The Aquatic Cycling of Mercury in the Everglades (ACME) project is a large multidisciplinary effort that is focusing on the processes that lead to mercury contamination of predatory game fish in the Florida Everglades. As such, this project required the establishment of a mercury analysis laboratory capable of low level mercury (Hg) analysis for water, sediment, and biota. In addition, to fully understand the major fate and toxic controlling process of mercury in the environment, it is necessary to perform these low-level analysis with speciation. However, aqueous samples from the Everglades present several problems for the analysis of total mercury (HgT) and methyl mercury (MeHg). Surface water samples are analyzed for HgT, reactive Hg, MeHg, and dissolved gaseous Hg (DGM) while porewater and biotic components (plankton, insects, zooplankton, fish) are analyzed for HgT and MeHg. Constituents such as dissolved organic carbon (DOC) and sulfide at selected sites present particular challenges due to interferences with standard analytical techniques. This is manifested by 1) the inability to discern when bromine monochloride (BrCl) addition is sufficient for sample oxidation for HgT analysis; and 2) incomplete spike recoveries using the distillation/ethylation technique for MeHg analysis. In our study, we found that for samples with DOC concentrations above 35 mg L-1, BrCl addition is insufficient for complete oxidation of HgT. Also, during high-water periods and anoxic conditions, porewaters and eutrophied surface waters have resulted in low spike recoveries for MeHg analysis. Thus, we present additions and modifications to current HgT and MeHg methodologies to deal with challenges associated with Everglades samples.

To overcome the incomplete oxidation of water samples, we devised a ultraviolet (UV) oxidation box for pre-oxidation of samples. The UV light box consists of three Spectroline X-series UV lights in a plastic box lined with aluminum foil for reflective purposes. Two of the lights are placed to shine directly into the sides of the sample bottles while a third is mounted on the inside of the lid of the box. Up to ten 500 ml bottles can be treated in the UV box at a time. Water samples from the Everglades generally require two to five days of pretreatment with UV light to remove the visible color. Once the samples clarify, the samples are oxidized using blanked BrCl (< 15 ng L-1) and placed into an oven at 50 oC overnight. Results from the UV box oxidation show an average increase in HgT concentration of 17 percent in the UV treated samples relative to the non-UV treated samples. This indicates that BrCl alone is not sufficient to release all of the dissolved Hg from organic matrices or ligands in the sample. Based on this evidence, all aqueous HgT samples from the Everglades for the ACME project are oxidized using the two step UV box-BrCl digestion.

Anoxic samples collected during high water periods from eutrophic areas of WCA 2A have shown the effects of a matrix interferant. Methyl mercury analysis of these Everglades water samples showed reduced spike recoveries between 25 to 35 percent. A temporary solution to the problem was first implemented by collecting a bulk sample, analyzing it and testing for the spike recovery interferant. If the interferant was present, the samples was then distilled in 50 ml aliquots rather than the usual 100 ml, and CuSO4 was added as a reagent in the distillation. If the interferant was not present, the sample was distilled and analyzed using typical 100 ml volumes.

Sulfide has not been shown to interfere with MeHg analysis because it is assumed that it is converted to hydrogen sulfide (H2S) after acidification (addition of H2SO4 during distillation setup) and lost during distillation [9]. Everglades samples are frequently high in sulfide, with concentrations as high as 300 mM observed in the porewaters of WCA 2A [6]. Since a noticeable H2S odor was present at the sites with spike recovery problems, an experiment was performed to determine the effects of H2S on MeHg spike recoveries. An Everglades surface water sample collected from site U3 in July 1996 with a MeHg concentration of 0.515 ng L-1 was split into five aliquots and the sulfide concentration was increased by 0, 100, 200, 300, and 500 mM using a saturated sodium sulfide (Na2S) solution. Samples were allowed to equilibrate overnight. Prior to distillation on the next day, samples were spiked with 50 pg of MeHg standard. An identical set of samples was also treated with CuSO4 (2 ml of 1 M) prior to distillation. Since CuSO4 was previously used to separate MeHg from matrices [20], it was added to the Everglades samples in this experiment. The sulfide-treated samples were analyzed for spike recoveries comparing them to the original sample concentration. Spike recoveries dropped from 118 to -110 percent with increasing sulfide in the samples not treated with CuSO4. The negative percent recoveries result from incomplete recovery of the original sample concentration. Meanwhile, the samples treated with CuSO4 yielded acceptable percent recoveries up to a sulfide addition of 300 mM. However, even with the addition of CuSO4, the sample spiked with 500 mM of sulfide yielded a spike recovery of only 67 percent.

These modifications to accepted analytical techniques represent substantial advances for these high DOC and occasionally sulfidic waters. These adaptations maybe be important for other wetland systems and porewaters.

Back to Project Homepage