In the design of commercial distilling columns, estimation of plate efficiency is a problem of considerable importance.
In order to solve this problem, the author defined the no. of stages (n') as Eq. (1), and the modified local efficiency (ELO) as plate efficiency at n'=1, and from these he derived a theoretical relation between them as Eq. (17).
As the relation -Eq. (3)- between local efficiency (EL) and NOG was obtained from WalterSherwood's study44) of NOG, it would be reasonable to correlate ELO directly with physical properties of binary mixtures in question.
Thus the author obtained an empirical equation for ELO as Eq. (25), by utilizing both his experimental data and those in literature.
To determine plate efficiency or X_n in practice, the author adopted the graphical solution as illustrated in Fig. (14), instead of employing Eq. (17) and a trial-and-error procedure.
As to the estimation of allowable vapor velocity in a bubble-cap plate column, the author developed Eq. (29), an empirical equation based on Peavy-Baker's data (35).
Of the correlation of the plate efficiency and allowable vapor velocity in plate columns, the observed values in literature and the calculated values by the present method showed fairly good agreements.

View full abstract

Studies were made on the rate of mass transfer to single spheres, with the aid of a diffusioncontrolled electrode reaction:
Microelectrodes used as anode were platinum spheres having 1.66mm and 2.74mm diameters respectively, each with PVC-coated platinum wire of 0.6mm diameter. The reference electrode used as cathode was a thin platinum plate of 50×120mm. The experimental apparatus employed was as shown in Fig. 1. To obtain uniform distribution of liquid velocity over the entire column cross-section, glass beads of 5mm diameter were packed to a height of 30mm, topped with 10 sheets of Saran screen of about 12 meshes. Below the packing and 8mm above the spherical electrode, a Saran screen of about 150 meshes was placed.
From the measurements of limiting current _{6), 8)} at various liquid flow rates, the mass transfer coefficients, kf, were calculated by the equation:
kf=i/nFAc (7)
The experimental results are shown in Table 1. It was found that neither the results thus obtained nor those by previous investigators_{3), 9), 10), 14), 16)} showed good agreements with the Ranz-Marshall's eq. (2)₁₄₎.
Thereupon, assuming that the empirical equation might be presented as:
Sh-2=kRe_pSc_q (8)
the authors determined the values of the constants k, p and q (Figs. 3, 4 and 5), obtaining an equation:
Sh=2.0+0.52Re^0.54Sc^0.35 (9)
The value of the exponent of the Schmidt group, 0.35, agrees well with 0.348 obtained by Frossling³⁾ from his experimental data.

View full abstract

In order to explore the film coefficient of a fluid adjacent to the inside wall of fixed beds, data were obtained on the rate of dissolution of β-naphthol, by passing water through the beds whose inside was coated with the same material.
Glass beads and sand were used as packings. The experimental range of the modified Reynolds number was from 1 to 2, 000, and the Schmidt number, from 1, 000 to 1, 700.
Table 1 presents the dimensions of the various solid particles and of tubes used in this work. Solution was analyzed by means of a spectrophotometer.
As the result, it was made clear that under the conditions employed in this investigation, the eddy diffusion in the radial direction had little effect on the coefficients of the wall-film transfer.
The experimental data and calculated results for typical runs are presented in Table 2.
Fig. 3 shows a summary of the wall-film coefficients, where Sh/Sc1/3 is plotted versus Rep.
In this plot the properties of the particles do not appear as a parameter. The following correlations are obtained from the lines drawn in Fig. 3.
Sh=0.60Sc^1/3Rep^1/2 Re_p<40
Sh=0.20Sc^1/3Re_p^0.8 Re_p>40

