The effect of the salts sodium chloride, calcium chloride and zinc chloride on the vapor-liquid equilibria (VLE) of three miscible binary system viz., benzene-pyridine, water-pyridine and methanol-benzene has been experimentally investigated at atmosphere pressure using a Smith and Bonner type of equilibrium still. The salts were dissolved in the solvent in which they are preferentially soluble, to concentrations upto 20 percent by weight wherever possible. In the case of benzene-pyridine system, the salts calcium chloride and zinc chloride cause the “salting out” of benzene. As for the azeotropic system, water-pyridine, while sodium chloride increases the pyridine content in the vapor phase, calcium chloride and zinc chloride bring about “salting out” of water and “phase splitting”. In respect of the third binary, methanol-benzene which is also azeotropic, shift in the azeotropic composition and ‘crossover effect’ are observed upon addition of calcium chloride and zinc chloride. The experimental VLE data have been correlated using Wilson and NRTL models after ascertaining their thermodynamic consistency.
The potential of mean force (PMF) of solute Na+ and Cl– ions approaching water/NaCl(001) and water/NaCl(011) interfaces is calculated by using classical molecular dynamics (MD) simulations. It is found that (1) Na+ and Cl– ions adsorb on the crystal surface either directly (direct adsorption) or with a water molecule interposed between the surface and themselves (solvent-separated adsorption), (2) both ions are adsorbed more stably on the NaCl surfaces under solvent-separated conditions in most cases, and they adsorb on NaCl(001) more easily than on NaCl(011) in the case of direct adsorption, and (3) Na+ adsorbs on the NaCl surfaces more easily than Cl– in the case of direct adsorption, but direct adsorption of Clndash; on NaCl(011) is impossible. These results indicate that the solute-surface and solvent-surface interactions are strongly affected by the lattice arrangement of the surface such that the PMF profiles largely depend not only on the size of solute ions but also on the lattice arrangement.
Supercritical CO2 can be used to remove phenol from waste water. For evaluation of various processing conditions, a model for the distribution coefficients of phenol is necessary, for example, when the aqueous phenol solutions are contacted with supercritical CO2. The range of distribution coefficients of phenol is 0.034 to 1.66, at 9.65 to 17.23 MPa, and 323 K. The objective of this paper is to extend the applicability of PRASOG to CO2 + water + phenol solution. The average deviations between the predicted and experimental distribution coefficients is 37.4%. For the four binary systems, i.e., phenol + water, phenol + CO2, m-cresol + CO2, and H2O + CO2, the average deviation of the pressure is 7.9%.
The power consumed by the Rushton and 60° pitch-blade impellers with Newtonian fluids and CMC shear thinning solutions have been obtained as a function of impeller rotational speed. The Power number with the Rushton impeller increased gradually with Reynolds number while it remained almost the same with the 60° pitched blade impeller with a sudden drop in power at a point where the discharge flow change from a radial flow to an axial one. A good correlation between the power numbers of the Newtonian fluid and CMC solutions in turbulent flow have been obtained with both impellers, in particular with the Rushton impeller, when a shearing constant of 54 was used. With the 60° pitch-blade impeller, the CMC solution had the effect of delaying the flow transition to a higher Reynolds number with higher concentration causing more of a delay.
The deterioration mechanism of reforming catalysts used for internal reforming molten carbonate fuel cells is studied through post-test analyses of the used catalysts. The catalysts studied are Ni/MgO-Al2O3. Deactivation of the catalysts is caused both by a decrease in Ni surface area of the catalyst and by loss of specific activity of the Ni surface (alkali poisoning). The decrease in Ni surface area is characterized firstly by accelerated sintering of Ni particles due to the existence of alkali metals (attached electrolyte). Secondly, sulfur poisoning also plays an important role, especially for sintered catalysts with small Ni surface areas. As for the effect of alkali poisoning, the specific activity of the Ni surface decreases to 15–60% of the original one. Blockage of pore structure by the electrolyte does not seem to be important for the loss of Ni surface area.
