The transient Back flow cell model (BFCM) has been presented as an alternative approach to describe the hydrodynamics of ozone bubble columns. Compared to the traditionally used models, transient BFCM provided a more flexible, reliable, and accurate tool to describe the backmixing in the liquid phase.

The two-phase steady-state axial dispersion model (2P-ADM) and BFCM have been developed for describing the performance of ozone bubble columns. Although, the 2P-ADM and BFCM have proven to be adequate for modeling the ozonation process, the BFCM was easier to formulate and solve.

A modified steady-state non-isobaric one-phase axial dispersion model (1P-ADM) has been developed. The 1P-ADM is composed of a single non-homogeneous linear second-order ordinary differential that was solved analytically. Compared to 2PADM, the 1P-ADM has provided reliable predictions of the dissolved and gaseous ozone concentrations.

An impinging jet bubble column, that utilized two venturi injectors, has been tested for the ozone mass transfer applications. The venturi injectors created impinging gas-liquid jets in the ambient fluid and that caused an increase in the turbulence produced in the ambient fluid and therefore, the gas-liquid mass transfer rates increased. The 1P-ADM was applied to analyze the measured dissolved and gaseous ozone concentrations. Excellent conformity between the measured and the predicted concentrations was observed. Therefore, the 1P-ADM model has proven to be an efficient, adequate, and easy-to-use tool for describing the ozonation in the impinging-jet bubble column.

Compared to conventional diffuser ozone bubble columns, the impinging jet design has provided comparable reduction efficiencies of the impurities present in Kraft pulp mill effluents. However, the impinging-jet bubble column was more effective due to its smaller reactor volume and lower off-gas ozone concentrations.

A 2-D laser particle dynamics analyzer was used to measure the characteristics of the gas bubbles in the impinging jet bubble column using clean deionized water and ozonated Kraft pulp mill effluent as the test liquid. The count mean bubble diameter (dB) and Sauter mean bubble diameter (d S) were smaller than those obtained in conventional bubble columns and that has led to a significant increase in the specific bubbles' interfacial area and gas hold-up.