-
Naoto Ohmura, Kunio Kataoka, Toshiyuki Kataoka, Yoshiyuki Naitoh
1997Volume 23Issue 6 Pages
741-748
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The generation of vortices and transitional properties to turbulent flow in a Taylor -Couette flow system with an aspect ratio of 3 were investigated in terms of the effect of the acceleration of the inner cylinder rotation speed by using numerical, electrochemical and flow-visualization techniques. The three-dimen-sional simulation based on the SMAC method clarified the generation mechanism of vortices and temporal and spatial properties in a wavy vortex flow. The electrochemical and flow-visualization techniques revealed the temporal and spatial properties in wavy, quasi-periodic wavy and weakly turbulent wavy vortex flows. Lower acceleration rates produced the two-cell primary mode, while the four-cell secondary mode was generated in the case of higher acceleration rates. In a wavy vortex flow, the single fundamental frequencies, which correspond to the passing frequencies of the travelling azimuthal waves, velocity and velocity gradient power spectra depend on the flow modes. In quasi-periodic and weakly turbulent wavy vortex flows, the temporal-spatial properties also largely depend on the cell modes. These results suggest that the geometric structure of vortex cells influences flow characteristics.
View full abstract
-
Masahiro Fujiwara, Keisuke Fukui, Syouji Hirota
1997Volume 23Issue 6 Pages
749-755
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Laminar vortex rolls which occur between horizontal parallel plates, of which the temperature is not uniform in the lateral direction normal to the forced main flow, are investigated by means of a flow-visualization technique and 3-D steady state numerical calculation. It was found that the vortex rolls were driven by the horizontal gravity currents, which were caused by density difference in the lateral direction, and that the generation mechanism of the vortex rolls could be classified into four types. The effects of vortex rolls upon the local Nusselt number were also investigated.
View full abstract
-
Yoshiro Inoue, Yushi Hirata, Eiichi Kunugita
1997Volume 23Issue 6 Pages
756-763
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The time evolution of a plane pattern in Bénard convection field is numerically simulated with the generalized Ginzburg-Landau evolution equation described with the pattern variable, and the obtained numerical results are compared with and discussed previous experimental and theoreticalesults. In spite of a single non-linear differential equation, the evolution equation derived phenomenally is able to express the various convection patterns : roll cells, hexagonal cells, zigzag-rolls, cross-rolls, and the wave number change observed in super-critical Rayleigh number. This approach makes the procedure for analysis of the phenomena accompanied with pattern formation extremely simplify.
View full abstract
-
Tatsuo Nishimura, Koji Kunitsugu
1997Volume 23Issue 6 Pages
764-771
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
We examine numerically and experimentally self-sustained oscillation and fluid mixing in grooved channels with different groove lengths. The critical Reynolds number for the onset of self-sustained oscillation decreases as the groove length becomes larger, but oscillatory flow is found to arise from the same Tollmien-Schlichting waves triggered by a shear layer above the groove. Momentum transfer due to the oscillating parts of the flow is analyzed by looking at the oscillatory stress and the production of oscillatory energy. Fluid mixing processes between the channel and groove flows are also explained by a particle advection procedure. The agreement between numerical and experimental results is satisfactory.
View full abstract
-
Hitoshi Habuka, Masatake Katayama, Manabu Shimada, Kikuo Okuyama
1997Volume 23Issue 6 Pages
772-779
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Transport of dopant gas, B
2H
6, for p-type silicon epitaxial film growth is numerically calculated using the equation of B
2H
6 mass conservation for developing a model to predict the concentration distribution of B
2H
6 on a substrate in a cold-wall horizontal single-wafer epitaxial reactor under a practical environment for preparing epitaxial silicon thin-films. The B
2H
6 concentration is shown to be nonuniform over the substrate surface, both stationary and rotating, because of a large thermal diffusive flux of B
2H
6 induced by the horizontal and vertical temperature gradients near the substrate, even when an uniform B
2H
6 concentration is imposed at the inlet of the epitaxial reactor. The ratio of the thermal diffusive flux to the molecular diffusive flux in the reactor is also evaluated to show the larger relative effect of thermal diffusion to molecular diffusion for B
2H
6 than SiHCl
3.