The purpose of this paper is to investigate the relationship between the properties and catalytic activity of the third phase in a phase transfer catalytic system. The conditions of formation of the third phase are investigated by changing the kind of phase transfer catalyst (PTC), organic solvents, and concentration of potassium hydroxide in the aqueous phase. When tetrabutylammonium bromide, (Bu)4NBr, is used as the phase transfer catalyst, the third phase is formed with both dodecane and toluene as organic solvents. On the other hand, when tetrahexylammonium bromide, (Hex)4NBr, is used as the phase transfer catalyst, the third phase forms with dodecane but not with toluene. The volume of the third phase, the base strength, the concentration of catalyst, water, and organic solvent in the third phase were measured. Solidification phenomena are observed for some cases of higher KOH concentrations in the aqueous phase. Dehydrohalogenation of 2-bromooctane in the organic phase was carried out in a batch reactor with the potassium hydroxide solution in the aqueous phase. The observed reaction rate constant (kobs), which is determined from the first order kinetics, increases remarkably with the increase in concentration of KOH in the aqueous phase (ξKOH) for the condition of third phase formation. However, for the solidification condition, the reaction rate drops sharply because only the external surface of the solid is effective for the reaction. The distribution coefficient of reactant between the organic phase and the third phase, (KA) and the reaction rate constant in the third phase (kthird) can be related to the concentration of water in the third phase (CH2O) independent of the catalysts and organic solvents. Since PTC exists in the third phase, it can be reused without any loss of catalytic activity.
The liquid mixing time of a bubble column with a draft tube is experimentally studied. The diameter and height of the column, the diameter and length of the draft tube, and the mode of sparging gas were changed and the liquid mixing time was measured. The lengths of bottom and top clearances have significant effects on the liquid mixing time. The liquid mixing time increases with increasing column aspect ratio of height to diameter. The liquid mixing time of the bubble column with a draft tube is longer than that of a standard bubble column when the column aspect ratio is relatively high. The liquid mixing time has a minimum when the diameter ratio of draft tube to column is 0.5 to 0.6. These effects of equipment dimensions are expressed by an equation. In the experiment using different solutions, the liquid mixing time becomes shorter with increasing liquid surface tension, or with increasing liquid viscosity.
If solvents in the liquid phase are incompatible each other, a gas-liquid-solid three-phase bubble column becomes a gas-liquid-liquid-solid four-phase bubble column. Deprotection of amino acid was carried out in this column. At lower gas velocity, deactivations occur due to the deposit of product on the catalyst. At higher gas velocity, however, the catalytic activity is stable. Gas and liquid holdups in the multi-stage bubble column were measured to investigate the effects of operating conditions on hydrodynamics. Gas and liquid velocities, the height of each stage and the weight of catalyst were varied. The ratio of the organic phase holdup to the aqueous phase holdup changes drastically with the increase in gas velocity. The organic phase becomes continuous at lower gas velocity and the aqueous phase becomes continuous at higher gas velocity. Gas holdup is almost independent of each stage height and the catalyst weight.
The preparation of composite Sr-Pb oxalate particles was carried out by using an emulsion liquid membrane (ELM, water-in-oil-in-water (W/O/W) emulsion) system, consisting of Span 83 (sorbitan sesquioleate) as a surfactant and D2EHPA (bis(2-ethylhexyl)phosphoric acid) as an extractant (cation carrier). Sr and Pb ions were extracted from the external water phase and stripped into the internal water phase, to make submicron-sized composite oxalate particles 0.2–0.8 μm in size, which were much smaller than those obtained in homogeneous aqueous solutions. The molar composition of the particles, (Sr/Pb)p, prepared in the ELM system was almost identical to that of the feed external solution, (Sr/Pb)f, and was greater than the molar ratio of the two metals moved to the internal phase, (Sr/Pb)m. This indicates that Sr oxalate precipitates more easily in the internal water droplets, and Pb oxalate is coprecipitated with Sr oxalate. XRD, XPS, thermal analyses, and lattice spacing and lattice strain studies indicate that the particles obtained at (Sr/Pb)f = 20–50 have a structure in which Pb oxalate is finely dispersed in SrC2O4·H2O. Calcination of the oxalate particles obtained gives SrPbO3 or Sr2PbO4, depending on the value of (Sr/Pb)p and calcination temperature.
