The accurate measurement of air discharge at low Reynolds numbers is important in many cases for chemical engineering researches. The quadrant nozzle is recommended by Koennecke as one of the most suitable fluid meters for low Reynolds numbers, because the coefficient of discharge can be maintained constant with low Reynolds numbers.
He gave the coefficient of discharge for quadrant nozzles having optimum values of the ratios r/d and m as shown in Fig. 6, as a result of his extensive experiments. However, owing to the fact that he used a steel pipe 40mm in diameter and incompressible fluid such as water or oil, it is questionable whether the results obtained by him are valid for the measurement of air flow through a pipe other than 40mm in diameter, because the effect of pipe diameter on the discharge coefficient may not be negligible. Furthermore, at present, no information is available on the expansion factors for air through quadrant nozzles.
Hence, the authors made a series of experiments on the quadrant nozzles with 1-in, and 2-in. steel pipes.
First, the discharge coefficient for a certain quadrant nozzle, No.1-St, was determined directly yb substituting water for air flowing through the nozzle, and the expansion factor was evaluated indirectly from tests with three air nozzles as shown in Fig. 2.
Finally, others as shown in Table 1 were calibrated individually with No.1-St.
From the experimental results, it may be concluded that: -
i) The coefficient of discharge for quadrant nozzles increases with an increase of the ratios m and r/d, but appears to decrease when r/d exceeds a certain limiting value.
ii) The agreement of the discharge coefficient with Koennecke's data is to some extent satisfactory with both the quadrant nozzles, No.1-5 and No.2-0, but the coefficient, as pointed out by Jorrisen et al., can not be maintained constant with Reynolds numbers quite as low as those asserted to have been reached by Koennecke.
iii) The expansion factor for air, although under the influence of m, is almost independent of r/d.
iv) The size of pipe diameter, as observed by Koennecke, affects the coefficient of discharge more or less, and at the same time changes the expansion factor.
v) Values of the discharge coefficient and the expansion factor for these devices are between the values for standard orifices and those for standard nozzles under identical conditions.
vi) A deviation from the prescribed values of m and r/d, which is due to the manufacturing procedure, causes changes within 0.4% in the discharge coefficient.
Furthermore, another experiment was conducted as follows. Two quadrant nozzles were installed in such a manner that the flat surface of the nozzle plate might face upstream, unlike in normal installation, and the authors succeeded in determining, as the result, the values of the discharge coefficient and the expansion factor.

抄録全体を表示

A new continuous method for preparing aniline from nitrobenzene by Béchamp's reduction method has been developed, using a liquid fluidized bed of iron particles induced by upward flowing liquid reaction-mixtures. Outline of the experimental apparatus is shown in Fig. 1.
The rate data obtained from this small-scale continuous apparatus in Fig. 3, when compared with those by batch reduction, showed that the latter is somewhat lower than the former. Possible reasons are presented for this lowering.
It has turned out from this basic survey that aniline of 99 to 99.9% purity may be prepared in industrial scale without much difficulty, using this new continuous method. Quantitative formulation of the behavior of the reactor has been developed, including reactor design of multistage type in series

抄録全体を表示

Film-coefficient of boiling heat-transfer on a horizontal flat surface has been measured experimentally, paying particular attention to the control over the ebullition ability of the heating surface.
The concept of an “ideal heat-transfer surface” is introduced to satisfy geometrically similar relation of bubble-formation on a heating surface. This newly defined particular surface has the property which is characterized by the fact that on it the number of active sites for bubble-formation per unit surface area remains unchanged during continued boiling, regardless of the difference of the liquids and the change in heat-flux and other operating conditions.
Artificial ebulliton points (A. E. P.) as shown in Fig. 2 are machined orderly on the heating surface in order to realize this particular surface experimentally. Experiments show that the surface with the number of A. E. P. equal to or greater than 32 per cm² behaves as an ideal one.
From the data obtained by changing the number of A. E. P. from zero to 64cm², the magnitude of the film coefficient is observed to change noticeably with the change in the number of active sites, and the comparison is made under a constant temperature difference of heat transfers (Cf. Fig. 5). Also, the data obtained on the surface with a definite number of active sites of 32 per cm², which behaves ideally, show that h changes with Δt almost in the same exponent with ordinary surfaces, i.e. h∝Δt^2.8 (Cf. Fig. 7). Experimental results obtained with this surface are correlated as shown in Eq. (4), where the concept of “interface factor” α is introduced as a measure of ebulliton ability of heating surfaces.
Using this concept to correlate the data on ordinary surfaces at atmospheric pressure, boiling liquids are grouped statistically into three classes, i.e., water and aqueous solutions, mainly alcohols, and others as shown in Fig. 16.

