Particle size effect on the filtration drag of fly ash from a coal power plant

In order to investigate the filtration properties of fly ash from a conventional coal power plant, the filtration drag across the dust cake over an absolute fiberglass filter element was measured. A fluidized bed column was utilized to obtain a well characterized particle stream. The cake resistance coefficient was analyzed by the equation proposed by Endo et al. [1998] in order to observe the effect of particle size and polydispersity. The filtration drag was measured for three kinds of particle stream having the geometric mean particle size of 3.15, 6.07, and 7.83 μm and the geometric standard deviation less than 1.44 in the practical operation conditions for the field applications of face velocity of 0.03–0.06 m/s and area dust load up to 0.2 kg/m². A dust cake of smaller particle size showed larger pressure drop even though the porosity was higher and presented high compressibility according to the face velocity. The particle polydispersity was also a dominant factor affecting the compressibility of the dust cake.     

Collective dynamics modeling of polydisperse particulate systems via Markov chains

This paper develops an efficient approach to modeling and analyzing the overall dynamics of polydisperse particulate systems, exemplified using a rotating drum with horizontal axis, under both constant and time-varying operating conditions. This approach captures the collective dynamics using stochastic models in the form of Markov chains. The characteristics of such dynamics can be obtained from the Markov chains operator. It provides a systematic way to the analysis of collective dynamical features of particle movements. The obtained operators are used to estimate the spatial particle distribution and the degree of particulate mixing as examples of collective dynamic features of polydisperse particulate systems. In this paper, Markov chains models were developed from discrete element method simulation results to show the effectiveness of the proposed approach.     

SOLUTE REJECTION IN THE ULTRAFILTRATION OF POLYDISPERSE ORGANICS FROM NATURAL PRODUCTS

Solute rejection in the ultrafiltration of solutions containing polydisperse solutes was modeled using a spherical solute/cylindrical capillary model, accounting for steric hindrance and wall drag effects. A power law relationship was used for the solute radius-molecular weight relationship. The three parameters in the nondimensional model are the ratio of mean pore radius/solute radius coefficient, the exponent in the solute radius versus molecular weight relationship, and the standard deviation of the logarithms of the pore diameters. Values of the parameters, obtained by fitting the model to rejection coefficient data for solutions containing dissolved organics from wood, were self-consistent and made physical sense. The model provides a useful tool for evaluating ultrafiltration membranes for specific solute fractionation applications.     

Methods for constructing analytic phase function for small spherical particle polydispersions

Two ways of constructing analytic phase function for a polydispersion of small spherical non-absorbing particles have been investigated. First one is a straightforward procedure emanating from the implementation of single particle scattering input into the defined polydisperse phase function. This results in an analytic phase function in terms of moments of the distribution. The second approach is a new strategy, based on the Lagrange mean value theorem. A clear understanding of the relationship between these two approaches has been developed. The efficacy and accuracy of the scattering phase functions is illustrated by applying it to a power-law size distributed sphere ensemble.     

Extinction condition of counterflow spray diffusion flame with polydisperse water spray

The effect of polydisperse water droplet size distribution on the burning behavior and extinction condition of counterflow spray diffusion flame was investigated experimentally in this study. N-heptane as liquid fuel spray and nitrogen as a carrier gas were introduced from the lower duct while water spray and oxidizer consisting of oxygen and nitrogen was issued from the upper duct. The burning behavior of spray flame for different fuel droplet size with and without water spray was observed and the extinction condition of counterflow spray diffusion flame was characterized by oxygen concentration at extinction. The results show that the minimum value of oxygen concentration at extinction for counterflow spray diffusion flame with water spray is similar to the extinction condition without water spray for higher mean droplet diameter of water. The minimum value of oxygen concentration at extinction shifts to the smaller fuel droplet size when decreasing the water droplet size. For fuel droplet size higher than 48 μm, the optimum of water droplet size for suppressing counterflow spray diffusion flame was smaller than gaseous flame. The explanation of optimum water droplet size based on the coupled effect of Stokes number and vaporization Damköhler number can be used for prediction of the effectiveness of water droplet on the suppression of counterflow spray diffusion flame.     

