This topic came up a while ago in a discussion about vinyl chloride in groundwater. A research hydrologist at a government agency offered some advice. I’ll paraphrase here for brevitiy:
As a gas, VC losses can be substantial if the container is not gas tight. Likewise sorption to stoppers can be a problem, so teflon coated stoppers are recommended. …flat septa (may be) unreliable for long-term storage of water samples for VOC analysis. Use teflon coated stoppers (Wheaton manufactures one) and crimp seals. Likewise, it takes experience with sealing VOA vials and personally analyzing those samples to recognize the vulnerability of such containers to poor container closure techniques, a particular concern for contractor collected samples. In other words, it is very easy to get an analytical result with little appreciation for the quality of that analytical result.
…volatile samples are best collected using modifications of techniques developed for collection of environmental oxygen analyses. The collection outlet (outflow tubing for pumped samples) should be placed in the bottom of the VOC container and the sample allowed to overflow for 3 times the container volume. Then the tubing is carefully removed, leaving an undisturbed convex meniscus of water. Preservative is added to the top of the meniscus with a dropper.The sample is then closed carefully.For VOC samples the volume of sample required to fill the container once is always compromised by off-gassing.
While our discussion was on the topic of preservation his point was that the way samples are collected can have a large affect on the results. It would be best if you consult with someone experienced in collecting samples for dissolved gasses before you start.