Richard F. Lee, Principal Investigator

Abstract
Stable water-in-oil emulsions, often formed after oil spills, contribute to the difficulties of cleanup due to their persistence and high viscosity. The primary objective of our studies was to determine the fate and effects of such emulsions after they enter estuaries. To achieve this objective, a stable emulsion formed from Kuwait crude oil, as well as nonemulsified oil, was added to estuarine mesocosms, followed by exposure of the estuarine grass shrimp (Palaemonetes pugio) to treated sediments. The mesocosm studies showed that emulsion treated sediments had polycyclic aromatic hydrocarbon concentrations (PAH) of 284, 198, 89, 49 and 7 µg/g sediment on after Days1, 14, 28, 42 and 56, respectively, while non-emulsified oiled sediments had concentrations of 271, 14, 3, 0.3 and 0.2 ug/g sediments on these same days, respectively. Female grass shrimp exposed to Day 14 emulsified oiled sediments did not produce embryos and exposure to Day 36 sediments resulted in reduced embryo production. In contrast, grass shrimp exposed to non-emulsified oiled sediments were not affected with respect to embryo production compared to controls. A second assay exposed grass shrimp embryos to sediment pore water followed by determination of embryo hatching rates and DNA strand breaks. Comparable results were found between the two assays (embryo hatching/DNA strand breaks versus embryo production) with sediments from Days 14 and 36. The persistence of PAHs in emulsified oiled sediments, relative to nonemulsified oiled sediments, is the likely explanation of the prolonged toxicity of the emulsified oiled sediments. Sediments after oil spills may have higher PAH concentrations (6000 to 100,000 µg/g sediment) than were present in our mesocosm studies. If these PAH decreases occur at the same rate observed in our studies, then emulsified oiled sediments would show toxicity for several months if emulsified oil entered an estuary during a spill. The results of the present project provide expected rates of PAH decrease in estuarine sediments. They suggest that if emulsified oil is not prevented from entering an estuary, water-in-oil emulsion can result in prolonged toxicity to grass shrimp. Grass shrimp can comprise up to 56% of the biomass of pelagic macrofauna in estuarine tidal creeks and are important in the diet of many estuarine fish. Methods to prevent formation of emulsions or emulsified oil from entering an estuary should be considered during an oil spill. The project also developed a pore water assay to rapidly assess the genotoxicity of sediment in large coastal areas and assess the presence of biological “hot spots” which require more attention. The sediment pore water assay can also be used to determine the recovery rate of grass shrimp in heavily oiled sites.

Keywords: Oil, Emulsions, Toxicity, Sediment, Comet Assay

Acknowledgements
Funding for this work was provided by the NOAA/UNH Coastal Response Research Center(Grant number 03-688). Karrie Brinkley, Ullrich Wartinger and Keith Maruya from Skidaway Institute of Oceanography were involved in various aspects of the project.

Table of Contents

1.0 Introduction
One of the consequences of oil spills in coastal waters can be the formation of stable water-in-oil emulsions with mixing energy provided by wind and tides (Daling et al., 2003; Fingas et al., 2001; Lee, 1999; Lunel et al., 1996). Stable water-in-oil emulsions are characterized by their persistence (5 days or longer), high water content (50 to 90%), high viscosity, small water droplets and higher density than the original oil (Brandvik and Daling, 1991; Fingas et al., 1994). The high viscosity results in a semi-solid or gel which contributes to the persistence of the emulsified oil and makes cleanup more difficult.

A variety of compounds and mixtures have been shown to promote and stabilize these emulsions, including sea water particulates, and fractions or compounds found in crude oil (Payne and Phillips, 1985; Lee, 1999 and references cited therein). It appears that compounds with higher solubility in the oil phase than in the aqueous phase (e.g., nickel porphyrins) are the emulsifying agents that can promote stable water-in-oil emulsions. Crude oils that form very unstable emulsions (e.g., Gullfaks crude from the North Sea) require weathering before stable emulsions are formed. The weathering may cause the formation of colloidal asphaltene particles and highly polar compounds that contribute to emulsion stabilization.