View full abstract
-
Kazuo Kondo, Mitsunori Yokoyama, Keisuke Fukui
1997Volume 23Issue 6 Pages
780-788
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Bump shapes are determined by the current distributions of the cathodes. These current distributions are discussed both by experiments and numerical computations.
The potentials are classified by linear, Tafel and diffusion control regions. At the linear region, the current distributions are uniform because of the large surface over potential. The growth rates are slow.
At the Tafel region, the distributions are not uniform because of large ohmic resistance. The cavity edges show hump for larger cavity widths of more than 100 μm. For smaller widths, the current distributions become uniform and the hump forms at the centers.
At the diffusion controlled region, the current distribution is determined by the vortices. The vortices act as the resistance of mass transfer to the cathode. For the 100 μm cavity width, the vortices grow at the up stream corner with high Peclet numbers of 1410 and 7311. The current distributions show hollows at the up stream corners. The vortices become smaller at down stream corners. These smaller vortices cause the increase in current distributions at down stream. For 30 μm cavity width, on the other hand, a large single vortex forms for the higher Peclet number of 44500 and a single hump in current distribution is achieved.
View full abstract
-
Takaji Inamuro, Akimasa Yamaguchi, Fumimaru Ogino
1997Volume 23Issue 6 Pages
789-795
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The effects of disk rotations and Prandtl numbers on fluid flow and heat transfer in a rotating cylindrical container with a counter-rotating disk at the fluid surface are numerically investigated. Flow and temperature fields are obtained for various rotational Reynolds numbers of the disk and for Prandtl numbers of the fluid. From the temperature fields the averaged Nusselt numbers on the walls are calculated and are compared with available experimental data.
View full abstract
-
Tohru Nagasako, Manabu Shimada, Kazuhiko Katayama, Kikuo Okuyama
1997Volume 23Issue 6 Pages
796-803
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The effects of addition of solvent on a polymerization reaction process were investigated taking into consideration temperature distribution, physical properties and conversion in the reactor. In the experiments, a mixture of styrene monomer, solvent and initiator for the reaction was fed into a vertically-standing tubular reactor equipped with electrical furnaces to measure the polymer conversion at the reactor outlet. When the initiator was added, the decrease in the conversion caused by the addition of the solvent was smaller than the case in which no initiator was added. The measured dependences of the conversion on experimental conditions were explained well by numerical simulations for momentum, heat and mass transfer accompanied by the polymerization reaction, which took account both for the change in the physical properties with temperature and conversion, and for the heat production by the polymerization reaction. The simulation results showed that the addition of the solvent caused the distributions of axial velocity and viscosity to be smaller. The results also indicated that such a change in the flow field in the reactor led to a change in the distributions in residence time and conversion of styrene in the reactor, which suggested the importance of considering the transport phenomena in a reactor for evaluating the effects of solvent on a polymerization reaction process.
View full abstract
-
Hiroshi Nogami, Jun-ichiro Yagi, Reiji Mezaki, Hiroaki Nagai
1997Volume 23Issue 6 Pages
804-810
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
A mathematical model is developed for a moving bed type coke gasifier based on the theories of transport phenomena, chemical reaction kinetics and two-phase flow. This model consists of conservation equations of momentum, heat and mass, and equation of continuity for reaction gas and coke particles together with reaction rate equations. The model estimates temperature and concentration distributions of gas and solid within the gasifier as well as flow characteristics. The validity of this model is confirmed by comparison of measured and computed temperature distributions in the gasifier.
Numerical experiments, which were performed to study the effect of furnace diameter, feed gas composition and feed rate of gasification agents, clarify the effect of these parameters on the furnace operation. Additionally, computation results show the importance of heat loss through the furnace wall for a gasifier with a water jacket.
