Takatoshi Miura

Numerical analysis of heat and mass transfer characteristics in the metal hydride bed

This paper presents a two dimensional mathematical model to evaluate transient heat and mass transfer in the metal hydride bed. Thermophysical properties of metal hydrides in this model depend on the reacted fraction. Hydriding and dehydriding kinetics are described by new model presented by Inomata et al. [1]. The result on velocity, temperature and composition distribution are presented and discussed. Moreover this paper discusses in detail the validity of several assumptions that the local gas temperature is equal to bed temperature and that the convection term in an energy equation for gas phase is neglected.

Measurement and modelling of hydriding and dehydriding kinetics

Abstract The rates of hydriding and dehydriding were measured for LaNi 5 H x under isothermal and isobaric conditions. Obtained data of hydriding rates indicated that the rate controlling step was the nucleation and growth process of the hydride phase and then changed to hydrogen diffusion through the hydride phase at the later stage. On the other hand, for the desorption of hydrogen, the rate controlling step was only the homogeneous chemical reaction of the hydride phase. Theoretical rate equations were derived on the basis of the kinetic model of hydriding and dehydriding reaction proposed in this study. A satisfactory agreement between the theoretical and experimental results was obtained.

Homogenization method for effective thermal conductivity of metal hydride bed

Abstract The effective thermal conductivity of the metal hydride bed is analyzed by the homogenization method. This method can represent the microstructure in the bed precisely. The referenced material is LaNi4.7Al0.3, which is practically used. Temperature and pressure ranges investigated here are from −80°C to +140°C and from 10−6 to 100 bar , respectively. Hydrogen, helium, nitrogen and argon are chosen as the filling gas. The validity of the homogenization method as the multi-scale analysis is confirmed by a good correspondence with the experimental data that the effective thermal conductivity depends on the pressure of the filling gas under these various conditions (Int. J. Hydrogen Energy 23(2) (1998) 107). The homogenization method can become a powerful tool for the estimation of the effective thermal conductivity of the metal hydride bed by considering the microscopic behavior such as the pulverization and the change of the contact area.

Modeling of eddy characteristic time in LES for calculating turbulent diffusion flame

Abstract We have constructed an eddy characteristic time derived from large-scale motion to calculate the combustion reaction rate using a new eddy dissipation concept (EDC) model, and estimated combustion characteristics in the combustor. The calculated temperature and CH 4 mole fraction distribution are in fairly good agreement with the experimental data. However, the calculated CO mole fraction distribution does not agree well with the measured result of CO mole fraction because of using a simple CO reaction mechanism. This study shows that the combustion simulation using LES with EDC model is effective for calculating the characteristics of turbulent diffusion flame.

Performance of numerical spray combustion simulation

Abstract The aim of this study is to examine the performance of numerical spray combustion simulation. A numerical simulation for the prediction of local properties of heavy oil spray flames stabilized by a baffle plate is described. Time-averaged governing conservation equations are solved to estimate the combustion gas flow, gas composition and temperature fields in the experimental combustor. The κ-e turbulence model is used to describe the turbulent flow field. The behavior of fuel droplets in the turbulent combustion gas flow is calculated by the Lagrangian method. The combustion rate of fuel vapor is estimated by the eddy dissipation model. The effects of radiation are accounted for by the six-flux model of radiation. The performance of the simulation is examined by comparison with measured data. In the isothermal (cold) case, the calculated flow pattern is compared with the data measured by LDA, and it is clear that the calculated results show quantitative agreement with the measured data. In the combustion case, however, the simulation cannot predict well the measured profiles of temperature, O 2 and CO 2 concentrations near the baffle plate. It is inferred that this simulation cannot estimate accurately the interaction between the recirculation flow induced by the baffle plate and fuel droplets.

Simulation of soot aggregates formed by benzene pyrolysis

Abstract Experiments and numerical simulations were carried out to analyze the soot aggregation mechanism during benzene pyrolysis. Soot was formed by the pyrolysis of 1 mol% benzene (in 99 mol% nitrogen) in an alumina tube, which was kept at 1573K with variations of residence time (0.03 to 0.5 sec). A new cluster-cluster aggregation model called the Aggregate Mean free Path (AMP) model was developed. This model simulated the cluster-cluster aggregation between soot particles and aggregates using the particles or aggregates mean free paths. The projected images of simulated soot were compared with the electron micrographs of experimental soot with the same magnification. Simulated shape, peripheral fractal dimension and size distributions of aggregates were in good agreement with experimental data.

