Mixing in a continuous vibrating mixer named “VIBRO MIXER®” has been observed with the aid of a visualization technique and a tracer response technique. The mixer consists of several mixing units having a spiral impeller going up and down. It has been found that homogeneous and rapid mixing can be obtained owing to the spiral structure of the impeller and the holes opened in the impeller. On the other hand, mixing between units is inhibited by a partition plate. Owing to the coupling of these two mixing effects, the mixer is regarded as a plug-flow type mixing device. When the mixer is operated under sufficiently high mixing Reynolds number and space velocity, the mixing can be described as a tanks-in-series model having the same number of tanks as the mixing unit.
The vortex depth in an agitated vessel with pitched blade paddle impeller was measured under various impeller dimensions and baffled conditions. The vortex depth measured was correlated well by the combination of the correlations of the vortex depth for paddle impeller by Kamei et al. (1999) and of the power input for pitched blade paddle impeller by Hiraoka et al. (1997) under both non-baffled and baffled conditions.
The effects of gas flow, temperature and concentration on the growth process of silicon epitaxial film in a horizontal single-wafer reactor of trichlorosilane (SiHCl3) were examined by 3-dimensional numerical simulation. For uniform growth over the silicon wafer, it is effective to grow the film on rotating susceptor under high SiHCl3 concentration with a high gas velocity in a reactor of low height. However, in cases of comparatively low gas velocities and low SiHCl3 concentration, uniform growth of silicon can be achieved by dividing the reactor inlet into three sections and by regulating each inlet width and gas velocity to obtain optimum conditions, and thereby of controlling the starting position and strength of buoyancy convection. Near the center of the wafer, a uniform growth rate cannot be archived by rotation of the susceptor, so a decrease in the gas velocity at the center of the reactor should be avoided.
For chars obtained from two different grades of coal, the variation behavior of the pore structure during the gasification process with CO2 and/or air was investigated in terms of gas adsorption methods. The increase in N2 surface area was significant for the lower-reactive chars. Attempts were made to analyze pore size distribution in terms of N2 ad-desorption. Air gasification (combustion) enlarged the pore size during reaction more remarkably than CO2 gasification. The softening coal char did not show the enlargement of pores as notably as the non-softening one. Correlation between variation of pore surface area and reaction rate during the gasification process was examined. The micropore surface area was proportional to CO2 gasification rate for the non-softening coal, and to air gasification rate for the softening one. There was also a correlation between only mesopore surface area and air gasification rate for some chars.
Stainless steel non-woven fiber filters with various of fiber diameters, basis weight and porosity were used to examine the effects of these variables on filtration characteristics and pore size distribution. The average pore size measured by the liquid permeation method was proportional to the fiber diameter and increased with the increase in porosity, but was independent of the basis weight change. The maximum pore size measured by the bubble point test increased as the porosity increased, and it decreased as the basis weight increased at the same porosity. The fiber distance distribution measured by microscope observation corresponded to a gamma distribution. The relationship between average fiber distance calculated from the gamma distribution and porosity agreed closely with the relationship between average pore size and porosity. Maximum particle size measured by the particle challenge test with latex beads, namely the size of particles of which 99% are blocked by the filter, increased as the porosity increased, but gradually decreased as the basis weight increased.
Sequential drying with microwave and convection was examined in a packed bed of glass particles wetted with aqueous polyvinyl alcohol solution. Samples with a higher initial moisture content than the critical moisture content gave a flat profile of polymer content except near drying surface after being dried. Such a profile had not been observed in the case of convective drying, in which minimum polymer content appeared at the bottom of the dried sample, or in the case of convective-conductive drying, in which minimum content appeared in the center. The profile was the result of enhanced evaporation of water inside the wet sample by the microwave. This sequential drying was effective for giving a new profile of polymer content within the dried sample.
The separation of water and organic solvents using a double Y-type microreactor was studied. Water and cyclohexane, which possess similar coefficients of viscosity, were supplied into a conventional double Y-type microreactor. Mixed liquid was obtained from both outlets, even though the laminar flow was observed under the microscope at the same flow rate. The incomplete separation is considered due to hydrophilicity of the glass channel wall. The separation was drastically improved by use of a microreactor in which plural partition walls are placed at constant intervals in the center of the channel.
Combined convective and microwave drying and convective drying alone were investigated in a glass particle layer wetted with aqueous polyvinyl alcohol (PVA) solution, styrene-butadiene rubber latex (SBRL), or water. Drying rate and polymer content profile within the dried layer were measured. Overall drying time for the combined drying decreased to 40-80% of the time required for the convective drying. Drying rate increased with a decrease in the initial moisture content, and the drying rate of SBRL was higher than that of PVA solution in the both combined and convective dryings, while polymer content near the drying surface within the dried layer was higher in the SBRL case. Polymer content profile within the dried layer tended to be flat in the case of the combined drying, and this may be another effect of the microwave assistance.