View full abstract
-
Shinichi Yuu, Kouki Nohara, Daisuke Futai, Toshihiko Umekage
1997Volume 23Issue 6 Pages
811-819
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Air and particle motions in a three-dimensional turbulent fluidized bed which contains more than 24 million small particles (
Dp= 310 μm) are numerically simulated. In this study, a stochastic scheme based on Direct Simulation of the Monte Carlo (D. S. M. C.) method was introduced into the calculation of particle collisions in order to treat a large number of particles. The locally averaged three-dimensional Navier-Stokes equations and Lagrangian type particle motion equations, where the drag and lift forces acting on particles, the collisions between particles and the mutual interactions between air and particles are taken into account, was simultaneously solved. Collisions between particles were calculated by using the hard sphere model (this means that a collision occurs between only two particles). Calculated results using D. S. M. C. method are in fairly good agreement with the experimental results of flow visualizations and air and particle turbulent characteristics. Calculated results described the formation of particle clusters well. Both calculated and experimental results quantitatively showed that turbulent motions of air and particles in a fluidized bed are remarkably enhanced by their mutual interactions.
View full abstract
-
Chun-Hua Luo, Shigeyuki Uemiya, Toshinori Kojima
1997Volume 23Issue 6 Pages
820-827
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
A mathematical model is developed to describe the gas transfer in the grid zone of a jetting fluidized bed. The model treats the mass exchange coefficient as a fitting parameter, and is used to calculate the concentration distribution of gaseous species. The results are compared with the experimental results measured in a bench-scale jetting fluidized bed at high temperature using inert He gas as a tracer. The experiments were carried out by varying gas velocity from the nozzle and bed temperature. The gas flow rate from the distributor was kept at the minimum fluidization level. Based on experimental results published in literature, a correlation for gas exchange coefficient is proposed. The prediction coincides with the experimental results when gas percolation from jet to annulus is taken into account in the model.
View full abstract
-
Takanobu Hanabusa, Shigeyuki Uemiya, Toshinori Kojima
1997Volume 23Issue 6 Pages
828-834
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Numerical simulation of polycrystalline silicon particle production by fluidized bed CVD using monosilane is carried out. The model consists of a Kunii-Levenspiel fluidized bed model and reaction kinetics of monosilane pyrolysis in the free space and on particles. Conversion of monosilane, conversion to fines (fines elutriation), and vertical temperature profile in the bubble are calculated using the present model. Effects of scavenging of the fines, deposition on the fines and the temperature profile on the conversions are discussed. The numerical results are compared with the experimental results of Hsu
et al, . The numerical results of the conversions in the case of deposition on the fines are larger than those without deposition, while other factors give lower conversions. Consideration of deposition on the fines is found to be essential to explain the experimental results. The diameter of secondary particles of fines agglomerates should be taken into account as a function of bed temperature. It is suggested that the experimentally observed effect of the bed temperature on the fines elutriation is possibly explained when the degree of scavenging or the diameter of secondary particles is increased with increasing bed temperature, both of which are caused by the increase in fines' adhesivity with temperature.
View full abstract
-
Meguru Kaminoyama, Kazuyuki Nishimura, Kazuhiko Nishi, Mitsuo Kamiwano
1997Volume 23Issue 6 Pages
835-843
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
We propose a method of numerically analyzing a polymerization process, extending for a few hours in a batch-type stirred vessel using modeling flow fields by zoning reaction process. We also propose a method of estimating quality of the product polymer, such as average molecular weight distributions and index, while at the same time simulating the reaction process. The methods are applied to thermal bulk polymerization of styrene in a stirred vessel equipped with a double helical ribbon impeller.
We verified that the methods proposed here have enough reliability by ascertaining good agreement of the simulated results with experimental values in the same system as that of the simulation.
View full abstract
-
Takashi Shimada, Noriaki Baba, Satoru Komori
1997Volume 23Issue 6 Pages
844-849
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Friction and pressure drags acting on two- and three-dimensional wavy walls are investigated using a three-dimensional direct numerical simulation (DNS). Two conventional methods for estimating the friction velocity from the log-law of the mean velocity and a peak value of the Reynolds stress were also examined applying the result of DNS. The results show that the ratio of the friction drag to the pressure drag is 0.86 and 2.9 for two-and three-dimensional wavy walls, respectively. Both conventional methods overestimate the friction velocity at about 10% for a two-dimensional wavy wall but they give good accuracy for a three-dimensional wavy wall.
