To investigate the collection of particles in column flotation, quartz particles 32 μm to 45 μm diameter were continuously flotaled by using a cationic surfactant as a collector. The axial dispersion of the particles was estimated by measuring the transient particle concentration at the bottom of the column as a function of flotation time in the absence of the collector. As a result, the value of the dispersion coefficient is found to be relatively small, and this is supported by previous investigators results. The value depended slightly on the varied gas flow rates. The continuous and dispersed phase particle concentrations and their removabilities are predicted by a series of mathematical equations which involve a net volumetric flotation rate constant, K and the particle dispersion coefficient stated above. The value of the rate constant for each collector concentration was experimentally determined by applying the predictive equations to the experimental continuous phase concentration profiles, formed along the column under steady-state condition. It was found that K depends strongly on the collector concentrations. When the particle surface is highly hydrophobized, the value of K could also be supported in the order of that based on the theoretical prediction of previous investigators.
A chromatography system for manufacturing a chiral antihypertensive drug, namely semotiadil fumarate, was studied. The racemate of semotiadil fumarate needs to be completely resolved because high purity is needed for the compound’s use as medicine. For this reason, high performance liquid chromatography (HPLC) was adopted. To supply the requirement, a continuous system is needed. In this case, a system having plural batch columns injected sequentially was employed. To design the system, the number of columns needed was predicted by simulation. The result was in good agreement with experimental results. To fractionate the racemate completely, it was found that the valves in the column outlets were related to their absorbance. It was found that these valves worked steadily when the disturbances in the absorbance detected were removed using first-order lag conversion. On the basis of the above findings, a system was designed and actually made. The eluate was passed through the detector only when it changed from one enantiomer to another, so the system was provided with only two detectors. The influences of some pipe puddles were large; however, with a wash system devised to complement the system, the racemate of semotiadil fumarate was separated continuously and completely. In this system, the purity, activity, and recovery are similar to that of a single column. However, because the throughput is larger, the system is useful for making the antihypertensive drug.
A new chemical heat pump designed to utilize high-temperature heat above 800°C generated from a high temperature gas nuclear reactor or other high-temperature industrial process is presented. Based on experimental results, a heat pump that uses a calcium oxide/lead oxide/carbon dioxide reaction system appears to be a suitable system. To demonstrate the validity of the heat pump, the equilibrium relationship and kinetics of calcium oxide/carbon dioxide and lead oxide/carbon dioxide, which comprise the reaction system of the heat pump, are studied experimentally. A study of the equilibrium relationship of lead oxide/carbon dioxide, which consists of a three-step equilibrium, reveals that the highest temperature equilibrium relationship of the three-step equilibrium is associated with optimal heat pump operation. The practical operation conditions of the heat pump are determined based on the equilibrium relationship and kinetic experiments. The proposed heat pump may be able to operate as a heat transformation-type heat pump, and is capable of storing heat above approximately 830°C and transforming the heat to a higher temperature of more than 870°C under subatmospheric pressure and thermal driving conditions with no mechanical work. The calculated mean heat output and heat output amount are 670 W/kg-CaCO3 and 1200 kJ/kg-CaCO3, respectively, at 870°C, 1 atm for 30 min. Thus, the new heat pump can be applied to a heat storage and heat transformation system for high temperature processes.
The dynamic ultrafiltration process of protein solution is experimentally investigated by measuring electrical conductivity within the filter cake. Three pairs of platinum wire electrodes were employed to measure the formation process of the filter cake on the membrane surface. The filter cakes formed from protein solutes behave as highly compressible cakes and shifting of pH from 5 to 4 or 9 results in a decrease in cake compressibility, as well as specific filtration resistance. The experimental results for bovine serum albumin and soy protein solutions coincide favorably with the calculated results based on the compressible cake filtration theory.
