Liquid-liquid extraction in continuous-flow agitated extractors
โ Scribed by A. W. Flynn; R. E. Treybal
- Publisher
- American Institute of Chemical Engineers
- Year
- 1955
- Tongue
- English
- Weight
- 520 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0001-1541
No coin nor oath required. For personal study only.
โฆ Synopsis
Abstract
Extraction studies were made with the systems toluene water benzoic acid and kerosene water benzoic acid in two dimensionally similar cylindrical extractors, one 6 in. in diameter by 6 in. high, the other 12 in. in diameter by 12 in. high. Each extractor contained four radial baffles, of width equal to 16.7% of the extractor diameter. The extractors were agitated by sixโbladed dimensionally similar turbine impellers, the diameters of which were in each case one third the diameter of the extractor.
When the extractors were full of liquid, i.e., in the absence of an airโliquid interface, the power required for a given impeller speed was independent of the rate of flow of liquid. However, such operation required more power than the equivalent openโtank agitator, and more baffling was required to produce fully turbulent flow.
Flow rates were varied from 0 to 4,000 lb./hr. and solventโtoโwater mass ratios from 3:1 to 1:6. Extraction stage efficiencies were found to be appreciable even at zero agitator speed.
The additional stage efficiency produced by agitation was found to be a function of the power input per unit volume of liquid flowing. This relationship permitted the correlation of all extraction efficiency data for a given system at a constant solventโtoโwater ratio over eightโfold range of power input. It was found possible to predict the performance of an extractor in continuous flow from batchโextraction measurements through the methods of MacMullin and Weber.
๐ SIMILAR VOLUMES
In an earlier paper (11), the continuous-phase mass transfer coefficients resulting from the continuous-flow contact of ethyl acetate and water in an agitated vessel, ethyl acetate dispersed, were reported to be substantially larger than might be expected from consideration of the data obtained from