The glassware preparation and mercury analysis methods are modifications of U.S.EPA Methods 1669 and 1631. Briefly, precipitation samples are collected and stored in 2 L, borosilicate glass bottles with Teflon-lined, phenolic resin caps. Bottles are first rinsed with 0.2M BrCl, then soaked in 30% HCl for 24 hours, and thoroughly rinsed with deionized-distilled water and dried. Each bottle then receives 20 + 0.5 mL of 0.12 M HCl (Hg < 0.5 ng/L). The bottles are enclosed in new polyethylene bags, and packed into polyethylene foam-lined shipping containers. The funnels and capillary tubes are cleaned by soaking in 30% HCl for 24 hours , followed by rinsing in deionized-distilled water and then dried in a clean oven at 100 degrees C for about 2 hours and cooled in a laminar flow hood. The funnel and capillary tubes are placed in separate new polyethylene bags and packed in the shipping container. These sampling supplies are sent to the MDN site operators each week.
Upon arrival at the laboratory, the sample bottles are unpacked in the clean air station, and low-mercury (< 0.05 ng/mL), 0.2 N BrCl in HCl reagent is added to each bottle to give a final BrCl concentration of 1%. This reagent oxidizes all of the mercury present in the sample to divalent mercury. The caps are replaced, and the bottles are shaken for at least 4 hours to remove adsorbed mercury from the bottle walls and to fully oxidize any suspended particles.
Weighed sample aliquots (50-100 mL) are poured into 125 mL Teflon bottles prior to analysis. 200 uL of 20% hydroxylamine hydrochloride is added to each aliquot to eliminate free halogens, and then the aliquot is immediately poured into a purge vessel. To reduce the divalent mercury back to Hg0 300 uL of 25% SnCl2 are added, and the sample is purged with ultra-pure nitrogen onto a gold-coated, silica trap. Mercury is removed from the traps by thermal desorption, concentrated and focused by dual gold trap amalgamation, and analyzed cold vapor atomic fluorescence. Quantification is by peak height. Average blank values are determined for each analytical run and subtracted from the total to determine sample mercury concentrations. The method detection limit for a 100 mL sample is approximately 0.1 ng/L (3 standard deviations of the reagent blanks).
Routine quality assurance procedures include analysis of laboratory bottle blanks, field blanks, and system blanks. Bottle blanks consisted of the preservative pre-charge solution (see sampling methods) added to the bottle with no exposure in the field. Field blanks are compiled from regular sampling weeks during which there is no rainfall and no sampler openings. The pre-charge solution is exposed to conditions in the sampler without receiving any precipitation and analyzed in the lab as a normal sample. Occasionally, the sampling train is rinsed with 300 to 600 mL of low-mercury deionized water after a week with no rain or sampler openings in order to evaluate passive contamination of the glassware during dry periods. These samples are referred to as system blanks and include both the bottle blank and any contamination picked up during dry exposure of the sampling train in the field.
Weekly analysis of a standard reference sample with a mercury concentration of 4.64 ng/L are run, along with weekly laboratory spike recovery tests.