View full abstract

The enthalpy-composition diagram, Fig. 1, has been constructed, with a view to facilitating the computation of heat balance in the manufacture of common salt. Although the method of construction adopted is substantially the same as that for ordinary binary system, special attention has been paid to the following points, because the sea water system is composed of many components as de- noted in Table 1.
1. Expression of composition: From many experiments, it has been found that the relative proportion of various salts in the sea water is constant anywhere on earth. At any concentration stage, the salts in the sea water may be conveniently expressed in a mass as "total salts" S, which includes salts both in the solution phase and the solid phase. Therefore, the sea water system may be denoted as a binary system of S and H2O as shown in Fig. 1. The weight per cent of S has been calculated from the following equation:
Per cent of S=(weight of salts in soln.)+(weight of deposited salts)/(weight of solution)+(weight of deposited salts)×100
2. State of system: For the calculation of enthalpy, the state of system or the composition of the solution and deposit must be known. This state was estimated by M. Ishibashi⁵⁾ Fig. 2) and Y. Onodera¹¹⁾ from the experiments on isothermal concentration at 1 atm, and by J.D'Ans¹⁾, from the researches in the solution equilibrium of ocean salts.
3. Enthalpy: The enthalpy of the system is given as follows:
H=H_l+H_s
H_s was obtained from the enthalpy of each deposited component.
H_l was approximately evaluated by the following equation:
H_l=n₁H₁+n₂H₂+……
where ni is a number of moles of component i in the brine and H_i, the enthalpy of i th single salt solution at the same molality as the total molality of the brine.
4. Lines of isobaric boiling point: The lines have been derived from the experiments by S. Uchida^{14), 15)}.
From the above calculations and data, Fig. 1 has been obtained. Although not free from faults, this diagram, aided by the pressure table for saturated steam, would meet the purpose of making easy the calculation of heat balance in the concentration process.

View full abstract

In the previous paper³⁾, we reported that the rate of extraction in liquid-liquid system, where a rapid chemical reaction was taking place, might be given by Eqs. (1) and (2). We thought perhaps this would also hold true in the case of chemical gas absorption, due to the concentration of the substance reacting with the solute in water layer, q.
But, to chemical gas absorption, these equations were said to be inapplicable. For instance, Sherwood and Pigford⁶⁾ compared the rate of absorption of NH₃ in water with the rate of chemical absorption of CO₂ in NaOH solutions above 1N, where KB was constant, (and this KB, according to Eq. (2), should be kB) and found that the former was much larger than the latter, so that the theory based on the infinitely rapid reaction failed to explain these results.
Considering these points, Eqs. (1) and (2) were re-investigated concerning liquid-lipuid extraction, using a vessel whose contents were gently stirred just as in our previous experiment. Results obtained may be summarized as follows:
1) Chemical extraction rates are constant, in the range above some critical value of q, for the systems of NH₃ (benzene)-H₂SO₄ (water), some organic amines (benzene)-H₂SO₄ (water), organic acids (b.)-NaOH (w.) or KOH (w.), and I₂ (b.)-Na₂S₂O₃ (w.) as shown in Table 1.
2) In the physical extraction, where K_B=k_B, k_B in the solute may be determined by physical extraction, when the distribution coefflcient m is very large with formic acid, monochloroacetic acid or ammonia in benzene-water system, as shown in Fig. 1. kB obtained from chemical extraction and from Eq. (2) were plotted versus D, and these coincided approximately on the same line of k_B^∝ D^2/3-0.7, as physical extraction for the system of large m (Fig. 1).
3) Gordon and Sherwood⁴⁾ suggested the method of determining the film coefflcients by plotting the left-hand side of Eq. (5) versus m, or the left-hand side of Eq. (4) versus l/m. By this method k_B of butyric acid (for instance) was determined at 4.17 and 4.12 from Fig. s. 2-3, and 4, respectively, and these values agreed with the kB of chemical extraction in Table 1.

View full abstract

Studies were made on the characteristic operation of an extraction column for pulsation flow, by extracting phenol with benzene from ammonia liquor, a byproduct obtained from the processing of coal into coke. The extraction column with an inside diameter of 8.0cm and a 1-2m packed height was filled with 1 inch saddle packing and 1/2-1 inch crushed coke. The operating condition was as follows:
Frequency of pulsation: 100, 150, 200 (c.p.m.)
Linear amplitude of pulsation in the column:1.5, 3.0, 4.5×10^-3(m)
The results obtained may be summarized like the following:
(1) Overall capacity factor increased with the pulsation intensity in a region where (amff²)< 2×10⁵[m/hrhr²], and at the optimum pulsation condition, the fractional increment of overall capacity factor of the pulsation condition to overall capacity factor of the steady condition was about 10.
In this region, the fractional increment of (K_Ra)p to (K_Ra)s was approximately expressed by:
(K_Ra)_p/(K_Ra)_s=c(a_mff²)ⁿ
where c=1.3×10^-3-3.0×10^-3 & n=0.6-0.65
(2) When the column height was in the vicinity of 0.8-2m, the effect of column height on (H.T.U.)OR was negligibly small, and the value of (H.T.U.)OR obtained from the experimentaldata was applicable to a large scale equipment, as well.
(3) Under the same flow and pulsation conditions, (H.T.U.)OR increased with the increment of fractional voidage of packing.
(4) Calculation of the flow condition at the flooding point was made possible by employing an approximate equation obtained from the experimental data.

View full abstract