The characteristics of the Karman vortex generated by a molecular dynamics (MD) simulation exhibit strong finite-size effects, and MD can only reproduce the experimental results qualitatively. Here, we seek the simulation conditions for quantitatively reproduce the vortex shedding frequency. We found that the finite-size effects are mainly caused by the interference of vortices and the nonuniformity of the fluid viscosity coefficient.

We investigated the Karman vortex behind a circular cylinder in a polymer solution by a molecular dynamics simulation. The vortex characteristics are distinctly different for short and long polymers. The solution with the long polymer exhibits a reduction in the vortex shedding frequency and broadening of the lift coefficient spectrum. On the other hand, the characteristics of the short-polymer solution are almost the same as those of the Newtonian fluid. These facts are consistent with the experiments. Because the distributions of the gyration radius and the orientational order of the long-polymer solution are highly inhomogeneous in the flow field, we conclude that the extensional property of the polymer plays an important role in changing the flow characteristics. Published by AIP Publishing.

The melting curve of the modified Lennard-Jones system, which has been obtained by equating the free energies between the solid and liquid phases, was reinvestigated using the nonequilibrium relaxation method. The latter method satisfactorily reproduced the previous result with much simplified procedures.

A molecular liquid GeI4 is a candidate that undergoes a pressure-induced liquid-to-liquid phase transition. This study establishes the reference structure of the low-pressure liquid phase. Synchrotron x-ray diffraction measurements were carried out at several temperatures between the melting and the boiling points under ambient pressure. The molecule has regular tetrahedral symmetry, and the intramolecular Ge-I length of 2.51 angstrom is almost temperatureindependent within the measured range. A reverse Monte Carlo (RMC) analysis is employed to find that the distribution of molecular centers remains self-similar against heating, and thus justifying the length-scaling method adopted in determining the density. The RMC analysis also reveals that the vertex-to-face orientation of the nearest molecules are not straightly aligned, but are inclined at about 20 degrees, thereby making the closest intermolecular I-I distance definitely shorter than the intramolecular one. The prepeak observed at similar to 1 angstrom(-1) in the structural factor slightly shifts and increases in height with increasing temperature. The origin of the prepeak is clearly identified to be traces of the 111 diffraction peak in the crystalline state. The prepeak, assuming the residual spatial correlation between germanium sites in the densest direction, thus shifts toward lower wavenumbers with thermal expansion. The aspect that a relative reduction in molecular size associated with the volume expansion is responsible for the increase in the prepeak's height is confirmed by a simulation, in which the molecular size is changed.

A model molecular crystalline GeI4 was examined using molecular dynamics simulation. The model was constructed in such a way that rigid tetrahedral molecules interact with each other via Lennard-Jones potentials whose centers are located at the vertices of a tetrahedron. Because no other interaction that can "soften" the intermolecular interaction was introduced, the melting curve of the model crystalline material does not exhibit the anomaly that was found for the real substance. However, the current investigation is useful in that it could settle the upper bound of pressure below which the model can predict properties of the molecular liquid. Moreover, singularity-free nature of the melting curve allowed us to analytically treat the melting curve in the light of the Kumari-Dass-Kechin equation. As a result, we could definitely conclude that the well-known Simon equation for the melting curve is merely an approximate expression. The condition for the validity of Simon's equation was identified.

We have proposed a modified Lennard-Jones (mLJ) potential to deal with problems, such as the accurate determination of the melting condition, in which an attractive interaction plays an essential role, but its range need not necessarily extend to infinity. An accurate phase diagram, including the triple and the critical points of the system characterized by the mLJ potential, has been investigated using mainly thermodynamic integration. To predict the thermodynamic behavior of the system, it is further desired to construct the equation of state (EOS) as accurately as possible. The modified Benedict-Webb-Rubin EOS was employed to this end. The 33 parameters involved in the equation were carefully determined in order for the EOS to be compatible with the temperature dependences of the virial coefficients as well as with an extremely large set of thermodynamic data obtained from our own molecular dynamics simulation performed over a wide fluid region. The resultant EOS was found to be not only sufficiently accurate at temperatures up to twenty times as high as the critical-point temperature but also effective in practical use.

The Lennard-Jones (LJ) potential has been widely used, even beyond fluid science, as a standard form to express interparticle interaction. However, numerical computation and simulation requires some convention for dealing with the potential's infinite tail. This delicate problem can be circumvented by adopting a truncated potential. The functional form of the latter should mimic that of the original potential as closely as possible, while simultaneously minimizing artificial effects caused by the resultant discontinuities. The advantage of choosing such a truncated LJ potential, called the modified Lennard-Jones (mLJ) potential, is that the thermodynamic behavior of fluids can be predicted from the extensive knowledge base regarding the LJ fluid. In this study, the properties of the mLJ fluid were treated as perturbations of the LJ fluid case, and the discrepancies were expressed as density series expansions. The second virial coefficient of the mLJ fluid was well reproduced by this method. Reproduction of the liquid-vapor coexisting envelope was also satisfactory, except in the vicinity of the critical point.

By adopting the isothermal-isochoric ensemble instead of the isothermal-isobaric one, the density of the modified Lennard-Jones system could be specified unambiguously even under extremely low-pressure conditions. The Gibbs free energy of the gaseous state could thereby be calculated with sufficient accuracy. For practical applications of the free energy, the lowest correction in the virial expansion was found to be sufficient under pressure conditions below 0.001 epsilon/sigma(3), where epsilon and sigma are the energy and length parameters, respectively, of the modified Lennard-Jones potential. Comparison of this free energy with those for the condensed phases obtained hitherto gave well-defined sublimation points. These points then defined the equilibrium solid-gas phase boundary. The present study thus confirmed the location of not only a sublimation point found previously under marginally stable conditions of an isobaric ensemble, but also the triple point; its temperature and pressure were speculated to be 0.61 epsilon/k(B) and 0.0018 +/- 0.0005 epsilon/sigma(3), respectively, where k(B) denotes Boltzmann's constant.

An investigation of the precise determination of melting temperature in the modified Lennard-Jones system under pressure-free conditions [Y. Asano and K. Fuchizaki, J. Phys. Soc. Jpn. 78, 055002 (2009)] was extended under finite-pressure conditions to obtain the phase diagram. The temperature and pressure of the triple point were estimated to be 0.61 epsilon/k(B) and 0.0018(5) epsilon/sigma(3), and those of the critical point were 1.0709(19) epsilon/k(B) and 0.1228(20) epsilon/sigma(3), where epsilon and sigma are the Lennard-Jones parameters for energy and length scales, respectively, and k(B) is the Boltzmann constant. The potential used here has a finite attractive tail and does not suffer from cutoff problems. The potential can thus be a useful standard in examining statistical-mechanical problems in which different treatments for the tail would lead to different conclusions. The present phase diagram will then be a useful guide not only for equilibrium calculations but also for nonequilibrium problems such as discussions of the limits of phase (in)stability. (C) 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4764855]