Numerical analysis of absorbing and desorbing mechanism for the metal hydride by homogenization method

Abstract Storage capacity and fast reaction rate are the most important properties for hydrogen storage alloys. To clarify the mechanism of the rate-controlling step, we analyzed the hydrogen diffusion coefficients of the hydride by the homogenization method, which considers the microstructure of the alloy at the microscale (0.1– 10 μm ) precisely. By applying nucleation and growth model, we have discussed the absorbing and desorbing mechanism of the metal hydride. Our results and analysis show that it is important for the hydriding or dehydriding reaction to consider the number of nuclei and the transfer of hydrogen atom through the α – β interface data. Estimation of the microstructure of the metal hydride by using the homogenization method is useful for the absorbing and desorbing mechanism.

Percolation model for simulation of coal combustion process

A pulverized coal combustion simulation model applying a percolation theory based on a Monte Carlo method has been developed to predict swelling and fragmentation behaviors during a coal combustion process. The shape of pulverized coal particles before reaction was assumed as a three-dimensional cube arranged in large number of small lattices. These lattices were classified into char, volatile, ash, or macropore, depending on the coals industrial analysis value, and they were arranged randomly in the cubic. The coal combustion processes were classified as devolatilization and char combustion, and they were assumed to occur simultaneously. In the devolatilization process, the dual-competing reaction model determined the devolatilization time of a volatile lattice. With coal devolatilization, the coal particles swelled due to increase in internal pressure. With char combustion, O 2 lattices were arranged around a coal particle and the random walk model was applied to represent the O 2 diffusion behavior. Furthermore, the char reaction model was applied to determine the char reaction time. Thus, a char lattice was lost when its total reaction time was longer than the O 2 diffusion time and the reaction time. With char combustion, ash agglomeration occured. Char combustion finished when all of char lattices were burned out or the char lattices were completely surrounded by ash lattices, preventing oxygen lattices from coming into contact with the char lattices. It was shown that this model well represents the difficulty of char burnout caused by increase in diffusion resistance in the latter period of the reaction. A particle temperature profile was determined by a model calculation in an environment in which atmospheric temperature was 1500 K. Using only the industrial coal properties, this model predicts detailed variations of reaction rate, porosity, and maximum relative particle diameter with particle conversion in the pulverized coal combustion process.

Estimation of thermal stress in lump coke

Abstract Permeability in the blast furnace is maintained by lump coke and is governed by the mean size and size distribution of the coke. Coke lump size is determined mainly by the growth of macro-cracks from the wall side to the centre of the coke oven charge, their formation behaviour depending on the property of the coal charged and the operating conditions. In the formation process, micro-cracks are simultaneously formed in all directions and their formation also plays a part in determining the strength and therefore the lump size of coke for the blast furnace. This study estimates the deformation behaviour of coke and the thermal stress distribution by assuming the size of lump coke formed in the coke oven (distance between macro-cracks) and the conditions required for fissure propagation, considering radiative heat transfer within the fissures. The tensile stress parallel to the oven wall is estimated to reach ∼ 10–20 MPa near the wall side. The growth of fissures arising from the wall side is successfully estimated. No fissure parallel to the oven wall is recognized, since only compressive stress is predicted.

Simultaneous estimation of thermophysical properties by periodic hot-wire heating method

Abstract A new unsteady heat flux method is proposed to measure thermal conductivity and diffusivity simultaneously, the principle based on an analytical solution for an infinite hollow cylindrical system with a periodic heat source in the center. Thermal conductivity and diffusivity are determined from the amplitude and phase lag of the temperature response within the cylinder. The method also makes it easy to measure the temperature dependency of thermal properties during a continous heating process. The measurement errors caused by the finite specimen size and the displacement of the thermocouple location are estimated numerically to confirm the accuracy of the measurement method. Effective thermal conductivity and diffusivity for the packed beds of aluminum oxide particles and potassium perchlorate particles are measured. The thermophysical properties measured by this method agree well with those measured by conventional methods such as the hot-wire, periodic heating, and continuous heating methods.