The rate of oxygen absorption by aqueous sodium sulfite solution containing cupric ion as a catalyst was measured, using a gas-liquid fluidized bed as shown in Fig. 1 and Table 1.
The results obtained are as follows:
1) Flow patterns can be roughly classified into three kinds according to the velocity of the gas flow:-(a) the streamline region, where the gas holdup, (1-ψ), increases linearly with the increase in velocity of the gas flow;(b) the foam bed or transition region, where (1-ψ) becomes maximum; and (c) the turbulent region, where (1-ψ) is nearly constant.
2) The gas holdup is a function of the velocity of gas flow, hole diameter and clear liquid height (L₀) when L₀ is smaller than 30cm, but it is nearly independent of the clear liquid height when L0 is larger than 30cm.
3) Pressure drop required for generating gas bubbles on the plate is expressed by Eq.(2)(Cf. Fig. 3).4) The rate of oxygen absorption by aqueous sodium sulfite solution is independent of cupric ion concentration, and is apparently of zero-order, with respect to the sulfite ion concentration under the experimental conditions.
5) The value of KGa' is closely related to the value of (1-ψ)/ψ, and is proportional to L₀-^1/3 when L₀ is smaller than 30cm (Cf. Fig. 6).
6) In the streamline region, KGa is a function of uG and d, and is independent of n and L0, but in the foam bed region, it is influenced by L₀ (Cf. Figs. 7-10).
7) The experimental equation (Eq.(12)) has been obtained for calculating the value of KG in the temperature range from 10 to 22°C (Cf. Figs. 5 and 11).
8) The rate of oxygen absorption is controlled by the liquid-side resistance.
9) In the foam bed and the turbulent regions, the mean contact area per unit volume of bed, a, as calculated from Eqs.(10) and (12) is found to be in the region between 600-1000m²/m³.

View full abstract

In order to control the operations of cooling crystallizers, it is desirable to know the relation between the rate of crystallization and that of cooling. In a batch operation, the relation between the change in concentration of the solution and the lowering of the temperature is given as follows:
where a dimensionless parameter, α, is calculated by
It is obvious that if the value of α is equal to (t_s-t)/(t-t_w1), the rate of crystallization becomes very great due to the infinite value of dX/dt, and if the value of a is equal to zero, the temperature of the solution is raised because of the negative value of dX/dt, even when the solution is subjected to cooling. The value of α, therefore, is a controlling factor for the process, but, if desired, it may be put under control to some extent by varying the cooling rate. A similar relation is obtained for continuous operations of a column-type crystallizer with a cooling jacket.
Based on the relations thus derived, a graphical method is presented to determine the change in concentration of the solution with the temperature, both for the batch and continuous operations.The results of the calculations for batch operations are given in comparison with the experimental data.

View full abstract

In dispersing one viscous liquid into another immiscible liquid, attempts were made to obtain strict droplet-size distributions by using a nozzle-type unit or a nozzle-diffuser type one, as shown in Figs. 1 and 2, respectively.
In the experiment with the nozzle-type unit, it was found that the average droplet diameter undergoes a change of four stages in accordance with the velocity of the flow of the syrup through the nozzle as shown in Fig. 4. Small droplets with a narrow size distribution were obtained in the third stage. In such a case, half of the total droplets could be brought to be of the sizes between 80 and 120 mesh.
Based on the results of the experiment with the nozzle-diffuser type unit, the authors succeeded in determining the dimension of a diffuser, most suitable for dispersing a syrup of 43c.p., as shown in Fig. 13. The degree of size uniformity obtained with this unit was so high that about 70% of the total droplets could be made to have the sizes between 80 and 120mesh. The empirical equation obtained for the average droplet diameter is as follows:
d_M=3.2(ν₀^1/3/V₂)
The chief feature of this device is that any droplet size between 60 and 200mesh can be obtained by adjusting the velocity of the flow of dispersant, independent of nozzle diameters.

View full abstract