Hello All--

I came up with a novel sampling apparatus. You do need a thermal desorption system (Markes, Gerstel, etc.) to make it work.

I bought some silicone tubing (VWR scientific, Sani-Tech, Cat# STHT-C-020-0). I cut 3 cm lengths and laced them onto some stainless steel wire (o.d. 0.020"). I cut the gizmo to 6 cm lengths. I conditioned them by baking them at 300 °C for 3 hours prior to use (N2 flowing around the devices using a Markes TC-20, 50 cc/min. flow of N2 through each channel of the system). The idea is you want to be able to poke the wire end through a Teflon-backed rubber septum to hold them in place during sampling liquids (or vapor space above solids) in a standard 22 mL headspace vial.

Sampling:

1) 20 mL of liquid in a 22 mL, square-bottom vial. Stir bars rotate more smoothly with square-bottom vials.
2) Add a Teflon-coated flea stir bar to each vial.
2) Poke hole in Teflon-backed septum with a syringe and poke the metal end of the sampling device through it so that the septum will hold the sampling device in place during sampling of the liquid.
3) Crimp the septum to the top of the vial (aluminum closure). Silicone tubing on the sampling gizmo completely immersed in the liquid.
4) Stir at 1,000 rpm for 1 hour at room temperature.
5) After 1 hr, removed the device and rinse with DI water. Pat dry with a Kimwipe and place in empty thermal desorption tube. Analyze.

My system is a Markes TD-100xr coupled to an Agilent 7890/5977B GCMS system. I have the chromatographic system configured to do heartcutting via an Agilent Dean's switch from nonpolar-to-polar column but in this case, I'm just cutting the entire precolumn separation to the analytical column so it's like a system that has a 30 m HP-5 coupled to a 30 m DB-Heavywax column (end-to-end).

I ran 3 unspiked samples (a 5% ethanol-based solution) and 3 standards (same beer fortified with 16, 33, and 66 ppbv [ng/mL] linalool). 6 samples total. The results are summarized in the graphic:

https://i.postimg.cc/9MqG2sk2/Linalool- ... phic-I.jpg

I did have to sort the TIC by m/z=93 to see a signal for linalool that I could integrate in the unfortified samples. The concentration is low in this sample.

If you don't want to buy some other type of sampling devices, you can perhaps get by less expensively using this type of gizmo. It worked surprisingly well. I was able to determine ~1 ppbv (ng/mL) of linalool in my sample. I don't see a reason that it wouldn't work for sampling the headspace as well. It's like a big SPME fiber. It was surprisingly easy to condition the silicone tubing to be substantially free from volatile materials.