SEARCH · Engineering Papers
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.
A study of polydisperse clusters of fluid drops at supercritical pressures has been conducted using a previously validated model of subcritical/supercritical isolated fluid drop behavior.
Using the results of direct, numerically exact computer solutions of the Maxwell equations, we analyze scattering and absorption characteristics of polydisperse compound particles in the form of wavelength-sized spheres covered with a large number of much smaller spherical grains.The results pertain to the complex refractive indices1.55 + i0.0003,1.55 + i0.3, and 3 + i0.1. We show that the optical effects of dusting wavelength-sized hosts by microscopic grains can vary depending on the number and size of the grains as well as on the complex refractive index. Our computations also demonstrate the high efficiency of the new superposition T-matrix code developed for use on distributed memory computer clusters.
The elements of the scattering or Mueller Matrix for three polydisperse systems of irregular, randomly-oriented particles have been measured in absolute terms as a function of scattering angle for several visible wavelengths. The samples consisted of commercially available silicon dioxide particles that fit three distinct lognormal size distributions. The measured matrix elements were compared with the matrix elements calculated for spheres that had the same refractive index and fitted the same distribution of projected areas. Correlations and discrepancies between the two sets of matrix elements will be discussed.
A study of the d(2) variation for isolated fluid drops and for fluid drops belonging to polydisperse clusters has been conducted at a high temperature and elevated pressures.
Particle size distribution determination by measuring light beam extinction in turbid medium with photometer
Intensity and polarization for meteorological spherical and nonspherical particle size parameters, comparing exact Mie scattering and ray optics
Review of the results of aerosol scattering calculations using models representative of the atmospheric aerosol for three laser wavelengths and two kinds of materials. These calculations show that measurements of scattering and polarization parameters can help significantly in characterizing the earth's aerosol.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The two-phase flow field of a small swirl-stabilized burner has been investigated, using a phase Doppler particle analyzer to obtain the velocity, drop size, and time of arrival components. Water/kerosene comparisons were made for the case of nonreacting flow. It is found that the spray flow field is strongly influenced by the aerodynamic flow field and by the reaction and fluid properties. It is noted that vortex shedding produces clusters of drops.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The dynamics of spray drop interaction with a turbulent coflowing air stream were investigated using a Phase Doppler Particle Analyzer that determines both drop size and velocity. Detailed measurements obtained included spray drop size, axial and radial velocity, angle of trajectory, drop Reynolds number, and size-velocity correlations. The gas-phase flow field was also ascertained via the behavior of the smallest drops. Also investigated were the drag coefficients of droplets in a turbulent air cross flow for both monodispersions and polydispersions. Most notable aspects of the coflow included the effect of air streams with velocities significantly different from the spray sheet. Local changes in number density were produced as a result of lateral convection and streamwise accelerations and decelerations of various drop size classes. The complexity of the spray field interaction illustrated by this data effectively describes the development of the spray due to the influence of the airflow. The droplet drag measurements showed similar behavior for monodispersed and polydispersed flows and similar trends to previously obtained data. The measurements also pointed out further studies which would assist in creating an improved drag law for polydispersed drops in a turbulent environment.
An explicit expression is obtained for the phase matrix in the I, Q, U, and V Stokes vector representation for a system containing a polydispersion of spherical particles. All of the symmetry relations derived by Hovenier using general arguments are established explicitly. Convenient algorithms are given for the computation of the phase matrix for a spherical polydispersion. Since this theory is so vitally important in radiative transfer, many researchers will need to compute these functions for realistic aerosols distributions. Therefore, results are presented for a haze L distribution so that other researchers will have a way of checking their programs which compute these quantities.
Computations of the intensity and linear polarization for single scattering by ice clouds have been made based on the assumption that the particles in ice clouds can be approximated by long circular cylinders which are allowed to be polydispersive as well as arbitrarily oriented in space. The results of two models of optically thin ice clouds are presented and compared with those for polydisperse ice spheres. The two models for ice cylinders are assumed to be either uniformly or randomly oriented in a horizontal plane. Four different wavelengths, 0.7, 3, 3.5 and 6.05 microns, are employed in the light scattering computations. It is found that, compared to ice spheres, long ice cylinders scatter more light in the region with scattering angles near 90 deg, at the expense of scattering in both the forward and backward directions. The glory and cloudbows, which occur in light scattered by spherical particles, are either lost (the glory) or largely reduced and distorted (the cloudbows in the case of cylinders). It is probable that for more irregular particles the cloudbows would also disappear. These differences in scattering by spherical and non-spherical scatterers therefore provide useful information for the differentiation between the ice and liquid phase of cloud particles.
