AmericanChemicalSociety.com

Progress and Results

Core Sampling, Dating, PAH extractions, OC and BC determination

We successfully completed field sampling of sediment cores from Oriole Lake (CA) in July 2007. The ACS-PRF work focused on one of the surface cores. Subsamples were sent out for ²¹⁰Pb determination of sediment ages, and were analyzed at URI-GSO for % moisture, organic (OC) and soot black carbon (BC) content and polycyclic aromatic hydrocarbons (PAHs). BC was isolated via a thermal oxidation method at 375 °C. OC and BC were also analyzed for d¹³C. Sediments were extracted for PAHs using an ASE 350. We homogenized three layers of the core, corresponding to 1905-2007 (1-16 cm), 1836-1905 (17-28 cm) and 1790 – 1836 (28-36 cm). These sediment layers were incubated with different masses of polyethylene (PE) samplers to derive porewater concentrations and sorption of PAHs at equilibrium.

Sorption of PAHs in Oriole Lake sediments

In a modification of our proposal, I decided to use the null hypothesis (H₀) that:

Black carbon and organic carbon are sufficient to explain the sorption of PAHs to Oriole Lake sediments.

According to H₀, the overall partitioning of PAHs can be attributed to absorption into the OC fraction (f_oc) and to adsorption onto the BC fraction (f_BC) via:

K_d = f_ocK_oc + f_BCK_BCC_w^n-1                                                                        (1)

with      K_oc the PAH's OC-water partitioning constant,

K_BC the PAH's soot BC-water adsorption constant,

n the Freundlich coefficient,

and C_w the dissolved concentration in µg/L.

Values of the Freundlich coefficient n, K_ocs and K_BCs were taken from the literature. C_w was calculated after equilibrating the sediments with the PE samplers.

Results

Our analysis of the top 50 cm of the cores showed minor fluctuations in BC content (around 0.5 %), and a very high OC content of ca. 20%. With the high amount of OC present, there is potential of charring, leading to erroneously high values of BC. A correlation of OC and BC values (r² = 0.26) indicates the potential for a minor contribution from charring. d¹³C values were ~ -29 ‰ (-28.6 to 30.5) for OC and ~ -27 ‰ (-24 to -29) for BC. There was always an offset between the two carbon pools, suggesting different fractions of carbon involved. While the d¹³C of the OC was fairly constant over time (-29 for last 100 years, -30 for deeper layers), the d¹³C of the BC was more variable, but also indicated a shift towards lighter C values in deeper layers. These results are consistent with some impact of industrial emissions on the BC, and possibly a natural origin in deeper layers.

The most recent sediments had highest fractions of black carbon, consistent with industrially-derived diesel soot deposition over the last few decades. In contrast, PAH concentrations were highest in the deeper layer, with the sum of 17 PAHs, (excluding retene and perylene) being higher by up to 50-fold than in the other layers. Exceptions were retene and perylene, two naturally produced PAHs, which displayed highest concentrations in the most recent sediment layer.

For the most recent sediments impacted by industrial emissions of diesel soot, sorption to black carbon is indeed necessary to explain the observed sorption in Oriole Lake sediments. In other words, organic carbon alone (K_d = f_ocK_oc ) was insufficient to account for the observed partitioning for the most recent sediment layer. The combined OC-absorption and BC-adsorption model using equation (1) using n=0.8 gave a satisfactory result for the sediments, implying that the charred sediments exerted no additional sorption.

In contrast, in the 17-28 cm layer, sorption of PAHs was stronger than could be explained using the combined OC and BC model. This suggests that additional sorption was occurring. As our BC isolation method is best at detecting soot-like BC, the presence of chars from natural fires could provide the additional sorption in these sediments. The very high concentrations of pyrogenic PAHs, at least 10 times higher than in the other sediments, provided further evidence for the occurrence of natural fire-derived material in these sediments. Lastly, tree-analysis from Oriole Lake confirms the presence of at least two major fires in the water shed during the time-frame of the sediments.

In the deepest layer for which the sorption of PAHs was analyzed, results differed slightly from the more recent sediment layers. First, there was some evidence of enhanced sorption, but the combined OC-absorption and BC-adsorption model was sufficient to explain the measured K_d values. Second, concentrations of PAHs were rather low, comparable to the more recent sediments. This seems to indicate the absence of major natural fire events, even though tree analysis indicated the presence of several natural fires in the watershed during the time accumulated in the sediments. A possible explanation could be that natural degradation processes have reduced the sorptive capacity of any chars present in the sediments.

We have sent out subsamples from these sediments for petrographic analysis, which will help us prove whether chars and other geosorbents were present in the deeper sediments, and could have affected the sorption of PAHs.

Our results suggest that natural fires resulted in both elevated PAH concentrations and enhanced sorption in sediments, probably due to fire-produced chars.

Training and education

The project provided training and education to two undergraduate students at URI-GSO: Julia Sullivan for 2 years and Kevyn Bollinger in the 1^st year only. Kevyn was involved in the field sampling, while Julia has been in charge of the OC, BC and PAH analysis, and the sediment equilibrations. Kevyn has moved onto being a graduate student in Ocean Engineering at URI, while Julia is considering joining URI-GSO as a graduate student next fall.