Abstract:
In this thesis, two-parameter Linear Free Energy Relationships (2p-LFERs) were developed for diverse set of environmentally-relevant biotic and abiotic phases, and enthalpy of solvation phases. Biological phases evaluated in this study include lipid-water, air-olive oil, liposome-water, and structural protein-water systems. Abiotic phases comprise of dissolved organic carbon (DOC)-water, and natural organic matter (NOM)-air systems. Two systems, air-octanol and air-water were considered for the modeling of enthalpy of phase transfer.
For biological phases, new 2p-LFER models successfully explained the partitioning variability in the datasets with R2 = 0.870 ̶ 0.954 and root mean square error (RMSE) = 0.435 ̶ 0.300 log units. For organic carbon systems, my models exhibited R2 in the range of 0.752 ̶ 0.958 and RMSE from 0.505 to 0.343. My model explained the variability in enthalpy of solvation data with R2 = 0.896 to 0.733 and RMSE from 9.84 to 7.29. All models developed in this study were tested for various criteria of internal and external validity. These tests show the models are statically robust.
The new models provide the mechanistic insight about how the attributes of hydrophobicity and volatility ̶ octanol-water partition coefficient (Kow) and air-water partition coefficient (Kaw) ̶ control the transfer of contaminants from one phase to another.
My models have an edge over the existing estimation approaches because the previous models are either parameter intensive, or parameters are not readily available for many compounds or in case of one-parameter LFERs, the estimation approach is less accurate. Also, my model is computationally cheap and can easily be integrated in EPI SuiteTM, which modules to estimate several environmental properties. Lastly, my model also provides mechanistic understanding about different parameters that control the transfer of pollutants from one phase to another.
