(533s) Joining Lab Experiments and CFD-Amozone Model for Deep Understanding of Batch Ozonation and Improve Experimental Method

Water treatment and (indirect) potable water reuse (IPR) is nowadays a topic of major attention due to the rising (and interconnected) concerns about water scarcity and micropollutants (MPs) such as pharmaceuticals and pesticides. Advanced tertiary treatments are increasingly adopted at water treatment sites to eliminate compounds with elevated biological resistance and produce safe water for human consumption (e.g. treatment to drinking water standards prior to aquifer discharge). In the US there are multiple examples (e.g. the HRSD SWIFT project (swiftva.com)) aimed at environmental protection from MPs. Conventional water resource recovery facilities (WRRFs) only partially contribute to these goals, and often in a rather unpredictable manner (Schaar et al., 2010).

Ozonation and its combination with other advanced oxidation processes (AOPs) is establishing more and more as an attractive IPR technology alternative for MP removal. Ozone has complex reactions with the effluent matrix. Some of those pathways lead to hydroxyl radical (HO*) formation. Both ozone and HO* will oxidise the MPs. However, depending on the natural Br- concentration in the water, the carcinogenic by-product bromate (BrO3) can be formed (Fischbacher et al. 2015). In addition to this, the specific ozone demand is strongly water-matrix-dependent, leading to significant differences in BrO3 formation, MP removal efficiencies, and ultimately operational costs.

Batch ozonation experiments have been used for decades as a method to evaluate the ozone demand of a given water matrix, on the suitability of ozone technology for micropollutants (MPs) removal, and even on its potential BrO₃ production. However, no standardized technique is currently available. Ozonation tests performed by different laboratories on the same water matrix show that the output of such experiments can be strongly biased by the way it is executed, especially when BrO₃ formation is a concern. The time and speed of injection, as well as the location where O₃ solution is supplied can drastically change the results. Even the O₃ stock solution concentration itself can influence the BrO₃ formation.

In this work, by coupling experimental tests and the use of the Amozone ozonation kinetic model integrated with computational fluid dynamics (CFD), important insights on the performance of batch ozonation tests were investigated for a secondary effluent matrix from the WRRF of Eindhoven (The Netherlands). The simulations in 3D and in millisecond time scale revealed the complex net of reactions involved in a batch experiment and how its execution can be better standardized to have more consistent results regarding the removal of MPs and the BrO3 formation and ensure data reproducibility.

Experiments were focused on testing the effect of ozone dose, stock solution concentration, and time of injection on MPs removal and BrO3 formation. The Amozone kinetic ozonation model was applied in CFD environment to replicate the experimental results and understand where the main improvements in the methodology were possible. The highest BrO₃ formation obtained in the batch ozonation tests was generally due to the higher local formation of OH* caused by either a faster injection or the use of a more diluted O₃ stock solution.

The joint effort between lab experiments and CFD simulations coupled with the Amozone kinetic ozonation model will contribute to set standards for the execution of batch ozonation tests and gain confidence in the applicability of ozone technology.

Methodological Approach

Batch ozonation tests

A total of 4 batch ozonation tests (Table 1) were planned in duplicates on a 24h composite water sample from the WRRF of Eindhoven (The Netherlands) in order to test the ultimate effect of the experimental protocol on the BrO₃ formation. The raw sample composition is reported in Table 2. In these tests O₃ residuals and HO* exposure (using pCBA probe component) were monitored at different moments in time with 10s time resolution. BrO₃ was measured both in the raw and ozonated sample.

Amozone model combined with CFD

Amozone is a kinetic model of ozonation and advanced oxidation processes that realistically predicts O₃ and HO* in real water matrices (Audenaert et al., 2019). The model describes the rapid and slow O₃ consumption, HO* production and scavenging, O₃ and HO* based MP oxidation, bromate formation mechanisms (von Gunten, 2003), nitrite quenching, chloramines formation and reactions, and several pH equilibria. This mechanistic model was embedded in the hydrodynamic simulation of the batch ozonation test replicating the experimental setup and its flow field in transient mode during the whole duration of each experiment. The kinetic model was calibrated using the experiment number 1 and validated by replicating the results of experiment 2 and 3.

