To improve the utilization efficiency of energy, development of technology including that for high-temperature gas tubine and MHD power generation systems, has been carried out. Exhaust gas from such high-temperature combustors contains high NO_x concentrations of several thousand ppm.
This report describes the application of a non-catalytic de-NO_x process by ammonia injection to exhaust gas that contains high concentration of NO_x.
The main results are as follows.
(1) High NO_x over 3000 ppm can be reduced by optimizing ammonia injection conditions, such as gas temperature at the ammonia injection point and ammonia injection ratio.
(2) Use of the multistage ammonia injection method is recommended in order to reduce unreacted ammonia in exhaust gas while reducing NO_x concentrations to less than 200 ppm.

View full abstract

Experiments in heating of a hollow steel device and its theoretical analysis as an unsteady 3-dimensional combined heat transfer problem of radiation, convection and conduction were carried out for a full-size heat treatment furnace equipped with radiant tube heaters to establish a strict design method for industrial furnaces from the viewpoint of heat transfer.
Experiments to clarify the difference in heating performance due to the input of heaters installed in the furnace and the emissivity of the device, both of which are very important factors in heat transfer design of the furnaces, were also conducted.
The results of heating experiments and theoretical analysis coincide closely, and this theoretical analysis of the full-size furnace can simulate the heat transfer sufficiently well to become a basic design method. As for the emissivity of the device, low emissivity makes the radiant heat transfer small, which results in slow heating of the device, but it also accelerates the heating of the inner walls. It was confirmed that heating of the device becomes slow and its temperature becomes uniform if the input of the heaters is decreased.

View full abstract

In solvent recovery by activated carbon adsorption, a serious degradation of adsorption capacity of the activated carbon often occurs, especially when ketones, such as cyclohexanone, are contained in the solvents. This is due to the alteration reactions of solvents adsorbed on the activated carbon, such as oxidation, decomposition and polymerization.
Ordinary solvent recovery processes include an adsorption step at normal temperature and a desorption step at 100 °C, using heat from atmospheric-pressure steam. The authors found by fundamental experiments that decreasing the temperature in the desorption step, for instance to 80°C, is very effective in preventing degradation of the activated carbon, and have developed an improved solvent recovery process utilizing a low-temperature (about 80°C) and reduced-pressure (about 47 kPa) steam desorption step. It is called the “Low-Temperature Desorption Solvent Recovery Process”.
A long-run experiment has been carried out for 142 days in a pilot apparatus based on this process in an existing conventional plant, to show that a very long life of activated carbon, i.e. 35 times longer than that in an ordinary process, is attainable.A cmomercial plant based on this process has been in operation since February, 1988.

View full abstract

A new mixer with a perforated screw has been developed for the reactor in which the viscosity of reactants varies in a wide range, from lower to much higher viscosity. This mixer was applied to the continuous polymerization of acrylonitrile in an aqueous medium. It was observed that the reactant slurry overflowing from the reactor with the perforated -screw mixer was less viscous than those from the reactor with a paddle mixer. The reduction in viscosity of the reactant slurry successfully improved the flexibility of operation. In particular, this reduction allowed the concentration of monomer in the feed to be raised to a higher level. Measurement of the sedimentation velocity and the volume of polyacrylonitrile particles, and the permeation rate of water through the sedimented bed of the particles, indicated that the polymer particles produced with the perforated-screw mixer are more compact than those with the paddle mixer. It is confirmed that this compaction makes the reactant slurries less viscous.

View full abstract

When colloidal solutions containing such materials as protein are separated and concentrated by ultrafiltration, colloidal particles thickened on the filtration membrane become gelatinous and exhibit very high filtration resistance.
Cross-flow electro-ultrafiltration has been developed to remove the bottleneck. The filtration flux is increased because the thickness of the gel layer formed on the membrane is decreased by electrophoresis and by the shear force caused near the membrane. Moreover, the filtration resistances of the gel layer and the membrane are apparently decreased by electroosmosis.
In this paper, a design method for cross-flow electro-ultrafiltration is proposed for the case of gelatin colloidal solution based on the design and experimental equations of filtration flux obtained previously and a design equation confirmed experimentally in this study.

