Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

Fast parallel algorithms for short-range molecular dynamics

Adsorptive separation is very important in industry. Generally, the process uses porous solid materials such as zeolites, activated carbons, or silica gels as adsorbents. With an ever increasing need for a more efficient, energy-saving, and environmentally benign procedure for gas separation, adsorbents with tailored structures and tunable surface properties must be found. Metal–organic frameworks (MOFs), constructed by metal-containing nodes connected by organic bridges, are such a new type of porous materials. They are promising candidates as adsorbents for gas separations due to their large surface areas, adjustable pore sizes and controllable properties, as well as acceptable thermal stability. This critical review starts with a brief introduction to gas separation and purification based on selective adsorption, followed by a review of gas selective adsorption in rigid and flexible MOFs. Based on possible mechanisms, selective adsorptions observed in MOFs are classified, and primary relationships between adsorption properties and framework features are analyzed. As a specific example of tailor-made MOFs, mesh-adjustable molecular sieves are emphasized and the underlying working mechanism elucidated. In addition to the experimental aspect, theoretical investigations from adsorption equilibrium to diffusion dynamics via molecular simulations are also briefly reviewed. Furthermore, gas separations in MOFs, including the molecular sieving effect, kinetic separation, the quantum sieving effect for H2/D2 separation, and MOF-based membranes are also summarized (227 references).

Selective gas adsorption and separation in metal–organic frameworks

Metal–Organic Frameworks for Separations

New materials capable of storing hydrogen at high gravimetric and volumetric densities are required if hydrogen is to be widely employed as a clean alternative to hydrocarbon fuels in cars and other mobile applications. With exceptionally high surface areas and chemically-tunable structures, microporous metal–organic frameworks have recently emerged as some of the most promising candidate materials. In this critical review we provide an overview of the current status of hydrogen storage within such compounds. Particular emphasis is given to the relationships between structural features and the enthalpy of hydrogen adsorption, spectroscopic methods for probing framework–H2 interactions, and strategies for improving storage capacity (188 references).

/pdf/hydrogen-storage-in-metal-organic-frameworks-9ols7xj8jh.pdf

Hydrogen storage in metal–organic frameworks

Hydrogen Storage in Metal–Organic Frameworks

For fluid-solid reactions, a random pore model is developed which allows for arbitrary pore size distributions in the reacting solid. The model can represent the behavior of a system that shows a maximum in reaction rate as well as one that does not, and it identifies an optimal pore structure for either of such systems. It is demonstrated that the new model subsumes several earlier treatments as special cases. By comparison with the grain model, a relationship is derived between the effective grain shape factor and a pore structure parameter defined here. When the variance of the pore size distribution is effectively zero, the new results approach those predicted by the Petersen (1957) model over a large range of conversions. The char gasification data of Hashimoto et al. (1979) are shown to produce correlations consistent with the expectations of the new model.

A random pore model for fluid‐solid reactions: I. Isothermal, kinetic control

The storage of gases in porous adsorbents, such as activated carbon and carbon nanotubes, is examined here thermodynamically from a systems viewpoint, considering the entire adsorption-desorption cycle. The results provide concrete objective criteria to guide the search for the "Holy Grail" adsorbent, for which the adsorptive delivery is maximized. It is shown that, for ambient temperature storage of hydrogen and delivery between 30 and 1.5 bar pressure, for the optimum adsorbent the adsorption enthalpy change is 15.1 kJ/mol. For carbons, for which the average enthalpy change is typically 5.8 kJ/mol, an optimum operating temperature of about 115 K is predicted. For methane, an optimum enthalpy change of 18.8 kJ/mol is found, with the optimum temperature for carbons being 254 K. It is also demonstrated that for maximum delivery of the gas the optimum adsorbent must be homogeneous, and that introduction of heterogeneity, such as by ball milling, irradiation, and other means, can only provide small increases in physisorption-related delivery for hydrogen. For methane, heterogeneity is always detrimental, at any value of average adsorption enthalpy change. These results are confirmed with the help of experimental data from the literature, as well as extensive Monte Carlo simulations conducted here using slit pore models of activated carbons as well as atomistic models of carbon nanotubes. The simulations also demonstrate that carbon nanotubes offer little or no advantage over activated carbons in terms of enhanced delivery, when used as storage media for either hydrogen or methane.

/pdf/optimum-conditions-for-adsorptive-storage-26opq4h9wj.pdf

Optimum conditions for adsorptive storage.

The kinetics of reaction between CO2 and lime is investigated in the range of 673 to 998 K with a view to examining the effects of product layer deposition and variations in the limestone calcination atmosphere. The reaction is initially rapid and chemically controlled and goes through a sudden transition to a much slower regime controlled by diffusion in the product CaCO3 layer. The magnitude of the estimated product layer diffusivity is in the range of 10−18 to 10−21 m2/s, the corresponding activation energy is 88.9 ± 3.7 kJ/mol below 688 K and 179.2 ± 7.0 kJ/mol above that temperature, suggestive of solid state diffusion. Plausible mechanisms are discussed.

Effect of the product layer on the kinetics of the CO2‐lime reaction

The discovery of periodic mesoporous MCM-41 and related molecular sieves has attracted significant attention from a fundamental as well as applied perspective. They possess well-defined cylindrical/hexagonal mesopores with a simple geometry, tailored pore size, and reproducible surface properties. Hence, there is an ever-growing scientific interest in the challenges posed by their processing and characterization and by the refinement of various sorption models. Further, MCM-41-based materials are currently under intense investigation with respect to their utility as adsorbents, catalysts, supports, ion-exchangers, and molecular hosts. In this article, we provide a critical review of the developments in these areas with particular emphasis on adsorption characteristics, progress in controlling the pore sizes, and a comparison of pore size distributions using traditional and newer models. The model proposed by the authors for adsorption isotherms and criticalities in capillary condensation and hysteresis is found to explain unusual adsorption behavior in these materials while providing a convenient characterization tool.

Recent advances in processing and characterization of periodic mesoporous MCM-41 silicate molecular sieves

The prior random pore model for kinetically controlled fluid-solid reaction (Bhatia and Perlmutter, 1980) is generalized to include transport effects arising from boundary layer, intraparticle, and product layer diffusion. Numerical solutions are presented for a variety of conditions. The results show that the rate and surface area maxima predicted in the kinetic regime are shifted to lower conversions as intraparticle or product layer diffusional resistances increase. In addition, with increasing temperature a decrease in the overall reaction surface is predicted, in agreement with experimental findings (Kawahata and Walker, 1962).

For reactions accompanied by an increase in the volume of the solid phase, it is shown that incomplete conversion may be expected, the ultimate conversion decreasing with an increase in the intrapellet diffusional resistance. Optimal temperatures for such reactions are also identified.

Suresh K. Bhatia

Papers

A random pore model for fluid‐solid reactions: I. Isothermal, kinetic control

Optimum conditions for adsorptive storage.

Effect of the product layer on the kinetics of the CO2‐lime reaction

Recent advances in processing and characterization of periodic mesoporous MCM-41 silicate molecular sieves

A random pore model for fluid‐solid reactions: II. Diffusion and transport effects