Calculations of light scattering by small particles are important in many diverse fields of science and engineering. In many cases of practical interest, scattering particles are nonspherical and are distributed over sizes and orientations. However, accurate light scattering computations for ensembles of nonspherical particles are difficult and time-consuming, and the literature in which such calculations are reported is rather scarce. In this paper, the T-matrix approach, as extended recently to randomly oriented particles, is used to calculate rigorously light scattering by size distributions of randomly oriented axially symmetric particles. To model the variation of particle sizes in real ensembles, we use a power law distribution typical of some terrestrial aerosols. Contour plots of intensity and degree of linear polarization for polydisperse prolate and oblate spheroids of different aspect ratios and effective equivalent-sphere size parameters from 0 to 10 are calculated and compared with calculations for equivalent spheres. The angular scattering behavior of nonspherical polydispersions is found to be greatly different from that of spheres, while the scattering properties of oblate and prolate spheroids of the same aspect ratio are similar. With increasing particle size, both intensity and polarization become more shape-dependent. In general, nonspherical particles are stronger side scatterers and weaker backscatterers than equivalent spheres. With increasing aspect ratio of nonspherical particles polarization tends to be predominantly positive. Possible effects of particle nonsphericity on optical remote sensing of atmospheric aerosols are discussed.
The introduction of non-intrusive measurement techniques in wind tunnel experimentation has been a turning point in error free data acquisition. Laser velocimetry has been progressively implemented and utilized in various wind tunnels; e.g. subsonic, transonic, and supersonic. The success of the laser velocimeter technique is based on an accurate measurement of scattered light by seeding particles introduced into the flow stream in the wind tunnel. Therefore, application of appropriate seeding particles will affect, to a large extent the acquired data. The seeding material used depends on the type of experiment being run. Among the seeding material for subsonic tunnel are kerosene, Kaolin, and polystyrene. Polystyrene is known to be the best because of being solid particles, having high index of refraction, capable of being made both spherical and monodisperse. However for high temperature wind tunnel testing seeding material must have an additional characteristic that is high melting point. Typically metal oxide powders such as Al2O3 with melting point 3660 F are used. The metal oxides are, however polydispersed, have a high density, and a tendency to form large agglomerate that does not closely follow the flow velocity. The addition of flame phase silica to metal oxide helps to break up the agglomerates, yet still results in a narrow band of polydispersed seeding. The less desirable utility of metal oxide in high temperature wind tunnels necessitates the search for a better alternative particle seeding which this paper addresses. The Laser Velocimetry (LV) characteristic of polystyrene makes it a prime candidate as a base material in achieving the high temperature particle seeding inexpensively. While polystyrene monodisperse seeding particle reported has been successful in a subsonic wind tunnel, it lacks the high melting point and thus is not practically usable in a high temperature wind tunnel. It is well known that rise in melting point of polystyrene can be achieved by a cross-linking technique. Since polystyrene already possesses all the desired characteristics for LV, to circumvent the low melting point, a cross-linking technique was investigated.
We present an experimental study of the dynamics of isolated chains made of super-paramagnetic particles under the influence of a magnetic field. The motivation of this work is to understand if the chain fluctuations exist and, if it does, how does the fluctuation affect chain aggregation. We find that single chains strongly fluctuate and that the characteristic frequency of their fluctuations is inversely proportional to the magnetic field strength. The higher the field the lower the characteristic frequency of the chain fluctuations. In the high magnetic field limit, chains behave like rigid rods without any internal motions. In this work, we used ferrofluid particles suspended in water. These particles do not have any intrinsic magnetization. Once a magnetic field is applied, a dipole moment is induced in each particle, proportional to the magnetic field. A dipolar magnetic interaction then occurs between particles. If dipole-dipole magnetic energy is higher than the thermal energy, the result is a structure change inside the dipolar fluid. The ratio of these two energies is expressed by a coupling constant lambda as: lambda = (pi(a(exp 3))(chi(exp 2))(mu(sub 0))(H(sub 0))(exp 2))/18kT Where a is the particle radius, mu(sub 0) is the vacuum magnetic permeability, H(sub 0) the applied magnetic field, k the Boltzmann constant and T the absolute temperature. If lambda > 1, magnetic particles form chains along the field direction. The lateral coalescence of several chains may form bigger aggregates especially if the particle volume fraction is high. While many studies