Mass transfer correlations for multiple-impeller gas–liquid contactors. Analysis of the effect of axial dispersion in gas and liquid phases on “local” [formula omitted] values measured by the dynamic pressure method in individual stages of the vessel

Abstract

Mixing intensity of the liquid phase, power consumption and volumetric mass transfer coefficients are presented for various single-, double- and triple-impeller configurations. The measurements were carried out in a flat-bottomed vessel of inner diameter 0.29 m. Twenty-eight combinations of various impeller types inducing radial, axial and combined liquid flow were used, viz. a Rushton Turbine (RT), a six Pitched Blade (PB), Techmix 335 (TX) impellers pumping downwards (D) and upwards (U), Lightnin A315 (LTN) and a Narcissus (NS) impeller. Water, 0.5 M Na 2SO 4 solution and solution of commercial thickener (Sokrat 44) were used as a liquid phase, representing coalescent, non-coalescent and viscous batches, respectively. Mixing intensity of the liquid phase was characterized by both exchange flows and homogenization times. The dimensionless mixing time, Nt m , does not depend on Reynolds number but on the direction of liquid flow induced by the impeller. The axial impellers (TXU, TXD, and LTN) provide more effective homogenization than the impellers with combined or radial flow (PBU, PBD, NS and RT). General correlations of gassed to ungassed power ratio (valid for all types of batches used) are presented separately for the bottom and upper stages of the vessel. Volumetric mass transfer coefficient k L a was measured in individual stages of the vessel by the dynamic pressure method (DPM). Analysis of the effect of liquid exchange flow between stages and the effect of axial dispersion in the gas phase on k L a values measured by the DPM has shown that the arithmetic mean of these “local” values depends neither on gas phase axial dispersion nor on the liquid exchange flows and is equal to the arithmetic mean of the true local mass transfer coefficients in individual sections of the vessel. Two types of correlations were used to fit the “local” and the average k L a values with specific total power dissipated, superficial gas velocity, power number (characterizing impeller type) and gassed to ungassed power ratio (characterizing impeller placement in the multiple-impeller configuration, bottom vs. upper stages). General correlations (valid for all impeller configurations) are presented for each type of liquid batch used (coalescent and non-coalescent non-viscous batches and viscous batch). 3RT, RT + 2 PBD and RT + 2 PBU are the most efficient impeller combinations for the mass transfer performance in the triple-impeller vessel. The two latter configurations require a 15% enhancement of agitator frequency to reach an equivalent power dissipation and performance when compared to the 3RT configuration.