermal–isobaric ensemble. The main result from a simulation conducted at a given temperature is a density probability distribution for a range of densities that includes both liquid and vapor states. Vapor pressures and coexisting densities are calculated in a straightforward manner from the probability distribution. The approach is demonstrated with the Lennard-Jones fluid. Coexistence properties are directly calculated at temperatures spanning from the triple point to the critical point.

We study the diffusive dynamics of a hard-sphere fluid confined between parallel smooth hard walls. The position-dependent diffusion coefficient normal to the walls is larger in regions of high local packing density. High density regions also have the largest available volume, consistent with the fast local diffusivity. Indeed, local and global diffusivities as a function of the Widom insertion probability approximately collapse onto a master curve. Parallel and average normal diffusivities are strongly coupled at high densities and deviate from bulk fluid behavior.

We report results from a molecular simulation study of the structure and dynamics of water near single carbohydrate molecules (glucose, trehalose, and sucrose) at 0 and 30 degrees C. The presence of a carbohydrate molecule has a number of significant effects on the microscopic water structure and dynamics. All three carbohydrates disrupt the tetrahedral arrangement of proximal water molecules and restrict their translational and rotational mobility. These destructuring effects and slow dynamics are the result of steric constraints imposed by the carbohydrate molecule and of the ability of a carbohydrate to form stable H bonds with water, respectively. The carbohydrates induce a pronounced decoupling between translational and rotational motions of proximal water molecules.

A new model for the n-alkane homologous series has been developed, parameterized to the vapor-liquid coexistence properties for a range of chain lengths.The model utilizes the Buckingham exponential-6 potential to describe the nonbonded interaction energy. Histogram reweighting grand canonical Monte Carlo methods were used to determine the model parameters. The new model reproduces the experimental saturated liquid and vapor densities and vapor pressure for ethane through octane to within average errors of 0.5%, 2.1% and 2.2% respectively. Critical temperatures and densities were also found to be in good agreement with experiment. Critical pressures are slightly overestimated for longer chain lengths. Comparisons were made to the TraPPE [J. Phys. Chem. B 1998, 102 2569] and NERD [J. Chem. Phys. 1998, 108, 9905] models. The two previous models reproduce the liquid properties with comparable accuracy to the proposed model, however the new model was found to describe the vapor pressures more accurately. Liquid densities were determined for the new model for chain lengths as long as C 78 . Agreement to experiment is within 1% at atmospheric pressure. Phase diagrams were calculated for mixtures of ethane-heptane, ethane-decane, ethane-eicosane, and octane-dodecane. The new model achieves near-experimental predictive accuracy for these mixtures.