NSF Award
2418755
Chlorine is used to kill harmful germs in drinking water. However, chlorine can also react with other substances in the water, like dissolved organic matter (DOM), creating unwanted byproducts called disinfection byproducts (DBPs). Phenolic compounds are key parts of DOM that often lead to the formation of DBPs. Because DOM is a highly complex mixture of different compounds, researchers study individual components of DOM like phenol and lignin to understand how DBPs form. This research has been ongoing since the 1970s, resulting in a large body of available scientific data. Machine learning (ML) is a powerful tool that holds great potential to identify trends in these data. The first objective of this project is to collate data from past studies and use ML to identify key factors in DBP formation from phenolic compounds. The second objective is to study DBP formation from specific phenolic compounds chosen by ML to help elucidate underlying mechanisms and direct the gathering of more data to improve ML accuracy. The final objective is to develop treatment strategies to reduce DBP formation in drinking water guided by the results of this study. Successful completion of this project will benefit society by providing guidance on DBP control to water treatment facilities to better protect human health. Additional benefits to society will include student education and training, with one graduate student and one undergraduate student mentored at South Dakota School of Mines and Technology and one graduate student at South Dakota State University.<br/><br/>Phenolic contents are major components of DOM that serve as a common precursor pool for a broad range of aliphatic DBPs. Because of the complexity and ambiguous structure of DOM, studying model compounds provides a more clear and distinctive profile of precursor chemistry and mechanisms responsible for DBP formation. Phenolic compounds (i.e., phenols) have been widely selected as representative structures within DOM to explore the kinetics and mechanisms of DBP formation. However, the chemistry involved in the process is not fully understood. Past studies of the oxidation of phenols were largely evaluated under relatively ideal experimental conditions (e.g., without natural organic matter and halides). Despite decades of effort, no accurate mechanistic model exists to predict DBP formation from phenols. The overall goal of this project is to identify critical characteristics of the transformation of phenols to aliphatic DBPs and minimize the production of intermediate halophenols in drinking water. This will be achieved using a multi-faceted approach to i) collect data from literature and our experiments, ii) conduct metadata analysis and develop ML models to reveal critical factors in the transformation of phenols to aliphatic DBPs, iii) investigate the formation of aliphatic DBPs and halophenols from phenols, and iv) develop effective treatment methods to minimize both aliphatic DBPs and halophenols. The outcomes of this project will help water utilities mitigate both HPs and aliphatic DBPs in drinking water treatment.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
