Abstract

Agrochemicals are a potential source of contamination in water resources. Whether an agrochemical poses an ecological or human health risk depends on its environmental concentration, which is a function of the rate of degradation, and its toxicity. One category of agrochemicals for which there has been increased use observed, potentially leading to increased environmental concentrations, are herbicides that are approved for use on commercially released genetically modified (GM) crops with the associated herbicide-tolerance traits. The GM tolerance trait for the herbicide isoxaflutole was introduced in commercially available crops starting in 2020. The addition of isoxaflutole-tolerance traits to GM crops has been identified by the US Environmental Protection Agency (EPA) as a beneficial tool to combat rising resistance to herbicides among weeds, especially for herbicides like glyphosate that have been used on GM glyphosate-tolerant crops for decades. The potential rise in the use of isoxaflutole due to the introduction of GM crops has resulted in increased concern of the fate of isoxaflutole post-application to fields. Paired with the high toxicity of isoxaflutole and its active form and hydrolysis product, diketonitrile, the transformation of both isoxaflutole and diketonitrile in the environment can determine the impact of applied isoxaflutole. The first objective of this dissertation was to investigate the persistence and transformation of diketonitrile present in drinking water sources in comparison to the previously demonstrated minimal reactivity of isoxaflutole with the drinking water oxidant, free chlorine. I found that diketonitrile reacts more rapidly with free chlorine than monochloramine, which are common oxidants used in drinking water treatment, but result in the formation of the same products (e.g., disinfection byproducts). I demonstrated that diketonitrile is a uniquely high-yield precursor for a highly toxic nitrogenous disinfection byproduct, dichloroacetonitrile (DCAN) in the presence of both oxidants. Using both kinetic and product data at multiple pH conditions, I proposed an updated reaction mechanism for the reaction of diketonitrile with both free chlorine and monochloramine, which is consistent with the high yield of DCAN. From this first objective, I concluded that the speciation of isoxaflutole and diketonitrile in drinking water sources will affect the way in which applied isoxaflutole poses a threat to human health. The second objective of this dissertation was to investigate the hydrolysis of isoxaflutole to diketonitrile to improve the prediction of the rate of hydrolysis in natural water sources. I demonstrated that isoxaflutole hydrolysis is particularly sensitive to catalysis by weak bases used in buffers in laboratory experiments to estimate hydrolysis in natural waters sources. I obtained the bimolecular rate constants for isoxaflutole hydrolysis with bases commonly present in laboratory buffers (i.e., carbonates, phosphates) relative to hydroxide, which is typically assumed to dominate the rate of base-catalyzed hydrolysis. Using the bimolecular rate constants I obtained, I constructed a linear free energy relationship, the Brønsted relationship, to assess the sensitivity of the reaction to base strength. I determined that rate constants previously measured and used in risk assessments to predict the rate of transformation in the environment overpredict the rate of hydrolysis, particularly near neutral pH. Lastly, I demonstrated that the bimolecular rate constants I obtained can be used to more accurately predict hydrolysis in natural water sources. The third objective of this dissertation was to investigate whether overprediction of hydrolysis occurs more widely (e.g., contaminants other than isoxaflutole). I conducted a review of literature to assess the typical conditions at which hydrolysis is measured in the laboratory to assess hydrolysis in real water sources (e.g., groundwater, surface water, drinking water treatment, agricultural water). I found that since 2014 there has been a trend of increasing reliance on the EPA and/or Organisation for Economic Co-operation and Development (OECD) guidelines for assessing hydrolysis, both of which cite a foundational publication from Mabey & Mill (1978). Notably, Mabey & Mill estimated that buffer effects are negligible within concentrations of 10 mM. I also found that only a small fraction of studies from 1971 to 2023 measured hydrolysis in real water sources. Among these studies there was a trend of faster hydrolysis in laboratory water, although the assumption that buffer effects were negligible led to most studies to look for causes of inhibition in the real water sample rather than sources of catalysis in the laboratory water. Using two linear free energy relationships, the Brønsted and Swain-Scott relationship, I demonstrated that consistent with our findings for isoxaflutole hydrolysis, buffer concentrations lower than 1 mM have the capacity to significantly contribute to the rate of hydrolysis. Based on these results, I provided updated recommendations for assessing the hydrolysis of contaminants in the laboratory that can be more accurately translated to real water sources. The fourth objective of this dissertation was to develop a method that would enable the quantification of isoxaflutole and diketonitrile in future field campaigns. While occurrence studies for isoxaflutole and its products have been conducted since the early 2000s at a variety of sampling locations, the effect that the introduction of GM isoxaflutole-tolerant crops have on the occurrence of isoxaflutole and diketonitrile in Midwestern water sources has not been investigated. In this objective, I outlined a field campaign to achieve this goal. Then, I determined the optimal method for the concentration of samples post solid-phase extraction of natural water samples. Lastly, I established a liquid-chromatography mass-spectrometry/mass-spectrometry (LC-MS/MS) method to quantify isoxaflutole and its products after sample processing. Overall, this dissertation (1) demonstrates that isoxaflutole, which converts to diketonitrile in water sources used for drinking water, poses a unique threat as a high-yield precursor for a toxic disinfection byproduct, (2) identifies the importance of assessing the contributions of weak bases present in buffers used to measure hydrolysis in laboratory conditions to prevent overestimation of the rate of hydrolysis in natural waters, not only for isoxaflutole but for any contaminant whose risk characterization is dependent in part on the rate of hydrolysis, and (3) establishes methodology required to assess the effects of the introduction of GM isoxaflutole-tolerant crops on the occurrence in real water sources. Thus, this dissertation contributes to an advanced and improved understanding of the transformation of both isoxaflutole and diketonitrile in natural and engineered water systems.

Committee Chair

Kimberly Parker

Committee Members

Brent Williams; Jason He; Richard Mabbs; Young-Shin Jun

Degree

Doctor of Philosophy (PhD)

Author's Department

Energy, Environmental & Chemical Engineering

Author's School

McKelvey School of Engineering

Document Type

Dissertation

Date of Award

8-6-2025

Language

English (en)

Download

Available for download on Thursday, August 05, 2027

Share

COinS