Systematic relations between porosity and compressional velocity Vp in the three component (sand, grains, clay and brine) systems (1) porous sandstone, (2) sands, and (3) suspensions, were obtained using experimental data and models. In Cemented Shaley Sandstones Vp was found to correlate linearly with porosity and clay content. The velocities in clean sandstones are about 7% higher than those predicted by the linear fit, indicating that a small amount of clay significantly reduces the elastic moduli of sandstones.

For uncemented shaley sand, a model for the dependence of sonic velocity and porosity on clay content and compaction was developed for sand with clay dispersed in the pore space and for shale with suspened sand grains. The model closely mimics the experimentally observed minimum for porosity and the peak in velocity versus clay content. The results explain much of the scatter in velocity data in-situ. Velocity in suspensions at ϕ = 39% of grains in brine is close to values predicted by the Reuss (Isostress) average. Velocity dispersion, as suggested by Biot (1956 a,b) is calculated and observed in coarser sediments such as sand, whereas velocities in the finer clay and silt follow Biot's low frequency value.

In total, our results provide the complete dependence of velocity on porosity in brine saturated sediment with clays, ranging from pure quartz to pure clay and water. Our results also highlight the crucial role of the critical porosity ϕ at about 39%, and the transition from cemented to uncemented sands.

Partially sintered silica gels were immersed in different NaCI solutions to deliberately introduce a small amount of sodium ions ranging from 387 to 9900 ppm. Dielectric measurements were carried out in the frequency range 12 to 107Hz under various relative humidity conditions. Two relaxation processes were observed in the impedance and modulus planes and were assigned to two separate regions in the hydrated silica surface. The top region is expected to be rich in hydrated sodium since its modulus relaxation time depends strongly on the relative humidity as well as the sodium content. The interfacial region is believed to be composed of structured water since its dielectric conductivity is affected mainly by the amount of water adsorbed on the pore surfaces.

In a colloidal suspension with attractive interaction, particles form aggregates that settle to the bottom of the container. As the concentration of particles is increased, the overlapping of the aggregates (flocs) produces a continuous network throughout the suspension before settling occurs and a colloidal gel is formed. Colloidal gels may be divided into hard gels and soft gels.1 Hard gels are those formed by fine particles such as silica or boehmite of about 0.01 µm in size with high interparticle attraction energies. Under small shear rates, there is little restructuring in the gel network. The flocs that pack to form the gel network still retain their fractal structure, which gives rise to the scaling behavior of various mechanical properties of a gel with respect to particle concentration. For example, the storage modulus G' of a hard gel remains constant at small strain and increases in a power-law fashion with particle concentration, (φ, as G' - φn. There can be two regimes, a strong-link regime and a weak-link regime. In both regimes, the exponent n can be expressed in terms of the fractal dimension of the flocs, D, and the fractal dimension of the backbone of the flocs, x, while the form of n depends on what regime the system is in. Furthermore, a gel may crossover from the strong-link regime to the weak-link regime as the particle volume fraction is increased2.

Small angle neutron scattering data of Vycor is analyzed using a chord distribution model describing the microporous structure. In addition, the same model is applied in interpreting the data taken when capillary condensation takes place in the porous medium. In both cases, the results are in good agreement with other measurements, and shows a promising potential in describing such a bicontinuous system.

A quasi-equilibrium model is developed for capillary condensation flow, in which surfacediffusion is included. The model is in excellent agreement with the experimental data for the flow of toluene in untreated Vycor glass and in Vycor glass whose pores were derivatized with C18H38. The observed increase in viscosity with decreasing pore size is ascribed to a stagnant monolayer adsorbed to the pore walls. The surface diffusion coefficient at 22ºC is 5 x 10-5 cm2/secfor both the derivatized and untreated samples.

Three of the most important elements of the diagenesis of sedimentary rocks are consolidation, porosity reduction, and pressure dissolution or cementation. These are discussed qualitatively with special emphasis on the central role of contact formation in controlling both the driving forces and the kinetics of morphological change. It is shown, in particular, that the equilibrium forces would drive the rocks towards collapse of their porosity earlier in diagenesis were they not prevented by kinetic barriers. At greater depths these barriers are effectively removed by pressure concentration at the small contacts. We propose an explicit, qualitative mechanism leading to the observed morphologies and identify specific problems for quantitative analysis.

