EPA 8270 with Appendix IX: 145 Semivolatile Compounds in 17 Minutes

 

Using an Rxi-5Sil MS 30m x .25 x .25 and a Restek Premium single taper WITH WOOL inlet liner, I ran an eight point 8270 calibration including the Appendix IX compounds. The calibration range was 0.5 µg/mL - 120 µg/mL, with individual points at 0.5, 1.0, 5.0, 10, 20, 40, 80, and 120 µg/mL. The attached Excel spreadsheet lists all the raw responses and relative response factors as well as compound by compound linearity evaluations (as %RSD) and is color coded to show failures based on 8270C and 8270D calibration criteria. SPCC and CC compound criteria were taken into account for 8270C evaluations.

The spreadsheet details compound linearity evaluations for four different ranges: 0.5 – 120, 0.5 – 80, 1.0 – 80, and 1.0 – 120. Only 7 of 138 target compounds did not meet 8270D %RSD linearity requirements in one of the four calibration ranges. Two additional compounds (a total of nine compounds) did not meet 8270C %RSD linearity requirements. By dropping high points and using linear regression as instructed in 8270C and D, method 8000 linearity criteria was met for all nine of the compounds. You can view the linear regression lines by clicking on any of the following chemical names: benzaldehyde, o-toluidine, benzoic acid, 2,4-dinitrophenol, 4,6-dinitro-2-methylphenol, diallate, 1,3,5-trinitrobenzene, atrazine, and 4-nitroquinoline-1-oxide.

The chromatogram on display at the top of the article shows a standard run at 10 µg/mL (IS/SS at 20 µg/mL). The peak shapes are excellent, with the vast majority of the compounds showing symetric peaks and little to no tailing. I've zoomed in on a few examples: Pyridine (m/z 79), 2,4-dinitrophenol (m/z 184) benzo[b]fluoranthene and benzo[k]fluoranthene, indeno[1,2,3-cd]pyrene and dibenz[a,h]anthracene. Notice the minimal tailing on the last few PAH compounds in the chromatogram section below, starting with 3-methylcholanthrene and ending with benzo[ghi]perylene.

For the really curious, I've uploaded detailed quant reports for the 10 µg/mL, 40 µg/mL and 120 µg/mL standard runs so you can how the peak shapes change as the column approaches overload. The calculated concentrations are not correct (the quant database hasn't been updated) but the integrations should be.

While the last compound is eluting at 17 minutes, the full instrument cycle time is closer to 26 minutes. The instrument run log shows that 15 injections were completed in just under six and a quarter hours. This should leave ample room for a full 20 sample batch as well as calibration and QA runs within the 12 hour tune window. Instrument acquisition parameters can be found here, and if you are running Chemstation ea.02.00 sp2 or later, you should be able to import this zipped method file.