Experimental modeling and uncertainty analysis of dispersed phase holdup at flooding in a pulsed disc-doughnut column, case study: Response surface methodology and Monte-Carlo simulation

Abstract This study aims to investigate, optimize, and simulate the dispersed phase holdup at flooding conditions for the standard physical systems in a pulsed extraction column with the disc-doughnut configuration. The interaction impacts for operational parameters (pulse intensity and organic and aqueous phase velocities) and interfacial tension (systems type) were examined using the response surface approach. A novel correlation for the dependent parameter, namely holdup at flooding based on the quadratic model, was developed with the central composition design methodology. A desirable agreement between actual data and calculated data from the proposed model was observed because of the high coefficient of determination. The maximized holdup at flooding was optimized at 0.43, when the interfacial tension, pulse intensity, organic phase velocity, and aqueous phase velocity were equal to 0.019 N/m, 3.97 cm/s, 2.93 mm/s, and 0.78 mm/s, respectively. The uncertainty analysis with the Monte-Carlo simulation was investigated to predict the sensitivity of input factors on the flooding curves in the pulsed column. Based on the simulation results, the pulsation intensity has the most critical impact on the output results. At the same time, the influence of the density and viscosity can be negligible on the dispersed phase holdup at flooding.