Mitsuaki Hirota

Study of the Wall Effect on Particle Packing Structure Using X-ray Micro Computed Tomography

Abstract The wall effect on particle assembling structure in a particle packed bed is an important issue for powder technology. However, the detailed and systematic experimental data about the wall effect on particle assembling structure in a particle packed bed is hardly adequate yet. The distribution of the local void fraction near the container wall in a particle packed bed is measured using X-ray micro computed tomography (CT). The wall of the cylindrical vessel affects the void fraction distribution, and the distribution can be expressed by the damped oscillation function that shows the relation between the distance from the container wall and the local void fraction. The damping factor, the amplitude and two factors in our damped oscillation function varied with the ratio of particle diameter and inner diameter of the cylindrical container. Our empirical equation can be applied for X-ray CT data and also compared with the traditional experimental data by Ridgway and Tarbuck.

Effect of size distribution on tapping properties of fine powder

Abstract As well known, a size distribution affects the packing properties of a fine powder, but no systematic data can be found about the effect of size or size distribution on packing density of fine powder under 50 μm. Fine powders with log-uniform size distribution were prepared by mixing of mono-sized silica or fly ash powder that was classified by Turbo classifier. The tapping property of the powder packed beds was measured using the Powder tester. From our experimental results, void fraction of the tapped powder beds is affected by size distribution in all tapping numbers. For log-uniform size distribution, the value of void fraction decreases with decreasing inclination of the size distribution. In the case of Andreasen (Gaudin–Shuhmann) distribution, the minimum void fraction was obtainable at the Fuller constant of 0.7–0.9. The experimental results were compared with the estimated values by our proposed model for a multi-component particle mixture.

The effect of the types of mill on the flowability of ground powders

It is empirically known that the flowability of ground powders varies with the types of mill. In this study, five kinds of test powders of the same size made from the same raw material were prepared by five different types of mills and classifiers. The flowability of these powders was measured and the difference between each result identified. Then, the relationship between the flowability of the powder and particle shape was investigated. The circularity, Wadells working sphericity, the fractal dimension of particle perimeters and the specific surface area ratio of the test powders were measured, and the effect of the particle shape on the flowability index was studied. The result was that the particle shape, especially the surface roughness of particles, had a very great effect on the flowability of the powders produced in different types of mill.

Effect of mechanical properties of powder on pneumatic conveying in inclined pipe

Abstract Pneumatic conveying of fine powder has merits, such as no dust pollution and wide flexibility of pipeline layout. Thus, pneumatic conveying is widely used in industry. However, there is no information about the relation between the pressure drop for pneumatic conveying of fine powder and the mechanical properties of powder. We explained that the pressure drop of pneumatic conveying of powder in a horizontal pipe could be estimated from the dynamic friction coefficient of the powder in the previous paper. However, the relation between the pressure drop of pneumatic conveying in an inclined pipe and the mechanical properties of the powder is not cleared yet. The effect of mechanical properties and the angle of an inclined pipe on the pressure drop for pneumatic conveying of fine powder was examined and compared with the calculated results by our model. Based on these results, it is cleared that the pressure drop for pneumatic conveying of fine powder can be estimated from the dynamic friction coefficient of the powder and the inclined angle of the pipe.

Effect of size distribution on the relation between coordination number and void fraction of spheres in a randomly packed bed

Counting the number of contact points on a particle in a packed bed is a time-consuming processes. So, usually, the number of contact points on a particle surface in a randomly packed bed, i.e. the coordination number, is estimated from the void fraction. Many researchers have proposed equations to estimate the coordination number from the void fraction in a packed bed, but all of these equations are obtained for a uniform sized sphere bed. Most actual powder has a wide size distribution, and the effect of the size distribution on the relation between the coordination number and the void fraction is very important. We investigated the effect of the size distribution on the relation using our model and a computer simulation. Based on our results, the coordination number is not changed with the size distribution of the particle bed; however, the void fraction becomes smaller for wider particle size distributions. It means that the value of the coordination number, as estimated by conventional equation for equal spheres from void fraction data, is overestimated for a randomly packed bed of multi-component spheres with wider particle size distributions.

