We constructed a 40 liter gel ice storage to search for how to design gel units as an ice storage substance. We studied the effects of size and shape of gel units on the ice production rate, ice packing factor and the cooling output under practical operation conditions. The following results were obtained. (1) To get a uniform temperature distribution within the gel units during ice production, we have to design the sizes and shapes of them so that the mean evacuation rate per evaporation surface area of the gel will be a value of under 5.7 g/ (m²·min). (2) A column gel unit of 30 mm in diameter showed uniform ice production at a practical speed. (3) From the experiments with 20 to 40 liters of column and sphere gel units, the followings were shown : a) The ice packing factor was 58% for the column gel unit. b) Water temperatures of 2.5 to 4°C were obtained at the outlet of the gel ice storage packed with 40 liters of column gel unit. c) The cooling output for the column gel unit was 0.58 kW, and this value is high enough for practical use.

View full abstract

To establish technologies for a refrigeration system of a temperature level below -20°C using hydrogen-absorbing alloys driven by solar energy, the simulation technologies capable of modeling the system's performance are developed, and the optimized control method driven by solar heat is studied.
The comparison between experimental results and calculated results demonstrates the validity of this model. The effectiveness of a new control method by using switching cycle time is reported with the simulation results.

View full abstract

This paper describes the performance characteristics of the absorber and the generator in an absorption refrigerator and heat pump, which can be driven by heat sources, using LiBr solution as working fluid. The absorber (generator) is composed of a vertical outer falling film column (height : 1 m and outer diameter : 15.7 mm), into which cooling (heating) water streams. Experiments are made at the LiBr solution flow rates from laminar to turbulent regime, and at three flow rates of cooling (heating) water. In the laminar flow region, the absorption (evaporation) rate |n_A| drops as the LiBr solution flow rate decreases, because the concentration of the falling solution changes greatly and the temperature driving force for absorption (generation) diminishes. On the other hand, |n_A| increases as the flow rate increases in the turbulent flow region, because the change of the falling solution concentration is small and the increasing flow rate enlarges the values of heat transfer coefficient h_L and mass transfer coefficient k_L, for the falling film. In the small flow rate region, the values of |n_A| for this column (height : 1 m) are less than those for a 0.58 m column, which we reported previously. Both the values of|n_A| for 1 m and 0.58 m columns, however, are the same in the large flow rate region. |n_A| decreases asthe flow rate of the cooling (heating) water decreases.
These experimental results agree well with the values obtained by the calculation method in which the temperature and the concentration at the vapor-liquid interface can be computed by using h_L and k_L of the falling film.

View full abstract

With a view to utilizing the cold heat of Liquid Natural Gas, candidates for cold storage media, which store cold heat in a temperature range between -20 to -50°C as latent heat, are investigated. Amine/water binary solutions are found to be promising candidates. A liquid/solution equilibrium experiment shows eutectic character of these solutions. Ametylamine/water solution is expected to be the lowest temperature cold heat storage media, and is examined in a phase change characteristic experiment. The solution generats dendrite crystals in the first phase change. The generated liquid/solid solution shows eutectic slurry character. Based on the experimental results, the possibility of using the eutectic solution as a fluid high-density cold storage media is discussed.

View full abstract

The heat output performance of a magnesium oxide/water chemical heat pump is studied using a newly developed reaction material which has high durability under repetitive reaction. A reactor bed having 1.8 kg of reactant is examined for the hydration and dehydration of magnesium oxide. The repetitive reactivity of the reactor is demonstrated through a nine cycles repetitive reaction. Although initial reduction in reactivity is observed, the following cycles maintain stable reactivity. From a high-temperature output experiment, output at 110 and 130°C is achieved using saturated steam at 70 and 80°C, respectively. These results suggest the possibility of steam supply system using the heat pump. The heat output performance of the heat pump is discussed from the experimental results.

