A new type of anchor impeller with original baffles, the NX Mixer, has been developed that provides good liquid mixing over a wide viscosity range and good heat transfer at the vessel wall. To evaluate the performance of the NX Mixer, the mixing time was measured by the discoloration method and found to be shorted compared to other impellers. The solid-liquid mass transfer coefficients of solid particles kL were measured by the electric conductivity method. The effect of power consumption per unit volume on the solid-liquid mass transfer was similar to that of the other large impellers. kL of the NX Mixer was larger than that of other impellers when gas was sparged into the vessel. The gas-liquid mass transfer volumetric coefficient kLa of NX Mixer was smaller than that of Rushton turbine impeller, because the gas hold-up of NX Mixer was smaller than that of Rushton turbine impeller. The observed mass transfer volumetric coefficients were correlated with a modified equation based on that of Hiraoka et al. (2001). In addition, the average mass transfer coefficient k and local mass transfer coefficient k at the vessel wall were measured by using ion transfer in dilute solution. The local mass transfer coefficients of the NX Mixer were larger than those of other large impellers at the upper part through the lower part of the vessel wall, because the NX Mixer has a narrower clearance between the impeller and the vessel wall.
An experimental study has been carried out to investigate flow characteristics of the liquid film flowing down on a vertical tube wound spirally with strip fins. We have developed this type of finned tube as an element of an absorber for absorption heat pumps. The liquid was supplied through a crescent feeding section to the outside top of the spiral finned tube. By introducing the parameter k=f/(16/Re), the friction factor through the feeding section was correlated within 20 percent error with an expression derived from Asino's analysis for a flow through the gap between an outer tube and an eccentric inner cylinder. The spirally wound strip fins could generate a dynamically oscillating falling liquid film with short wave length. In the range of film Reynolds number from 500 to 2500, the frequency of the fluctuating liquid thickness was 0.3 to 3 Hz or one-tenth of that on a bare tube, and its wave length was 25 to 30 mm, which is shorter than the wave length of 55 to 120 mm of the falling liquid film on a bare tube. Hence, the number of the waves per unit vertical length is dense, being 1.8 to 4.8 times that on a bare tube. The average falling liquid film velocity and wave velocity have been expressed as functions of film Reynolds number. With increasing film Reynolds number, the ratio of the latter to the former approaches 1.3 for single spirally stacked fins and 2.0 for double spirally stacked fins. The dimensionless film thickness representing the holdup on these finned tubes has been correlated by Eqs. (19)–(21) as functions of film Reynolds number.
A fast and stable solution of matrix equations formed by a finite volume method as a discretization scheme was investigated in steady flow calculation, especially with non-computational cells and cyclic conditions often used in cylindrical coordinate systems. To simulate a computational domain having complex geometry, a method was developed to form a coefficient matrix discretized by a finite volume method. The matrix equations were iteratively solved by the Bi-Conjugate Gradient Stabilized method with polynomial preconditioning (Bi-CGSTAB), and the calculated results were compared with those given by the Tri-Diagonal Matrix Algorithm (TDMA). Convergence and stability of the iterative calculation and under-relaxation factors are also discussed. No differences between TDMA and Bi-CGSTAB for a solution of matrix equations were found in the calculated results. With a greater number of computational cells, faster convergence was obtained by Bi-CGSTAB than by TDMA. Furthermore, larger under-relaxation factors could be employed for the calculation of Bi-CGSTAB than that of TDMA. Bi-CGSTAB is superior to TDMA giving in fast and stable convergence as a solution of the matrix equations discretized by a finite volume method. Therefore, stable calculation and fast convergence can be obtained using the present method.
p-Aminobenzenesulfonic acid is a potential impregnant on activated carbon for the chemisorption of acetaldehyde vapor. However, because of its low solubility in water, it is difficult to prepare the impregnated activated carbon with a high concentration of p-aminobenzenesulfonic acid. In the present work, activated carbons impregnated with various p-aminobenzenesulfonic acid salts were prepared by soaking three types of activated carbon in aqueous solution of p-aminobenzenesulfonic acid with various alkali compounds as solubilizing agents. Ammonium hydrogen carbonate was found to be the most suitable solubilizing agent for the impregnation of p-aminobenzenesulfonic acid. Alkali metal hydroxides reacted with p-aminobenzenesulfonic acid forming p-aminobenzenesulfonic acid alkali metal salts which had no reactivity with acetaldehyde, whereas ammonium hydrogen carbonate produced p-aminobenzenesulfonic acid ammonium salt which had high reactivity with acetaldehyde following the same stoichiometry as that of p-aminobenzenesulfonic acid.
