Experimental work was carried out on re-entrainment and pressure drop in an up-flow mesh mist eliminator.
Hold-up of collected mist was measured by using a water-seal technique. Surface tension of mist was in a range between 45 × 10^-3 and 77 × 10^-3 Newton/m. Volume fraction of the mesh was between 0.01 and 0.069. The experimental result indicated that performance of the mesh mist eliminator can be classified into three manifestations by using loading point and re-entrainment point. Collected mist was apt to coalesce and formed two different liquids. One of them was increased with fiber length per unit volume. The other was the loading liquid, which was freely held within the mesh.
Empirical equations were formulated for prediction of the re-entrainment point.
Pressure drop was assumed to consist of pressure drops due to the loading liquid, the fiber-attracted liquid with the loading liquid and the fiber-attracted liquid without the loading liquid.From the above consideration, equations for calculation of pressure drop were obtained.
Experimental results verified the assumption and equations were formulated to estimate pressure drop.

View full abstract

The mechanism of electrical dislodging of a dust layer is analyzed. Two models are proposed, one static, the other vibrating. Dust-dislodging force and voltage in each model are obtained as a function of variables such as electrode diameter, ratio of pitch to diameter of the electrode, distance from the electrode to the dust layer surface, and phase of applied electricity. Analytical solutions of the static model are in good agreement with experimental results. Six-phase electricity is more effective than 2- or 3-phase. The optimum diameter of electrode wire is obtained, and is found to increase with increase of distance from the electrode to the dust layer surface, and with decrease of the ratio of pitch to diameter of the electrode. The optimum value is estimated theoretically to be a few millimeters. On the other hand, it is noticed that the return of particles to the original surface becomes important when the frequency of applied electricity is higher.

View full abstract

Theoretical and experimental studies of the error in sampling particle concentration of stationary air were conducted. By determining critical trajectories of particles to sampling probe inlet by a photographic method, the experimental sampling efficiencies have been obtained. Theoretical results obtained under the assumption of viscous flow are as expected in good agreement with the experimental data. It is also found that numerical results of viscous flow agree with Levin's equation within 5 percent deviation when the gravitational parameter G is less than 0.05 and Levin's parameter k less than 0.03, and that Levin's equation is in good agreement with the point sink numerical results when Levin's parameter k is less than 0.5. The efficiency of downward sampling is higher than that of upward sampling over a certain range of the inertial parameter, but drops sharply in the larger inertial parameter region.

View full abstract

A purification model for decreasing benzoic acid content in naphthalene crystals is proposed on the assumption of diffusion of benzoic acid through holes in the crystal, and a correlative equation between elapsed time and amount of benzoic acid in naphthalene crystals is derived.
This equation is characterized with two constants, one corresponding to diffusion coefficients and the other related to the condition of crystallization.
Purification tests of crude naphthalene crystals were carried out by keeping the crystals at equilibrium conditions and the data were correlated by the proposed equation. The two characteristic constants in the equation were also obtained.
Rotary-drum purifying crystallizers were devised and their design was studied from the standpoint of the proposed equation.
Operating conditions of this type crystallizer are discussed relative to purity of product crystals.

View full abstract

A fluidized bed with a rotational disk was applied to the reduction of very fine Mt. Newman iron-ore particles of diameter smaller than 0.074 mm. Hydrogen was used as a reducing gas. The reduction temperature ranged from 490°C to 935°C. The rotational speed of the disk was varied from zero to 240 rpm.
By rotating the disk, agglomeration of particles was prevented completely even for such fine ore. Furthermore, the utilization factor for hydrogen, i. e., the ratio of consumption rate to feed rate of hydrogen, improved with increase of rotational speed, and this factor was found to be affected significantly by bed height.
Coarse particles of 1 to 2 mm diameter were also tested. For these particles, the utilization factor became very high as bed height was increased.

View full abstract

Simple relations between relative intensity of drying and critical moisture content are derived by using previous analytical method for a semi-infinite model.
Transient changes of weight (and average moisture content) of beds of wet sand or brick powder during the initial stage of drying were measured by an electronic transducer. Changes of surface moisture content were also measured simultaneously by an IR hygrometer.
Comparing the results of the experimental critical moisture content with the theoretical values, it is shown that the critical moisture content under high-intensity drying conditions can be estimated by the present model.

View full abstract

Flashing flows at high flow rates in a single-stage flash evaporator were observed through a front glass. Their temperature distributions and non-equilibrium temperature differences were simultaneously measured. The results clarified the following.
Flashing flows in a simple chamber were classified into two patterns, i. e. bubbly boiling and splash boiling. In this study, both patterns were observed at 40°C. At that temperature, bubbly boiling was apt to take place when the liquid level was increased and the liquid flow rate was decreased. When the temperature was higher than 50°C, splash boiling was always observed.
Non-equilibrium temperature difference was discontinuously changed by the transition of the flashing pattern. The measured data in bubbly boiling were 1.52.0 times that in splash boiling.
A chamber having a trapezoidal bottom was effective in reducing the temperature difference.

View full abstract

An experimental study was made of the periodic fluctuation of torque of the drive shaft in a horizontal stirred vessel with baffle plates. Experiments were performed with three similar vessels of different size.
On the torque wave measured, sharp peaks of torque appeared periodically. These torques were of great importance in the mechanical design of the axis and impeller. The wave pattern was determined by the Froude number of the liquid in the vessel. The magnitude of the maximum torque was also a function of the Froude number. The ratios of maximum value to average value were 2-6 for 2-blade impeller and under 2.5 for 4-blade impeller.