View full abstract
-
numerical simulation of springtime high-concentration
Seok Jea Kang, Hiromasa Ueda
1997Volume 23Issue 6 Pages
850-860
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Air quality in East Asia could deteriorate considerably as a consequence of accelerated development of fossil fuel systems and high economic and population growth rates. Frequently, on clear spring days large scale wind systems develop and transport pollutants from the Asian continent towards the Northwestern Pacific Ocean. On these high-concentration days, a yellow sand event shows good evidence of long range transport from the continent toward the Northwestern Pacific Ocean occurs. In the present paper, an Eulerian transport/chemistry/deposition model of regional-scale is developed to obtain the concentration changes of air pollutants over this area and a series of numerical simulations are carried out for clear springtime high-concentration air pollution epiosodes, mainly from 31 March to 3 April 1993. A field observation is also carried out to measure the concentration of air pollutants at three sites in Yantai (China), Cheju island (Korea) and Kagoshima (Japan) in 1993, 1994 and 1996.
At first, from a detailed comparision of the simulation results with the field observation and monitoring data in the East Asia, the validity of the present numerical model is confirmed. Secondly, by numerical simulation, the formation of a highconcentration air pollution belt in East Asia, connecting the eastern area of mid-China, the southern area of Korea and the western area of Japan is shown clearly. In particular, the trospheric ozone concentration in this belt area is found to be remarkably high and a high ozone, concentration at the level of Japanese Environment Standard or more, resuits in Japan and the Northwestern Pacific Ocean through long range transport from the continent.
View full abstract
-
Comparisons of Two-Phase Flow Models and Higher-order Up-wind Difference Schemes
Kenya Kuwagi, Hiroyuki Ozoe
1997Volume 23Issue 6 Pages
861-869
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
A dispersed flow model a two-fluid model for a gas-liquid two-phase flow problem. This dispersed flow model is compared with a single pressure model used generally by adopting higher-order up-wind schemes for the advection terms. They are Kawamura-Kuwahara, UTOPIA, QUICK and the second order up-wind schemes. Analyses were carried out for a bubbly flow injected from a small circular central bottom wall of a cylindrical vessel.
The computation for the dispersed flow model is more stable than the single pressure model. The QUICK scheme gives the best agreement with the measured results of the four schemes.
View full abstract
-
Kenji Nakashima, Shinichi Yuu
1997Volume 23Issue 6 Pages
870-877
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
In the simulation studied in this paper, the fifth-order upwind finite difference scheme and second-order Runge-Kutta method are used for the convective terms and the time developing term, respectively, in Navier-Stokes equations to simulate directly free air jet flow. The Reynolds number is 1200 and the calculated flow field ranges from the nozzle exit to the developed turbulent region. Since the computational cell size is smaller than Kolmogorov microscale in the developed region, calculated results can represent motions including micro-scale eddies. Calculated turbulent characteristics are compared with experimental results. Calculated results are in fairly good agreement with experimental data. The results show differences between high and low Reynolds number free jets. The calculated power spectra express experimental results, including high frequency spectra, well. Our computational visualizations show the unique phenomena of a low Reynolds number free jet, such as radial projecting eddies. It is concluded that the direct numerical simulation describes large-scale to microscale eddies in a low Reynolds number free jet, well.
View full abstract
-
Jun Fukai, Yuji Shiiba, Mitsuru Tanaka, Osamu Miyatake
1997Volume 23Issue 6 Pages
878-884
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Fluid flow and heat transport phenomena of a liquid droplet colliding with a substrate are modeled. The fluid flow model accounts for the deformation of the droplet. The heat transfer model accounts for heat transfers within the droplet and the substrate. The mathematical models are numerically solved using a finite element method.
Calculations are performed for the case where a molten tin droplet impacts various substrates. As a result, the following conclusions were obtained :
1) The periphery of the spreading droplet is more rapidly cooled than the central region.
2) The effects of initial temperatures and thermophysical properties of the droplet and the substrate on the temperature at the liquid-solid interface can be assessed using the analytical solution for the case where two infinite solids of different initial temperatures are contacted.
3) Heat transfer rate at the liquid-solid interface is dominated by the heat conduction within the substrate.
4) The heat transfer rate at the liquid-solid interface increases with increasing preimpact velocity of the droplet. The increase of the heat transfer rate is due to enlargement of the heat transfer area rather than improvement of convectional heat transfer within the droplet.
