Finally! A True Peak Capacity Increase for GCxGC

An ongoing controversy in comprehensive two-dimensional gas chromatography (GCxGC) is whether a true peak capacity increase has ever been achieved. You can read all about it in the article cited below, but in summary, Blumberg et. al state that a properly optimized one-dimensional (1D) GC system produces higher peak capacities than an improperly optimized GCxGC setup with the same primary column. Furthermore, they indicate that essentially every GCxGC report in the literature is on non-optimized systems. You see, we GCxGCers often sabotage our 1D separations to broaden peaks before they enter modulators so we can get at least 3 slices across first dimension peaks for second dimension (2D) separations. To some extent, this is because of speed limitations in modulation and detection technologies when GCxGC is done with typical 30m x 0.25mm 1D columns. That is, peaks are relatively narrow on properly operated 30m columns and that means very fast (and impractical) modulation cycles are necessary for GCxGC.

Comparison of one-dimensional and comprehensive two-dimensional separations by gas chromatography. L.M. Blumberg, F. David, M.S. Klee, and P.Sandra. Journal of Chromatography A, 1188 (2008), 2-16.

My colleague Mark Merrick of LECO Corporation and I decided to approach this problem another way: go slow! Perhaps it’s my Oklahoma drawl that gave me this idea…

We collected data on 120, 150, and eventually, 60m, columns that “naturally” produce broader peaks for modulation with today’s technology. We were able to meet (or closely approach) Blumberg’s suggestions for using Speed Optimized Flow and Optimal Heating Rate (see previous blog) such that our 1D separations were optimum, and then we modulated appropriately for the second dimension separation. To some extent, any separations achieved in the 2D can be considered “gravy” for the meat-and-potatoes 1D separation. Put that terminology in your GCxGC lexicon book…

An example of this is shown in the figure below by contrasting 15m and 120m primary column GCxGC separations. The 120m column is operated very efficiently and we maintain primary column separations, important given that the fog oil we analyzed is full of isomers. We’re also using the second dimension very nicely to spread things out, in this case with the orthogonal 2D column, a high thermal-stability, shape-selective stationary phase developed under Frank Dorman’s direction by Mike Wittrig. You can check out more on GCxGC of fog oil via the article below.

Characterization of military fog oil by comprehensive two-dimensional gas chromatography. A. Kohl, J. Cochran, and D.M. Cropek Journal of Chromatography A, 1217 (2010) 550-557.

Finally, and you regular readers knew this was coming. I have to tie peak capacity into my ongoing South African trip somehow and this time it’s through a leopard. The leopard’s coat is beautiful, with nicely patterned rosettes that are somewhat evenly spaced throughout. Think of those rosettes as peaks fully occupying a contour plot (the GCxGC chromatogram) and there you have the perfect use of peak capacity. Conversely, imagine the leopard with one diagonal line on its body (non-orthogonal column selection). Or how about lots of run-together spots in both dimensions (inefficient 1D separations)? Probably not near as purty, eh? Same in GCxGC!

Enjoy my Mala Mala leopard photographs by clicking on the thumbnails to expand and then use the browser back button.

Next blog: how to use a GCxGC Calculator to help maximize peak capacity in comprehensive two-dimensional GC.