Application of piston-driven ultra-rapid pressure swing adsorption (URPSA) to CO2 recovery from a stack gas was investigated by using a model stack gas. Specific objectives of this work are to experimentally achieve high capacity CO2, reduction and to elucidate the effects of moisture contained in the stack gas on CO2 recovery by the URPSA using hydrophilic and hydrophobic zeolites. Also, CO2 recovery using URPSA was investigated by numerical simulations on the basis of a simplified mathematical model. CO2 recovery from a model stack gas is successfully demonstrated by piston-driven ultra-rapid PSA. The moisture contained in a model stack gas was accumulated in hydrophilic zeolite and drastically reduced CO2 adsorption capacity. However, since the effects of accumulation of H2O are negligibly small in the case of hydrophobic zeolite, hydrophobic zeolite should be employed for this purpose. The production capacity of the URPSA was about one order of magnitude higher than those of conventional PSAs. Still, the performance of CO2 recovery of the URPSA was low. It is known that decreasing the pressure drop of the adsorbent and optimizing the sequence control are needed to improve the efficiency. Numerical simulations using a simplified mathematical model were successful, and the model developed will be useful for further investigations.
To elucidate the antibacterial mechanisms of the ceramics magnesium oxide (MgO), calcium oxide (CaO) and zinc oxide (ZnO), damage to bacteria caused by these powder slurries are studied on the basis of change insensitivities to antibiotics, of which the primary inhibitory actions are understood well. Four kinds of antibiotics, penicillin G, chloramphenicol, nalidixic acid and rifampicin, were used as the selective reagents. The MgO and CaO powder slurries increased the sensitivities of Escherichia coli to rifampicin and chloramphenicol. Though the MgO and CaO powder slurries have high pH values, changes in the sensitivities by the MgO and CaO powder slurries were obviously different from those via alkaline treatment. The ZnO powder slurry enhanced the sensitivity of the E. coli to chloramphenicol. This result suggests that the antibacterial actions of MgO and CaO powder slurries were different from those of ZnO powder slurry.
A separation process for rare earth elements (Pr/Nd and Y/Er) by liquid-liquid extraction with bis(2-ethylhexyl) phosphinic acid is investigated. Extraction equilibrium formulations for these elements are established and extraction constants are determined employing kerosene and toluene as diluents, in which MR3(RH)3 is formed at low loading, and then the larger aggregated species, (MR3)2(RH)4, appears with increasing loading ratio. The proposed formulations and constants successfully express the extraction behaviors over the whole range of loading ratio up to 0.4 in single element systems. These are valid for binary element systems of Pr/Nd and Y/Er and also for the distribution behaviors in scrubbing treatment. The separation is enhanced at high pH region, that is, at high loadings. Scrubbing is also effective for separation. Simulation of Pr/Nd separation by continuous counter-current mixer-settler cascades shows that separation is enhanced by employing appropriate pH control and scrubbing treatment.
Adsorption characteristics of organic compounds on an octadecylsilyl (ODS)-silica gel for two mobile phases, i.e, methanol/water (70/30, v/v) and acetonitrile/water (70/30, v/v) are compared. It is confirmed that the values of adsorption equilibrium constant, isosteric heat of adsorption, and activation energy of surface diffusion are smaller in 70 vol % acetonitrile than in 70 vol % methanol. On the contrary, the values of surface diffusion coefficient, Ds, are larger in 70 vol % acetonitrile than in 70 vol % methanol. These results indicated that the interaction between ODS ligands and solute molecules was weaker in 70 vol % acetonitrile than in 70 vol % methanol. The ratio of Ds to molecular diffusivity, Dm, is about 0.3 irrespective of chromatographic conditions, i.e., the type of organic modifiers and sample compounds, and temperature. It is possible to calculate Ds from Dm, which can be estimated by various correlations previously proposed in the literature. The almost constant ratio of Ds to Dm suggests the presence of a kind of correlation between surface and molecular diffusions.
