Author ORCID Identifier:

https://orcid.org/0009-0008-4892-8840

Date of Graduation

5-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Civil Engineering

Advisor/Mentor

Fairey, Julian

Committee Member

Wahman, David

Second Committee Member

Chimka, Justin

Third Committee Member

Speir, Shannon

Keywords

environmental sampling

Abstract

Diffusive gradients in thin-films (DGT) passive samplers are an alternative to grab sampling for quantifying time-weighted average (TWA) concentrations of per- and polyfluoroalkyl substances (PFAS) in aqueous systems. DGTs may enable long-term PFAS site characterization in a more time and cost-effective manner compared to grab sampling, but critical information is lacking in PFAS diffusion coefficient (DGel) uncertainty, PFAS quantitation limits, and environmental impacts on DGT performance. This work provides a thorough evaluation of DGT performance for a suite of 32 PFAS through (i) determining gel-layer PFAS diffusion coefficients and their associated error (i.e., 95 % confidence intervals (CIs)), (ii) assessing low- and sub-ng L–1 PFAS quantitation and investigating the effects of DGT deployment time ranging from 5–91 days, and (iii) the influence of environmental conditions (e.g., competing anions, natural organic matter (NOM), and hydrodynamics) on DGT effectiveness. A non-steady state finite difference model was used to interpret diffusive flux in two-compartment diffusion cell tests with the 32 PFAS and determine DGel ± 95 % CIs, which were ± 10 % of their PFAS free water diffusivities. The error in DGel was propagated into the DGT-determined TWA PFAS concentrations, CDGT, assessed in batch reactors dosed at 1–200 ng L–1 and continuous flow reactors dosed at 0.02–75 ng L–1. The findings showed accurate determinations in 48–97 % of quantified PFAS compared to the TWA concentrations of grab samples (CTWA) or the continuous PFAS dose (CDose) when deployed in buffered reagent water (pH 7). DGTs were assessed at a 50 ng L–1 PFAS dose in the presence of mg L–1 levels of common anions and NOM. The impact of common anions was negligible in synthetic groundwater and seawater, evidenced by up to 93 % PFAS with CDGT values indistinguishable from CDose, but CDGT was underestimated for 48–78 % of PFAS in NOM-amended waters likely due to adsorption inhibition and/or steric hindrances. Last, the diffusive boundary layer thickness (DBL) was investigated under varying low-mixing conditions and in a field study using a linear regression method (LRM) enabled by deploying DGTs with different gel layer thicknesses and an empirical model. Relatively large uncertainty in the LRM favored the use of the empirical model to estimate δDBL and maintain or improve the accuracy of CDGT relative to CDose. Collectively, this work establishes a framework for deploying DGT passive samplers in aqueous laboratory and field settings for the purpose of reliable PFAS quantitation at ng and sub-ng L−1 levels, and it provides an assessment of DGT performance under competitive matrices and low-flow conditions.

Download

Share

COinS