In order to analyse the solid-gas reaction in a moving bed, we must know the rate of reaction distribution of particle elements against time under unsteady state, or the distribution of elements in the axis direction under steady state. However, it is essentially difficult to obtain direct readings by inserting some kind of measuring element or sampling pipe directly into the bed because the particles in a moving bed are always in motion. The purpose of this paper is to study an expedient method to overcome this disadvantage.
Let us suppose a moving bed in which the lower zone has been differentiated from the upper zone by coloured particles. When this moving bed is operated and the particles are moved down and discharged through the turning table, the two zones will become mixed with each other. Plotting the mixing ratio of coloured particles according to the order of discharging in weight, we were able to obtain a mixing ratio curve S (x).
Experiments have assured this mixing curve to be inherent in the dimension of moving bed and property of particles (Fig. 3).
At the same time, let us suppose that we obtained the reaction rate curve h(xi) according to the order ofdischarging when solid gas reaction is effected in the same dimension moving bed and with the same particles. Then, making use of this S(x), the reaction rate curve in the bed g(x) would be expressed Eq.(1).
We have described two methods to calculate this Eq.(1). First was to develop the integral numerically, and reduce it to a set of simultaneous linear algebric Eq.(7). Second was to give g(x), h(xi) and S(x) some approximate mathematical equations, and decide the characteristics constants of these equations.
Futhermore, we have shown an example in which this method was applied to a pilot plant data (Fig. 4).

View full abstract

This work was undertaken to investigate the effect of gas chamber volume, V_c′. and number of holes, m, on the size of bubbles formed from perforated plates. The experimental apparatus used is shown in Figs. 1, 2 and Table 1.
The experimental results have been represented by using geometric mean diameter d and standard deviation S. because the bubbles are almost in conformity with logarithmic probability distribution, as shown graphically in Fig. 3. Replacing V_o with V_c′/m, which is gas chamber volume per unit hole, d of gas bubble from perforated plates coincides with the experimental equations (3)-(6) which is obtained for single orifices. Standard deviation S is greater than that of single orifices to some extent and shown with a solid line in Fig. 5 independent of m and V_o′.

View full abstract

Physical properties of ternary system acetic acid-water-butyl acetate were determined experimentally. Solubility and tie-line equilibria, density and refractive index on the biaodal curve and iso-refractive index line in the homogeneous region were observed at 30°C. Solubility and tie-line equilibria at boiling point were measured by the new method whose principle is based on the fact that there is a sharp change in temperature drop when adding the third component to the original boiling binary system.
Several tie-line correlations were examined and it was concluded that the linear correlation represented by one straight line gives errors in smoothing equilibrium data especially for the region of smaller solute concentration. Plait point at 33°C was determined by Hands' method and found to be that the weight percent of acetic acid, water and butyl acetate are 38.23, 25.53 and 33.24 respectively. Middle points of tie-line on the refractive index or density concentration diagram at 30°C lies on the straight line which connects the plait point and the middle point of the binary conjugate line. In other words, it may be said that the law of rectilinear diameter exists in the correlation of tie-line concerned with refractive index or density.

View full abstract

Fractional crystallization has not been as widely used commercially as its potential value as a powerful separation tool justifies, because equipment has remained undeveloped and inefficient and distillation has usually been used. In exploring a new approach to the development of more efficient equipment, a crystallization purification column was developed which produces a high purity product by continuous countercurrent treatment of the crystal mass with some of the melted crystal product.
We investigated this countercurrent crystallization purification apparatus, and analysed the various phenomena in the purification processes: i. e. the relation between the falling rate of porous piston and the heat supplied at the bottom of the column, the effect of pressure on the purifying capacity of the equipment, temperature distribution and concentrations of products. The transfer processes of heat and mass in this column were analysed by a simple model for the crystallization process, and experimental results were compared with the theory. From these results the transfer coefficient and H.T.U. of this column were estimated.

View full abstract

Experiments were carried out on the rate of crystal growth in a seeded solution of CuSO₄·5H₂O within a metastable zone where crystal may grow without forming new nuclei.
The mass transfer coefficient k_G and the over-all growth coefficient K_G are experimentally determined by the dissolution and crystallization in a batch agitated vessel containing about 0.70 lit, of solution at a constant temperature of 20, 30 and 40°C, respectively.
The values of k_d and K_G obtained at each temperature are plotted against the Reynolds number (=ND_i²/ν) as shown in Figs. 1 and 2, and K_G (C-Cs) vs. (Ci-Cs) at 30°C is plotted in Fig. 3. From the latter plot the value of n is found to be unity, and Eq.(4) is reduced to Eq.(13). The former plot indicates that kd and K_G are approximately proportional to Re^1.7 for Re>4, 000-6, 000, where the particles of crystal are suspending. Therefore, 1/K_G and 1/k_d vs. 1/Re^1.7 are plotted to obtain the value of k_r as an intersect with the ordinate as shown in Fig. 4. k_r thus obtained at each temperature is expressed by Eq.(14) as shown in Fig. 5. Fig.
6 shows a comparison of the experimental and calculated values of k_d by plotting a value of (k_dd_p/D_L)(Sc) _L^-1/2 and a modified Reynolds number, Re₀, recommended by Oyam and Endoh (CA, 51, 2338 d).

View full abstract