The effects of filtering velocity u and dust concentration c on pressure drop parameters, i.e., specific resistance of dust cake α and specific resistance of filter cloth with residual dust ζd, were examined for three kinds of dust in a bag filter. The measured specific resistance α for a dust cake with a void fraction larger than 0.5 was almost proportional to the 1/2 power of the filtering velocity and to the −1/3 power of the dust concentration. Comparing the model equation of the filter cloth with dust to the Kozeny-Carman equation, the value of (1−ε)/ε3was obtained from the measured value of α and correlated empirically with the operating conditions and the physical properties of dust. The resultant correlation gave rise to the following correlation of α, α=1.15×1011(Dv/Ds)3Dv-3/4ρs-2/3c-1/3u1/2μ1/2 The estimated values of α from the correlation coincided well with the measured ones within the experimental error of ±25%. The measured specific resistance of filter cloth with residual dust ζd was almost independent of the filtering velocity and the dust concentration. The dependency of α on the variation of the filtering velocity u with time during a filtering operation in a multi-compartment bag filter gave a slightly larger value of the pressure drop than that at the constant value of α through the operation, but did not affect the time-dependent behavior of the pressure drop.
A patched cleaning model was proposed for the pressure drop in a bag filter, in which the filter cloth had two surface areas, i.e., the area with residual dust with the specific resistance ζD and the area fraction εD, and the perfectly cleaned area with ζC and εC. The measured pressure drop under the batch filtering operation coincided well with the calculated one with the patched cleaning model through the operation time. Also, the measured pressure drop in a three-compartment bag filter under the continuous filtering operation coincided well with the calculated one with the proposed model by using the specific resistance of dust cake α, the specific resistances of filter cloth ζD, ζC and the area fraction εc that were determined experimentally for the batch filtering operation, though the specific resistance of filter cloth with residual dust ζD needed to be corrected for the continuous filtering operation. The velocity dependency of the specific resistance of dust cake α did not affect the calculation of the pressure drop under the batch filtering operation, though the velocity-dependent α gave a slightly smaller calculated value of the specific resistance of filter cloth with residual dust ζD in the three-compartment bag filter under the continuous filtering operation than the constant α. However, the evolution of the pressure drop under the continuous filtering operation was almost independent of the velocity-dependency of α.
Preparation and stabilization of particle sizes of microsuspensions of the practically insoluble drug griseofulvin (GF) and hydrogenated soybean lecithin mixtures were investigated to improve drug dissolution. Particle sizes in the suspension gradually decreased with increasing pass number through an interaction chamber, and reached an average of 60 nm at the pass number 20, when the suspension became transparent. Polydispersion index (dw/dn) also decreased with increasing pass number to a value of 1-2. The microsuspension could be preserved for about 1 month in a refrigerator, but it gradually became opaque after two weeks of strage at room temperature. Particle sizes in a rehydrated freeze-dried microsuspension were practically restored to those of the original microsuspension, when a saccharide such as xylitol, treharose or sucrose was added at more than 5, 3 and 3 wt%, respectively, before the freezing procedure. The restration of nano-size particles restoring is related to the molecular interactions among saccharides, drug and water molecules. The results obtained above provide valuable information for proposing practically liquid dosage forms for drug delivery.
Encapsulated particles for controlled release generally show a diffusion-type release pattern. However, a release curve of sigmoid type could be useful for various applications. Therefore, a structural bed was prepared by filling voids among encapsulated particles with fine particles dispersed in the filler in order to obtain sigmoid-type release curves. Then, the effects of the volume fraction of fine particles, the packing structure of encapsulated particles and the structural bed height on the release rate were studied. As a result, it was possible to obtain sigmoid-type release curves with the structural bed. The initial release rate decreased with decreasing volume fraction of fine particles and increasing voidage among encapsulated particles, but it was independent of the bed height.
Experimental and numerical studies have been conducted on the effect of blade angle in a mechanical impact mill. The particle trajectories around the rotor were calculated for a long simulation time by use of the CFD software Fluent and were compared with the experimental results obtained previously. The product fineness was related to the amount of normal impact energy of particles colliding with the blade and stator walls while the particles pass through the pulverizing zone. It was found that fine particles with a narrow range of size distribution are obtained when blades are inclined forwards at an angle of 30 degrees.
IGCC processes have been actively developed domestically and overseas to reduce carbon dioxide emission. Gasifiers in these processes are operated at high temperature and pressure. For example, the pressure is about 3 MPa and the temperature peaks at 2000 K. In order to discuss the properties of rapid pyrolysis, which is the initial reaction in coal gasification, rapid pyrolysis tests were performed for four coals using a pressurized drop tube furnace facility (PDTF) operated at high temperatures and pressures. The results indicated that pyrolysis temperature rather than pressure had a great influence on char reactivity of the product char. The higher the pyrolysis temperature, the lower the gasification reactivity for some coals. The reactivity of these chars had a good correlation with the graphitization of carbon in char. On the other hand, the pyrolysis temperature had little influence on the reactivity for coal with high ion-exchangeable calcium content, which maintained its reactivity at high temperatures.