Fractal dimensions of particle projected shapes

In order to quantitatively determine a particle shape, the profile data of particle-projected figures were fed into a personal computer using a digitizer, and then the fractal dimensions of the particles were calculated using the divider method. The fractal dimensions of 28 different kinds of sample particles were compared with Wadells working sphericity and the degree of circularity (or the perimeter ratio) based on the same particle-projected figures. The results show that the fractal dimensions of samples are correlated with circularity. The fractal dimension becomes a useful means for the quantitative representation of a particle-projected shape as well as circularity.

Proposal of an approximation equation for the yield locus to evaluate powder properties

The yield locus (YL) of powder bed can be used to determine many mechanical properties of a powder such as cohesion, unconfined yield stress, stress ratio, etc. Generally, the YL of powder beds is obtained by fitting the results of shear tests to linear approximations based on the Coulomb equation or to curved approximations based on the Warren–Spring equation. Meanwhile, the yielding characteristics of a powder bed are expressed by the Roscoe condition diagram. In this diagram, the YL appears orthogonal to the normal stress axis at both ends corresponding to tensile and compressive strength. However, the YL approximated by the Coulomb or Warren–Spring equations is not orthogonal to the normal stress axis at both ends, and is not the same shape as the YL shown in Roscoe condition diagrams. Thus, the abovementioned mechanical properties obtained from the YL of a powder bed are likely to be affected by the approximate expression for the YL. Despite this, no one has investigated how the mechanical properties of powder beds such as stress ratios are affected by the approximation method for the YL. In this paper, we propose a new approximation equation for the YL that conforms both to the shape of the YL in the Roscoe condition diagrams and experimental results. Then, these YL obtained by our equation, and by the Coulomb and Warren–Spring equations are used to determine the mechanical and flow properties of powder beds. These values are compared with each other in order to discuss the validity of our equation.

The effect of the mechanical properties of powder on the characteristics of dense phase horizontal conveyance of powder

The characteristics of the horizontal conveyance of dense phase flow of powder were experimentally investigated using a progressive cavity pump. The results show that the pressure drop depends on the dynamic friction coefficient of the powder, i.e. dynamic wall friction coefficient or dynamic internal friction coefficient. At the same time, the friction factor also depends on the dynamic friction coefficient. Next, the correlation between the friction factor and the dynamic friction coefficient of powder was discussed. The pressure drop and friction factor were affected by the flow pattern of the powder. These should be estimated by the dynamic internal friction coefficient for plug flow, but in other cases these are estimated by the dynamic wall friction coefficient.

Shape Effects of the Yield Locus on the Rankine Coefficient

The Rankine coefficient is the ratio of the vertical pressure (maximum principal stress) and horizontal pressure (minimum principal stress) acting on a powder bed. The value of the Rankine coefficient is usually determined from the yield locus, which comes from the shear test results either by linear approximation using the Coulomb equation or by curve fitting using the Warren–Spring equation. Since the shape of the yield locus is obtained by approximation, the Rankine coefficient might change with the estimation method. The Rankine coefficient is an important coefficient for estimating the pressure distribution in a powder bed. However, no studies about the effect of the yield locus estimation method on the Rankine coefficient can be found in the past literature. In this paper, first we study the effect of three approximations, i.e. the Coulomb equation, the Warren–Spring equation and an equation proposed by the authors, on the Rankine coefficient. Based on these results, we investigate how the selection of the yield locus estimation method affects the Rankine coefficient. Then we compare the Rankine coefficients obtained from consolidation tests and from each of these yield loci, thereby demonstrating the validity of our proposed approximation equation.

Wear Properties of DLC and Plasma Sprayed WC Structure Coating

Diamond-like carbon film (DLC) with an interlayer of plasma sprayed tungsten-carbide (WC) was prepared on an aluminum alloy substrate (A5052) by a hybrid process of plasma-based ion implantation and deposition using hydrocarbon gas. Typical thicknesses of DLC and WC films were 1 μm and 100 μm, respectively. The hardness and friction coefficient of DLC were typically 15 GPa and 0.15, respectively. The durability of DLC/WC/A5052 system was evaluated from the measurement of the friction coefficient by a ball-on-disk friction tester in which the loaded ball was drawn repeatedly across a sample and the load was increased with each traverse. For the DLC/A5052 system, which has no WC interlayer, the DLC film was broken quickly because of distortion of the substrate. For the DLC/WC/A5052 system, on the other hand, the DLC film was excellent in durability for long running. The wear rate of rubber rotor to the metal rotor was measured by a roller-pitching-type wear testing machine, showing large reduction in wear rate using DLC-coated metal rotor.