Carbothermal reduction and nitridation of titanium dioxide was investigated by X-ray diffraction and gas chromatography. The results indicated that anatase was transformed to rutile phase before it reacted to produce TiN. The solid products of the reaction are TiN and TiC. The fraction of TiN in the solid product may be increased by increasing the reaction temperature or decreasing the initial molar ratio of C/TiO2. The gas products of the reaction are mostly CO and trace amount of CO2. The concentration of CO2 in the evolved gas can be decreased by increasing the reaction time or increasing the reaction temperature.
Numerical solutions of the 3-dimensional convective diffusion equation are used to obtain a quantitative measure of concentration polarization for Dean vortex flow of an aqueous salt solution in a spiral reverse osmosis system. The velocity fields for total spanwise, radial and streamwise directions are incorporated into the convective-diffusion equation to obtain the concentration profiles. The concentration polarization for flow in spiral (with Dean vortices) and flat (without Dean vortices) membrane channels are calculated under similar flow conditions. The rate of concentration build-up in a spiral membrane channel with axial distance is significantly inhibited when Dean vortices are present. Dean vortex flow promotes rapid mixing and inhibits the growth of the solute concentration boundary layer, hence a module with Dean vortices is more productive at larger downstream distances than one without such instabilities.
A field-test of a 20 kW apparatus for adsorptive desiccant air conditioning has been carried out at the Biochemical Engineering Laboratory, Kumamoto University. In the system, operated in a ventilation mode, supply fresh air is dried in a thermal swing honeycomb rotor dehumidifier, and cooled by a regenerative heat exchanger and an evaporative cooler. Return air is used as a cooling medium in the regenerative heat exchanger, and then heated by hot water in a finned coil heater and discarded through the honeycomb rotor dehumidifier after desorbing the moisture adsorbed from the supply air stream. The performance of the present cooling system is examined at regeneration air temperature as low as 80°C, the amount of enthalpy of about 16.5 kJ/kg can be reduced from the room with the COP (thermal coefficient of performance) of 61% at the superficial air velocity of 1 m/s. Changes in state of air during processing are discussed in terms of a psychrometric diagram.
This paper describes the colloidal interactions of preformed positively charged iron oxyhydroxide particles: zeta potential, coagulation, and flocculation in the presence of nonionic, cationic and anionic polyelectrolytes. In salt-coagulated systems, the zeta potential shown as a function of the pH and of the KCl salt concentration is the controlling factor in the formation of particle aggregates. In polymer-flocculated systems, a moderately charged anionic polyelectrolyte was used to adsorb onto particle surfaces and then induced the bridging flocculation of oppositely charged iron oxyhydroxide particles, in which maximum flocculation was observed near zero zeta potential. It is pointed out that the relative importance between polymer bridging and charge neutralization mechanism in the coagulation-flocculation processes of colloidal particles is pH dependent. Correlation of the magnitude of zeta potential with the interparticle interaction energy is simulated by using DLVO theory. Moreover, the present study extends the conventional analysis through considering the temporal variation of the fractional surface coverage of colloidal particles by polymers. The theoretical predictions on the stability coefficients are compared with the experimental results.