Payne et al. (1983) added Prudhoe Bay crude oil to a large wave tank and found that emulsification did not occur until a significant amount of evaporation had taken place to remove lower molecular weight compounds. Payne and Phillips (1985) found that most oils that formed emulsions in real spill events also formed stable emulsions in the laboratory. Spills involving oils with relatively high asphaltene concentrations (e.g.Torry Canyon, Tanker Arrow and IXTOC-1) produced stable water-in-oil emulsions (Payne and Phillips, 1985). In contrast, Ekofisk crude, which is low in asphaltenes (0.03%), formed unstable emulsions (Audunson, 1978). The work reported here supports the earlier studies of Payne and Phillips(1985) concerning the importance of asphaltenes in the oil, particularly porphyrins, in producing stable emulsions. We further showed that for Prudhoe crude oil that particles in the surrounding ocean water are also important in producing emulsions

Stranded oil is toxic to a variety of marine invertebrates including crustaceans, polychaetes and mollusks(Glemarec and Hussenot, 1982; Jackson et al., 1989; Lee et al., 1981; McGuiness, 1990; National Research Council, 1985; Peterson, 2000; Sanders et al., 1980). While many of the spilled oils formed emulsions, it is not clear from these field studies if exposure to emulsified oil produced different effects than exposure to non-emulsified oils.

The studies covered in this report include determination of the stability and properties of water-in-oil emulsions formed by five crude oils, photo-oxidation products in emulsions after solar exposure, and fate and effects of an emulsified and non-emulsified oil added to estuarine mesocosms.

Since emulsions often form after oil spills, they present major problems for effective cleanup after the spill, due to their persistence and physical properties. The results of this project provide information on the effects and fate of emulsified oil entering which can be used when decisions are made concerning cleanup and use of dispersants after an oil spill has occurred.

2.0 Objectives
The primary objective of this project was to determine the fate and effects of water-in-oil emulsions formed after oil spills in estuaries. Specific objectives to achieve this overall objective included determination of the: 1) stability and properties of water-in-oil emulsions formed from different crude oils; (2) changes over time of the concentrations of polycyclic aromatic hydrocarbons (PAHs) in estuarine sediments after the addition of emulsified/non-emulsified oils; and (3) changes over time in the toxicity (reproduction/embryo development/ DNA damage assays) of estuarine sediments to grass shrimp after the addition of emulsified/non-emulsified oils.

3.0 Methods
Preparation of water-in-oil emulsions
Water-in-oil emulsions were prepared by rotating 500 ml cylindrical separatory funnels containing 300 ml of estuarine water and 30 ml of Kuwait crude oil. The rotating apparatus run at 30 rpm for up to 24 hrs was similar to the one described by Hokstad et al. (1993). Water contents of the emulsions were noted at 0.5, 1, 2, 4, 6, 8, 12 and 24 hours and the data used to calculate an uptake t½ [defined by Hokstad et al. (1993) as the time needed to pick up half the maximum water content]. Water droplet sizes in the emulsion were determined with an Olympus microscope (Model 1X50) equipped with an image analyzer system and software (Image Pro Plus; Silver Springs,MD). Both white and blue fluorescence light were used in determining water droplet diameters.

Addition of emulsified and non-emulsified oil to mesocosms
The mesocosms used were 1m deep with a volume of 300 m3 and each contained approximately 1000 kg of estuarine sediment (Skidaway River estuary). A water-in-oil emulsion (360 ml of emulsion) made from Kuwait crude oil as described above was added to a 30 cm x 30 cm section in the center of the mesocosm. To a second mesocosm, non-emulsified Kuwait crude oil (360 ml) was added. A third mesocosm, where no oil was added, served as a control.

Analysis of PAHs in sediment
Sediment cores were taken from the control mesocosm and the oil treated mesocosms at Days 1, 14, 28, 42, 56, 70 and 84. Cores were sectioned at 0-3 cm and 3-6 cm. Procedures for the analysis of sediment sediment cores for PAHs were similar to those described earlier (Maruya et al., 1997). Sediment samples (10g) from each core were freeze dried, extracted with methylene chloride and extracts passed through a silica gel column. The PAH fractions from the silica gel column were analyzed with a Hewlet Packard 6890 Series Plus gas chromatograph coupled to a 5973 mass spectrometer. Twenty three individual PAH analytes were quantified with a detection limit of 10 ng/g for each analyte. Except for naphthalene, recovery of spiked PAH standards ranged from 89 to 95%, while naphthalene recovery ranged from 55 to 67%. Procedural blanks, NIST sediments, spiked matrices and replicate samples were analyzed using the same procedures outlined above. For PAH spiked sediments, recoveries were between 90-100%.