The steady state filtration flux in electrically-enhanced crossflow microfiltration is estimated using a correlation equation proposed for several kinds of suspensions. Baker's yeast and Rhodotorula glutinis were used as model samples of microbial cells, and PMMA particles were used as samples of non-living solids. Application of the electric field in crossflow microfiltration is a useful method for improving the filtration flux of these samples. High flux levels for the cells were achieved when an electric field above 3000 V/m was applied. The effect of the electric field in increasing the filtration flux of the steady state was analyzed theoretically using a force balance model where the viscous drag force, FJ, the electrophoretic force, FE, and the re-entraining force, FB, were considered to act on a particle on the membrane surface under a steady state of filtration, respectively. From force balance analysis, it is found that on application of an electric field, the electro-osmotic effect can be neglected in the present study, so that the filtration flux of the steady state, JES, can be presented by, JES = UEPE + J0S where UEP is the electrophoretic mobility of particles and E is the electric field applied. J0S is the filtration flux in the absence of an electric field, which is correlated with the operating parameters for suspensions tested.
In the present paper, an incinerator of waste liquids is developed so that atomization takes place upwards on the centerline of the burner in a riser. Waste liquids are modeled by using three liquids, namely, water, water-ethanol and a solution of water-ethanol-ammonium sulfate. The stable combustion region of a methane-air diffusion flame with the atomization of liquids, and the behavior of the droplets and NOx emission in the riser are studied. The stable combustion region expands somewhat to a higher equivalence ratio when material with heat generation like ethanol is contained in the liquid. Conversion from nitrogen in ammonium sulfate (fuel-N) into NOx falls with increasing atomization rate. Preferable fittings between the data in the present work and Fenimore’s correlation are obtained. It is found that almost all evaporation of droplets occurs in the flame, and the behavior of the evaporation rate of atomized droplets is described well by the behavior of the NOx emission rate.
The flow pattern of aerosol and sheath air in a cylindrical annulus of a Differential Mobility Analyzer (DMA) is critical to determine the classification performance of the DMA. In the present work, a new technique to check aerosol flow inside the DMA is proposed and the influences of aerosol flow distortion and broadening on the classification performance are investigated by means of the Tandem DMA method. As a result, it is found that (1) the annular flow inside the studied DMA is a fully developed laminar flow with a negligible mixing between aerosol and sheath air when the aerosol flow rate is less than 3 l/min and the sheath air flow rate is 17.5 l/min, (2) a larger width of aerosol entrance slit causes uneven distribution of aerosol flow in the circumferential direction, and (3) both the uneven distribution and broadening of the aerosol flow lead to a broader band of classified particle electrical mobility, as well as a shift of peak electrical mobility to a smaller range. These results suggest that aerosol flow inside a short-type DMA is more sensitive to the flow rates and its structure compared to the conventional TSI type DMA.
The influence of liquid flow on the growth of red beet hairy roots is investigated by using a single column reactor operated at 25 °C under a fixed value of superficial velocity of medium. In the cultures of the hairy roots for 172–190 h, average growth rates of 2.8 × 10–2, 1.6 × 10–2 and 0.98 × 10–2 kg/(m3·h) are obtained at superficial velocities of 4.2 × 10–3, 7.8 × 10–3 and 11.1 × 10–3 m/s, respectively. Using the packed column with the hairy roots at void fractions of 0.49 to 0.71, it is found that pressure drop through the root bed increases with an increase in superficial velocity of medium elevated up to about 14 × 10–3 m/s. The effective diameter of the hairy roots is estimated at 1.5 × 10–4 m, considering the root hairs found on the surfaces of the main roots. Pressure drop through the root bed in the column can be correlated with superficial velocity of the medium by using an extended Ergun’s equation expressed in terms of effective diameter of the hairy roots and void fraction in the bed. When the hairy roots are kept for 50 h in the column under pressure drop values of 8.4 × 103 to 3.1 × 104 Pa/m, the reduction in viability of growing points at the hairy root tips is observed with increasing pressure drop in the range.
A hybrid system of latent heat storage and spray flash evaporation is devised in an attempt to develop an energy saving system, which stores intermittent thermal energy such as solar heat, waste heat or heat from the surplus steam of a power station at night and utilizes the stored energy for the generation of steam for industries, power stations and domestic air conditioning. Experimental results of the transient discharge characteristics of the heat storage column packed with the phase change material and the transient flash evaporation characteristics are reported. It is also shown that by using the numerical results of the discharge characteristics and the empirical equation of the efficiency of flash evaporation, the amount of generated steam can be predicted with sufficient accuracy. The utilization of more than 97% of the amount of stored energy for the generation of steam, substantiated experimentally, confirms the high efficiency of the present system.