An experimental study on mass transfer in heterogeneous distillation was made with wetted-wall columns of different lengths for an ethanol-benzene-water system under total reflux conditions. The observed local diffusion fluxes obtained from the data of shorter distillation paths agree well with the theoretical values of mass transfer in a tube, but the average diffusion fluxes of longer distillation paths deviate from the theoretical ones. Simulation of heterogeneous distillation with a wetted-wall column is carried out using the theoretical diffusion flux in a tube. The predicted concentrations and reflux flow rates at the top of the column showed good agreement with the observed data of longer distillation paths obtained by a longer column. The simulation results show that a large concentration driving force causes a longer distillation path and results in a large variation along the column. Also the deviation of the average diffusion flux from the theoretical value is demonstrated and explained by being made dimensionless with the average or non-local concentration driving force. Furthermore, the length of the distillation path is affected considerably by the bottom concentration and the vapor phase Reynolds number.
A separation process for gallium and indium in chloride solution with organophosphorus acids as extractants is investigated. Extraction equilibrium formulations are established up to high loading ratios, and extraction equilibrium constants are determined, employing bis(2-ethylhexyl)phosphoric acid (D2EHPA), 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester (EHPNA), and bis(2-ethylhexyl)phosphinic acid (PIA-226) as extractants, and kerosene as a diluent. A feature of the present extraction systems is the inclusion of chloride and hydroxyl ions in the extract entities to compensate the positive charge of the metals, especially at high loadings. With EHPNA and PIA-226, mixed complexes containing both metals in a mole ratio of 1:1 are formed at high loading ratios. This is not the case with D2EHPA, which makes D2EHPA most suitable for separation of gallium and indium. Simulation work on separation of gallium and indium by counter-current mixer-settler cascade with D2EHPA shows that separation and recovery of gallium and indium are enhanced under the conditions of high pH values of aqueous feed solution and large values of ratio of the flow rate of organic to aqueous phase, E/F. Effective separation is realized by counter-current cascade of 15 stages, and an E/F ratio of 3 with gallium purity of 99.8% and indium purity of 97.4%.
Using a salting-out technique with alcohol, precipitation of borax from a stripping solution containing boric acid after the solvent extraction process is investigated. From the low solubility of borax and small temperature effect on the solubility, ethanol was selected as a salting-out agent. The X-ray diffraction results show that the precipitates were borax decahydrate. 74% of boric acid present in the organic phase is recovered as precipitates when the volume ratio of the organic phase to the stripping phase is 2.
A cold model experiment using semi-cylindrical spouted bed with a draft tube was carried out to improve the performance of a spouted bed-type coal gasifier. The effects of the gas flow rate, tube-cone clearance, included angle of conical base, particle diameter and bed weight on the solids flow behavior in the unit are investigated. The gas velocity in the draft tube depended on the gas feed rate. The gas velocity in the annulus increases with increasing in the tube-cone clearance, included angle of conical base and the mean particle diameter but decreased with increasing the bed weight. The particles circulation rate increases with increasing in the tube-cone clearance, the mean particle diameter and the bed weight. Tendencies of the change of particle circulation rate with gas feed rate for 5π/18 and 6π/18 of included angle are different from those for 7π/18 and π of included angle in the region of high gas feed rate. Correlations to describe the annulus gas velocity and particle circulation rates are obtained.
The hydrodynamics of a circulating fluidized bed with a bubbling bed section separated by an inner rim baffle are examined both by experiment and mathematical analysis. To maintain sufficient solids holdup in the upper section, the inner rim baffle was set up in the middle of the riser. It is found that the solids holdup in the section above the baffle is greatly increased. Investigation into the baffle dimensions shows that the variation of the baffle clearance has little effect on the solids holdup in the upper section when the clearance is greater than a certain value, whereas the variation of the baffle height has a significant effect on the solids holdup in the upper section, i.e. the lower the baffle height, the higher solids holdup. A hydrodynamic model is developed based on the mass balance and the pressure balance of the solids circulation loop. With respect to gas-solids flow behavior in the upper section, the clustering suspension and the core-annulus flow are taken into account; for the lower section, a bubbling fluidized bed model is included. By comparing the calculated results with the experimental ones, the model is found to predict fairly well the gas-solids flow behavior of the compact bubbling-circulating fluidized bed.