抄録全体を表示

i) A rigorous solution of the concentration distribution in the horizontal type thermal diffusion cell is obtained by the Laplace transformation in the form of the series of error functions as given by Eq. (13).
ii) From Eq. (13) the working equation is derived as follows.
Y=-(dn/dξ)_ξ=1/2/n₀(1-n₀)P=2erfc(1/4√r)
iii) This equation gives the universal relation between Y and r, (Fig. 1), and it is shown that the limitation of the applicability of this equation is r<0.1.
iv) Utilizing this universal relation, several methods are now proposed so as to determine the diffusion coefficient and the Soret coefficient from the relatively short-time experiments.
a) If the inflexion points (y_i, t_i) on the y vs. t curve are indicated, D and δ directly and independently calculated from Eqs. (19) and (20)
b) In such a case where the points measured in the neighborhood of the inflexion point are approximated by a straight line and the inflexion point is not evident, the inclination of this approximate line (dy/dt)_i and the intercept (t_p) can be obtained (Fig. 2). Using these values, D and σ can be computed from Eqs. (21) and (22).
c) When the two points, whose tangents have the equal inclination (t₁, t₂) are known. D can be calculated from Eq. (23).
The values estimated by the above mentioned methods are plotted in Figs. 3 and 4. From these results, the authors have concluded that these methods are useful for estimating D and δ.
v) The concentration changes in the chambers of the two-chamber type cell are solved and n_t and n_b are given separately. The solution by Jones and Furry gives only the difference between n_t and n_b.

抄録全体を表示

The diffusions of CO₂, C₂H₄ and C₃H₆ through the pores of active carbon were studied respectively, and the effective diffusivities of these gases were measured by the constant-volume method at 0, 30 and 50°C.
Experimental
In the vessel A shown in Fig. 3, was packed 1.8gr (about 20 pieces) of the granular cylindrical active carbon (4mmφ and 10mm 1g) whose cylindrical surface was coated with epoxide resine.
The indicial response curves of adsorption were measured for the step-changes of gas pressure in the range between 0 and 700mm Hg at 0, 30 and 50°C.
In this experiments adsorption took place only on the two parallel surfaces of the cylindrical active carbon, and the diffusion equation, Eq. (1), based upon Fick's law was applicable. A solution of Eq. (1) is shown in Eq. (3) by Wilson. The relationship between the change of pressure and adsorbed amount in adsorption system is given by Eq. (2).
The effective diffusivities of these gases were obtained from the indicial response curves of pressure change in adsorption and Eqs. (1), (2) and (3).
Results Obtained and Discussion
1) The effective diffusivities of these gases were big in the order of C₃H₆, C₂H₄ and CO₂, and they varied widely with temperature and adsorbed amount as shown in Fig. 5.
2) The Knudsen diffusivities calculated from Eq. (11), by using the mean pore radius determined by Eq. (10), were not in agreement with the effective diffusivities observed.
3) The diffusivities obtained from Eq. (12)', by using the life time of adsorbed molecule determined by Eq. (13), were in good agreement with the effective diffusivities observed.

抄録全体を表示

Mean mass transfer coefficients for naturally developed compressible turbulent boundary layer were obtained by measuring, in 2.0cmφ pipe of blow down pressure wind tunnel, the rate of sublimation of camphor to CO₂ stream through 0.6cmφ camphor cylinder. Experimental flowing ranges were Mach number of 0.55≤M_∞≤0.93 and Reynolds number based on the distance from the leading edge, 2.91×10⁶≤Re_∞≤6.33×10⁶.
Present data were correlated as a graph of Sh or J_D vs. Re in Figs. 4 and 5 for properties of fluid CO₂ based on the free stream temperature and the effective temperature, respectively. In Fig. 5, present data based on the effective temperature give values of J-factor, which show rather good agreement with C_F/2 line corresponding to Blasius' equation but are 4060 percent lower than the values obtained from Prandtl-Schlichting's equation.
Eq. (19) which has been derived by the approximate analysis in which effective temperature method is used shows the effect of Mach number and temperature ratio, T_ω/T_∞ on mass transfer rate.

抄録全体を表示

Radial heat transfer rates in the fixed bed have been obtained by several investigators from the radial temperature traverses measured. Results are reported in terms of effective thermal conductivities and wall heat-transfer coefficients of the fixed bed. These values can be divided into two groups, one independent of fluid flow and the other dependent on the lateral mixing of the fluid in the bed interstices. The effective thermal conductivity caused by the lateral mixing of fluid has been expressed by Eq. (14), which is derived from the concept of a “random walk” theory, and the general correlation of effective thermal conductivity in the fixed bed has been obtained in the form of Eq. (16). Further, the authors have carried out the experiments of heat transfer in the bed of solid particles mixed by means of a paddle-typed impeller and have measured the values of effective thermal conductivities and wall heat-transfer coefficients of four kinds of powders, such as sand, coke, silica-gel, and cement clinker. It has been found that the heat transfer mechanisms in the bed of solid particles mixed are asociated with two processes operating in parallel, one of which is the heat transfer in the quiescent bed of solid particles, taking place in the same manner as in the fixed bed with motionless fluid and the other the heat transfer due to the mixing of solid particles corresponding to the heat transfer caused by lateral mixing of fluid flow in the fixed bed.
Accordingly, by following the general correlation of effective thermal conductivity in the fixed bed and by introducing the apparent diffusivity D_s of solid particles mixed, in place of the effective diffusivity (E)_td in the fixed bed, the correlation of effective thermal conductivity in the bed of solid particles mixed has been obtained in the form of Eq. (20). The correlation of wall heat-transfer coefficient in the bed of solid particles mixed has been expressed by Eq. (25) and a new expression for estimating the wall heat-transfer coefficient in the quiescent bed of solid particles has been presented as Eq. (23).
Applying these correlations to the experimental values of effective thermal conductivities and wall heat-transfer coefficients, it has been found that these correlations are in good agreement with the experimental values.
The authors have ascertained that the results obtained in both the fixed bed and the bed of solid particles mixed can be handled with the help of the same concept of a “random walk” theory.

抄録全体を表示