View full abstract

Heat and mass transfer characteristics of an adsorber, composed of an ethanol Adsorption Heat Pump (AHP) of Plate Super Activated Carbon (PSAC), and a heat transfer sheet, are evaluated numerically. The relation of ethanol adsorption range and cycle time for effective AHP operation are also studied.
Numerical results of temperature change and amount adsorbed, by use of a proposed model of one-dimensional mass balance and two-dimensional heat balance, are well fitted to the results of experiments on account of a contact resistance between PSAC and sheet. Cold heat output under several temperature conditions can be estimated by use of this model. Cycle time for maximum output under several ethanol operation ranges are found. An AHP with PSAC has high cold heat generating power as much as silica-gel/water AHP, which is enough for practical use, more than 0.3 kW·kg^-1.

View full abstract

An energy network system, in which low level waste heat from industry region is recovered by endothermic methanol decomposition into H₂/CO, is proposed.
In this investigation, a plate type catalyst for methanol decomposition applicable to such systems is developed. Plate catalysts were prepared under various conditions by impregnating Pd into porous layer of anodized aluminum plate and activity and selectivity of each catalyst were examined by a labo-scale reactor.
Those catalysts prepared by directly impregnating Pd using aqueous solution formed a large amount of dimethyl ether as by-product. Such inferior selectivity is supposed to originate in the acid sites of Al₂O₃, and therefore silica coating, addition of alkali metals on anodized layer, and/or impregnation of Pd by acetone solution are tested to decrease the formation of by-products by reducing the acidity of the support layer.
Test results show that the catalysts prepared by these procedures have higher selectivity and Pd impregnation by acetone solution, with Na addition into the support layer combined, is most effective to improve the selectivity.
Further, it is found that the activity of the catalyst is dependent on the Pd surface area, and turn over frequency (TOF) is scarcely affected by Pd particle size.

View full abstract

As a new technology for effective utilization of electric energy, a chemical heat pump unit for night-electric heat-storage and cold/hot heat-recovering in daytime air-conditioning/hot-water-supply is proposed and a prototype is made. The basic heat-releasing characteristics and the effects of reactant particle amount and heat recovering condition on the characteristics are studied experimentally.
It is found that the electric energy stored as densely as 1.86 MJ/kg-CaO in the form of thermochemical energy is released in the form of thermal energy as cold heat at around 273 K and hot heat at around 500 K. The heat is recovered as cold heat for air-conditioning at around 275 K and hot-water at around 350 K. The total amount of reactant particles should be determined with a balance of heat-loss, although heat releasing is improved by using a shallow bed with a small amount of particles. Furthermore, the recovered heat amount is found to be enlarged because of increased reactivity of reactant particles when the heat-recovering rate is properly controlled.

View full abstract

A chemical heat pipe system using the endothermic heat of mixing of solid-liquid (s-l) or liquidliquid (l-l) mixable materials seems very attractive for cold heat generation and accordingly, evalua-tion of the most suitable mixing-pairs in terms of cold heat capacity is carried out. In s-l pairs, urea/ water and ammonium nitrate/water are found to generate large amount of endothermic heat, about 8.16 kW·kg^-1, as soon as the materials are mixed. Similarly, irrespective of the initial temperature, isobutanol/acetonitrile and 1-butanol/acetonitrile are found to produce cold heat of about 0.841 kW-kg^-1. For regeneration, drying and distillation processes are reviewed for s-l and l-l systems, respectively, and it is found that heat at 353 K can be used for regeneration. As a result, the authors propose a heat pipe system using low level waste heat energy for cold heat generation based on the endothermic heat of mixing.