Grass shrimp reproduction assay
For this assay, approximately 1 kg of sediment taken from each mesocosm (control, treatment with emulsified Kuwait crude oil addition and treatment with non-emulsified Kuwait crude) was subdivided into three parts and each part added to one of three aquaria, followed by addition of estuarine water (40 L to each aquarium) and grass shrimp (20 shrimp to each aquarium) collected from the nearby estuary (Skidaway Island, GA). Sediments for testing were taken on Days 14, 36 and 58. The salinity of the water was 25-27 ppt and the temperature of the aquaria was maintained at 27oC with heaters. The grass shrimp were fed frozen Artemia sp. and kept under a l2 h light/12 h dark regime. Every 5 days the following parameters were determined in each aquarium: (a) mortality; (b) number of females with mature ovaries; and (c) number of females with attached embryos. Grass shrimp were exposed to sediments for 58 days with no water change. An egg bearing female from each of the three aquaria were removed and 48 embryos (stage 8) from each female were transferred to 48 well polystyrene plates with each well containing 1.1 ml of estuarine water and 1 embryo. These culture plates were kept in the dark at 27oC and the number of embryos hatched determined by daily examination of the plates under a dissecting microscope. Hatching was generally completed within 48 h after transfer to culture plates. Data is reported as mean ± standard deviation (n=3; 3 aquaria for each test sediment)

Pore water assay
Pore water was collected from oil treated and reference mesocosm sediments (Days 1, 7, 14, 28, 49 and 87) by centrifugation at 1500 x g. Approximately 3 ml of pore water was collected from 40 g of sediment. Forty eight stage 7 (undergoing organogenesis) embryos [Winston et al. (2004) described grass shrimp embryo stages] were used for the pore water assay. Grass shrimp embryos were removed with forceps from egg bearing females. One grass shrimp embryo (24 embryos for each time period) was placed in each well of a polystyrene tissue culture plate with each well containing 1.1 ml of diluted sediment pore water [1:10 diluted with filtered (0.25 µm filter) estuarine water]. Plates were kept at 27oC in an incubator in the dark. Hatching rates [percentage of embryos which hatched into the zoea stage (stage 12)] were determined by daily examination of embryos in the plates under a dissecting microscope.

Comet assay for pore water assay
Twenty embryos (Stage 7) from the same female used for the hatching test described above were added to a beaker containing 20 ml of diluted pore water (1:10 diluted with estuarine water) from each mesocosm and incubated at 27oC for 24 hours. The embryos were removed from the beaker after 24 hours and used for the comet assay. The procedures for the comet assay were a modification of those described by Singh et al. (1998) and Steinert et al. (1998). Embryos were placed in a microcentrifuge tube on ice, followed by addition of 1 ml of 4oC filtered estuarine water to each tube. Embryos were homogenized in a 1 ml ground glass tissue homogenizer and left to settle at 4oC for 5 min. During this settling, individual cells remained in suspension while intact tissue pieces and fragments sank to the bottom. A portion of the supernatant (800 µl) was transferred to a second microcentrifuge tube, followed by centrifugation at 14,000 x g for 2 min at 4oC. After removal of the supernatant, the pelleted cells were resuspended in 70 µl of 0.65% low-melting point agarose in Kenny’s salt solution (0.4 M NaCl, 9 mM KCl, 0.7 mM K2HPO4, 2 mM NaHCO3), cooled to ~25oC after melting. This suspension was pipetted onto a slide coated with 1% agarose and a coverslip applied. The agarose was allowed to harden on ice-cold trays for 3 min. Slides were transferred to lysis buffer (2.5 M NaCl, 0.1 M EDTA, 0.01 M TRIS-HCl, 10% dimethyl sulfoxide, 1% Triton X100, pH 10). After incubation overnight at 4oC, slides were rinsed with water (4oC). After this water rinse, slides were placed in unwinding buffer (0.2 M NaOH, 1 mM EDTA, pH>13.5, 4oC) for 15 min. Electrophoresis of slides in the unwinding buffer was carried out at 300mA and 25V for 20 min. At the end of the electrophoresis run slides were neutralized by rinsing them in 0.4 M TRIS-HCl (pH 7.5) and gels were fixed by rinsing in 100% ethanol (4oC) for 5 min. Once dry, cells were stained with 20 µl of ethidium bromide (20 µg/ml). DNA strand breaks were determined by quantitation of fluorescence (excitation at 540 nm) in the head and tail of 50 randomly chosen cells from each replicate slide using an inverted fluorescent microscope (Nikon Eclipse E400), high sensitivity charged coupled device (CCD) camera and an image analysis system (Komet 4.0 - Kinetic Imaging Ltd.; Nottingham, UK). Data is reported as the mean (percent of DNA in the tail) ± SEM (n=5) where 50 cells/slide were counted. Datasets were compared using 1-way ANOVA followed by Tukey’s post-test. Statistically significant differences were expressed as P < 0.05 or P < 0.01.