Polycyclic aromatic hydrocarbons (PAHS) are carcinogenic, mutagenic, and teratogenic environmental contaminants. But not all PAHs are created equal, so the European Food Safety Authority (EFSA) suggests analysis of the PAH4, benz[a]anthracene, chrysene, benzo[b]fluoranthene, and benzo[a]pyrene, to represent the majority of the toxicity in any food sample. Unfortunately, some of these PAHs coelute with other isobaric PAHs on typical “5% phenyl-type” and other GC columns leading to erroneous quantification and overestimated toxicity. Two of the most common coelutions are triphenylene and chrysene (m/z 228) and benzo[b] and benzo[j]fluoranthenes (m/z 252). This led Restek to develop the Rxi-PAH GC column, which separates the EFSA PAH4 and other important PAHs.
While the “5% phenyl-type” is the most common GC column for environmental priority pollutant-type analysis, including for the 16 PAHs suggested for analysis by the US EPA, for those studies that require a different selectivity and higher efficiency, the 60m x 0.25mm x 0.10µm Rxi-PAH GC column may be an option. Don’t expect ultra-fast analysis times, but the separations are excellent for comprehensive congener-specific PAH analysis (CCSPAHA).
In fact, they are so good in one-dimensional GC that I can report here the first simultaneous GCxGC separation for triphenylene and chrysene, and benzo[b]fluoranthene, benzo[k]fluoranthene, and benzo[j]fluoranthene. I even threw in benzo[a]fluoranthene as a bonus. The triphenylene and chrysene separation, especially, isn’t trivial here, but with a True Peak Capacity Increase GCxGC approach, i.e. a longer first dimension GC column, optimum flow rate in the first dimension, optimal GC oven program rate, a relatively short second dimension column, and fast modulation times, I’m able to pull it off!
To illustrate the point about quantitative error on PAHs, I use a Spring Creek soil extract that contains PAHs (kindly provided by my colleague Frank Dorman at Penn State University) and contrast the results using an Rxi-5Sil MS and those from the Rxi-PAH GCxGC analysis. The Rxi-5Sil MS, like all 5% phenyl-type columns has a triphenylene/chrysene coelution and a benzo[b]fluoranthene/benzo[j]fluoranthene coelution, leading to high bias for chrysene and benzo[b]fluoranthene. The Rxi-PAH, in contrast, provides accurate numbers because it separates the isobaric interfering PAHs.
Next up: one dimensional Rxi-PAH work on environmental samples.