View full abstract
-
Tetsuya Yamamoto, Kengo Okanishi, Shinya Watanabe, Jieff Ma, Eiji Shin ...
1997Volume 23Issue 6 Pages
885-891
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
In order to obtain the information of the high-performance cyclone classifier, three-dimensional numerical simulation and experimental studies are conducted on the fluid flow and particle trajectory in a cyclone classifier with and without blow-down. By use of the blow-down method, the overall separation efficiency of a cyclone is higher than the case without blow-down. From the results of simulation, the reason for this is found to be the fact that downward velocity components and circumferential velocity components near the conical wall are stronger in the case of blow-down compared to the case without blow-down.
It is also found by the simulation that regardless of the use of blow-down method, as the initial position of particles at the cyclone entrance is lower, the partial separation efficiency becomes higher.
It is also made clear by comparing the calculated partial separation efficiency and experimental results that in the case of blow-down method, the repulsion probability of particles near the wall becomes lower than the case without blow-down.
View full abstract
-
Fumiaki Yoshikawa, Susumu Nakashima, Norihisa Shiraishi
1997Volume 23Issue 6 Pages
892-898
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
In order to use coke oven gas (COG) as a chemical feedstock and/or city gas, it is necessary to reduce or eliminate various impurities, especially sulfur compounds. An adsorbent to remove a sulfur compound, carbon disulfide (CS
2), at around room temperature is investigated. Calcium silicate impregnated with diethanolamine is found to show high adsorption activity in the low temperature range of 4050°C. In COG containing large amounts of hydrocarbons and carbon dioxide, the adsorbent removes 100% of CS
2 for 280 days at 45°C and a space velocity of 99 h
-1. By using this adsorbent and the previously reported K
2CO
3 impregnated alumina catalyst, more than 96% of CS
2 and 76% of carbonyl sulfide, which account for most organic sulfur compounds in COG, are removed for 300 days at 45°C under atmospheric pressure
View full abstract
-
Kazuhiro Kido, Nakamasa Sato, Masaya Shimoji, Eiji Nakanishi
1997Volume 23Issue 6 Pages
899-905
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
A feedforward/feedback control system was developed to maintain the temperature profile of air in an air conditioning system for photosensitive material coating and drying. The feedforward control was attained based on a heat exchanger dynamic model using the flow rate of hot water as the manipulated variable and the air temperature as the controlled variable, while feedback control was performed by means of optimum control theory based on a linearized heat exchanger dynamic model. To evaluate the performance of the control system developed in this study, simulation and experimental studies were carried out where a stepwise change of set values is performed in order to maintain uniform production quality for each product grade. It is shown that the control system under consideration successfully controls air temperature in an air conditioning system for photosensitive material coating and drying.
View full abstract
-
Toshitaka Yoshino, Masashi Momonaga, Katsuhiko Shinozaki, Hisatoyo Yaz ...
1997Volume 23Issue 6 Pages
906-913
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Most medical supplies which are obtained as purified crystals are produced in the form of tablets mixed with additives before dosing. The most important property for such medical supplies is to control the elution rate of tablets appropriately. The elution rate changes according to the internal structure, crystal size, affinity with additives and other factors. To develop such a complex, compound tablet, crystallization of Octotiamine was carried out and the crystallization method and conditions that produced the optimum tablet are established. Furthermore, we examin scale-up factors for this crystallization process.
View full abstract
-
Combustion System (RRX)
Yasuo Hirose, Hitoshi Kaji, Norio Arai
1997Volume 23Issue 6 Pages
914-919
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
We have developed a rotary regenerative combustion burner (RRX burner) as an energy saving type, which is combined with a regenerative air preheater and can be operated continuously.
In this burner, motive steam or air is provided through a jet nozzle located coaxially outside of the fuel gas nozzle to keep the flame sharp. By changing the flow rate of motive fluid, combustion characteristics of the burner can be easily changed. Test results indicate that, by increasing the flow rate of the motive steam, the temperature profile along the flame becomes more uniform and NO
x concentration therefore becomes lower.
