According to Whitman, on passing mixed furfural vapor and steam onto Cr-Zn-Mn oxide catalyst, furan is obtained in good yield. The present authors studied certain factors involved in the preparation of this type of catalyst aswell as some of the operating conditions affecting the yield of furan.It has been found that when keeping the reaction temperature at 415℃ and concentration of furfural around 8.5 mole per cent, the yield of furan increases to a maximum as the space velocity is increased to the order of 980. Alternatively, when the space velocity and temperature are kept constant, the yield of furan goes up as the concentration of furfural is decreased. On the other hand, if space velocity and concentration are constant, the decarbonyla-tion takes place most smoothly at 415℃. When the ZnO-0.53Cr203-0.25MnO2 catalyst is prepared by precipitation at 60℃ and calcined at 550℃, a furan yield as high as 86.7% is obtained on passing the raw material at 130℃ without the use of porcelain filler. The space velocity in this experiment is 980, the reaction temperature 415℃,the concentration of the mixture 8.5 mole per cent (near the azeotropic mixture of furural-water system).

Various mixtures of synthetic apatite, magnesium oxide and silica oxide in various proportions have been fused at 1400℃, for thirty minutes. The citric acid solubility of the products has been determined. At SiO2/MgO ratios of 1 to 2.6 by weight, the percentage of synthetic apatite in the mixtures corresponding to the highest citric acid solubility of the products, has been found to be within the range of 40% to 46%. It has been found that, if the SiO2/MgO ratio in the mixtures is greater than 3, the citric acid solubility of the products decreases suddenly. If the R2O3 content in the mixtures is not too high, for example, 5% R2O3, the citric acid solubility of the product is slightly higher than that in the absence of R2O3.

The authors made a theoretical analysis regarding slot opening for a bubble cap in operation. Experimental work was carried out on air-water system for a vertical, single, rectangular slot to correlate slot opening with air flow rate. Experiments were carried out on the basis of the work of Cross and Ryder, but with the experimental conditions extended to different slot widths and different immersed depths. Experimental conditions are as follows: air flow rates, 80~16,300 liters/hr; slot widths, 2.06~13.95 mm; immersed depths 0~100mm. The present work confirms the statement of Cross and Ryder that the slot opening formula derived by Rogers and Thiele, V, is only applicable to high air flow rates. The authors further point out that the discharge coefficient u in the above formula is a constant only when the air flow rate is high and its value is 0.63. For low flow rates, the value of u is no longer constant but is a function of Reynolds number. Data obtained also indicate that the slot opening is independent on the immersed depth. In the continuous bubbling region, the relationship between discharge coefficient u and Reynolds number is as follows: 200 3250 : u = 0.63 Since Fraude number = , therefore : 200 3250 : Fr = 0.353 The formula thus obtained will be applicable to any slot widths and any air flow rates ordinarily used providing bubbling is continuous.

Among common industrial solvents, toluene, mixed xylenes and chloro-benzene have been considered as the most suitable extractants. Data of distribution of l-ephedrine between the above mentioned solvents and pure water (and aqueous sodium hydroxide solution) at different temperatures were obtained. The distribution coefficient, m=(concentration in solvent)/(concentration in aqueous solution), for each solvent is practically constant over the concentration range investigated, increasing with rising temperature from 20 to 60℃. The coefficient is also increased with the addition of sodium hydroxide in the aqueous phase. The relationship between the pll value of the leach liquor and the percentage of alkaloids present in free form was also investigated. It shows that only at pH≥ll.5 practically all alkaloids can be set free and therefore extractable. For the system toluene-leach liquor (with the addition of NaOH to pH=12)-mixed alkaloids present in Ephedra Sinica Stapf, the values of m at 20°,40° and 60° were found to be 2.1, 2.6 and 4.4; and at 60 when chlorobenzene was used m = 5.2. These values of m are smaller than those for the corresponding pure systems. Using mixed xylenes as solvent m at 60° was found to be 5.4, which is somewhat greater than that for the pure system. Alkaloids in the organic solvents can be easily and completely re-extracted into aqueous solution to form hydrochlorides by hydrochloric acid. Thus the solvents can be recovered and returned to the cycle for extraction of leach liquor.

Extraction of ephedrine from Ephedra Sinica Stapf leach liquor by three organic solvents was studied in an experimental column of 100 mm. diameter packed with 19 mm stoneware Raschig rings. When toluene was used as the dispersed phase, the mass transfer coefficient Koca obtained at 40° was much greater than that at 20°, whereas the difference between Koca obtained at 60° and at 40° was much less pronounced. The values of Koca can be correlated by the following expressions: at 40? --- = 9? + 3.0xlO-2 Koea Vc Vd at 60? --!- = -^j- + 2.0x10-3 K0ca Vc Vd When mixed xylenes were used, the mass transfer coefficient was somewhat smaller than that for toluene, e.g., at Vc ? Vd ?< 15 m3/(m2) (hr), K0ca for toluene was 5% higher than that for xylenes. The maximum throughputs were nearly the same in both cases. Although chlorobenzene has a distribution coefficient almost as high as mixed xylenes and much higher than toluene, it is far inferior to the latter two due to very small mags transfer coefficient, very low throughput andliability of forming permanent emulsion with the crude leach liquor. The application of pulsation in the present case resulted only limited improvement in mass transfer process. With amplitudes of 5 ~ 7 mm. and frequencies of 240 ~ 370 cycles/min., the reduction of (H.T.U.)OC was about 20~35%. The leach liquor with 0.5% of suspended solid matter could be successfully treated in the packed column without any trouble under prolonged or intermittent extractions. According to Pratt[2] packed column cannot handle any liquid containing more than 0.1% of solid, this view may be worth reconsidering if the solid is highly dispersed and does not settle readily. The data of re-extraction of alkaloids from organic solvents by hydrochloric acid were treated according to Sherwoods method[12] for gas absorption with rapid second- order irreversible chemical reaction, and so far as there was some appreciable free acid present in the aqueous phase the mass transfer coefficient was independent of the acid concentration. The following correlations were obtained: (1) Packed column with acid as continuous phase and toluene dispersed, (2) Spray column with acid as continuous phase and toluene dispersed, (3) Spray column with acid as continuous phase and mixed xylenes dispersed, Vw = 2.27 m3/(m2) (hr), A new flow diagram for the extraction plant was proposed.