The problem of healing of microscopic pores and cracks in the temperature field and understatic and dynamic pressures attracts considerable attention. Indeed, regeneration of continuity leads to a significant improvement of physical and mechanical properties of the materials where micropores were formed as a result of long—term loading (high—temperature creep, fatigue), surface deformation (friction, wear, grinding), specialconditions of processing (casting, powder metallurgy, condensed films) [1– 10]. For instance, the elimination of porosity allows longer lifetimes [1– 7] and higher ultimate strength [8], elastioonstants [91], and also electric and magnetic properties [10]to be achieved.

A physical model is presented to describe solid—phase sintering of pressed ferromagnetic powders under conditions of self—organization.

The complex microstructure of hardened cement paste is produced by hydration reactions between cement particles and the water in which they are suspended. In recent years, algorithms like the diffusion-limited aggregation (DLA) and Eden models have demonstrated that simple growth rules can result in complex aggregated structures. The model described in this paper simulates, via simplified growth rules, the microstructural development ofhydrating cement paste. This model has similarities to DLA, but with the additional novel features of dissolution of solid particles, and a free-space nucleation probability. The percolation aspects and transport properties of the model's pore space are computed and discussed.

We report here the use of nuclear magnetic resonance imaging in the observation of liquid-phase fraction distributions in a partially-wetted sample of glass beads. By combiningboth large- and small-scale imaging techniques, we can study the transition from local-averaged saturations to large-scale averaged saturations. The image data allows us to assess the utility of the large-scale measurements We comment on the reliability and generality of the measurements for our specific system.

This paper is concerned with two related problems: (1) the construction of geometrical models of porous media relevant to granular composites and (2) the description of transport processes in these model systems. We will show that a variety of interesting porous media can be generated by the packing and subsequent modification of spherical grains. This modification may involve a change in either the grain's size, shape, or both. Steady state transport processes such as the flow of electrical current or viscous fluids are controlled by the distribution of pore throat sizes and, within the presentframework, can be studiedefficiently by random walk simulations of diffusion. The techniques developed here are also of interest in connection with dynamic transport processes such as the filtration of fine grained particles into consolidated granular networks. Wediscuss briefly the modeling of such processes as well as the interaction of steady state and dynamic transport.

The microgeometry of the pore space influences the membrane potential Em. and theDC electrical conductivity σ of a shaly sand in a similar manner, independent of the details of the geometry: Em and σ being related via the conductivities of cations and σanions;σ=σcation + σ onion, and Em α σ cation/(σcation + σanion). This explicit relationship is used to investigate the role of the geometrical factors which influence both Em and σ in a related manner. The dependence of σ on the water conductivity σw can be well approximated with four geometrical parameters which can be obtained from the slopes and the interceptsof σ vs. σw curve at high and low salinities. We show that these geometrical factors appear in the expression for Em a well. These geometrical parameters (one of them is the formation factor) vary from rock to rock, and any trend in these parameters depend on the local geology.

A hybrid discrete/continuum Monte Carlo technique combining random walk simulations with first passage time (FPT) concepts is developed here in order to estimate diffusion properties of randomly-assembled sintered porous structures. This work combines the creation of realistic porous solid structures with controlled pore size, shape, and tortuosity features with the application of an efficient algorithm for calculating effective diffusivities in all diffusion regimes (Knudsen, transition, and molecular). The hybrid simulation technique consists of creating a “protective” boundary layer surrounding solid surfaces within which discrete random motion simulations are performed while continuum FPT results are used in the remaining pore space. The boundary layer thickness reflects a characteristic length scale, of the order of a few mean free paths, over which the FPT approximation breaks down. This procedure significantly reduces the computations required to cover statistically representative regions of the porous structure,a serious shortcoming in previous studies of gas diffusion through porous solids; it leads to effective diffusivity estimates that are in excellent agreement with experimental measurements.