Another characteristic of this burner is that the whole amount of high temperature combustion flue gas returns to the burner and is exhausted to the flue through the burner after exchanging heat with incoming fresh air.
We tried to reduce the amount of hot return gas to the burner by providing a new flue on opposite side of the burner wall so as to exhaust some of combustion gas through it.
As a result of the above, it is observed that by decreasing return gas, the concentration of NO
x is lower and the flame is longer.
In addition, it is proven that fuel oil can also be burned with this RRX burner without any serious problems.
View full abstract
-
Hiroaki Kuroki, Naoki Fujiwara, Masayuki Horio
1997Volume 23Issue 6 Pages
920-927
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Combustion behavior of CWP (Coal Water Paste) fuels made of 15 types of coals was observed by using a small atmospheric bubbling fluidized bed combustor made with a quartz tube. Ignition delay, flame combustion time and char combustion time were measured and compared with those of dry coal. From these observation, the combustion behavior of CWPs was classified into three different modes. Some CWP aggregate broke up already during drying and their coal particles burned separately as in the case of dry coal combustion (mode 1). Combustion aggregates that maintained their size during the flame combustion period were classified as mode 2. Some CWP aggregates bursted into small pieces during the flame combustion period (mode 3). Burn out times of mode 2 and mode 3 CWPs were much longer than those of the dry coal combustion. It was found that the factor governing the combustion behavior of CWPs is the crucible swelling number of coal. The relation between the combustion modes and the prescription for CWPs was also studied.
View full abstract
-
Takeshi Kachi, Noriyuki Kobayashi, Norio Arai, Yoshinari Kato, Kinya K ...
1997Volume 23Issue 6 Pages
928-933
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
This paper describes the effects of SiC coating on the oxidation resistance of C/C composites in combusting fields, which are expected to be applied to high temperature structural materials at over 1770 K. The coating methods employed were CVD and pack cementation. The time changes in weight loss of the specimens were measured at temperatures of 1770 K and 1900 K under the equivalence ratio of 0.9 generated by methane-air combustion, and the surface of the specimens before and after the experiment was observed by SEM.
Although the weight loss of the specimens coated by the CVD method was minimal, the coating layer was easily peeled off from the substrate.
On the other hand, the layer of the specimens coated by the pack cementation method was stable and adhered to the substrate, but the substrate was degraded because of penetration of oxygen through the pores in the layer. To cover the pores, the specimens were additionally coated with glass materials comprised of metal-Si, B
4C, SiC and ZrSi
2. No weight loss occurred for the double-coated specimens, and this coating technique seems to be promising to enable C/C composites to resist oxidation in combusting fields.
View full abstract
-
Masaru Noda, Shinji Hasebe, Iori Hashimoto
1997Volume 23Issue 6 Pages
934-941
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The optimal reflux operation which minimizes energy consumption is derived for rectifying, stripping, and total reflux columns separating binary components. The batch distillation columns are modeled as differential and algebraic equation (DAE) systems. First, each DAE system is converted to a set of algebraic equations using orthogonal collocation on finite elements. Then, the problem of finding the optimal reflux policy is formulated as a nonlinear programming problem, and it is solved using successive quadratic programming techniques.
The results of the example problems show that the separation performance of the total reflux column can be improved considerably by optimizing the reflux flow rate, though the separation performance of the rectifying column can not be improved so much, even if the reflux ratio is optimized. As a result, the energy consumption of the total reflux column is equal to or less than that of the rectifying column in most cases. The reason why the separation performance of the stripping column is worse than those of the other columns is made clear using the characteristics of vapor liquid equilibrium.
View full abstract
-
Issei Fujiyoshi
1997Volume 23Issue 6 Pages
942-946
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Sodium ion and ammonium ion were adsorbed by the ion exchange method on granular porous vycor glass at 298 K. An equation to estimate the amount of ion exchanged was introduced, assuming that solid acid contains mutiple kinds of ion exchange sites which have different acidities is and ion exchange equilibria respectively. A numerical analysis is carried out in order that the estimated values are fitted in with the experimental values for the amount of ion exchanged. It reveals two kinds of ion exchange equilibria, and give the apparent ion exchange equilibrium constants and effective amounts of acid. Then, the amounts of sodium ion and ammonium ion exchanged can be estimated. The apparent ion exchange equilibrium constants of sodium ion corresponding to the acidity index, that is, pKa of 6.41 and 8.72, suggest that porous glass is a considerablely weak dibasic solid acid. It is considered that such acidity of porous glass is due to geminal-type and vicinal-type of silanol groups and that isolate types do not contribute to ion exchange.