Deformation and rotation process of an interface between two fluids with the same properties has been studied in a three-dimensionally bent channel, which consisted of upstream and downstream square channels connected via a circular channel. A straight fluid interface parallel to the sidewall of the upstream channel was formed in the central region by introducing two fluids with the same properties at an equi-volume rate from the branches of a Y-shaped channel into its stem, which was used as the upstream channel. The fluid interface thus formed was visualized in the upstream and downstream channel cross-sections by the Laser Induced Fluorescence (LIF) method, and its deformation and rotation were investigated. In the case of Reynolds number larger than 2, the interface was curved and stretched by a secondary flow generated in the flow from upstream to downstream via the circular channel, whose length increased linearly to the logarithm of Reynolds number. In the case of Reynolds number less than 2, the interface rotated but was deformed a little from its initial straight shape while the fluid passed through the bent channel. The rotation angle of the fluid interface in the exit cross-section perpendicular to the flow direction was almost equal to the crossed axis angle between upstream and downstream channels. CFD calculations of the position, shape and rotation angle of the fluid interface were in good agreement with the experimental results. The CFD analysis for a rectangular channel demonstrates that the linearized fluid interface rotates almost by the crossed axis angle of a three-dimensionally bent channel, and that the deformation of the fluid interface is small provided that the channel aspect ratio is less than one or the channel height is larger than the channel width.
Monodisperse silica particles were prepared by hydrolysis and condensation of tetraethylorthosilicate (TEOS) at various impeller speeds up to 600 rpm to study the effects of impeller speed on particle size distribution. Ammonia-catalyzed reactions of TEOS were carried out in an ethanol–water solution at a water concentration of 11.0 kmol/m3, ammonia concentrations of 0.5 and 1.0 kmol/m3, and TEOS concentrations of 0.2 and 0.4 kmol/m3 in the presence or absence of an electrolyte (KCl). When the final particles prepared were small, the particle size hardly depended on impeller speed; but, as the size of final particles increased, the particle size more strongly depended on impeller speed, and an increase in impeller speed increased the size of final particles. This impeller speed dependence is considered to be caused by sheer flocculation, which dominates over Brownian flocculation in the case of large particles. When particle size was increased by the addition of KCl, agitation at high impeller speeds brought about excessive coagulation, resulting in generation of hard aggregates and disappearance of spherical particles.
Adsorption kinetics of CO2 gas on Zeolite 13X at high temperature and pressure (~600 K, ~1 MPa) has been investigated experimentally and numerically for application to a CO2 storage system in a CaO/CaCO3 type high-temperature chemical heat pump (CHP). The analytical model adopted reasonably simulates the empirical adsorption profiles on the assumption that mass transfer is controlled by pore diffusion and heat involved by adsorption is transferred within the particle by conduction. The adsorption of CO2 on Zeolite 13X proceeds very fast under the present conditions, and the rate of adsorption was verified to be sufficient to store the CO2 gas produced by decomposition of CaCO3 in the proposed CHP. However, since the adsorption in the early stage is greatly hindered by the heat of adsorption, removing the heat involved by adsorption is very important for the practical use.
Conventional methods for treating waste and ash generated in incineration furnaces are to melt and solidify the ash at high temperature. Though various systems have been installed and operated, they have some critical problems with their high operation cost including fuel cost, technical difficulties in melting mixed materials contained in ash, and practical incapability in melting iron materials and clinkers. The system proposed here is designed to reduce fuel consumption by utilizung the combustion heat of unburned carbonaceous materials contained in discharged ash from a waste incineration furnace to assist melting the incineration ash. The furnace system has oxygen burners for burning carbon and for melting iron materials and clinkers. Test operation in a 12 t-ash/day test plant revealed that the proposed system is capable of melting the ash containing 20–30 wt% of iron and clinker rumps of the size of around φ15 cm without auxiliary fuel supply other than unburned carbon residue in ash.
Experiments were performed to examine Cd removal from waste water by sulfide precipitation using Na2S. It was shown that Cd is precipitated via two reaction paths: direct CdS precipitation, and indirect CdS precipitation through Cd(OH)2 due to hydrolysis of Na2S. In order to realize suitable operation conditions, a dynamic model was built based on the experimental results. Modeling results were in good agreement with experimental results. Furthermore, a pH control method for Cd precipitation and prevention of redissolution of Cd was proposed.
The effects of the mixing ratio of magnesium nitrate hexahydrate on the reaction rate of magnesite reformation were experimentally investigated in order to reform magnesite to magnesium oxide at a lower temperature. The results demonstrated that increasing the ratio of magnesium nitrate hexahydrate increased the reformation rate, and magnesite was successfully reformed to magnesium oxide at a temperature of ~300 K lower.
The influence of O2 concentration on the decomposition of halide gases (CCl4, CHF3, CHClF2) by non-thermal plasma was investigated using a wire-in-tube type reactor. Results showed that the decomposition ratio of CCl4, CHF3 and CHClF2 in the plasma reactor was lower in O2–N2 atmosphere than in N2 atmosphere. The major reaction products detected from the decomposition of CHClF2 in O2–N2 atmosphere were HCl, Cl2, HF, F2, CCl2F2, COCl2, COF2 and CO2. In the case of non-thermal plasma decomposition with in-situ absorption of Ca(OH)2, the decomposition ratio of CHClF2 was increased, and halogen byproducts were scavenged by Ca(OH)2, independent of reaction atmosphere. With an increase in O2 concentration, the fraction of CO2 produced by the plasma decomposition of CHClF2 was increased in both the reactors with and without Ca(OH)2 absorbent.