View full abstract

The adsorption equilibria of water vapour on structural active carbon (SAC) prepared either by carbonizing or by carbonizing and activating polyvinyl formal (PVF) under various production conditions were measured in the temperature region of 303 to 343K at reduced pressure. The results were compared with previous works and the applicability of SAC to the adsorption heat pump was studied.
1) The SAC produced had a sufficient water vapour adsorption capacity.
2) The adsorption equilibrium curves were roughly classified into two patterns. Those for SAC prepared by carbonizing or carbonizing and low activation (Group 1) took the shape of rising in a low region of relative vapour pressure (φ) and showed a loosely rising tendency with an increase in φ. On the other hand, those prepared by high. activation (Group 2) showed a sharp rise at φ =0.5.
3) In Group 1, except for the SAC prepared by carbonizing only, the reversibility of adsorption was excellent, and in applying these SACs to the adsorption heat pump the temperature difference for pumping up grew larger. Consequently, these SACs can be used as the heat storage material in the adsorption heat pump.

View full abstract

The purpose of this paper is to develop a model for predicting the effective thermal conductivity of fibrous insulation. A heat transfer model was proposed, considering both the conduction and the radiation forms of heat transfer, based on a unit cell model. The effective thermal conductivities at room temperature for a ceramic fibrous thermal insulation blanket composed of Al₂O₃ ; 0.92 and SiO₂; 0.08 were obtained by the guarded hot plate method in the range of air pressure from 1 to 10⁵ Pa and packed densities ρ from 50 to 183 kg/m³. These experimental results were satisfactorily correlated by the proposed heat transfer model, in which the fitting parameters were independent of air pressure as well as packed density.

View full abstract

This paper describes one approach to predicting radiative heat flux in fibrous insulations. The spectral extinction coefficients and scattering albedo in the wavelength range 1.5 < λ < 27 micro-meter at 293K and 723K for three kinds of commercial fibrous insulations composed of Al₂O₃ and SiO₂ were determined from results of optical transmission and reflection measurements and the four-flux radiative transfer calculation. Extinction coefficients of radiative heat transfer in the temperature range 350-800K were predicted from the spectral optical properties in the case where the radiative heat flux can be calculated according to Fourier's law. The radiation extinction coefficients were obtained from effective thermal conductivities in vacuum in the temperature range 373-773 K as measured by the guarded hot plate method, where the conduction form of thermal conductivity was predicted by the heat transfer model earlier proposed by the authors. The radiation extinction coefficients obtained by heat transfer experiments were compared with the predicted ones, and quantitative agreement was obtained.

View full abstract

A Petroleum pitch produced by the Eureka process contains small mesophase spherules which are dispersed in an optically isotropic pitch. A precursor of generalpurpose carbon fiber is available from the isotropic pitch which was filtered off the mesophase spherules.
The filtration of viscous molten pitch under constant pressure was experimentally studied by using an experimental apparatus equipped with a cylindrical filter medium.
After the experimental study, it was confirmed that the process of constantpressure filtration of the molten pitch consists of two stages : an initial filtration process that follows Ruth's cake filtration law, and a final filtration process that follows Hermans-Bredee-Grace's standard blocking law. Concerning the reason for the occurrence of blocking in the final filtration process, it is conceivable that the phenomenon may originate in the blockage of trace infusible materials contained in the molten isotropic pitch.

View full abstract

It is important to know the dynamical properties of a heat storage unit for effective use of energy at low temperature for air conditioning, hot water supply or dehydration. In this report, a neat system with air and packed bed of small-scale metallic capsules in which the phase change material (PCM) was encapsulated for the improvement of transient response was investigated, where heat could be stored with high density and used in a rather narrow temperature range.
Dynamical heat transfer problems in the storage unit were analyzed and the results of calculation were compared with experimental ones obtained with a packed bed of 4400 aluminium capsules, where Rose's metal was used as PCM. Agreement was fairly good, and good performance was confirmed.

View full abstract

Assuming linear combination among parameter deviations involved in system models, this paper is concerned with analysis and design methods for flexible linear systems. We focussed our special interest on the system having equality constraints explicitly, and discovered a few properties to be taken into account in applications.
In these developments, we recommended that soft variables must be distinguished from decision variables according to whether they are alterable around their standard variables or not. On the condition of this distinction, it was shown that we can obtain rational results for actual utilization.
Combining the present results with the previous ones, which assumed independent parameter deviation, we can formulate the problem simply in a similar framework regardless of difference of imposed deviations and models. Relying on the special structure of the problem, we also examined approaches which enable us to deal with problems having many uncertain parameters and/or large scale systems.

