In situ sampling with solid-phase microextraction (SPME) was coupled with laser-induced fluorescence (LIF) in an effort to develop a simple field-portable method to determine total dissolved PAH (polycyclic aromatic hydrocarbon) concentrations in sediment pore water. Glass fiber rods with a 50 microm coating of optically clear polydimethylsiloxane (PDMS) were inserted directly into sediment/water slurries. After 1-140 h (typically 18 h), the coated rods were recovered, rinsed with water, and their LIF response was measured with excitation wavelength (308 nm) and emission wavelengths (350-500 nm) chosen to monitor 2- to 6-ring PAHs. SPME-UF response was independent of sediment sample size, as is required for equilibrium sampling methods to be used in situ in the field. Potential interferences from high and variable background fluorescence from dissolved organic matter were eliminated by the use of the nonpolar PDMS sorbent. The detection limit in pore water was ca. 2 ng/mL (as total PAH-34), which corresponds to ca. 0.2 EPA PAH toxic units. Good quantitative agreement (r2 = 0.96) for total PAH-34 pore water concentrations with conventional GC/MS determinations was obtained for 33 surface sediments collected from former manufactured gas plant (MGP) and related sites. Quantitative agreement between SPME-LIF and GC/MS total PAH-34 concentrations was also good for 11 sediment cores (r2 = 0.87), but the predominance of 2-ring PAHs (compared to the other sites) resulted in a lower relative SPME-LIF response compared to the surface sediment samples. The method is very simple to perform, and should be directly applicable to field surveys.
We describe the development of a novel method for real-time in situ characterization of polycyclic aromatic hydrocarbons (PAHs) in submerged freshwater sediments. Laser-induced fluorescence (LIF) spectroscopy, a mature technique for PAH characterization in terrestrial sediments, was adapted for shipboard use. A cone penetrometer-type apparatus was designed for probe penetration at a constant rate (1 cm/s) to a depth of 3 m. A field-portable LIF system was used for in situ measurements in which the output of a pulsed excimer laser was transmitted by optical fiber to a sapphire window (6.4-mm o.d.) in the probe wall; fluorescent emission was collected by a separate optical fiber for transmission to the spectrometer on deck. Four wavelengths (340, 390, 440, 490 nm) were selected via optical delay lines, and multiple-wavelength waveforms were created. These multiple-wavelength waveforms contain information on the fluorescence frequency, intensity, and emission decay rate. Field testing was conducted at 10 sites in Milwaukee Harbor (total PAH concentrations ranged from approximately 10 to 650 microg/g); conventional sediment core samples were collected concurrently. The core samples were analyzed by EPA methods 3545 (pressurized fluid extraction, PFE) and 8270C (gas chromatography-mass spectrometry, GC-MS) for PAHs. A partial least-squares regression (PLSR) model wasthen created based on laboratory LIF measurements and PFE-GC-MS of the core samples. The PLSR model was applied to the in situ field test data, and 13 of the 16 EPA-regulated PAHs were quantified with a relative error of <30% overall (the remaining three PAHs were found at levels insufficient to quantify). We additionally describe preliminary source apportionment relationships that were revealed by the PLSR model for the in situ LIF measurements.
Field screening of fuel-contaminated soils using laser-induced fluorescence is a cost effective and timely method of characterizing contaminated sites. Data collected with laser-based screening tools are often extensive and difficult to interpret. Pattern recognition algorithms can be utilized to enable less highly trained personnel to identify contaminants. In this work, fluorescence intensity of various hydrocarbon fuels deposited on various soil types was measured as a function of emission wavelength and decay time, generating wavelength-time matrices (WTMs). The data were arranged into a three mode array and subjected to trilinear decomposition (TLD). The results ofthe TLD were then utilized in pattern recognition schemes, specifically, linear discrimination and classification and hierarchical cluster analysis. Classification rates and clustering results indicate that these techniques can be very valuable tools in site characterization.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.