Percolation theory in SOFC composite electrodes: Effects of porosity and particle size distribution on effective properties

A percolation model, accounting for polydispersion of powders and presence of pore formers (i.e. porosity), is presented to predict effective properties of composite electrodes for solid oxide fuel cells, such as the three-phase boundary length and the mean hydraulic radius. Porosity affects both numbers of contacts (so probabilities of connection) and number of particles per unit volume. Both these effects, together with granulometric distribution, are accounted for the estimation of effective properties. As a consequence, the theory can predict numbers of contacts, coordination numbers and therefore effective properties of the electrode for multicomponent polydisperse mixtures.Model simulations show that the three-phase boundary length sharply decreases as porosity increases while the effects of polydispersion of powders are less pronounced, although significant, suggesting that these features should be considered in SOFC electrode models.     

Polydisperse granular flows in a bladed mixer: Experiments and simulations of cohesionless spheres

The flow and segregation of polydisperse, spherical particle mixtures in a bladed mixer was investigated using experimental and computational techniques. Discrete element simulations were able to reproduce the qualitative segregation profiles and surface velocities observed experimentally. For a binary system with a 2:1 size ratio, segregation by size occurs due to a sieving mechanism. Segregation in the binary system is fast, with a fully segregated system observed after just 5 revolutions. However, the numerical simulations showed that the extent of segregation in the bladed mixer can be reduced by introducing intermediate particle sizes in between the smallest and the largest particles. Addition of intermediate particle sizes increases convective and diffusive particle motion promoting a mixing mechanism that reduces segregation via the sieving mechanism. Void fraction within the bladed mixer increases as the degree of polydispersity is increased allowing the particles to move more freely throughout the particle bed. Higher void fractions also increase the ability of large particles to penetrate deeper into the particle bed. Normal and shear stresses are also affected by particle size distributions, with lower average values obtained for the system with the largest number of particle species. Differences in the amount of stress generated by each particle species were observed. However, the difference in stresses is reduced as the number of particle species in the system is increased.     

Experimental and numerical investigation of effects of particle shape and size distribution on particles’ dispersion in a coaxial jet flow

In this study, an experimental and a numerical investigations are performed to investigate the effect of particle’s shape and size distribution on its dispersion behavior. Firstly, particle dispersion of pulverized coal and spherical polymer particles is observed by Particle Image Velocimetry (PIV) technique in the experiment. Secondly, a simulation is performed to analyze the particle dispersion in detail. Spherical and spheroidal motion models are applied to particle’s movement to investigate the shape effect. Furthermore, monodisperse and polydisperse for particles are applied to investigate the size distribution effect on the dispersion. Experimental results show that in the jet turbulence flow, pulverized coal particles, which have complex shapes and various sizes, have quite different dispersion behavior compared to spherical particles. In terms of the results of the simulation, this difference is mainly caused by the size distribution effect. Although particle’s shape affects the dispersity, it is weakened by the size distribution effect.     

Deposition and coagulation of polydisperse nanoparticles by Brownian motion and turbulence

Turbulent and Brownian coagulation rates as well as deposition coefficients of polydisperse nanoparticles were measured experimentally. The coagulation rates were obtained from the change in the total number concentration of polydisperse sodium chloride aerosols, with geometric mean diameters ranging from 30 to 120 nm, in a closed chamber at atmospheric pressure. The geometric standard deviations of the experiments were in the range of 1.55–1.65. The experimental coagulation rates took deposition rates into account, because coagulation and deposition occur simultaneously in a closed chamber. As a result of deposition, it was shown that the experimental deposition coefficients of polydisperse aerosols were agreed well with the theoretical data of Park et al. [(2001). Wall loss rate of polydispersed aerosols. Aerosol Science and Technology, 35, 710–717]. It was shown that the effect of the coagulation was much greater than that of the deposition in the high particle number concentration. In addition, the results represented that bigger turbulent coefficients, caused by higher fan rotation speeds, make the turbulent coagulation process become stronger. In the small particle size range, however, the coagulation rates tend to converge though turbulent coefficients are different. In conclusion, it was shown that experimental coagulation rates followed the values of Lee and Chen [(1984). Coagulation rate of polydisperse particles. Aerosol Science and Technology, 3, 327–334], which were calculated for polydispere aerosols in the gas-slip regime and free-molecule regime.