Natural circulation of air in a loop cylinder induced by heating a part of one of the vertical arms was investigated.
The method of analysis was the same used previously for the natural convection of air in vertical cylinders. From the energy equation of fluid flow combined with the continuity equation, the temperature distribution in the loop was obtained as Eqs.(5) and (6). These functions were applied to the momentum equation and were integrated over the total volume of the loop. The result is given by Eq.(13) or, after a few approximations, by Eq.(16), by which the rate of the circulation, or the Reynolds number, can be estimated from the wall temperature and the geometric conditions of the loop.
Fig. 1 illustrates the structure of the loops and Table 1 gives the conditions and the main results of experiments. In Fig. 2, the mean reduced temperature of air at the upstream end of the heating section is given as a fuction of a for the conditions of L/r₀=191 and l/r_o=10.4. Thus, under the conditions shown in Table 1, the temperature of air at the inlet of the heating section can be assumed to be very close to the room temperature.
Fig. 3 illustrates the relations of aα vs. T_w-T_oo, Figs. 4 and 5 illustrate the wall temperature of the heating section required to get a given value of a when the length of the horizontal arms is varied, and Fig. 6 illustrates the effect of the height of the heating section on the values of Good coincidences were observed between the experimental and the estimated values.
The effect of the curvature of the bends on the natural circulation was discussed under the view of White.

View full abstract

The authors carried out experiments similar to the work of Tin and McCabe. Tin and McCabe found two groups of supersolubility curves for MgSO₄·E7H₂0 in continuously cooled, stirred, and seeded solutions. The location of these curves wa shown as dependent on several factors, but no attempt was made to derive any quantitative relations among them.
In our experiments, unseeded and seeded solutions of copper sulfate were allowed to cool in a batch agitating vessel with cooling jacket at constantstirring speed. The flow rate of cooling water was kept constant, without contrriling the cooling rate.
The ranges of experimental variables were as follows:
Stirring speed in r. p. m. 140-780
Rate of cooling water (w) in g/sec 10-80
g. of seed crystals (CuSO₄·5H₂0) per 100g. soln.(S) 0-10
Initial sat'd. concn. of anhydrate in wt. per cent 22-32
Corresponding initial sat'd. temperature in °C 38-70
Observations were made on the point where the first nucleus appeared and the point where sudden crystallization took place, while simultaneously, the relations of time with temperature and concentration were determined.
The point where temperature and concentration producedthe first appearance of crystals was named“first supersaturation point” and the point wheresudden crystallization occurred was called“second supersaturation point.”
The temperature and concentration values at these points were determined from an analysis of the curves and the results of observation obtained under experimental conditions. The first supersaturation point thus obtained was plotted in Fig. 4, and the second supersaturation point in Figs. 2 to 3e.
Fig. 4 shows an unique supersolubility curve which is obviously independent of the variables tested here. This fact differs from the results of Tin and McCabe. On the contrary, Figs. 2 to 3e indicate clearly the effects of stirring speed, theamount of seed crytals and the flow rate of cooling water. This shows a coincidence with their work qualitatively.
The authors, therefore, attempted to correlate the data of the second supersaturation point, and obtained an experimental but dimensionless relation, as follows:C_II′/C_SII′=1.18(wc_w/UA_k)^0.1·exp.(-0.028S)
where C_II′ and Cs_II′ are the second supersaturated and the saturated concentrations of CuSO₄ at the same temperature, U is the over-all heat transfer coefficient, A_h, is the heat transfer area of agitated vessel, and c_w, , is the heat capacity of cooling water, respectively.
It is shown in Fig. 7 that the above relation coincides with the results of experiments within an accuracy of 10 per cent.

View full abstract

It is well known that the separating agent for extractive distillation improves the relative volatility of the original binary system. But the function of the agent is not well understood yet. The authors considered that the function of the agent is divided into the following two parts.
1) Raising the boiling point of the original binary system.
2) Improving the relative volatility of the original binary system.
In some systems the first function is disadvantageous for improving the relative volatility. For example, in the binary system carbon tetrachloride-benzene, if the relative volatility is represented by the ratio of the vapor pressures of two pure components, this ratio decreases with increasing temperature.
In order to illustrate the above two functions of the extractive distillation agent, the vaporliquid equilibrium of the ternary systems carbon tetrachloride-benzene-solvents have been studied experimentally and shown in Tables 2, 3, 4, 5 and Figs. 1, 2. The solvents used were toluene, xylene, monochlorobenzene and o-dichlorobenzene.
Adding the above results, previous works were also studied as shown in Fig. 2. The ternary systems carbon tetrachloride-alcohols illustrate the first function of the agent and the ternary systems carbone tetrachloride-benzene-benzene derivatives illustrate the second function of the agent.

View full abstract

Based on the new research result obtained in the laboratory process of manufacturing hydroxylamine by three stages, oxidation of ammonia, absorption of nitrous anhydride gas by ammonium sulfite aqueous solution and reduction of the solution by sulfur dioxide, a pilot plant test was done and the following results were obtained.
(1) The yield of hydroxylamine from ammonia in this process is 5-10% higher than that of the process in which ammonium carbonate solution is used as a solvent, and the amount of byproducts, ammonium nitrate and ammonium salt of sulfamic acid, is 2/3-1/2 of that in the latter process.
(2) In the oxidation of ammonia, Pt-Rh catalyst showed the average yield of 86.4% and this yield affects the yield of the following absorption tower.
(3) In the absorption of nitrous anhydride gas, the composition of the gas p is the important factor which affects the yield, and the analysis was done by using the constant of trimolecular reaction velocity observed by Bodenstein.
As the results, it is observed that the oxidation of NO in the absorption tower can not be ignored and that the optimum p is less than 0.40.
(4) In manufacturing hydroxylamine, the most effective factor is the Iodine value ratio of the absorved solvent, the optimum ratio of which is 21%.
(5) The apparent heat of reaction between nitrous anhydride gas and ammonium sulfite solution in this process was measured and the average value of 34.7kcal/(NH₄)₂SO₃mole was obtained.

View full abstract