PhD Dissertation Defense: Oindrila Ghosh
Oindrila Ghosh, PhD Candidate
Advisor:
Dr. Upal Ghosh
Title:
Abstract:
Polychlorinated biphenyls (PCBs) are hydrophobic organic contaminants (HOCs) that have persisted in the environment for decades after production was banned by the USEPA in 1979 and industrial use phased out. PCBs are known carcinogens and bioaccumulate into organisms through the aquatic food web, thereby posing threats to ecological and human health even at trace concentrations in the environment. The freely dissolved concentration in the aqueous phase (Cfree) is a useful indicator of chemical activity and is critical to determine exchange between environmental phases and bioaccumulation in organisms. Polymeric passive sampling (PS) provides time-integrated Cfree measurements in surface waters and sediment porewaters, unlike grab sampling that provides a snapshot of the concentration at one point in time. PS, as a technology, has been used for bioavailability and risk assessment studies over the last two decades by academic and government researchers and multiple guidance documents for correct implementation of PS methods have been developed. Broader implementation of equilibrium PS of HOCs as an effective analytical tool requires adequate mechanistic understanding of the diffusive uptake of contaminants in PS, development of USEPA or ASTM standard methods and successful adoption of the standardized methods by commercial analytical laboratories.
The primary objectives of this research were: (1) to understand the true nature of the time-integrative property of equilibrium passive samplers under fluctuating ambient surface water concentrations, (2) to design PS devices (by manipulating their major rate-limiting zones for mass transfer so that approach to equilibrium is expedited) and develop field demonstrated PS prototypes that can perform (i) short-term measurements of HOCs in surface water during a storm event or (ii) address the challenges associated with Cfree porewater measurements for strongly hydrophobic compounds, (3) to develop standardized methods for impregnating performance reference compounds (PRCs) into PS (that helps determining the extent of a sampler’s approach to equilibrium) and calculating the cost of preparing PRC impregnated low density polyethylene (PE) samplers.
The theoretical and modeling framework of exchange kinetics in PS developed in this study, indicate that the timing of fluctuations, hydrophobicity of the analyte, and sampler physical characteristics impact the time-integrated Cfree measurement provided by equilibrium PS. This mathematical framework describing mass transfer in time-integrated PS was utilized to develop and optimize the designs of PS devices for long- and short-term deployments in the field for accurate pore-water and surface water Cfree measurements.
In high flow conditions, when diffusion is controlled by the sampler side, using a thin sheet sampler can help achieve faster equilibrium and reduce sampling time scales down to 24-48 hours. These sampling time scales are compatible with the duration of storms in field environments and can thus help in estimating the contribution of a storm event to the Cfree levels in surface water. Thin sheet passive samplers (18 mm PE) were used to perform short-term Cfree measurements over a stormflow event lasting for 42 hours in a PCB contaminated tributary to an urban river in Washington DC. Results indicated that surface water concentrations predicted by long-term measurements over a period of 115 days were similar to the short-term measurements revealing evidence that the tributary concentration was not diluted by high storm flows and was a major source of PCB loading to the main river during stormflow conditions.
In compact sediments, diffusion is controlled by the water boundary layer (WBL) external to the sampler. Mechanical disruption of the WBL outside the sampler by introducing periodic vibration on the sampling platform greatly enhanced the approach to equilibrium for more hydrophobic contaminants. Field demonstrated vibration devices showed that the amplitude of vibration of a powerful motor was dampened by almost 50 times when buried inside the sediment. But even with the limited vibration achieved within compact sediment, PS reached at least 20% equilibrium in 8 days even for the more hydrophobic PCBs and dioxins/furans, as demonstrated by the depletion of PRCs from the samplers. Performing PRC corrections to account for non-equilibrium for the more hydrophobic compounds with confidence for correctly estimating Cfree in sediment porewater was possible.
Finally, an in-depth analysis of PRC loading methods revealed that accurate cost of one loading batch greatly depends on precise measurements of the partitioning behavior of PRC compounds into PE from a complex mixture of solvents like methanol-water and the corresponding fractional mass transfer ratios of each PRC. PRC stock solutions in nonane when spiked to aqueous solution created a multiphase suspension leading to non-homogenous loading into PEs. Non homogenous PRC loading can lead to error prone estimations of fractional PRC losses used for correcting for non-equilibrium conditions. The costs of preparing 1 g of PE at 200 ng/g from water-only and 80:20 methanol-water solutions were similar ($8 for both cases).