Predicting photolytic removal of CECs in wastewater and stormwater treatment assets using a chemoinformatic tool

Predicting photolytic removal of CECs in wastewater and stormwater treatment assets using a chemoinformatic tool

Current tools do not allow utilities to predict the removal of chemicals of emerging concern (CECs) in wastewater treatment lagoons; thus utilities are not able to assess risk associated with the treated water, and optimise treatment conditions.


Furthermore, climate change impacts and reduced fresh water sources force utilities to utilise alternative water sources within the concept of the circular economy (including but not limited to storm water). To address compliance with new demanding regulations for varying sources coupled with budget restrictions, utilities need a cost-effective tool to identify the most problematic CECs and their removal in natural treatment processes.

Photolysis, especially indirect photolysis, has been shown to be an important removal mechanism the extent of which varies with each CEC and conditions. It was found that the photodegradation of CECs was correlated to a number of selected lagoon wastewater quality parameters. However, models developed based on limited data using logarithms or linear functions are not able to produce accurate results, especially when the relationships between variables are highly complex and non-linear. Therefore, to further demonstrate the correlations between the lagoon wastewater and plant performance in terms of photolytic removal of CECs, it is proposed that machine learning be adopted as it performs better with complex datasets. 

Project description schematic:


The overall aim is to develop a wastewater lagoon photolysis model to enable prediction of the photolytic removal of CECs during lagoon wastewater treatment.

This will be achieved via the following objectives:

(i) To determine 2nd order reaction rate constants for CEC moiety groups and each of the 4 PPRIs experimentally.

(ii) To use the information from (i) to develop QSAR algorithms for the prediction of 2nd order reaction rates between the PPRIs and different moiety-containing CECs, e.g. olefins, amines, phenols.

(iii) To undertake long term monitoring of the wastewater characteristics (E2:E3, DOC, nitrate, turbidity) in situ using a spectral analyser and determine PPRI concentrations experimentally weekly.

(iv) Use a big data approach to develop algorithms or machine learning models to further build correlations between the wastewater characteristics, temperature, sunlight irradiance and PPRI concentrations. 

(v) To trial the model at Western Treatment Plant (WTP) over several months to determine the robustness of the approach.

The long-term objective is to extend the work to include more CEC moieties, a longer term onsite study at WTP or other plant, and trials at lagoon systems at other WWTPs (particularly in different climate zones). The final product of this project would be a model which covers both internal (CEC-related) and external factors (plant-related) influencing the photolytic removal pathway of CECs of interest during lagoon treatment.

WaterRA Contact

Dr Arash Zamyadi | Research Manager 

Amount being sought


Due Date

30th Sep, 2020