Evaporation from a plane, free-liquid surface into a turbulent air stream

Abstract

In the past the experimental study of evaporation, and the interpretation of the results obtained, have shown in general only meagre reference to the aerodynamics of the problem. The present treatment, which is concerned with the evaporation from plane, free-liquid surfaces of relatively small dimensions into a tangential air stream, demonstrates the importance of the type of boundary layer flow. The rates of evaporation under the influence of a turbulent boundary layer are then tested against a hydrodynamical theory due to O. G. Sutton. Sutton's theory assumes that the turbulent transfer of any entity is determined by the momentum interchange coefficient, which is shown to involve the kinematic viscosity of the diffusing medium, and which leads to a functional form for evaporation which has been shown previously to be in good agreement with experimental data. Developed into a computable form, and tested against the present experiments on the evaporation of bromobenzene and against experiments by Elias on the analogous problem of convective heat transfer, the theory is now shown to predict the absolute rate of turbulent transfer in a satisfactory manner. An extension of the analysis to the relative rates of evaporation of various liquids, as determined in the present experiments and in recent experiments by Wade, shows that the theory specifies inadequately the variation of rate of evaporation with type of liquid. In the absence of a precise theoretical argument, an empirical generalization of Sutton's theory is set forth, in which the turbulent interchange coefficient is modified by the molecular diffusion coefficient appropriate to the entity undergoing transfer. The range of physical characteristics covered in the present evaporation experiments, and in those performed by Wade, is sufficient to demonstrate the closer agreement provided by the generalized form of the theory. A more general test, against previous investigations for which the aerodynamic conditions can be estimated with reasonable confidence, shows that the absolute rate of evaporation may be predicted correctly in order of magnitude. In all cases considered the observed rates are in excess of the theoretical values, but it is significant that the discrepancy decreases as the experimental, conditions conform more closely to the ideal conditions assumed in the theoretical treatment.