SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Reactants released by crushing or puncturing. Agglomerated gas-filled spheres hexagonally close packed and sintered or glued together into rods strung together at ends. Rods fed into crushing machine to release material in spheres as needed.
Spheres sized and treated for desired sieve properties. Filter constructed from densely packed spheres restrained by screens. Hollow gas-filled plastic or metal spheres normally used. Manufactured within one percent or better diameter tolerance. Normally, all spheres in filter of same nominal diameter. Filter used as sieve to pass only particles smaller than given size or to retain particles larger than that size. Options available under filter concept make it easy to design for specific applications.
Here, we present a statistical mechanical description of randomly packed spherical particles, where the average coordination number is treated as a macroscopic thermodynamic variable. The overall packing entropy is shown to have two contributions: geometric, reflecting statistical weights of individual configurations, and topological, which corresponds to the number of topologically distinct states. Both of them are computed in the thermodynamic limit for isostatic and weakly underconstrained packings in 2D and 3D. The theory generalizes concepts of granular and glassy configurational entropies for the case of nonjammed systems. It is directly applicable to sticky colloids and predicts an asymptotic phase behavior of sticky spheres in the limit of strong binding.
Explore the source record for details and available documents.
Not provided.
The capillary pressure in the pores defined by equidimensional close-packed spheres is analyzed numerically. In the absence of gravity the menisci shapes are constructed using Surface Evolver code. This permits calculation the free surface mean curvature and hence the capillary pressure. The dependences of capillary pressure on the liquid volume constructed here for a set of contact angles allow one to determine the evolution of basic capillary characteristics under quasi-static infiltration and drainage. The maximum pressure difference between liquid and gas required for a meniscus passing through a pore is calculated and compared with that for hexagonal packing and with approximate solution given by Mason and Morrow [l]. The lower and upper critical liquid volumes that determine the stability limits for the equilibrium capillary liquid in contact with square packed array of spheres are tabulated for a set of contact angles.
Packed spherical particle beds have wide applications throughout the process industry and are usually analyzed using an appropriate combination of laminar and turbulent flows with empirically derived coefficients of which the Ergun (1952) relation is probably the best known. The 3-D complexity of the void distribution within the beds has precluded detailed studies of sphere clusters. Numerical modeling and flow vector visualization for the basic tetra- and hexa-sphere clusters and two hyper-sphere clusters are presented at two Reynolds numbers, 400 and 1200. Cutting planes are used to enable visualization of the complex flows generated within the sphere clusters and are discussed herein. The boundary conditions and flow fields for the simple clusters are also compared to the hyper-clusters with larger variations found for hexa-clusters.
Unique sphere-packing mesophases such as Frank–Kasper (FK) phases have emerged from the viable design of intermolecular interactions in supramolecular assemblies. In this work, a series of C n -G2-CONH 2 dendrons possessing an identical core wedge are investigated to elucidate the impact of peripheral alkyl chain lengths (C n ) on the formation of the close-packed structures. The C 18 and C 14 dendrons, of which the contour lengths of the periphery L p are longer than the wedge length L w , assemble into a uniform sphere-packing phase such as body-centred cubic (BCC), whereas the C 8 dendron with short (L p < L w ) corona environment forms the FK A15 phase. Particularly in the intermediate C 12 and C 10 dendrons (L p ≈ L w ), cooling the samples from an isotropic state leads to cooling-rate-dependent phase behaviours. The C 12 dendron produces two structures of hexagonal columnar and sphere-packing phases (BCC and A15), while the C 10 dendron generates the A15 and σ phases by the fast- and slow-cooling processes, respectively. Our results show the impact of peripheral alkyl chain lengths on the formation of mesocrystal phases, where the energy landscape of the dendrons at L p /L w ≈ 1 must be more complex and delicate than those with either longer or shorter peripheral alkyl chains.