An additional (virtual) test was performed using the same experimental conditions of test 3, removing NH₄⁺ from the sample composition to appreciate the buffer effect of NH4-N on the BrO₃ formation.

Results and Discussion

The BrO₃ formation in the different experiments (Figure 1), despite the use of the same water matrix and laboratory tools, can significantly differ depending on the procedure used. The stock solution concentration had the strongest impact on BrO₃ formation due to the higher sample dilution with the addition of more stock solution to reach the same O₃ dose, e.g. in test 3 the same O₃ dose as test 1 was delivered by adding the double of the stock solution. The dilution effect lowered the OH* scavenging by the water matrix causing higher OH* exposure, as confirmed by the probe component pCBA (Figure 2). The time of injection also had a significant impact on the BrO₃ formation.

Combining the Amozone chemical kinetics with the hydraulic model in CFD was possible to understand how the several reactions with the water matrix involving O₃, OH*, and Br-intermediates evolved. During the first instants of the injection, reactions involving O₃ and organics occur, leading to consumption of O₃ and OH* formation. The presence and the ratio of O₃ and HO* decide the extend of the BrO₃ formation together with the Br^- availability.

Local high O₃ concentrations can be observed shortly after injection (Figure 3, top) while at the same time, the generation of OH* causes the depletion of the probe component pCBA (Figure 3, bottom). For the first few seconds after the injection, the O₃ stock solution spreads unevenly in the beaker from the injection point and a cloud of highly concentrated O₃ starts to react with the water matrix. The virtual experiments confirmed that the use of the less concentrated stock solution caused less HO* scavenging in experiment 3 as compared to experiment 1, and generated BrO₃ via the more efficient pathway through the BrO* disproportionation (data not shown).

Comparing the experiment 3 (model validation) with the results of a virtual test in absence of NH₄⁺ (while maintaining the same experimental method and sample composition), higher BrO₃ (55 ug/L) would be formed in case the WRRF would have completely removed the NH₄⁺.

Conclusions

Local O₃ and OH* concentrations are severely impacted by the way that ozone batch tests are performed, and they directly impact the final BrO₃ formation. This can lead to biased decisions when it comes to assessing the suitability of ozonation for a specific water. Therefore, a standardized method of performing ozone batch experiments need to include detailed guidelines on the speed of injection and stock solution dose. In addition to this, the magnetic stirrer speed could also have an important impact on the results. Further virtual tests will investigate mixing effect among other experimental conditions for different water qualities in order to provide solid ground for the standardization of batch ozonation tests.

The power of coupling experiments with virtual testing is threefold: i) investigate with unprecedented spatial and temporal detail chemical mechanisms which are water matrix-specific, ii) test conditions which are impossible to test in reality (e.g. removing NH₄⁺ without altering the sample), and iii) replicate the same procedure across experiments maintaining the same water quality.

The joint effort between lab experiments and CFD simulations coupled with the Amozone kinetic ozonation model will contribute to set standards for the execution of batch ozonation tests and gain confidence in the applicability of ozone technology.

References

Audenaert, W., Bellandi, G., Pearce, R., Takács, I., Buehlmann, P., Hogard, S., Salazar-Benitez, G., Rehman, U., Nopens, I., Wilson, C., Bott, C.B., 2019. A novel kinetic ozonation model for prediction of bromate formation, bromate mitigation and trace organic contaminant removal, in: 24th World Congress of the International Ozone Association.

Fischbacher, A., Löppenberg, K., Von Sonntag, C., Schmidt, T.C., 2015. A New Reaction Pathway for Bromite to Bromate in the Ozonation of Bromide. Environ. Sci. Technol. 49, 11714–11720. https://doi.org/10.1021/acs.est.5b02634

Schaar, H., Clara, M., Gans, O., Kreuzinger, N., 2010. Micropollutant removal during biological wastewater treatment and a subsequent ozonation step. Environ. Pollut. 158, 1399–1404. https://doi.org/10.1016/j.envpol.2009.12.038

von Gunten, U., 2003. Ozonation of drinking water: Part I. Oxidation kinetics and product formation. Water Res. 37, 1443–1467.