View full abstract

A liquid-phase methanol synthesis process, which combines two-step reaction involving methyl formate is studied to recover wasted or unused discharged heat from industrial sources for the thermal energy demands of residential and commercial areas by chemical reactions. To evaluate the effect of development condition of a plate-type Raney copper catalyst on the over-all chemical reaction rate with transport process within catalyst involved, the chemical reaction rate of methanol synthesis by hydrogenolysis of methyl formate is measured using the catalyst in an autoclave reactor. The reaction rate is obtained by measuring the decrease in pressure due to chemical reaction. We focus on the effect of Raney copper catalyst thickness on the liquid-phase chemical reaction by varying the development time of the catalyst. Investigation results of the catalyst such as surface area, pore radius, lattice size, and photographs of scanning electron microscope (SEM) are also given. It is observed that the over-all reaction rate increases with catalyst thickness for thickness under 56μm due to enhancement of activation of the catalyst, then decreases with it for higher thickness over 56 μm due to the decrease of activation of the catalyst and the effect of diffusion process within the catalyst. It is considered that the optimum of activation arises from the proper size, and the regular shape of catalyst particles.

View full abstract

The paraldehyde chemical heat pump is combined with a chemical reaction and a vapor-compression heat pump. The paraldehyde chemical heat pump generates cold thermal energy with depolymerization of paraldehyde (2, 4, 6-trimethyl-trioxane). The cooling rate of the paraldehyde chemical heat pump depends on the catalyst activity of the paraldehyde depolymerization. This study investigates the effect of the acid strength of the solid catalyst on the paraldehyde depolymerization rate, the cooling rate, and the coefficient of performance of the heat pump. The acid strength distribution of Amberlyst 15 E was measured for estimation of the catalyst activity of the paraldehyde depolymerization. The depolymerization rate was measured with Amberlyst 15 E of modified acid strength distribution. The high activity range of the acid strength is-7.1 <pKa<-4.3 for the paraldehyde depolymerization with Amberlyst 15 E. The coefficient of performance, the cooling rate and the catalyst weight were estimated with the acid strength. The weight of the catalyst can be reduced to 0.05 g with 1.0 g·s^-1 of the flow rate in the system when the cooling rate and COP are 0.9 kW and 6, respectively.

View full abstract

Alcohols are important solvents with a variety of applications. This paper describes cost estimations of pervaporation (PV) and vapor permeation (VP) membrane separation systems for ethanol-water mixtures. Experiments using a homogeneous cellulose membrane are performed at higher ethanol feed concentration (above 90 wt%), since conventional distillation is economical at lower concentration. The fluxes observed are smaller, and the selectivities are larger in VP compared to PV. However, the differences in PV and VP decrease with increasing ethanol feed concentration.
Simulated calculations are carried out for the estimation of membrane area under three operational conditions for a commercial-scale plant, based on these experimental data. As a result, it has been clarified that membrane area steeply increases at higher concentrations. The low flux involveddemands a larger membrane area in VP compared to PV. A conceptual design of the membrane separationsystem is presented, and the costs of membrane module and other main components are estimated.
As a result of this economic assessment, it is clarified that capital cost is significantly affected by themembrane area, and that cost is lower in PV compared to VP over a wide scale of plants. But contradictory results are obtained from other researchers' data using the GFT membrane. The economics of membrane separation systems. strongly depend on separation characteristics of the membrane and the range of concentration.
Next, the authors discuss the selection of an applicable membrane from various PV data showing “trade-off” relations between α and flux. It is clarified that high flux is a very importantcondition, rather than high selectivity for industrial applications.

View full abstract

The performance of artificial neural networks on building process models for estimating the quality of petroleum products, research octane number of gasoline, boiling point of gas oil of a topping unit, and flash point of bottom product of a naphtha splitter, are examined in this study. Three types of artificial neural networks models are developed in this study ; back propagation neural network, radial basis function neural network and Wave-net. It is shown that radial basis function neural network model and back propagation neural network model are superior to the other neural networks models on building a steady state model. Wave-net is useful in building a dynamic model for time series data.

