In substitution of path integral isomorphism of the quantum particle, an effective polymer ring model is proposed in the density functional calculation for hydrogen adsorption in single-walled carbon nanotubes. The excess intrinsic Helmholtz energy for quantum particles includes contributions from hard-sphere repulsion, interatomic bonding and soft attraction. The first two contributions are considered through the method developed by Yu and Wu [J. Chem. Phys. 117, 2368 (2002)], and the last contribution is obtained from mean field approximation using Weeks–Chandler–Anderson potential. The theoretical predictions are in good agreement with Monte Carlo simulation data for the density distributions of the hydrogen molecule inside the tube. In addition, the proposed model is applied to the calculation of the adsorption isotherms of hydrogen at 100 and 150 K. The present model is simpler than the current existing theories for quantum fluids.

A density-functional approach and canonical Monte Carlo simulations are presented for describing the ionic microscopic structure around the DNA molecule immersed in mixed-size counterion solutions. In the density-functional approach, the hard-sphere contribution to the Helmholtz energy functional is obtained from the modified fundamental measure theory [Y.-X. Yu and J. Z. Wu, J. Chem. Phys. 117, 10156 (2002)], and the electrostatic contribution is evaluated through a quadratic functional Taylor expansion. The new theory is suitable to the systems containing ions of arbitrary sizes and valences. In the established canonical Monte Carlo simulation, an iterative self-consistent method is used to evaluate the long-range energy, and another iterative algorithm is adopted to obtain desired bulk ionic concentrations. The ion distributions from the density-functional theory (DFT) are in good agreement with those from the corresponding Monte Carlo (MC) simulations. It is found that the ratio of the bulk concentrations of two species of counterions (cations) makes significant contribution to the ion distributions in the vicinity of DNA. Comparisons with the electrostatic potential profiles from the MC simulations show that the accuracy of the DFT becomes low when a small divalent cation exists. Both the DFT and MC simulation results illustrate that the electrostatic potential at the surface of DNA increases as the anion diameter or the total cation concentration is increased and decreases as the diameter of one cation species is increased. The calculation of electrostatic potential using real ion diameters shows that the accuracy of DFT predictions for divalent ions is also acceptable.

An equation of state (EOS) extended from statistical associating fluid theory (SAFT) has been proposed recently to describe the thermodynamic properties of non-polar fluids across the critical point (CP) (Fluid Phase Equilib. 186 (2001) 165). In thiswork, thisEOSis modified further to describe the thermodynamic properties of polar fluids across the CP by taking into account the dipole-dipole interaction. The dipole–dipole term proposed by Twu and Gubbins (Chem. Eng. Sci. 33 (1978) 863, 879) is used in this work in which the effective dipole moment is approximated as a liner function of density (Ind. Eng. Chem. Res. 35 (1996) 4727).We call this series of EOS with improved results across the CP as SAFT-CP. The new EOS that contains the dipole–dipole term includes five parameters for each polar fluid: the segment number m, the segment non-spherical parameter α, the segment volume v00, the segment interaction parameter u0/k and adjustable parameter c for effective dipole moment. Twelve polar fluids including hydrogen sulfide, ammonia, chloromethane, 1,1-difluoroethane, 1,1,1-trifluoroethane, 1,1,1,2-tetrafluoroethane, diethyl ether, ethyl acetate, acetone, 2-butanone, 2-pentanone, and 2-heptanone are used as examples. Polar fluids with associating sites are not included in this work, such as water and alkanols. The SAFT-CP EOS reproduces the saturated pressure and liquid density date with an average absolute derivation (AAD) of about 1%. Critical temperatures, pressures and densities are calculated with the AAD less than 2%. In the one-phase region including supercritical region, the SAFT-CP represents the experimental values of density with an AAD of about 2%. The comparison of the calculated critical exponent β with the experimental one for ethyl ether shows the improvement in the region of temperature approximately up to (TC-2) K.

The equilibrium selective adsorption and fluxes of oxygen/nitrogen binary gas mixtures through carbon membranes are investigated at 303 K, respectively, using a grand canonical Monte Carlo simulation and a dual control volume grand canonical molecular dynamics method. The carbon membrane pores are modeled as slit-like pores with a two-dimensional structure where carbon atoms are placed according to the structure of graphite layers. The effect of the membrane thickness, bulk pressure, and pore width on the equilibrium selective adsorption and dynamic separation factor is discussed. Meanwhile a new iteration approach to calculate the flux and dynamic separation factor of binary gas mixtures through membranes is proposed, by which we can simulate the permeation and fluxes of gases through the membranes in the presence of pressure gradient and consider the effect of pressure and composition of low-pressure side in the meantime. The simulated results show that bulk pressure and membrane thickness have no effect on the equilibrium selectivity, but they have a great effect on the fluxes and dynamic separation factors of gases. The pore width impacts the equilibrium selectivity and dynamic separation factors strongly, especially when the pore width is very small. Molecular sieving dominates the separation of oxygen/nitrogen in non-equilibrium simulations. But due to the comparable molecular size of oxygen and nitrogen, we have to modify the carbon membranes in order to improve dynamic separation of atmosphere.

Solid–liquid equilibria in mixed electrolyte aqueous solution have been investigated using available thermodynamic data for solids and for aqueous electrolyte solutions. The mean spherical approximation (MSA) modified by Lu et al. [Fluid Phase Equilib. 85 (1993) 81] is used to calculate the mean ionic activity coefficients in mixed electrolyte solutions at saturation conditions. Solid-liquid equilibria of seven mixed electrolyte systems at 298.15K are successfully predicted using the modified MSA method with the parameters obtained from activity coefficient data of corresponding single electrolyte solutions. The total average absolute deviation between predicted and experimental values is 5.58%. Furthermore, the predicted results of solid-liquid equilibria for four mixed electrolyte solutions over a range of temperature indicate that the modified MSA method can fairly be used to predict solid-liquid equilibria for mixed electrolyte aqueous solutions at various temperatures.