To clarify the mechanism of soot formation is very important from the viewpoint of reduction of environmental load and energy saving. Despite the large number of experimental and theoretical studies made on soot production, the mechanism of soot formation remains unclear. In this article, benzene pyrolysis was carried out in an alumina tube heated by plasma, and a numerical simulation was performed with a kinetic model that takes into account the reactions of benzene pyrolysis, nucleation formation, surface growth and coagulation and was constructed by adopting the Discrete-Sectional Method. The effects of residence time and initial benzene concentration on soot formation rate were examined. The results show that soot particles have a larger diameter and a narrower size distribution with the increase in residence time and reactive gas concentration.
The effect of valid particle diameter of activated carbon on dioxins removal efficiency was explicated using a fixed bed column and 1, 2, 3, 4-tetrachlorinated dibenzo-p-dioxin (T4CDD). In the adsorption process of T4CDD on to activated carbon, the mass transfer rate of T4CDD in a boundary film on activated carbon could be evaluated by the valid diameter calculated from the pressure drop through the packed bed. The mass transfer rate of T4CDD in the pores of activated carbon increased with increasing roughness of the surface of activated carbon particles. The dioxins removal efficiencies of a bag filter with activated carbon injected into the inlet flue gas and an adsorption tower filled with activated carbon were estimated quantitatively, using the mass transfer model and the linear-isotherm system (Henry’s law). The adsorption coefficient of activated carbon with a surface area of 181 m2/g for pores of 2-50 nm in diameter was estimated to be 6.1×105 m3/kg at 170°C.
As an alternative method to the expensive and complex conventional methods for detoxification of dioxins (DXNs) and heavy metals in fly ash from municipal solid waste incineration, fly ash pellet reburning, an inexpensive and easy method to decompose DXNs and stabilize heavy metals, was investigated in terms of the heavy metal behavior during and after the reburning treatment in the incinerator. After pelletization of the fly ash, cement and water mixture with or without an additive (Na3PO4), the pellets were reburned in a laboratory-scale bubbling fluidized bed incinerator. Although the Pb landfill standard was not met without the reburning in the case of non-additive fly ash pellets, all the fly ash pellets were able to meet the landfill standards for heavy metals with the reburning. In the case of non-additive pellets, reburning at temperatures in the range of 700-900°C also considerably reduced the elution of heavy metals. With the reburning, Ca in the fly ash pellet was found to react with Al, Si, Mg and Cl to form mineral, which reduced the Ca solubility and accordingly reduced the pH of the leachate from fly ash pellet. Pelletization of fly ash and a small addition of Na3PO4 to the pellet were found effective to reduce emission of heavy metals during reburning.
The mass transport characteristics of sodium in cathode-side electrodes strongly influence the overall output performance of the alkali metal thermal-to-electric converter (AMTEC). At temperatures above 1100 K, the rapid degradation of molybdenum electrode performance with time has been reported. On the other hand, a high initial performance may be maintained for much longer at lower operating temperatures. In this paper, measured performance of the molybdenum electrode in the temperature range of 900 to 1050 K and time dependence of the performance are discussed. The sodium exposure test cell (SETC) was used to measure the electrode morphological dimensionless factor, G, which can be calculated from the limiting current in the current-voltage curve. G was observed to become approximately 20 after 150 h of operation at 904 K. The analysis based on the experimental results shows that the molybdenum electrode operating at 900K has a predicted output power density of 0.12 W/cm2 and its performance is about 50% higher than that of the conventional titanium nitride electrode.
Hydrothermal treatment was investigated as a new pretreatment method for biodegradability improvement of wastewater containing vegetable proteins. Soybean milk was used as a model wastewater. After 7.4 min of hydrothermal treatment, BOD1 of a sample treated under the reaction conditions of 300°C and 9 MPa was twice as high as that of raw sample. Moreover the D-BOD5/D-CODCr ratio of the treated sample reached 0.80, while that of the raw sample was 0.54. The results of GPC analysis indicated that the biodegradability improvement of the treated sample was due to a shift of the molecular weight distribution to a low molecular weight range. Next, tofu industrial wastewater, an actual wastewater, was tested under the reaction conditions of 300°C and 9 MPa that were optimal for the treatment of soybean milk. The ratios of D-BOD5/D-CODCr of treated sample and raw sample were 0.70 and 0.49, respectively. Thus, hydrothermal treatment can be a useful pretreatment to improve biodegradability of wastewater before a conventional biological treatment process.