and applications deal with the rheological properties and the structural changes of these dipolar fluids, this work focuses on the understanding of the chain dynamics. In order to probe the chain dynamics, we used dynamic light scattering (DLS) in self-beating mode as our experimental technique. The experimental geometry is such that the scattering plane is perpendicular to the magnetic field. Therefore, only motions in this plane are probed. A very dilute sample of a ferrofluid emulsion with a particle volume fraction of 10(exp -5) is used in this experiment. We chose such a low volume fraction to avoid multiple light scattering as well as lateral chain-chain aggregation. DLS measures the dynamic structure factor S(q,t) of the sample (q is the scattering wave vector, t is the time). In the absence of the magnetic field, identical particles of ferrofluid droplets are randomly distributed and S(q,t) reduces to exp(-q(exp 2)2D(sub 0)t). D(sub 0)=(kT/(6(pi)(eta)(a)) is the diffusion coefficient of Brownian particles (where Xi = (6(pi)(eta)(a)) is the Stokes frictional coefficient of a spherical particle in a fluid of viscosity eta). If interactions or polydispersity can not be ignored, an effective diffusion coefficient is introduced. Formally, D(sub eff) is defined as: D(sub eff) = - q(exp -2) partial derivative of (ln(S(q,t)) with respect to time, as t goes to 0. D(sub eff) reduces to D(sub 0) if no interactions and only a few particles size are present. Therefore, we can use DLS to measure particle size. The particle radius was found to be a=0.23 mu m with 7% of polydispersity. In this case, if we vary the scattering angle theta (and so q) we do not have any change in the measured diffusion coefficient: it is q-independent. When a magnetic field is applied, particles aggregate into chains if lambda > 1. We first studied the kinetics of the chain formation when lambda = 406. At a fixed scattering angle, we measured diffusion coefficient D(sub eff) as a function of time. Experimentally, we find that D(sub eff) decreases monotonously with time. Physically, this means that chains are becoming longer and longer. Since we are only sensitive to motions in the scattering plane and since chains have their main axis perpendicular to this plane, the measured diffusion coefficient is the trans-verse diffusion coefficient. We can relate D(sub eff) to the mean number of particles per chain N(t) at a given time and to the diffusion coefficient of an isolated particle D(sub 0) as D(sub eff)=f(N(t))D(sub 0). Since f(N) is known from other recent work, N can be expressed as a function of the time. We found a square root dependency: N(t) proportional to the square root of t. As expected for very low volume fraction, this behavior is characteristic of a diffusion-limited aggregation as suggested by several authors and by our previous work. In this study, we focus on the dependence of the effective diffusion coefficient on the scattering angle and the magnetic field strength. After the magnetic field is applied (lambda = 406) for a long time, typically 6 hours, kinetics of chain formation becomes very slow. Chain size does not vary much over the next hour period. Thus, we can perform different interesting experiments. First, at a fixed magnetic field, we measure the effective diffusion coefficient as a function of the scattering angle (from 5 to 130 deg). Our results show that the measured diffusion coefficient increases linearly with the scattering angle: D(sub eff) proportional to q. If we do the same experiment for different lambda values, D(sub eff) depends on lambda as D(sub eff) proportional to lambda(exp -1/2). We also find for different lambda values that the same asymptotic D(sub eff) value is obtained when q approaches zero. The angle dependency of D(sub eff) suggests that an additional motion exists besides chain drifting. Chain size is constant during experiment, which was verified by measuring the same diffusion coefficient at the beginning and at the end of the angle switching. If chains are rigid, D(sub eff) is independent of q. Therefore, we found that D(sub eff) not only measures the motion of the entire chain but also its internal fluctuations. These internal motions are the fluctuations of the particles in the chain. To understand the q dependency of D(sub eff), let us look at the probing length used. In our study, the characteristic length scale probed is l=2pi/q which is in the range of 0.9<l/a<20. When l is much larger than the particle radius (l/a going to infinity) we are mainly sensitive to the center of mass diffusion of the chain. D(sub eff) is then the diffusion of the entire chain and depends only on N but not on q and lambda. The value D(sub eff) allows us to obtain the number of particles per chain N. In the opposite limit (i.e. l/a < 1), we are sensitive to motions on the size of individual particles. The main contributions to the measured diffusion coefficient come from internal motions of the chains (i.e. particles' fluctuations inside the chain). We investigated also the effect of the chain size on D(sub eff). For the same value of the magnetic interaction lambda and different chain sizes, we found that the slope of D(sub eff) versus q decreases when N increases as a power law D(sub eff) proportional to N(exp -0.7). This is an important behavior because this means that the longer the chain the slower its internal fluctuations besides the whole chain motion.