Very low concentration measurements of
sulfur hexafluoride and chlorofluorocarbons in air collected from firn and ice.
(update 12/2003)
Recent studies on cristaline rocks (Harnisch and Eisenhauer, 1998) have revealed the natural
occurrence of the trace gas sulfur hexafluoride (SF6) as well as some other trace gases which
so far have believed to be entirely of anthropogenic origin. This discovery has been confirmed by
the findings of high concentrations of SF6 in old groundwater (Busenberg and Plummer, 2000; own work).
Using electron capture (ECD) gaschromatographic techniques we are able to detect very low
concentrations of SF6 and CFCs in air samples. We have done this also for antarctic
firn air samples provided by David Etheridge and Paul Fraser (CSIRO, Melbourne).
Here are some analytical details for an old firn air sample which I analyzed on a gas
chromatograph at the Scripps Institution of Oceanography. The chromatogram below shows
peaks for SF6 and CFC-12. Injection is 130 ml of firn air which is cryofocused on two consecutive traps.
The results for this analysis are concentrations for SF6
of 6.3 ppq (parts-per-quadrilion or parts in 10E15) expressed as dry air mole fraction
and for CFC-12 of 1.1 ppt (parts-per-trilion or parts in 10E12). This sample is from Law Dome
and dates back to about 1935. At that time CFCs were already released to the atmosphere by
human activities and the observed concentration might be entirely of anthropogenic origin.
SF6 on the other hand is believed to not have been emitted to the atmosphere in large
quantities until the early 1950s
which means that the measured concentration may partially be of preanthropogenic origin.
The figure below can also be downloaded as PDF file.
In addition to the samples provided by David Etheridge we have
also analyzed firn air samples provided by James Butler (CMDL/NOAA) which were collected
at South Pole in 2001. Even though the air
in these samples is about 100 years old, deriving a natural
SF6 concentrations is hampered by contamination of the sample with ambient air during collection
as assumed from the measured CFC concentrations.
However assuming that the contamination of the samples is of typical SH background CFC and
SF6 signatures at the time of collection
our analysis shows residual SF6 values which are close to those measured in the Law Dome sample.
In summary we have not been able to prove the absence of CFCs or SF6 in preanthropogenic air. However
recent analysis of large volume air samples from outcropping ice in Greenland (samples
provided by Vas Petrenko and Jeff Severinghaus, SIO) which dates back to the Holocene,
seem to confirm our earlier findings of a
preanthropogenic SF6 concentration of about 5 to 20 ppq.
We are still looking for some cleanly sampled old
firn air samples. We have not found any samples in which the CFCs were absent. This is probably
due to some sampling contamination but also, the possibility of preanthropogenic CFC signal
should not be excluded.
For the analysis of air samples on the SIO SF6 GC, we estimate analytical detection limits
of 1.4 ppq for SF6, 0.04 ppt for CFC-12, 0.03 ppt for CFC-11 and 0.08 ppt for CFC-113.
These detection limits could probably be improved on by injecting larger air samples.
The analytical technique is described in detail by
Vollmer, M. K. and Weiss, R. F., 2002.
Simultaneous determination of sulfur hexafluoride and three
chlorofluorocarbons in water and air.
Mar. Chem., 78, 137-148.