Applying forward mixing model in an L-shape pulsed packed extraction column to investigate the influence of drop size distribution on mass transfer

•Drop size distribution in the investigated column is best followed by the log-normal distribution.•Mass transfer coefficient of the large droplets is larger than that of the small droplets.•Volumetric overall mass transfer coefficient of the small droplets is greater than that of the large droplets.•Pulsation intensity and interfacial tension are the most important factors on mass transfer coefficient. The objective of this work is to model the effect of drop forward-mixing on the mass transfer efficiency of a two phase countercurrent extraction process. Based on the flow mechanism and drop size distribution in extraction columns, a novel model with a simplified sequential algorithm is developed. Unlike the other models that use mean diameter of the dispersed phase droplets, this model accurately considered the effect of drop size distribution on mass transfer efficiency. On the base of this model the volumetric overall mass transfer coefficient has been investigated in a pilot plant of L-shape pulsed packed extraction column by using two liquid systems of toluene/acetone/water and butyl acetate/acetone/water. It is found that, although the mass transfer coefficient related to the large droplets is larger but the corresponding volumetric overall mass transfer coefficient is less than that related to the small droplets. Therefore, any factor that reduces the drop size can improve the efficiency of mass transfer. Furthermore, the effect of operational variables and physical properties including the dispersed and continuous phases flow rates, pulsation intensity and interfacial tension have been considered on mass transfer coefficients. It has been found out that the pulsation intensity and the continuous phase flow rate have seriously affected on mass transfer coefficient, however, the dispersed phase flow rate has a weaker effect. Finally, new correlations are proposed to accurately predict the mass transfer coefficient, axial mixing and drop size distribution. Good agreement between predictions and experiments was found for all operating conditions that were investigated.