Our understanding about the presence, behavior, and toxicities of atmospheric persistent organic pollutants is limited by our ability to accurately measure them. This dissertation details the development and characterization of a model for the determination of an accurate sampling rate (Rs), and effective sampling volume (Veff), for polyurethane equipped passive air samplers (PUF-PAS), and the subsequent application of PUF-PAS sampling methods towards novel applications studying polychlorinated biphenyls (PCBs).
The user friendly mathematical model resulting from this work, published as a Matlab script, predicts Rs and Veff as a function of local hourly meteorology and the physical-chemical properties of the target analytes. The model was first developed using active sampling methods in urban Chicago, where good agreement was found between the PUF-PAS and high volume active samplers: Active/Passive = 1.1 ± 1.2. The model was then expanded and calibrated globally using the dataset from the Global Atmospheric Passive Sampling (GAPS) network. After this global calibration we found acceptable agreement between modelled and depuration-determined sampling rates for an independent dataset, with several compounds having near zero mean percent bias (±6%). The globally applicable model is the best alternative for locations experiencing low average wind speeds or cold temperatures, and is particularly useful for the interpretation of samples with long deployments, deployments conducted under warming conditions, and compounds with high volatility. An interactive web-based graphical user interface for the sampling rate model was developed. Users input sampler locations, deployment dates, and target chemicals, in the web-interface and are provided with a sample and compound specific Rs and Veff.
The sampling rate model was examined for use in the indoor environment and it was found that both the experimentally calibrated (1.10 ± 0.23 m3 d-1) and modeled (1.08 ± 0.04 m3 d-1) Rs agreed with literature reports. Correlating sample specific wind speeds with uptake rates, it was determined that variability of wind speeds throughout the room significantly (p-value < 0.001) affected uptake rates. Despite this, the PUF-PAS concentration measurements using modelled Rs values were within 27% of the active sampling determined concentration measurements.
Using PUF-PAS samplers, PCBs 47, 51, and 68 were found to account for up to 50% of measured indoor sum PCB concentration (2700 pg m-3). Direct surface measurements were conducted to identify finished cabinetry to be a major source, as a result of the decomposition of 2,4-dichlorobenzoyl peroxide used as an initiator in free-radical polymerization of polyester resins. While this phenomenon has been detected at trace levels in other polymer products, it has never been shown to be a significant environment source of PCBs.
PUF-PAS samplers were similarly used to study the presence of airborne hydroxylated polychlorinated biphenyls (OH-PCBs) and PCBs in the metropolitan Chicago area. While OH-PCBs have been hypothesized to be an important removal mechanism for atmospheric PCBs, they were not directly measured in the air until recently. The two most frequently detect OH-PCB congeners in this study, 2OH-PCB2 and 6OH-PCB2, were detected at levels comparable to a previous report of atmospheric OH-PCBs utilizing active sampling methods, suggesting the viability of PUF-PAS methods to study atmospheric OH-PCBs. One sampling site detected as many as 50 OH-PCBs but uncertainties with sampling and laboratory methods prevent any strong conclusions from being drawn.
| Identifer | oai:union.ndltd.org:uiowa.edu/oai:ir.uiowa.edu:etd-7825 |
| Date | 01 May 2018 |
| Creators | Herkert, Nicholas John |
| Contributors | Hornbuckle, Keri C. |
| Publisher | University of Iowa |
| Source Sets | University of Iowa |
| Language | English |
| Detected Language | English |
| Type | dissertation |
| Format | application/pdf |
| Source | Theses and Dissertations |
| Rights | Copyright © 2018 Nicholas John Herkert |
Page generated in 0.0057 seconds