View full abstract

For the purpose of practical application of the adsorptive desiccant cooling system using a honeycomb rotor dehumidifier, a 20 kw field-test apparatus has been carried out. Effects of operating conditions and ambient air conditions on performance of the adsorptive desiccant cooling system are investigated, and the following results are obtained ; 1) The optimum air temperature T₈ for regeneration of the adsorbent rotor is found to maximize the thermal coefficient of performance. When the temperature and absolute humidity of ambient air are T₁=28.432.3°C, X₁=8.811.7 g/kg, the optimum air temperature for regeneration is around T₈=6585°C. 2) The cooling capacity is found to decrease with increasing absolute humidity of ambient air. 3) At regeneration air temperature above 100°C, the thermal coefficient of performance is improved significantly when the flow rate ratio of regeneration air to supply air through the dehumidifier is selected as one-half. 4) The cooling performance is estimated to be improved by using a water spray heat exchanger instead of the conventional evaporative cooler.

View full abstract

The mixing and segregation characteristics of model wastes such as a wood cylinder (ρ_s=649 kg/ m³) and an acrylic sphere (ρ_s=1, 120 kg/m³) in a fluidized bed with an inclined distributor are investigated. In this experiment, the plenum chamber was divided into 12 sections so that the an flow rate distribution can be arbitrarily adjusted, and that particle circulation can be induced. After throwing a batch of model wastes, fluidization was kept for a given operation time. Then, the distribution of model wastes in the bed was investigated. Under the present experimental conditions, the model wastes were dispersed into the bed most evenly when the ratio of the higher superficial gas velocity to the lower u_0s/ u_0w was 2 and u_0w=u_mf.

View full abstract

Characteristics of power consumption in an agitated vessel with an “EGSTAR” agitator, which has been practically used as a bioreactor, are clarified experimentally. The power consumption at nonbaffled condition was well correlated with the friction factor and the modified Reynolds number, in a similar manner to Kamei et al. (1995) for a paddle impeller. In this work, the parameter γn_p^0.7b/H for the paddle impeller is modified to 1.57γ (d/H) (h_d/ d) ^0.59 for the EGSTAR agitator. The power consumption under baffled conditions was also correlated well with the correlation method, similar to Kamei's method. The fully baffled condition and the corresponding power number for EGSTAR agitator are presented by the following equations.
(BW/D) n_b^0.8≥0.74 (h_d/d) ^0.12
N_{Pmax, ε}=3.7 (h_d/d) ^0.59 (1-ε) ^-1.1

View full abstract

A continuation of our previous study, the goal of this study is to express the dispersion conditions of particles in a dispersed material as a fixed quantity, and to clarify their impact on thermal diffusivity.
The test method used is the same as that for the previous report. For the study, the thermaldiffusivity of a bulk sample was measure, and a test piece was taken from the sample. A microphotograph was taken of the test piece, which was put through an image analyzer. Using fractal dimensions F, we evaluate the dispersion conditions of the particles, and then investigated the relationship between the dispersion conditions and thermal diffusivity.
Fractal dimensions F proved to be an effective parameter for expressing the dispersion conditions of the particle. As for the effects on thermal diffusivity, while changes in fractal dimensions F are a matter of course, the effect on thermal diffusivity is also large, depending on whether the ratio of the thermal diffusivity of the base material α_c and the thermal diffusivity of the dispersed particles α_d (α_d^*=α_c/α_d) is greater or less than 1.0.

View full abstract

The shearing and friction characteristics of a powder bed consisting of porous particles under conditions of various liquid content have been evaluated by means of the apparent viscosity coefficient η_a. In the present case, Rumpf's model has been applied to estimate the shearing force H at the contact point between two particles measured by the intrusion method. According to the tendency to change inη_aand H with the ratio of liquid to powder weight w, the states of the powder can be classified into thethree regions. In the region I (0≤w≤0.89) η_a and H have similar values in the dry state, respectively, i. e. no change in surface condition due to absorption of the liquid into the pore spaces of the particles. In the region II (0.89≤w≤1.50), the powder property changes from the pendular to the funicular state, so η_a and H increase with increasing w, and H does not change in the void fraction ε. In the region III (w≥1.50), the powders are in the capillary states, η_a and H decrease with increasing w, and H increases with decreasing ε.

