Modelling gas-liquid mass transfer in wastewater treatment: when current knowledge needs to encounter engineering practice and vice versa.

Gas-liquid mass transfer in wastewater treatment processes has received considerable attention over the last decades from both academia and industry. Indeed, improvements in modelling gas-liquid mass transfer can bring huge benefits in terms of reaction rates, plant energy expenditure, acid-base equilibria and greenhouse gas emissions. Despite these efforts, there is still no universally valid correlation between the design and operating parameters of a wastewater treatment plant and the gas-liquid mass transfer coefficients. That is why the current practice for oxygen mass transfer modelling is to apply overly simplified models, which come with multiple assumptions that are not valid for most applications. To deal with these complexities, correction factors were introduced over time. The most uncertain of them is the α-factor. To build fundamental gas-liquid mass transfer knowledge more advanced modelling paradigms have been applied more recently. Yet these come with a high level of complexity making them impractical for rapid process design and optimisation in an industrial setting. However, the knowledge gained from these more advanced models can help in improving the way the α-factor and thus gas-liquid mass transfer coefficient should be applied. That is why the presented work aims at clarifying the current state-of-the-art in gas-liquid mass transfer modelling of oxygen and other gases, but also to direct academic research efforts towards the needs of the industrial practitioners.

Considering the plug-flow behavior of the gas phase in nitrifying BAF models significantly improves the prediction of N2O emissions.

Nitrifying biologically active filters (BAFs) have been found to be high emitters of nitrous oxide (N2O), a powerful greenhouse gas contributing to ozone layer depletion. While recent models have greatly improved our understanding of the triggers of N2O emissions from suspended-growth processes, less is known about N2O emissions from full-scale biofilm processes. Tertiary nitrifying BAFs have been modeled at some occasions but considering strong simplifications on the description of gas-liquid exchanges which are not appropriate for N2O prediction. In this work, a tertiary nitrifying BAF model including the main N2O biological pathways was developed and confronted to full-scale data from Seine Aval, the largest wastewater resource recovery facility in Europe. A mass balance on the gaseous compounds was included in order to correctly describe the N2O gas-liquid partition, thus N2O emissions. Preliminary modifications of the model structure were made to include the gas phase as a compartment of the model, which significantly affected the prediction of nitrification. In particular, considering gas hold-up influenced the prediction of the hydraulic retention time, thus nitrification performances: a 3.5% gas fraction reduced ammonium removal by 13%, as the liquid volume, small in such systems, is highly sensitive to the gas presence. Finally, the value of the volumetric oxygen transfer coefficient was adjusted to successfully predict both nitrification and N2O emissions.