View full abstract
-
Shuichiro Hirai, Taichi Taniguchi, Toshimi Takagi
1997Volume 23Issue 6 Pages
947-953
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Direct numerical simulation of a two-dimensional mixing layer is conducted and mass transport phenomenon are discussed based on the database obtained from the simulation. Retardation of mass transport in the initial vortex formation region of the mixing layer is pointed out and its mechanism is discussed by comparing the terms that make up the transport equation of mass flux
v'c'. Retardation of mass transport originates through the effect of the turbulent diffusion term-∂ (
v' v' c') /∂
y suppresses the mass transport. This phenomenon is discussed in detail by considering the instantaneous turbulent diffusion term. At the initial vortex formation region of the mixing layer, interaction between (1) large velocity induced by a secondary flow formed by the interaction between the vortices and (2) tail-type instantaneous concentration profile is considered to have a large effect on the instantaneous turbulent diffusion term, retarding mass transport.
View full abstract
-
Guo-Qing Lu, Hee-Joon Kim, Ichiro Naruse, Kazutomo Ohtake, Mitsushi Ka ...
1997Volume 23Issue 6 Pages
954-961
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The combustion time history of a biobriquette is divided into two stages in which volatile matter content mainly evolves and burns prior to char combustion. The volume model for the former stage and the shrinking-core reaction model for the latter stage are employed to calculate the combustion rate of the biobriquette. In this study, the influence of briquette shape, compression pressure and types of coal on the combustion rate are investigated by using the two models and experimental data.
The shapes of spherical and cylindrical biobriquettes hardly influence the combustion behavior of the biobriquette (if they have same diameter and the length of cylinder is equal to its diameter). Compression pressure has no effect on the rate of the former stage but gave obvious effect on the latter stage through the density change of the biobriquette when it is under about 245 MPa. The reaction rate constant and its order vary linearly with the fuel ratio of coal briquette in the former stage. The type of coal and compression pressure have a larger influence on the rate of the latter stage due to different porosity in the ash layer. Finally, the combustion rate can be analyzed by the obtained models and semi-empirical equation.
View full abstract
-
Effect of Intraparticle Diffusion
Takashi Aida, Ryoichi Kobayashi, Hiroo Niiyama
1997Volume 23Issue 6 Pages
962-968
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Periodic operation of NO-CO reaction over Pt/Al
2O
3 was conducted by using an egg-shell-type catalyst and an egg-white-type catalyst in order to investigate the effect of intraparicle diffusion on the reaction performance. The former catalyst held platinum in the outer layer of the particle and the latter held in the inner layer. Time-averaged conversion of NO under periodic operation exceeded that under steady operation in a certain range of periods, showing a maximum versus period for both catalysts. In the case of the egg-white-type catalyst, the conversion rose and declined with a lasser dependence on the period than in the case of the egg-shell-type catalyst. Under periodic operation with periods up to 360 s, the egg-white-type catalyst showed reaction performance higher than that under steady operation. A model calculation reproduced the tendency and revealed that intraparticle diffusion slowed the rate of desorption of CO, and that the slow desorption of CO resulted in the experimental observation for egg-white-type catalyst.
View full abstract
-
Setsuro Hiraoka, Noboru Kamei, Yoshihito Kato, Yutaka Tada, Hyeong-Gyu ...
1997Volume 23Issue 6 Pages
969-975
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The power consumption for a pitched blade paddle impeller in both baffled and non-baffled agitated vessels was measured over a wide range of Reynolds numbers from laminar to turbulent flow regimes. Power consumption for a non-baffled vessel was well correlated with the same correlation as that for a paddle impeller by modifying the coefficients in the correlation with blade angle. Power consumption in a baffled vessel was also well correlated with a correlation method similar to Kamei's one for a paddle impeller. The fully-baffled condition and the corresponding power number were presented by the following equations.