View full abstract

This paper deals with dead zones in two cases of flow patterns of bulk solids. One is seen in hoppers or silos where dead zones are formed near the side wall since the exit is provided in the center of the bottom, leading to the converging flow of particles. The other is that represented by the deadman formed in the center of an ironmaking blast furnace as a result of coke particles descending towards the surrounding side wall in a diverging manner.
Slip line analysis is used for predicting the dead zone boundaries in the two cases. The state of stress field, active or passive, is determined by the flow pattern of bulk solids. The internal and the wall frictions are also examined both theoretically and experimentally, and thus comparison becomes possible between theory and experiment. The results are satisfactory.

View full abstract

Synthesis of SiC ultrafine particles was carried out by employing the thermal CVD method where SiH₄ and a hydrocarbon gas were used as raw gas. Two hydrocarbon gases-CH₄ and C₂H₂-were used and compared with each other. Hollow SiC particles of diameter 40 nm were produced in the former case, solid SiC particles of diameter 17 nm in the latter case. A reaction temperature above 1400 °C was required for the formation of pure SiC in the former case whilst 1200 °C was sufficient for the latter case. The effect of mixing mode between SiH₄ and hydrocarbon was examined. Diffusion-type mixing could realize a higher conversion rate of SiH₄ to SiC than premix-type mixing under the condition of stoichiometric mixing ratio at the same reaction temperature.

View full abstract

Power consumption in a horizontal gas-solid mixed-type polymerization reactor with two parallel axes and a spacious upper part was investigated. Instantaneous torque was measured under operating conditions for various vessel dimensions, rotational speeds, clearance between blade tip and vessel bottom, and properties of solid particles.
The results show that the torque is periodically varied by a rotating blade and it increases when a blade enters a particle bed. Averaged torque is independent of both the internal and wall friction. Dimensionless averaged power consumption is well correlated with the Froude number and dimensionless height of the particle bed. The ratio of maximum torque to averaged torque is between about 2 and 5 and is also correlated with the Froude number and the dimensionless height of the particle bed.

View full abstract

A CFM-bed (Combined Fast Fluidized Bed and Moving Bed) gas-solid contactor was applied to a solid fuel combustor. This combustor was proposed as a promising solid-fuel combustor with high efficiency and a countermeasure against environmental pollution. The combustor is composed of a set of double pipes. A fast fluidized bed and/or an entrained bed is formed in the inner pipe where partial combustion of coal particles may take place. A moving bed is made at the outlet of the annular part wherechar combustion may happen.
In the present study, Loy Yang coal and coke were used as a typical low-grade solid fuel. Rich and lean combustion limits were experimentally investigated with a single-staged combustion mode by changing the primary air ratio. It was confirmed that this combustor could be operated successfully in a wide operational range of primary air ratio from 0.3 to 2.4. The principal ignition mechanism of the fule in this combustor was made clear by considering not only internal particle circulation but also the exhaust heat recirculation from the annular part to the inner pipe.

View full abstract

A new system of bulk liquid membrane has been developed in order to increase the permeation rate. The droplets of liquid membrane were dispersed into water phases to increase contact area. The transport of _L-isoleucine by liquid membrane containing Aliquat336 was used as a model system of the present study to demonstrate the performance of the system. The following results were obtained.
The permeation rate of _L-isoleucine in the new system was much higher than that in the conventional bulk liquid membrane. In addition, the permeation rate was dependent on agitating rate and size of droplets. This new system was favorable to operation and preparation of liquid membrane.

View full abstract

A simulation model is presented for predicting the dust load distributions in fibrous mat and granular bed filters. The calculations are based on the interception model proposed in the previous papers.
The increment of a single-fiber or single-sphere collection efficiency due to dust loads is derived on the assumption that the interception caused by captured particles works to increase the collection efficiency as the predominant mechanism.
It is confirmed that the simulation results are similar to the experimental results qualitatively. Quantitatively, however, the effect of re-entrainment of captured particles seems to cause some difference, which is confirmed experimentally to be dependent on the inlet dust loads and the filtration velocity.

