The SF6 mixing ratio in air and the solubility of this gas in water is low. A 15 cm3 sample of air or unsaturated zone gas is needed for gas analysis. A minimum of a 1 liter of water is required to date ground water with SF6.
Bottle Ordering InformationA true 1-Liter sized, plastic safety coated, amber glass bottle with a polyseal cone lined cap is used.
You can order these 1-Liter bottles from Qorpak Inc. at 1-800-922-7558. Catalog number GLA-00959 is for a case of 12 bottles or catalog number GLA-00958 is for a case of 30 bottles.
The caps can be ordered from Brad-Pak Enterprises at 908-233-1234. Catalog number CPS-33E-B-PC-PS is a 33-430 size polyseal cone lined cap which fits the bottle.
Field Collection of SF6 Water Samples
- Purge well
- Place tubing from pump in the bottom of 1L bottle Additional Bottle Info and Pictures
- Fill bottle and allow it to overflow from the neck. Allow three total liters of sample water to flow through the bottle with the tubing at the bottom of the bottle.
- Slowly remove tubing from the bottle while water is still over flowing
- Cap bottle tightly (do not leave any headspace) and tape cap in a clockwise direction with electrical or similar tape.
- Collect two bottles per site.
- Keep bottles in a cooler and not in the sun. If the bottles heat up, the water will expand and may crack the bottle.
Gas Sampling Procedures
Gas samples can be collected in stainless steel cylinders that are equipped with inflow and outflow stainless-steel bellows valves. This design allows the cylinders to be flushed by the pumped air stream. The procedures that are used to collect gas samples in stainless steel cylinders are described in detail by Busenberg et al. (1991) and Thompson et al. (1985). The cylinders were filled in the laboratory with ultra-pure SF6-free N2. The N2 in the cylinders is released prior to sampling by opening the outflow valves. The inflow valves are then opened to allow air flow through the cylinders for several minutes to purge the N2. After the cylinders are completely flushed with air, the outflow valve is closed and pumping continues to pressurize the cylinder to about 2 atmospheres before closing the inflow valve.
Gas samples can also be collected in 100 mL borosilicate glass ampoules using procedures developed for sampling for CFCs (Busenberg and Plummer, 1992). The gas samples are heat-sealed in borosilicate glass ampoules with a torch.
Shelf Life of SF6 Samples
No changes in concentrations of SF6 were observed after storage of 3 months.
Modification of SF6 by Microbial Activity
No perceptible degradation of SF6 was observed in highly reducing waters.
Unsaturated Zone Processes
When the unsaturated zone is relatively thin, the unsaturated-zone air composition tracks that of the troposphere (Oster et al., 1996). It is reasonable to assume unsaturated-zone SF6 concentrations closely track tropospheric concentrations to unsaturated zone depths of less than 10 m (Weeks et al., 1982; Busenberg et al., 1993), in deeper unsaturated zones, there is a lag time for diffusive transport of SF6 through the unsaturated zone. The time lag is largely a function of the tracer diffusion coefficients, tracer solubility in water, and soil water content (Weeks et al., 1982; Cook and Solomon, 1995).
Dissolved Gases in Ground Water
It is recommended that some dissolved gas analyses be performed to determine the recharge temperature and the amount of excess air present in the ground waters. This can be done by submitting samples to the N2 / Ar dissolved gas lab.
Uncertainty in recharge temperature of 1 to 2°C introduces no significant error in the SF6 model ground water age (±0.5 year) because of the rapid increase in the atmospheric mixing ratio of SF6 of about 7 % per year (Geller et al., 1997).
Excess air is introduced into ground water when air bubbles dissolve during a rapid rise of the water table. The addition of excess air to ground water increases the SF6 concentration of the groundwater above the air-water equilibrium concentration. If the presence of excess air is not considered in the calculation of an SF6 model age, then the apparent age will be too young. If the excess air present in the ground water was not known, and was underestimated by one cm3/kg of water at STP for typical U.S. ground waters, the age of the ground water will be under estimated by 1 to 2.5 years. In all cases, the error in the apparent date of recharge was higher for the waters that were recharged at the higher temperature. Excess air concentrations of 0 to about 2 cm3/liter were found from N2-Ar measurements on shallow groundwater recharged by aerial infiltration through sandy soils, however, concentrations as high as 10 cm3/liter can be found in ground waters from some semi-arid regions.