The common spherical mesophases observed from chemically distinct compounds have promoted extensive studies of packing structures in soft-matter self-assemblies. In this report we present a new approach for controlling sphere-packing phases by modulating homologous copolymer chain configurations. To this end, three sets of sample pairs, compositionally asymmetric polydimethylsiloxane-b-poly(2,2,2-trifluoroethyl acrylate) (PDMS-b-PTFEA) diblock copolymers (diBCPs) and their homologues containing the corresponding triblock copolymer (triBCP), are synthesized via atom transfer radical polymerization and designed with the consistent volume fractions of the core-forming PTFEA block, f PTFEA = 0.20, 0.18, and 0.15. We identify the formation of C15 and hexagonally close-packed (HCP) phases in the triBCP-containing PDMS-b-PTFEAs, whereas the sigma and C14 phases are formed in their diBCP analogues, respectively. These phase shifts into C15 and HCP, due to molecular configuration differences, reveal a significant yet unrealized role of chain configurations and the associated chain size effects in determining the stable sphere-packing phases in BCP systems.
Since the discovery of the Frank–Kasper (FK) phase a decade ago, single-component block copolymer systems have offered unique opportunities to explore the structural and phase-transition properties of topologically close-packed FK phases composed of point-particle motifs. Here, we present the formation of complex low-symmetry phases and their thermodynamic stability using high-χ and high-ε linear diblock copolymers, where χ and ε are the Flory–Huggins interaction parameter and conformational asymmetry between the two blocks, respectively. In this work, a series of polydimethylsiloxane-b-poly(2,2,2-trifluoroethyl acrylate) (PDMS-b-PTFEA) copolymers are prepared by tuning the volume fraction ($f_{\text{PTFEA}}$) of a minor PTFEA block. The packing structures of PDMS-b-PTFEAs are mapped in the space of temperature versus $f_{\text{PTFEA}}$ using X-ray scattering measurements. In addition to the packing structures reported earlier, we identify another A15 phase at higher $f_{\text{PTFEA}}$ close to a hexagonally packed cylinder (HEX) phase and establish the phase sequence of HEX–A15–σ–C14–BCC with decreasing $f_{\text{PTFEA}}$. Theoretical calculation of relative free-energy densities confirms the experimental phase sequence of the composition-dependent sphere-packing structures. Interestingly, the domain sizes of three-dimensional (3D) sphere-packing structures are nearly temperature-invariant compared to that of a two-dimensional HEX structure, and we attribute this dimensional stability to the strong enthalpic effects in the formation of 3D packing structures and the geometrical characteristics of spherical domains. Our results demonstrate that the interaction parameter χ, together with the known role of conformational asymmetry ε, serves as the primary thermodynamic parameter for selecting and stabilizing various low-symmetry packing structures in sphere-forming linear diblock copolymer systems.
Artificial neural networks offer a highly nonlinear and adaptive model for predicting complex interactions between input-output parameters. However, these networks require large datasets which often exceed practical considerations in modeling experimental results. To alleviate the dataset size requirement, a method known as physics guided machine learning has been applied to construct several neural networks for predicting propeller tonal noise in the time domain over a broad range of flight conditions. Three space-filling designs, namely, Latin-Hypercube, Sphere-Packing, and Grid-Space, were used to distribute points throughout the input parameter space encompassing nondimensional flight conditions and observer geometry. Each neural network’s performance was validated by conditions outside of the training set and compared to the Propeller Analysis System tool from the NASA Aircraft Noise Prediction Program. Compared to the Grid-Space input design, the Latin-Hypercube and the Sphere-Packing designs provided a better representation of the domain for training. Regarding the network archetype, a fully connected perceptron was found to outperform the partially connected perceptron in their ability to predict tonal noise for small datasets. The black-box nature of these neural networks was also explored to understand how the networks constructed the waveform and understand why some network designs produce better models.