The volumetric over-all mass transfer coefficient Kca and the volumetric dispersed phase mass transfer coefficient kda in a mixer-settler extraction column were measured. In the measurement of Kca, iodine was extracted from I2-KI aqueous solution into heptane, while iodine was back-extracted from a heptane solution of iodine into sodium thiosulfate aqueous solution to obtain kda. By using specific interfacial area determined from the estimated Sauter mean diameter of the dispersed drops and the estimated holdup of the dispersed phase, the mass transfer coefficients for the dispersed and the continuous phases were determined from the volumetric coefficients Kca and kda. The coefficient kd for the dispersed phase coincided with the theoretical value based on diffusion within a rigid sphere using the residence time distribution of dispersed drops. On the other hand, the coefficient kc for the continuous phase agreed well with the correlation of mass transfer around a rigid sphere by assuming that the relative velocity of the dispersed drop to the continuous phase was given by the terminal settling velocity of a rigid sphere having the same diameter and density as the drop.
3D periodical turbulent flow induced by a low shear hyperboloid stirrer in a fully baffled stirred tank reactor is experimentally studied at a constant Reynolds number, 5600, by using a diode fiber laser-Doppler anemometer. The ratio D:T:H of the hyperboloid stirrer diameter, D = 50 mm, the inner tank diameter, T, and the filled fluid height, H, is 1:3:3. Three low stands of hyperboloid stirrer clearance, D/10, D/7 and D/5, are also targeted to evaluate influence on flow structure. The experimental results are applied to verification of CFD-code based on a standard k–ε model. The flow field close to the hyperboloid stirrer is characterized by a series of vortex jets that are trailing periodically down within a radial extension of 0.5D to 0.7D. The maximum measured tangential velocity was 96% of the tip stirrer velocity. In this “near-of-flow area”, the kinetic energy is extensively dissipated itself due to impingement of velocity components. The axially-symmetric separation zone in the spatial form of λ-character induces two large-scale vortices above and below the separation. A highly rated kinetic energy lies within the head stream of λ-separation. Variation of the clearance at a factor less than D/5 sensibly influences flow structure and kinetic energy distribution. It is considered that computing on a turbulent hyperboloid flow field based on a k–ε model at a relatively high Reynolds number may result in a discharge of kinetic energy induced within the “near-of area”.
A calculation approach to discuss the solvent effect on liquid phase adsorption has been developed by using the molecular dynamics simulation method and solvophobic theory. The distinctive feature of the approach is that the calculations of the potential energy changes related to the solvent effect can be achieved on the basis of the molecular structure only, without using any experimental physicochemical property of the solute. This approach is applied to adsorption of five berberine alkaloids contained in Coptis japonica Makino onto the graphite surface from six solvents. Detailed analysis of the potential energy changes due to the solvent effect derived by molecular dynamics calculations yields an outline of the relative adsorption capacities of the alkaloids. Furthermore, the prediction of the preferential adsorption of alkaloids onto a graphite surface from the liquid phase and the eluent strength of solvents for desorption of the alkaloids from graphite surface were obtained. the calculated results are in good agreement with experimental data.
Computer simulations have been performed to study the effects of surface heterogeneity on gas permeation through slit-like carbon membranes by using a novel simulation technique, μVT ensemble Non-Equilibrium Molecular Dynamics. The energetically heterogeneous surfaces are made by randomly removing a certain number of carbons on the first layer of graphite basal planes. Methane (1) and ethane (2) are used as model gases for pure- and mixed-gas permeations. A small heterogeneity on the pore surface is found to result in a large decrease in the permeation fluxes due to an increase in frequency of molecular collisions onto heterogeneous sites. The densities in equilibrium with feed gases decrease almost linearly with an increase in the surface heterogeneity, though the decrease is not so large as that in permeation fluxes. In the case of the permeation of a binary mixture, permselectivity of a more adsorptive substance (ethane), which is much larger than the ratio of pure gas permeances, P^2/P^1, is found to change only slightly with an increase in the surface heterogeneity.