Steam ejectors can have various kinds of performance characteristics, but when high vacuum for mixing chamber or high compression ratio for entrained fluid is required for their performances, as is usual with such a case, the mixture of driving steam and entrained fluid should be kept in supersonic and fulfilled flow state at the diffuser throat inlet. The characteristics of steam ejectors in this flow state are called high vacuum characteristics.
In this flow state, the entrainment performance of an ejector, as expressed by the relation between the pressure of mixing chamber and the entrained flow rate, is dctermined independently of the back pressure of the diffuser. Therefore, a method of calculating the entrainment performance has been proposed, taking into consideration the drag force of steam jet from the nozzle and the entrained fluid passage area as measured cross-sectionally at the convergent part of the diffuser.
This desirable flow state depends largely on the outside conditions of the ejector. When the back pressure of the diffuser is raised above the limit value-the critical back pressure-, the high vacuum characteristics cannot be observed and the vacuum of mixing chamber will suddenly decrease. For the compression performance of the diffuser, another calculation method has been proposed for the purpose of obtaining the critical back pressure, taking into consideration the mixing of driving steam and entrained fluid with the mean velocities of the mixture relative to flow rate, momentum and energy, though, in the previous calculation methods the mixing process was neglected.
From the calculation method presented in this paper, Eq. (24) has been obtained for the entrainment ratio which is the ratio of the flow rate of entrained fluid to that of fluid taken as a basis or at a base state. Comparison of Eq. (24) with the experimental results by Holton and Schulz^{3, 4)} has proved that the drag coefficient of steam jet χ' is independent of the molecular weight of entrained fluid, but has a tendency to increase with the temperature rise of entrained fluid. If the drag coefficient χ' and χ are kept constant, the following formula may hold which represents the fact that the flow rate of entrained fluid will be proportional to the square of dimension ratio, when ejectors are geometrically similar.

View full abstract

Observations on evaporation from liquid surface under a high and a medium vacuum have revealed the existence of two new surface, phenomena, the former of which we call icefield-torpid surface and the latter, concave-evaporating surface.
Measurements of the effect of vapor pressure and of the pressure of the inert gas on the evaporation rate have served to clarify the mechanism of mass transfer in an evaporating space under a high and a medium vacuum.
From the above the following conclusions may be drawn:
1. Molecular evaporation consists of two mechanisms, one of which is molecular-projective-evaporation, where the vapor molecules pass relatively freely through the inert gas, and the other is molecular-burst-evaporation, where the escaping vapor pushes the inert gas along before it.
2. Mass transfer in the evaporating space under a medium vacuum may be understood as a slipdiffusion as shown by Eq. (10), when the rate does not reach the maximum.

View full abstract

Separation process of falling-film molecular distillation has been clarified through the study of mass transfer in distilland film, and a theoretical solution for the case where volatilities of components greatly differ from each other has been derived. The variation of concentration of light component within liquid film is given by Eq. (3), and the distillation rate of both light and heavy components at any cycle, by Eqs. (6) and (8).
An experiment was conducted in the molecular distillation of sperm oil, using the apparatus shown in Fig. 1. On the supposition that this oil was of two groups of components, -wax and glyceride-, the results obtained by calculation, employing the above-mentioned Equations, were compared with the experimental results as shown in Table 2. It was proved that there was satisfactory agreement between the two. Based on this, it has been concluded that these Equations sufficiently explain the falling-film molecular distillation in binary system or two-group system.
By making use of the Equations obtained, some considerations have been given to the effect on the distillating capacity of the apparatus (diameter and length of the column) and operating conditions (flow rate, temperature and repeated cycle), and also to the determination of the appropriate condition when the molecular distillation of this type is employed.

View full abstract

Mechanism of vacuum drying (press. range: 4-300mmHg) was studied, using SHIGARAKI clay as testing material.
Drying by radiation heating was conducted, using apparatus A (Fig. 1). Data obtained are summarized in Table 1. Drying curves and characteristic curves of drying are shown in Figs. 2 and 3. When the water content was equal, the ratio of the decreasing drying rates for various drying conditions were found to be equal to the ratio of the constant drying rates. Thickness of the material was observed to have no influence on the decreasing drying rate (Fig. 4), therefore, drying time may be considered to be proportional to the thickness of the material. Water content gradients in the material were measured at various drying times (Fig. 5). Obviously, these curves were different from those under atmospheric pressure.
From these facts, we may conclude that the drying of clay in vacuum is typical of the drying caused by the capillary action.
Surface evaporation coefficient, k_g', defined by Eq. (3) was calculated and shown in Table 1 and Fig. 13.
Drying by conduction heating with air leakage was carried out, using apparatus B (Figs. 6 and 7). Data are summarized in Table 2.
Drying curves and characteristic curves of drying are shown in Figs. 10 and 11.
The air was leaked quantitatively to the vessel and the water vapor pressure in the vessel was measured and calculated precisely.
During the constant drying period, the surface evaporation coefficient, kg, defined by Eq. (5), was calculated, and it is shown in Table 2.
The relation between k_g and total pressure, π, was correlated for various drying conditions, as follows:
(6)
The values of k_g' could be expressed by Eq. (6) within the range of our experiment.
Thus, we can design the vacuum chamber dryer and decide the pump capacity, by using Eqs. (4), (5), and (6).

View full abstract