We examine random packs of discs or spheres, models for ammonium-perchlorate-in-binder propellants, and discuss their average properties. An analytical strategy is described for calculating the mean or effective heat conduction coefficient in terms of the heat conduction coefficients of the individual components, and the results are verified by comparison with those of direct numerical simulations (dns) for both 2-D (disc) and 3-D (sphere) packs across which a temperature difference is applied. Similarly, when the surface regression speed of each component is related to the surface temperature via a simple Arrhenius law, an analytical strategy is developed for calculating an effective Arrhenius law for the combination, and these results are verified using dns in which a uniform heat flux is applied to the pack surface, causing it to regress. These results are needed for homogenization strategies necessary for fully integrated 2-D or 3-D simulations of heterogeneous propellant combustion.
Sphere packing is an ancient problem. The densest packing is known to be a face-centered cubic (FCC) crystal, with space-filling fraction Φ FCC = π / √ 18 ≈ 0.74 . The densest “random packing,” random close packing (RCP), is yet ill defined, although many experiments and simulations agree on a value Φ RCP ≈ 0.64 . We introduce a simple absorbing-state model, biased random organization (BRO), which exhibits a Manna class dynamical phase transition between absorbing and active states that has as its densest critical point Φ cmax ≈ 0.64 ≈ Φ RCP and, like other Manna class models, is hyperuniform at criticality. The configurations we obtain from BRO appear to be structurally identical to RCP configurations from other protocols. This leads us to conjecture that the highest-density absorbing state for an isotropic biased random organization model produces an ensemble of configurations that characterizes the state conventionally known as RCP.
Systems are described for using multiple closely-packed spheres. In one system for passing fluid, a multiplicity of spheres lie within a container, with all of the spheres having the same outside diameter and with the spheres being closely nested in one another to create multiple interstitial passages of a known size and configuration and smooth walls. The container has an inlet and outlet for passing fluid through the interstitial passages formed between the nested spheres. The small interstitial passages can be used to filter out material, especially biological material such as cells in a fluid, where the cells can be easily destroyed if passed across sharp edges. The outer surface of the spheres can contain a material that absorbs a constitutent in the flowing fluid, such as a particular contamination gas, or can contain a catalyst to chemically react the fluid passing therethrough, the use of multiple small spheres assuring a large area of contact of these surfaces of the spheres with the fluid. In a system for storing and releasing a fluid such as hydrogen as a fuel, the spheres can include a hollow shell containing the fluid to be stored, and located within a compressable container that can be compressed to break the shells and release the stored fluid.
A videotape presentation of flow in a packed bed of spheres is provided. The flow experiment consisted of three principal elements: (1) an oil tunnel 76.2 mm by 76.2 mm in cross section, (2) a packed bed of spheres in regular and irregular arrays, and (3) a flow characterization methodology, either (a) full flow field tracking (FFFT) or (b) computational fluid dynamic (CFD) simulation. The refraction indices of the oil and the test array of spheres were closely matched, and the flow was seeded with aluminum oxide particles. Planar laser light provided a two-dimensional projection of the flow field, and a traverse simulated a three-dimensional image of the entire flow field. Light focusing and reflection rendered the spheres black, permitting visualization of the planar circular interfaces in both the axial and transverse directions. Flows were observed near the wall-sphere interface and within the set of spheres. The CFD model required that a representative section of a packed bed be formed and gridded, enclosing and cutting six spheres so that symmetry conditions could be imposed at all cross-boundaries. Simulations had to be made with the flow direction at right angles to that used in the experiments, however, to take advantage of flow symmetry. Careful attention to detail was required for proper gridding. The flow field was three-dimensional and complex to describe, yet the most prominent finding was flow threads, as computed in the representative 'cube' of spheres with face symmetry and conclusively demonstrated experimentally herein. Random packing and bed voids tended to disrupt the laminar flow, creating vortices.
