Welcome to the New Restek.com! Take a tour of the new site and feel free to send us your feedback.

Milk Matters – Using QuEChERS extraction approach with GCxGC determination

  • Michelle Misselwitz
  • #Blogs
  • Share:

Last week I had the honor to present my work at the 31st International Symposium on Halogenated Persistent Organic Pollutants (Dioxin 2011). While it is still a work in progress, Jack Cochran and I evaluated the QuEChERS extraction approach with a silica SPE cleanup for PCBs and BFRs in cow milk and human breast milk. The QuEChERS extraction used hexane:acetone (1:1), as an alternative extraction solvent to acetonitrile, and the QuEChERS Unbuffered packets .  The hexane:acetone provided a good extraction of the fat in the cow milk samples, which is important since both PCBs and BFRs are lipophilic. To remove the coextracted fat, part of the QuECheRS extract was loaded  on a 500mg Silica SPE cartridge and eluted with 5 mL hexane, which gave good recoveries of the PCBs and BFRs in cow milk (see table).  The samples were analyzed by comprehensive two-dimensional gas chromatography (GC x GC) with electron capture detection (ECD).  The use of a 30m x 0.25mm x 0.25µm Rxi-XLB in the first dimension and a 1m x 0.15mm x 0.15µm Rxi-17Sil MS in the second dimension allowed simultaneous determination of both the PCB and BFR compounds of interest. (View more GC x GC products here) The ECD was used for two reasons: first it allowed low pg detection of the multiple halogenated compounds, and second, it keeps the Cheap in the QuEChERS approach.

Everything seemed to be going pretty smoothly (or so we thought), so we tried this methodology on a NIST Certified Reference Material of fortified human breast milk (NIST SRM 1954).  The fat content was equivalent to the fat content in the cow milk, so no adjustments were made in the cleanup.  Analyzing these samples on the GCxGC-ECD gave results that were about 25% off from the certified values (removing two outliers that were biased high from coelutions).  We wanted to see if we could get even closer results by using GCxGC-TOFMS and all I can say is wow! Sometimes they say ignorance is bliss, using the selective ECD we thought all was well, but when using a more universal detector, the fatty acids reared their ugly heads (see image below)! Later I found out that even though the fat content is essentially the same for cow and human milk, the fatty acid content is not.  There is a much higher concentration of conjugated linoleic acids in human milk that aid in the brain development of the infant. So now we are trying to add a quick PSA cartridge SPE cleanup in addition to the silica cleanup to remove the fatty acids in human milk.  Sometimes when you think you are at the end of the road, there is a bend and you see you have much farther to go!  So I will continue this journey and hopefully we will get to the end with a method that is environmentally friendly, multi-residue, quick, easy and cheap! (One can dream, right?)

If you are new to GC x GC check out the Restek GC x GC Column Combination Guide!

Separation of BFRs and Aroclors using Rxi-XLB and Rxi-17Sil MS
Separation of BFRs and Aroclors using Rxi-XLB and Rxi-17Sil MS

Average (n=3) % recoveries of spiked cow milk
Average (n=3) % recoveries of spiked cow milk

Fortified Human Milk SRM 1954 on GCxGC ECD and GCxGC TOFMS


Tue, Sep 06, 2011

Hi Michelle, very nice. What ECD did you use here? I recall not so long ago some ECD work on a capillaries showed in "tailing" because of the design of the ECD-Cell. That made me think that very fast peaks may be also a challemnge. The ECD peaks look very nice, so I assume thereisi no issue at all using this type of ECD for GCxGC work

Tue, Sep 06, 2011

Hi Jaap, Thank you! That is a very good question. The LECO GCxGC-ECD system uses an Agilent 6890 with a µ-ECD. You are correct that a larger ECD cell would cause some peak distortion of the very narrow peaks from GCxGC.