View full abstract

The authors measured the heat transfer coefficient at the wall of a mixing vessel (10 cm and 19 cmφ) with various types of impeller and baffle conditions. The distribution of local heat transfer coefficient along the vessel wall changes according to the shape and height of impeller.
The mean heat transfer coefficients are correlated with the term εD⁴/ν³, which is obtained from dimensional analysis and represents flow behavior rather than Reynolds number. Irrespective of non-baffled and baffled conditions, all data can be correlated by the following equation within an error of ±20%.
hD/λ=0.512 (εD⁴/ν³) ^0.227Pr^1/3 (d/D) _0.52 (b/D) _0.08
h : heat transfer coefficient, D : vessel diameter, λ : thermal conductivity, ν : kinematic viscosity Pr : Prandtl number, d : impeller diameter, b : impeller width

View full abstract

The boundary layer equations for a purely viscous non-Newtonian fluid in cross flow around a circular cylinder are integrated by a method similar to that which was given by Kármán-Pohlhausen. The approximate solutions of thermal boundary layer equations given by Dienemann are extended to a non-Newtonian fluid. Both power law model and Prandtl-Eyring model are employed as the rheological model of pseudoplastic fluid. The calculated results reveal the zero defect of power law model near the separating point and give the profile of local Nusselt number both for power law model and Prandtl-Eyring model.

View full abstract

The experimental results of heat and mass transfer for a circular cylinder in pseudoplastic fluid flow are presented. The results of pressure distribution and separating point measurements are also shown. Local heat transfer coefficient was measured in water solutions of carboxymethyl cellulose. Mass transfer experiments were performed by an electrochemical method in electrolytic solutions with carboxymethyl cellulose. It was found from the flow measurements that the pseudoplastic fluid flow around a circular cylinder was almost the same as the Newtonian fluid flow. The correlating equations of local heat transfer rate at the front stagnation point and over-all heat transfer rate were obtained both for power law model and Prandtl-Eyring model.

View full abstract

The transport of electrolyte in an aqueous solution composed of electrolyte and non-electrolyte was analyzed from the viewpoint of structure of solution.
A relationship between diffusion coefficient of electrolyte and radii of solute molecules was derived. Taking into account the fact that electrical conductivity is affected adversely by the increase of non-electrolyte concentration, the number of ions associated per unit volume of solution with non-electrolyte was assessed from the measurement of ionic conductivity.
An equation is presented to permit the estimation of diffusion coefficient of electrolyte in the aqueous solution, provided the radii of ions and non-electrolytic molecules are given.

View full abstract

The solubilities of oxygen in 11 aqueous electrolyte solutions were measured by a volumetric method at 1 atm. of oxygen partial pressure and at a temperature of 15, 25 or 35°C. The solubilities of oxygen in most electrolyte solutions were correlated by the Setschenow equation over the whole concentration range of the electrolytes. However, the equation was not applicable to some electrolytes.
The solubilities calculated by the van Krevelen and Hoftijzer method for all measured systems of oxygen and aqueous electrolyte solutions except NaNO₃ deviate considerably from the measured values. The deviation amounts to about 10% even at 2 mol/l ionic strength in solution.

View full abstract

Molecular diffusivities of carbon dioxide in aqueous solution of 15 electrolytes were measured by gas absorption method using a laminar liquid jet at 25°C and partially at 15 and 35°C.
In the concentration ranges of the measured electrolytes, a linear relationship between the diffusivity and the concentration of electrolyte was obtained. The experimental results at 25°C were correlated for most systems by

-d/dC_s (D_L/D_L0) =0.020+0.483 (-Δ_φ/φ₀)
with high accuracy.
Furthermore, the diffusivities were correlated with the liquid viscosities by
D_Lηⁿ=const.
where n=1 for 1-1 electrolytes and n=0.50 for the electrolytes containing bivalent ions.

View full abstract

The oxidation velocity of aqueous solution of sodium sulfite with a catalyst of cobaltous chloride by gaseous oxygen was measured by gas absorption in a wetted-wall column.
The concentration ranges of sodium sulfite and cobaltous chloride in the experiment were from 0.02 to 0.6 mol/l and 10^-7 to 10^-4 mol/l, respectively. The partial pressure of oxygen was 1 atm. and the temperature was 15, 25 or 35°C.
The following experimental results were obtained : The reaction mechanism changes with narrow concentration ranges of sodium sulfite. Since the effect of cobaltous ion concentration on the oxidation velocity changes with the different ranges of reaction mechanisms, the concentration of sodium sulfite at the boundary of two ranges changes with the concentration of catalyst.
Furthermore, an equation of the apparent reaction velocity was obtained.

View full abstract

Densities and viscosities of solutions were measured over the range from undersaturation to supersaturation.
Prior to the measurements, the solutions were filtered and kept higher than the saturation temperature. These treatments prevented nucleation and made measurements possible even in highly supersaturated states.
Empirical quadratic equations of density are derived from the experimental data as a function of a given temperature and the saturation temperature of the solution.
Neither densities nor viscosities of a given solution showed sudden change near the saturation temperature.
These results suggest that the structure of a supersaturated solution is similar to that of an undersaturated solution, as in the case of potassium chloride solution in the authors' previous paper.

View full abstract