View full abstract

Among the support effects on metal catalyst, strong interaction between metal and support (SMSI) is an attractive and up-to-date subject. In the study of this effect, specific activity at a definitive reaction condition is usually used. But for more precise examination, an analysis using rate parameters, such as rate constant, equilibrium constant, activation energy and heat of adsorption is required. In this study, the hydrogenation of propionaldehyde with nickel catalysts supported on various oxides was studied kinetically over a wide range of reaction conditions. The effect of SMSI on the reaction was tested. The supports were Al₂O₃, SiO₂, TiO₂, Nb₂O₅ and Ta₂O₅. At the condition in which the minor reactions were negligible the rate equation was derived and the rate parameters were determined. The experimental results for all catalysts used were well represented by this rate equation. In addition, the activation energy and the heat of adsorption were the same even for the SMSI state of the catalyst.

View full abstract

To investigate the reasons for variation in column performance with the composition of the test mixture, distillation experiments are carried out under similar conditions to those used by Liang et al., who reported remarkable variation in N_OG with composition. The present results also indicate that N_OG varies greatly with the composition of the test mixture.
To take into account the effect of surface tension on the performance, the surface tension of this test mixture is measured up to near the boiling point. The result shows that the surface tension of this test mixture at the boiling point has a maximum value at a certain composition and that the surface tension characteristics change from negative to positive at this point. Good correlation between N_OG and the stabilising index proposed by Moens was observed.

View full abstract

To enhance convective heat transfer, injection air was added through a slit into the laminar boundary layer formed on a flat plate. The effect of injection on heat transfer and skin friction characteristics was investigated experimentally.
Friction factor and heat transfer coefficient were measured for air flow with Reynolds number Re=u_∞·l/ν between 3.0×10⁴ and 6.4× 10⁵ and with injection rate F=ρ₀υ₀/ρ_∞u_∞ between 0.005 and 0.1.
Conclusions were obtained as follows : The heat transfer coefficient and skin friction factor increased over the whole range behind an injection slit and the point of transition moved upstream with increasing injection rate.
The extent of augmentation of the heat transfer coefficient and skin friction factor could be varied by varying the injection rate. This characteristic may have application in automatic control systems. On this point, the injection method has an advantage compared with a fixed separation-inducing obstacle that provides a heat transfer augmentation which is more or less invariable. When a performance evaluation for enhanced heat transfer was made under the condition of constant pumping power, the maximum values of Stanton number St/St₀ for F=0.1 and F=0.05 were 1.39 and 1.29 respectively.

View full abstract

For the development of a new type of waste water treatment, a semicontinuous filtration fermentation process which consists of the combination of a ceramic microfiltration membrane with a methane fermentor was investigated.
Optimum membrane pore diameter, cleaning conditions of the membranes, maximum organic substance loading for the fermentor and filtered stability at various loading conditions were studied experimentally.
A membrane having pore diameter of 0.04 μm permitted very stabilized permeate flux all through a continuous long-time filtration. Back washing by sodium hypo-chlorite solution renewed this membrane completely to the early permeability. These tendencies could be explained by the relation with size of fine particles in the ferment suspension.
The maximum permissible loading rate for the fermentor was about 3kg·m^-3·d^-1. In this loading, TOC removal and excess sludge rate per unit. removed TOC were about 90 % and 30 %, respectively. Above this loading, TOC removal and production gas rate decreased with an increase in loading, and propionic acid was gradually accumulated in the fermentation solution.
It was also found that the membrane permeability was almost constant independent of the loading changes.

View full abstract

A numerical analysis was performed to determine the heat transfer characteristics of axial laminar flow between a square array of cylinders with a uniform wall heat flux.
The velocity distribution is introduced according to the analytical solution obtained by Sparrow et al., and the energy equation is solved numerically to obtain the axial variation of temperature distribution by means of a forward-marching, implicit finite-difference method with iteration.
The numerical results for the local Nusselt number are presented graphically for the local Graetz numbers up to 10⁶ and spacing-to-diameter ratios of 1.01 to 4.0. From the results, approximate expressions suitable for predicting the local Nusselt number are derived.
The numerical results are compared with those obtained previously for a triangular array of cylinders. It is realized that the local heat transfer coefficient for the triangular array is larger than that for the square array under the same volume fraction of cylinders.

