A Better Way for VOCs in Soils?

Martin Perkins

16th February 2016

Dan Carrier, Methanolic Extraction, Purge and Trap, Richard Davis, Soil, Static Headspace, Video, VOC Analyser, VOCs, Volatile Organic Compounds,

Virtually every lab (at least in Europe) running volatile organic compounds in soils does so by static headspace.

Over the years, Element has sold many VOC analysers for soil analysis but we have always had doubts about the usual way of doing things.

The standard way of working is to weigh a portion of soil into a vial, then add water and salt.  The idea is that the salt solution removes the VOCs from the soil and the VOCs partition into the headspace.

The problem with this approach is that the aqueous phase has two conflicting requirements:

  1. It needs to have a high affinity for VOCs to free them from the solid matrix.
  2. It needs a low affinity for VOCs to enable them to partition strongly enough into the headspace.

In the real world, a compromise is always struck, but we have always felt that there is a better way to get dependable data.

In the USA, where purge and trap sampling is widely used for soils analysis, methanolic extraction is a common approach.

Soil samples are extracted with methanol (which has a high affinity for organic volatiles), and then an aliquot of the extract is added to water in a VOA vial.  The methanol disperses in the water, resulting in a liquid phase with low affinity for the VOCs, which then partition nicely into the headspace.

This is a more elegant approach, but attempts to adapt methanolic extraction to static headspace sampling have failed due to the limits of detection being a bit too high for UK requirements.

With the advent of the current Agilent 5977A GC-MSD, more MS sensitivity is available and so the methanolic extraction of soils should now work ok.

What is more, we now have all of the tools needed to automate the whole process.  This could well make automated methanolic extraction for VOCs in soils (and some other solid matrices) a better option than the three-phase headspace methods that are currently ubiquitous.

Dan Carrier and Richard Davis configured a system for automated methanolic extraction, built a method and have run some tests to see if they could hit the 2ppb LOD that would be needed for this to be viable.

The 2ppb limit for VOCs in soils was achieved, with very little optimisation.  If you follow this link, you can see a video of the methanolic extraction method running.

This preliminary work and more needs to be done.

We think this is very promising.  What do you think?