(Bω/D) nb0.8≥0.44
Npmax,
θ0.2 (2θ/π)
0.72Npmax,
θ=8.3 (2θ/π)
0.9 (
np0.7bsin
1.6 θ/
d)
View full abstract
-
Sachiko Kagawa, Hidehiro Kamiya, Masao Azegami, Masayuki Horio
1997Volume 23Issue 6 Pages
976-982
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
To obtain fundamental information on the combustion and emission characteristics of refuse derived fuel (RDF) in fluidized bed combustors, the behavior of NO
x emission was investigated with a special focus on the effects of CaO and Cl on it. Several model RDFs, in which CaO and Cl were present or not, prepared from the mixture of news paper, bean-curd refuse and PVC. The molecular composition of CaO, Cl, C, H and N in the model RDFs was adjusted to a composition almost the same as those of commercial RDFs prepared from municipal solid wastes. The model RDF containing both CaO and Cl had higher conversion of fuel-N to NO
x than that of the model RDF containing only either one of the two. The conversion of fuel-N to NO
x of model RDF containing either CaO or Cl did not differ from that of model RDF with no CaO nor Cl. The effect of CaO-Cl coexistence on NO
x emission increase was intensified either when stoichiometric air ratio was decreased or CaO content was increased. The catalytic activity of CaO for NH
1 oxidation to NO
x was found to be much increased by the existence of Cl in RDF. Furthermore, it was confirmed that preparation process, the density and shape of RDF and the difference of Ca form (i.e. CaO or Ca (OH)
2) are the minor factors concerning the conversion of fuel-N in RDF to NO
x.
View full abstract
-
Hiromasa Komatsu, Masayuki Nakamura, Yasuhiro Yamashita, Chirima Hirai
1997Volume 23Issue 6 Pages
983-992
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
At atmospheric pressure, vapor-liquid equilibrium relationships for five binary systems of methanol-tert-butylalcohol, methanol-tert-butylmethylether, tert-butylal-cohol-tert-butylmethylether, water-tert-butylalcohol and water-tert-butylmethyleth-er are observed. The Wilson parameters for the binary systems have been obtained. Comparisons between the Wilson method, The ASOG method and the UNIFAC method are made. Vapor-liquid equilibrium relationships of the quaternary reactive system producing tert-butylmethylether by use of methanol and tert-butylalcohol are also observed, and it is confirmed that the Wilson parameters resulting from the data of binary systems can be used to estimate the vapor-liquid equilibrium data for the quaternary reactive system of methanol-tert-butylalcohol-tert-butylmethylether-water.
View full abstract
-
Atsushi Ikari, Yasuo Hatate, Ryoko Aikou
1997Volume 23Issue 6 Pages
993-997
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
Vapor-liquid equilibria of a watertasu+ 1-butanol system containing a trace amount of furfural were measured at atmospheric pressure by use of a Iino-type still for systems of limited miscibility.
Vapor-liquid compositions for the major components (water and 1-butanol) are shown to be nearly coincident with those of the binary system.
In the partially miscible region, the vapor-liquid equilibrium ratios of the trace component (furfural) at bubble point were found to be 2.5 and 0.46. Consequently, the partition coefficient of the trace component between the two liquid phases is 5.4.
The equilibrium ratio curve of the trace component is presented, in which the calculated curve within the partially miscible region is shown to be coincident with the experimental data.
View full abstract
-
Tomoaki Satoh, Yosuke Yamamoto, Mikio Konno
1997Volume 23Issue 6 Pages
998-1000
Published: November 10, 1997
Released on J-STAGE: November 12, 2009
JOURNAL
FREE ACCESS
The sol-gel method is applied to the synthesis of Pb (Zr
0.5Ti
0.5) O
3 particles with various post treatments. After the reaction of Pb (OAc)
2, Zr (OBu)
4 and Ti (OEt)
4 in ethanol solvent at 73°C over different reaction times, complex alkoxide is subjected to ammonia water and acetonitrile as a cosolvent at various water concentrations. The selection of reaction conditions enables the production of highly monodispersive particles with an average diameter less than 100 nm.
View full abstract