View full abstract

To prevent deterioration of supported liquid membrane (SLM) which is caused by wash-out of extractant from the SLM, a new method is proposed in which. Teflon porous plane film laminated with nonwoven polypropylene fibers is used as the support. The capillarity of the laminated polypropylene layer plays an important role in supplying membrane liquid to SLM. To investigate the applicability of this method, five extractants having different solubilities in water, from LIX64N to MIBK (methyl isobutyl ketone), were tested for long-term operation. It was found that even in the MIBK system, which has the greatest solubility in water in these systems, a long-life SLM can be realized with this method.

View full abstract

A flow -type apparatus was used to measure the vapor pressures of tetralin, 1-naphthol, and biphenyl in the temperature range from 308 to 383 K, 348 to 453 K, and 328 to 453 K, respectively. Most of the data reported in this paper were newly measured. The data are believed to be correct within 0.5 % margin of error from error analysis. The data are in good agreement with those reported in the literature. However, the deviation for 1-naphthol is relatively large. A significant deviation is also found for biphenyl in the solid state.

View full abstract

The control scheme recently developed by the authors for distributed parameter systems of a parabolic type was applied to a heat conduction process having a boundary input. Performance of the control system was experimentally examined. It is shown the controller is robust and works satisfactorily for the system with a boundary input.

View full abstract

A comment on the two modified numerical methods proposed by Yamada et al. is described. They proposed these methods under an understanding that Takata's central finite-difference method could not give a stable solution at higher Peclet numbers. But this understanding is questionable from the viewpoint of the author's published results, because they fixed a space increment for numerical calculation at 0.05 only and took no account of the significant effect of the space increment on the accuracy of numerical values. It is actually shown that Takata's method can give much more stable solutions than Yamada's two modified methods when the space increment is set at 0.01, even at a high Peclet number as high as 100.

View full abstract

Problems in the application of the digital PI or PID predictive control to practical control systems are studied. Predictive control is expected to be effective for a process having large dynamics and large time delay. This control system deals with a prediction of controlled variable. Practical application to a real process is very easy, because a conventional PI or PID controller can be applied.
The consecquent control performance is superior to that with the conventional PI or PID control. The usefulness of predictive control is confirmed through digital simulations and experiments.

View full abstract

The Smith compensator is very interesting for a large dynamical system. However, it needs a precise process model, and then it is very difficult to apply for practical control. Here, a very simple model which is of the first order and time delay is introduced. Smith compensating configuration with this model gives control performance superior to that with a conventional PI or PID control. This control system is applied for temperature control of heaters.
The proposed method is very easy to desing and may be considered as a practical and effective way to improve control performances.

View full abstract

Thermal desorption of mixed organic solvent from activated carbon was studied to determine solvent vapor concentrations in the work environment. Desorption was conducted at a low temperature, e.g., below 473K, to avoid thermal decomposition of solvents. A method of estimating desorption curves and desorption efficiencies is proposed on the basis of mass balance and extended Freundlich isotherm. As the estimated values of desorption efficiency agreed well with the experimental values, solvent vapor concentrations in the air are considered to be determined from the experimental values of the amount of solvent desorbed and those of the desorption efficiencies estimated.

View full abstract

The liquid-side mass transfer performance in an annular bubble column with a rotating cylinder was studied experimentally for various operating conditions and rotor diameters. The limitation for the mass transfer varied from an aerationcontrolling one at relatively high gas rates to an agitation-controlling one at relatively high rotor speeds, in a manner similar to the case of an aerated mixing vessel. The volumetric mass transfer coefficients increased with increasing rotation speed and diameter of the rotor in the agitation-controlling region and were larger than those for a bubble column and lower than those for an aerated mixing vessel at the same power consumption.

View full abstract

Numerical simulations of bench -scale reactor behaviour were carried out for methanol conversion to light olefins over a modified ZSM -5 type of zeolite catalyst. This catalyst has high selectivities and high yield properties for ethylene and propylene in methanol conversion. The kinetic model proposed consists of reactions in which olefins having 25 carbons are formed and cracking reactions. It was demonstrated that the calculated concentrations of species at the exit of the catalyst bed and the temperature profile in the bed were in good agreement with the observed data.

View full abstract

Ca⁺⁺, Na⁺ and a small amount of K⁺ ions were found to dissolve from chlorite particles into water. Concerning the Ca⁺⁺ and Na⁺ ions, measurements were made of the ion dissolution rates and the saturation concentrations at various temperatures.

View full abstract