We present optical evidence in the visible and mid-infrared for the coal to graphite transformation proceeding via a nucleation mechanism. In the visible region, we show that changes in reflectivity that occur with increasing coal rank can be understood by an increase in the volume fraction of sub-microscopic “graphite-like” nuclei and a subsequent increase of the complex part of the dielectric constant with increasing coal rank. Optical phenomena associated with the granular nature of coal is also observed in the mid-infrared; specifically, measurements of absorption across the mid-infrared show a significant increase of absorption with increasing coal rank. This increase can be ascribed to absorption arising from a plasma resonance “graphite-like” material of the granular composite. A broad range of the optical repsroopnearntciees foofr hitghhe rank coals in the visible and mid-infrared can thus be readily understood and quantitatively modeled in terms of a granular medium consisting of nuclei of “graphite-like” material imbedded in an insulating organic matrix. If validated by further experimentation, this model should greatly expand current capabilities in the opticalcharacterization of high rank carbonaceous materials.

Granular metal films consisting of small Sn, Bi, and Pb particles, typically from less than 100 Å to several 1000 Å in size and embedded in a SiO2 matrix, have been fabricated over a range of metal compositions by RF sputter deposition. Two different film geometries have been prepared, homogeneous films and multilayer films consisting of alternating layers of granular metal and SiO2. These films have been characterized by x-ray diffraction and transmission electron microscopy and their melting behavior studied by differential scanning calorimetry. As the concentration of the metal component is decreased, the average particle size decreases and the particle size distribution becomes more narrow. When the solid-liquid transition is studied, the melting temperature has been found to be increasingly depressed as the particle size is reduced. In the smallest particles the relative reduction in the melting temperature is greater than 10 percent. No strong evidence for melting point enhancements, due to pressureeffects arising from the different thermal expansions of the metal particles and the SiO2 matrix, has been observed in either the homogeneous or multilayer films.

We present results of theoretical calculations on the thermal characteristics of small granular particles. By using the selfconsistent Einstein model with a Morse interaction potential, it is shown that Debye temperature drops significantly for particle sizes below 50Å, a phenomenon observed experimentally. Melting temperature of small particles is calculated using a generalized Lennard-Jones-Devonshire theory. Depression of the melting point by as much as 60% is obtained from particles <10Å in size. Consideration of the constraining effect of the elastic matrix show that under the condition of tight coupling with the matrix the melting point of small particles may be significantly enhanced.

The deactivation / reactivation behavior of fresh granular Pt metal catalysts supported on alumina is used to obtain information about the nature of catalytically activesites. It is found that relatively fresh catalysts can in some cases be reactivated by evolving only ∼1/100 of a monolayer of molecules off the Pt metal. This implies that either less than ∼1/100 of the surface is catalytically active with a site turnover frequency of greater than ∼100 reactions/(site-sec) orthat the reactivation process does not evolve small cracked molecules from the catalyticsites and involves some other physical or chemical change of the catalyst surface.

The absorption spectrum of small MoS2 particles in colloidal dispersions differs significantly from that of bulk material. In particular the amplitude and position of exciton absorption peaks shifts when small particles are suspended in a dielectric fluid. Our results are compared with predictions of a Maxwell-Garnett effective medium theory for ellipsoidal particles. We find that particle aspect ratio is an important parameter for interpreting these changes.

Semiconductor nanocrystals have received an extraordinary degree of attention during thepast few years. The electronic, optical, photochemical, and thermodynamic properties of these materials are strongly size dependent, providing a rich area of research, with many potential applications1. It remains true that the range of phenomena that can be studied in nanocrystals is limited by the quality of available samples. In this paper we describe two advances in the preparation of nanocrystalline semiconductor samples. First is the preparation of CdS nanocrystals covalently bound to a metal surface; in this configuration it is possible to study the nanocrystals by photoemission. Second, we report the organometallic synthesis of GaAs nanocrystals.

Films of Co/SiO2 composites were prepared by sequential sputtering from Co and SiO2 targets in Ar. As-grown films as well as films annealed in H2/N2 atmospheres consist of hcp Co particles of uniform size embedded in the SiO2 matrix. The measured optical absorption spectra of these composites are consistent with the Maxwell-Garnett theory except for the absence of the predicted structure that arises from the interband transitions of cobalt. Composites consisting of semiconducting Co3 O4 particles embedded in SiO2 were prepared by post-growth annealing of Co/SiO2 films in air. The optical properties of these composites were found to be comparable to those of CO3O4.