Spherical silica microcapsules whose average diameter was 3.6 μm were synthesized by the sol-gel method in W/O emulsions containing an anionic surfactant at a high water to surfactant molar ratio. Special agitational conditions are needed to prepare spherical silica microcapsules. The properties of the resulting silica microcapsules were determined by Fourier-transform infrared spectroscopy and scanning electron microscopy. As the spherical silica particles have a hollow core structure, hydrolysis and condensation of TEOS occur at the micellar periphery, and the diameter of the water droplets in the W/O emulsion at a high water to surfactant molar ratio (Wo) are micron sized.
Solid circulation behavior in a cylindrical spouted bed of 0.205 m diameter with a draft tube is studied based on the power spectral analyses of the time series data of pressure fluctuation and optical signals measured by optical fiber probes in the draft tube and the entrainment region. The effects of solid particle size, bed mass and the length of the entrainment region on solid circulation are examined. In the low gas velocity region, solid particles move from the annulus into the gas jet to form clusters and are carried upward in the draft tube with a frequency of 7–10 Hz. In the high gas velocity region, solids flow is in a stable pneumatic convey state. As the particle size increases and the bed mass decreases, the frequency of clusters is reduced by the promotion of gas bypassing, leading to a decrease in solid circulation rates. In the pneumatic transport regime, the larger particle system exhibits a higher solid circulation rate. The extent of the length of the entrainment region decreases the frequency of clusters but increases the solid circulation rate by virtue of increasing the size of clusters.
We aim to establish a robust CSTR control system with uncertain dynamics by combining linearizing control and H∞ control systems. The procedure presented in this study demonstrates practical usefulness for achieving a stable control system design for chemical plants such as a CSTR (Continuous Stirred Tank Reactor) which has not only strong nonlinearity, but also uncertain dynamics.
An adaptive predictive control method using multivariable bilinear model is proposed and evaluated experimentally. A recursive parameter estimation algorithm is used to determine the model parameters and one-step ahead prediction is used in the control law. To verify the performance and effectiveness of the proposed control method and to demonstrate the possibility of the application to a real processes, both computer simulations and experimental studies were performed. Results of computer simulations and experiments show that the proposed adaptive predictive control method using multivariable bilinear model exhibit robust and satisfactory control performance.
Protein denaturation processes under stress conditions such as pH, denaturant and heat are evaluated by employing several approaches such as aqueous two-phase partitioning and circular dichroism (CD) measurement in order to determine the conformational states and surface properties of proteins. It is suggested that the manner of conformational change of protein from the native state to the molten globule one does not depend on these sorts of stresses if both the strength of stress and surface properties of proteins are normalized between the above two states. By using these normalization parameters, eight proteins used in this study were simply divided into two groups namely monomeric proteins and oligomeric one. On the other hand, the strength of stress which inducing protein denaturation up to the molten globule state greatly depends on the sort of stresses or protein species. From these considerations, the possibility of using effective stress-mediated separation process of protein is suggested as an application of the analysis of the stress responsive behaviors of proteins.
Human keratinocytes were cultured on an inert polymeric film support in serum-free medium to produce subconfluent autologous skin grafts for healing of burn wounds and chronic leg ulcers. The growth of keratinocytes in Petriperm with such a hydrophilic film bottom as a culture support was superior to that in T-flask owing to better cell adhesion. Based on this principle, an autonomous modular bioreactor, KERATOR for large scale production of skin grafts up to 5280 cm2 in size was constructed. It is computer controlled, and operations such as cell seeding and medium change are automated. The cell growth rate profiles in the bioreactor and Petriperm are similar, evincing the significance of KERATOR. The harvested skin grafts consist of the polymeric film covered by subconfluent sheets of non-differentiated keratinocyte cells. Such grafts may be applied to wounds in “upside-down” fashion.