View full abstract

Lipase from Candida cylindracea was immobilized in the gel beads prepared by copolymerization of N-isopropylacrylamide, N, N′-methylenebisacrylamide, and acrylamide. The optimal conditions of immobilization of lipase, for example molar ratio of monomer, cross-linking reagent, polymerization initiator, and stirring speed, are investigated. The esterification reaction between oleic acid and glycerol was carried out at 37°C by using the immobilized lipase or free lipase. It is found that selectivity of monoolein synthesis is higher when using immobilized lipase in gels compared with free lipase. The gel beads undergo thermal phase transition. The gel swells below lower critical solution temperature (LCST) and shrinks above LCST. An esterification reaction by lipase immobilized in gel was carried out by swinging the temperature between 37°C (above LCST) and 5°C (below LCST). It is found that the mass transfer rate in the gel is enhanced by phase transition.

View full abstract

Twomey's non-linear iteration method is applied to retrieve a particle size distribution from the sedimentation curve measured with a sedimentation balance method. The optimum power, p of weighting function and capability of shortening the measurement time are investigated numerically based on the log-normal and bimodal distributions. It is found that the appropriate value of power, p is 0.72 in most cases. Furthermore it is clarified that the non-linear iteration method can predict size distributions to a certain extent in the case of shortening a measurement time. In order to predict the log-normal distribution accurately, a measurement time longer than the settling time of the mass median diameter of tested powder is required.
Based on these results, the size distributions were measured by a sedimentation balance for fly ash and glass beads. As a result, we can predict size distributions of these powders accurately after less than 30 minutes measurement.

View full abstract

In order to develop a life cycle assessment method, which can be widely applied to electrical appliances and office automation eguipment we have investigated resource depletion, energy consumption, and environmental pollution in the manufacturing process of circuit boards, and we have identified the common units, such as 228 Mcal/kg of energy consumption and 15.979 kg-C/kg of carbon dioxide emission. Using these common units, we have calculated the categorized impact for the environment (resource depletion, global warming, acid rain, and atmospheric pollution) using a laptop personal computer, whose processing speed (PS) is improved by 5 times, weight is decreased by 47%, and energy-consumption is cut by 44%, compared with previous type. We have also shown that the life cycle environmental impact (EI) per personal computer decreases by 38% in resource depletion, life cycle EI per processing speed decreases 87% in all categories, and environmental efficiency (PS/EI) in all categories improves by a factor of approximately eight (factor value:8).

View full abstract

Precipitation of calcite is studied by agitated mixed bed of calcium carbonate intermediate and glass beads in a batch crystallizer. In test results obtained by addition of glass beads of 1.0×10⁵ particles, a calcium carbonate intermediate was transformed to the mixture of calcite and vaterite by addition of small glass beads of 105 and 155 μm average diumeter, but calcite precipitated dominantly by addition of large glass beads of 390, 575 and 791 μm average diumeter. Precipitated number of calcite increases with the cube of the average size of glass beads. The energy generated by collision of glass beads is considered to promote the transformation from intermediate to calcite. Precipitated number of calcite increases with square of the cube root of the number of added glass beads, and was affected by agitation speed of an impeller. Equilibrium concentration of calcium carbonate intermediate was 3 × 10^-3 mol/ l and growth rate of calcite was 0.9 μm/min when enough amount of calcium carbonate intermediate cnexists.

View full abstract

For better understanding of particle mixing behavior in fluidized beds, the concentration of tracer particle was measured in a rectangular fluidized bed to examine horizontal and axial particle mixing characteristics in a fluidized bed. The experiments were performed both by taking images at the bed wall with a CCD camera and by analyzing the images with an image digitizer. Two kinds of lime stone particles were selected to examine the influence of particle size on the mixing. As a result, a large scale particle circulation developed in the bed largely affects on solid mixing in bed wall region. The mixing in the center region and bottom region of fluidized bed are governed by movements of bubbles.