The close-packed mesocrystal structures from soft-matter assemblies have recently received attention due to their structural similarity to atomic crystals, displaying various sphere-packing Frank–Kasper (FK) and quasicrystal structures. Herein, diverse mesocrystal structures are explored in second-generation dendrons (G2-X) designed with identical wedges, in which the terminal functionalities X = CONH 2 and CH 2 NH 2 represent two levels of the strong and weak hydrogen-bonding apexes, respectively. The cohesive interactions at the core apex, referred to as the core interactions, are effectively modulated by forming heterogeneous hydrogen bonds between these two functional units. For the dendron assemblies compositionally close to each pure component of G2-CONH 2 and G2-CH 2 NH 2 , their own FK A15 and C14 phases dominate other phases, respectively. We show the existence of the wide-range FK σ including the dodecagonal quasicrystal (DDQC) phases from the dendron mixtures between G2-CONH 2 and G2-CH 2 NH 2 , providing an experimental phase sequence of A15–σ–DDQC–C14 as the core interactions are alleviated. Intriguingly, the temperature dependence of particle sizes shows that the high plateau values of particle sizes are maintained equivalently until each threshold temperature (T th ), followed by a prompt decrease above the T th . A decrease in T th by alleviating the core interactions and its composition dependence suggest that the more size-dispersed particles, the more susceptibility to chain exchange with increasing temperature. Here our results on the formation of supramolecular dendron assemblies provide a guide to understand the core-interaction-dependent mesocrystal structures toward the fundamental principle underlying the temperature dependence of their particle sizes.
The flow experiment consisted of an oil tunnel, 76 x 76 mm in cross-section, packed with lucite spheres. The index of refraction of the working fluid and the spheres were matched such that the physical spheres invisible to the eye and camera. By seeding the oil and illuminating the packed bed with planar laser light sheet, aligned in the direction of the bulk flow, the system fluid dynamics becomes visible and the 2-D projection was recorded at right angles to the bulk flow. The planar light sheet was traversed from one side of the tunnel to the other providing a simulated 3-D image of the entire flow field. The boundary interface between the working fluid and the sphere rendered the sphere black permitting visualization of the exact locations of the circular interfaces in both the axial and transverse directions with direct visualization of the complex interstitial spaces between the spheres within the bed. Flows were observed near the surfaces of a plane and set of spheres as well as minor circles that appear with great circles and not always uniformly ordered. In addition to visualizing a very complex flow field, it was observed that flow channeling in the direction of the bulk flow occurs between sets of adjacent spheres. Still photographs and video recordings illustrating the flow phenomena will be presented.
The study of solute transport in porous media is of interest in many chemical engineering systems. Some example applications include packed bed catalytic reactors, filtration devices, and batteries. The pore scale modeling of these systems is time consuming and may require large computing resources, for this reason computational fluid dynamics (CFD) simulations are not practical if a large number of simulations is required, like in multiscale modeling, where a model at a large scale calls for pore scale simulations. It has been shown that neural networks can be trained with a dataset of flow simulations and then predict fields orders of magnitude faster, and with less computational resources, in new domains. However, it is crucial to provide the neural network with an effective description of the domain and the undergoing operating conditions to be able to train models that generalize accurately in unseen samples. Therefore, research is needed to employ neural networks in new complex systems. The appropriate training of a network for predicting coupled flow and solute transport processes is an outstanding problem due to the complex interplay between geometry and operating conditions. In this work, we train a multi scale convolutional neural network (MSNet) with a diverse dataset of simulations of transport and chemical reaction in porous media to predict the local concentration fields in images of porous media. Our dataset contains a wide diversity of sphere pack arrangements under different operating conditions (Péclet and Reynolds numbers). Further, we train a robust model by employing different input descriptors that represent the medium and the different operating conditions of each system. Our trained model is able to provide nearly instantaneous predictions, compared to around twenty hours of the CFD workflow, with less than 3.5% error on new geometries and transport conditions. Thus the model could be easily integrated in a multiscale workflow where fast response is needed.