A rapid and convenient means of immobilizing biological membranes is described. Sonication for formation of small vesicles, diffusion of the vesicles in support beads, and freeze-thawing is carried out in an immobilized multi-enzyme system. Large amounts of brush border membranes from bovine kidney were immobilized in Sepharose CL-6B, Sephacryl S-500, and Sephacryl S-1000 and increased in that order. For the latter two beads, the amounts were further increased with repetition of freeze-thawing up to a maximum of three times. Particle distribution analysis reveals that small vesicles are enlarged by the freeze-thawing repetition, suggesting an immobilization mode in which enlarged vesicles are physically entrapped in the beads. The bead-immobilized membrane vesicles were relatively stable, exhibiting constant enzyme activities for at least 6 h under packed column operation.
A fundamental study of NOx removal using a surface discharge induced plasma chemical process (SPCP) is presented. SPCP involves high-frequency surface discharge from strip-like electrodes attached to the surface of a ceramic tube to a film electrode attached to the opposite side of ceramic tube. This surface discharge has the following properties: (1) relatively low energy consumption for discharge generation, (2) operation at atmospheric pressure, and (3) high-energy electrons. These properties show that SPCP is suitable for the removal of environmental pollutants such as NOx, SOx, and soot emitted from diesel engines and combustion furnaces. In this study, experiments on NOx removal were performed by introducing N atoms, which were produced by the addition of N2 into the surface-induced plasma, into an NO-containing gas. Furthermore, the concentrations of N and O atoms were measured by ultraviolet (UV) resonance absorption spectroscopy in order to investigate the reaction mechanism. The following results are obtained. (1) NOx removal is observed by introducing N2 gas activated by SPCP, and the amount of NOx removal increases with increasing N2 concentration in the surface-induced plasma. (2) NO reduction mainly proceeds according to the reaction: N + NO → N2 + O. (3) The amount of O atoms generated rapidly decreases due to the cyclic recombination reaction via NO2.
The oxidative reaction of OH radicals or O atoms with carbon black (CB) particles is studied by using a silent discharge reactor operated at room temperature and atmospheric pressure. The CB particles are oxidized by OH radicals or O atoms to produce approximately corresponding amounts of CO and CO2 through the silent discharge reactor. The OH radicals and O atoms are generated from the dissociation of H2O and O2 by silent discharge, respectively. The concentration of OH radicals is measured by resonance absorption spectroscopy while that of O atoms is estimated from a computer simulation by taking into consideration molecular dissociative reactions by electron impact. The amount of CB particles oxidized gradually decreases in the high H2O concentration range in spite of an increase in OH radical concentration. It is considered that the surface reactivity of the CB particles decreases upon adsorption of H2O on their surfaces. On the other hand, the amount of CB particles oxidized with addition of O2 is almost independent of the O2 concentration. The CO and CO2 concentrations produced are correlated with that of O atoms, which is estimated from O3 concentration. These results indicate that the O atoms are much more reactive than the O3 molecules for oxidation of CB particles. The sticking coefficient for oxidation of CB particles by OH radicals or O atoms is calculated from the experimental results. The OH radicals are more reactive than O atoms when H2O concentration is under 0.6%. However, beyond this value, the sticking coefficient of OH radicals suddenly decreases, and then becomes lower than that of O atoms.
The rotting of municipal solid waste is one hazard in the RDF production process. In this study, the effect of CaO addition on rotting prevention is experimentally examined. A mixture of soybean protein and distilled water was used as the model refuse. Escherichia coli was adopted as a rotting indicator. CaO powder was uniformly added to the model refuse, and the changes of pH and cell number were measured. The rotting was prevented by CaO addition because CaO addition made the alkalinity in the sample higher. The effect of CaO addition on rotting prevention depends on moisture content.