Sample Containers & Preservation for Mercury Analysis in Waters
By Tim Crowther, Regional Client Services Manager, ALS Environmental – Canada
On August 15, 2013, ALS Canada will begin supplying our clients with borosilicate glass containers with Teflon® lined caps for the collection of total and dissolved mercury in all water samples. Hydrochloric acid (HCl) preservative will also be supplied. Recent literature and ALS experimental test results indicate a glass container with HCl preservation is the most effective method for reducing mercury losses following sample collection. The sample bottle and preservative pictured overpage will be the recommended container for low-level total and dissolved mercury (≥10 ng/L), which were previously collected in a high-density polyethylene (HDPE) ‘plastic’ bottle with nitric acid preservation. HDPE containers are not suitable for ultra-trace level (0.2 – 10 ng/L) mercury analysis. Ultra-trace mercury sampling requires more sample volume, as well as the use of cleaner sample handling and analysis procedures.
The British Columbia Ministry of Environment (BC MoE) will require the use of Teflon® or borosilicate glass containers with HCl preservation for the collection of water samples for mercury analysis effective November 15, 2013. The United States Environmental Protection Agency (US EPA) and the Ontario Ministry of Environment have already prescribed the same. Various other agencies are considering similar changes, including the Canadian Council of Ministers of the Environment (CCME) and Alberta Environment.
Additionally, ALS recommends that filtration for dissolved mercury analysis be conducted within one hour of sample collection using a suitable in-line filter or 0.45 μm syringe filter supplied by ALS.
Rationale Behind ALS Recommendation
Until recently, the most common container type and preservative used for mercury analysis of waters (with the exception of ultra-trace levels) was an HDPE plastic container and nitric acid. The U.S. EPA Method 245.11, which was first published in 1972, was the cited source for this recommendation. Even today, the EPA continues to recommend HDPE and nitric acid (HNO3) preservation for mercury at levels greater than 100 ng/L, although most recent literature now recommends better practices.
In 2002, US EPA Method 1631, Revision E2 was published for the determination of ultra-trace mercury with levels between 0.2 – 100 ng/L. The new “ultra-trace” level was established in support of the U.S. Clean Water Act, and also specified the use of borosilicate glass or Teflon® and the “clean hands-dirty hands” procedure for sample collection3. EPA Method 245.7 (2005) also requires the use of borosilicate glass or Teflon® and HCl preservation.
Over the past two decades, regulatory authorities recommended many combinations of container type and preservative for the collection of samples for mercury analysis, but there are several current studies which demonstrate that glass or Teflon® containers with HCl preservation provide the most reliable conditions for stabilizing mercury at all levels and water types.
Several studies over the last ten years have reviewed the loss of mercury from water in Teflon®, borosilicate glass, and HDPE containers4 . A significant decrease in mercury concentration is often observed in water from HDPE containers. However, mercury is stable in Teflon® and borosilicate glass at ambient and refrigerator temperatures, as long as the samples are preserved with (1:1) hydrochloric acid.
Recent publications have compared the effects of preservation on the analysis of mercury and other trace metal elements using inductively coupled plasma mass spectrometry (ICP-MS)5. These studies demonstrated that nitric acid was not sufficient to prevent significant losses of mercury at levels protective of human health levels (1 μg/L) in HDPE containers and that the addition of hydrochloric acid or chloride to the sample greatly increased the stability of the samples.
Further studies4 were performed with radioactive labeled inorganic mercury (197Hg) to trace mercury loss over time. In unpreserved lagoon and river water samples, the mercury concentration decreased as much as 50 percent after 12 days. The decrease was more pronounced at room temperature (20°C) than refrigerated conditions (5°C). Conversely, there was no significant decrease in labeled mercury concentration in unpreserved lake and marine waters at either room temperature or refrigerator temperature (5°C), apparently due to the presence of levels of chloride to stabilize the mercury.
Although there were no significant losses of mercury from lake and marine waters, the study identified that the chemical and physical properties of the water determined the stability of the mercury concentration in solution and for this reason, ALS recommends preservation with (1:1) hydrochloric acid for all types of water.
A study by Guevara and Horvat (2013) demonstrated that the rate of mercury transfer from solution to particulates was rapid and varied significantly among all the types of water studied. In all cases, at least 25 percent of labeled mercury transferred from the solution to the particulates within the first hour, reaching equilibrium in one to several days.
After filtration and preservation with hydrochloric acid, none of the natural or lagoon waters studied showed a decrease in mercury concentration throughout the 12 days of storage.
ALS now recommends the use of a pre-cleaned borosilicate glass container for the collection of all water samples requiring mercury analysis at all concentrations. If dissolved mercury is required, the samples should be filtered within one hour of collection using a suitable in-line filter or a 0.45 μm syringe filter supplied by ALS. All samples should be preserved with (1:1) hydrochloric acid.
For routine low-level mercury analysis (down to levels of 10 ng/L), a 40 mL pre-cleaned borosilicate glass container preserved with 0.5 mL (1:1) hydrochloric acid is required. This will be the required container type and preservative for mercury results submitted to BC MoE effective November 15, 2013.
For ultra-trace mercury analysis (levels between 0.2 and 10 ng/L), a 120 or 250 mL pre-cleaned borosilicate glass container is required. Additionally, ALS recommends that our clients use the US EPA “clean hands / dirty hands” sampling protocols when conducting ultra-trace mercury analyses3. Regardless of whether or not these sampling protocols are used, we strongly recommend frequent use of Field Blanks (i.e. Equipment Blanks) for ultra-trace mercury analysis, which is the only way to verify that trace level sources of contamination are controlled.
1 U.S. Environmental Protection Agency, Method 245.1, Revision 3.0 (1994)
2 U.S. Environmental Protection Agency, Method 1631, Revision E (2002)
3 U.S. Environmental Protection Agency, Method 1669: Sampling Ambient Water for Trace Metals at EPA Water Quality Criteria Levels (EPA 821-R-95-034)
4 H. Louie, C. Wong, Y.J. Huang and S. Fredrikson, Anal Methods, 2012, 4, 522
5 S.B. Guevara and M. Horvat, Anal Methods, 2013, 5, 1996-2006