View full abstract

A vertical cylinder maintained at room temperature is located right above a horizontal circular heated plane to constitute a narrow air space between the plane and the cylinder bottom surface. Natural convection heat transfer in the space is experimentally investigated.
Average heat transfer coefficients of the heated plane are presented with the variation of space distance and Rayleigh number, and are compared with the predictions of the correlation equations which have been proposed for the space between two infinite parallel plates.
Visualized flow patterns above the heated plane are also shown. The relation between the flow pattern and the heat transfer coefficient is discussed to clarify the mechanism of heat transfer in the narrow space. As a result, a heat transfer correlation is proposed, which is applicable to a wide range of space distances.

View full abstract

The pressure drop and heat transfer of solid-liquid two-phase flow, in which the ratio of diameter of solid particles to that of a pipe is large, were investigated experimentally. The solid particles used were spherical particles whose density is close to that of the liquid. Measurements were made on friction factor, heat transfer coefficient on the pipe wall, and on number density of particle distribution in thepipe. The result of visualization of flow indicates that the particles stay at an equilibrium position forsmaller Reynolds numbers and smaller concentration of particles. The equilibrium position of particles approaches closer to the pipe wall for larger flow rates, and closer to the pipe axis for smaller pipe diameters. The ratio of the friction factor of two-phase flow to that of Poiseuille flow increases gradually with increasing Reynolds number in the region of low Reynolds number, and increases steeply in the region of high Reynolds number. The ratio of friction factors increases with increasing fraction of the volumetric flow rate of the particles, and decreases with increasing ratio of diameter of the particles to that of the pipe. The ratio of heat transfer coefficient of two-phase flow to that of Poiseuille flow has same tendency as the friction factor.

View full abstract

The local gas holdup of carboxymethyl cellulose (CMC) solution-air system in the annular section of a double-tube bubble column is measured by a probe technique. The local gas holdup decreases with increasing CMC concentration. The empirical equation developed for a water-air system to estimate the maximum gas holdup in the annular section is modified by the relative viscosity of the solution to water. Compared with the experimental results, it is demonstrated that the modified correlation is appliable to a CMC solution-air system.

View full abstract

Model geothermal brines based on a component of dissolved salts in the actual geothermal brines were used, and the effect of dissolved salts on the silica removal performance by a seeding method using silica gel is experimentally examined.
Potassium and sodium ions increase the silica supersaturation degree by their salt out effects, and as a result promote silica removal. Calcium and ferric ions promote silica removal by production of their silicates in addition to the salt out effect. Boric acid has little effect on silica removal.

View full abstract

The effects of the height of packed bed in a downcomer on the liquid mixing time in a bubble column with a draft tube are experimentally investigated, while varying gas flow rate and packed bed height under conditions of two types of gas dispersion gas dispersed into the inner section of the draft tube, and into the annular section between the column and the draft tube.
In either type of gas dispersion, the liquid mixing time in the presence of the packed bed increases with increasing superficial gas velocity and decreases with increasing packed bed height. It shows that the degree of increase in liquid mixing times is proportional to the superficial gas velocity raised to thepower of 1/3 and the packed bed height raised to the power of 1/4, regardless of the type of gas dispersion.

View full abstract

Changes in water content and viscosity for A- and C-heavy oil during emulsion formation are experimentally examined. The A-heavy oil hardly forms a W/O emulsion. For C-heavy oil with sea water, the rate of emulsion formation and water content for the same shaking time are lower than those of C-heavy oil with distilled water. Regardless of the kind of water phase, the relationship between water content and viscosity of the emulsion of C-heavy oil is well expressed by the Mooney equation.

View full abstract

The gas-liquid mass transfer volumetric coefficient k_Lα for an EGSTAR agitator, which has been practically used as a bioreactor, is measured. The effect of geometries (i.e., hole fraction, pitched paddle width, draft tube height) on k_Lα is clarified experimentally. k_Lα depends only on the power consumption per unit volume (P_av+P_gv), regardless of the hole fraction and the pitched paddle width in the power region of (P_av+P_gv) =0.54.0 kW/m³. As the draft tube height of EGSTAR agitator is lower, the power consumption at the transition point from bubbling controlling to agitation controlling is lower.

View full abstract