Eight Tips for Easy GC×GC

I love to attend the Multidimensional Chromatography Workshops organized annually in Toronto by Eric Reiner, an early adopter of comprehensive two-dimensional gas chromatography (GC×GC), who works for the Ontario Ministry of the Environment (MOE). The workshop has over 150 attendees, an impressive number given that Eric arranges it by himself and that it takes place in Canada – in January. The meeting features an excellent range of GC×GC presentations, from theory to instrumentation to applications to data processing, in an informal atmosphere that promotes discussion and stimulates collaboration. The audience includes people who have never used GC×GC, but may be considering it to solve their own separation problems, for example, in environmental research where the matrices are complex and the residues are often present at trace levels.

Just over two years ago, while attending Eric’s workshop, I was sitting in the back row of the MOE auditorium with a person who pioneered the use of GC×GC with time-of-flight mass spectrometry (ToF MS) for the practical purpose of determining (and discovering) environmental chemicals for human biomonitoring. We listened to one of our colleagues describe the complexity of GC×GC method development in a lecture: two columns, two stationary phases, two lengths, two inner diameters, two film thicknesses, two flow rates, two oven temperatures, variable modulation times and modulator temperatures, two sample loading capacities... Oh, and all of those parameters may be independent of each other… I heard my colleague mutter, “No, no, no”, and when I looked over he was shaking his head. I asked him what the problem was and he replied that we – the community – should not present GC×GC as being that complicated to those considering whether to employ it (or not) for the first time. I emphatically agree!

In my view, GC×GC needs to be presented as simply as possible to increase the number of practitioners, or more appropriately put, to entice potential users to take advantage of its power for separation problem solving. I’m not talking about a disingenuous representation of GC×GC as a push-button technique. My suggestion is to recommend columns and operating conditions based on sound one-dimensional GC principles. Those principles include efficient carrier gas flow and optimal heating rate as presented by Leon Blumberg and Matthew Klee in several publications (1–3). So, here are eight simple tips on how to start method development for GC×GC-ToF MS, especially when doing broad semi-volatile compound screening work (for example, metabolomics, petroleum-omics, emerging environmental compound work):

1. Use 5% phenyl-type × 50% diphenyl-type or 50% diphenyl-type × 5% phenyl-type stationary phases. Or pick the set that offers the largest spread of compounds in first and second dimensional space.
   
2. For the dimensions of the first column, use 30m x 0.25mm x 0.25µm (a commonly used GC column in a wide variety of stationary phases). 60m length may be even better since it has higher peak capacity.
   
3. For the dimensions of the second column, use 0.60m x 0.25mm x 0.25µm. Use a non-restrictor, so that the first dimension separation is two to three times faster than if using 0.10mm. A column of 0.25mm also offers better sample loading capacity versus 0.10mm. If you selected a 60m first dimension column, use 1.3m x 0.25mm x 0.25µm in the second dimension.
   
4. Use helium as the carrier at a constant flow of 1–2 mL/min for efficient first dimension separation.
   
5. Use an oven program rate of 10°C / holdup time (minutes) – optimal heating rate maximizes first dimension peak capacity.
   
6. Use a modulation time of 1-3 sec (maximum). Slice the first dimension peak approximately three times to preserve the first
       dimension separation.
   
7. Use a modulator temperature offset of 15°C higher than the second dimension oven temperature offset (think of the modulator as the inlet for second dimension column – the inlet should be hotter).
   
8. The second dimension oven temperature offset should be 5-10°C or more (control wraparound, as necessary).

Another trick is to use highly selective columns in the first dimension that are known to separate isobaric congeners. Generally, we need selectivity and efficiency to separate structurally similar compounds. The second dimension should be chosen to move interfering matrix compounds out of the way.

I like to call the approach “true peak capacity increase GC×GC”. And I often conclude my presentations on the subject with: “If somebody from Oklahoma can do GC×GC, anybody can...” Let’s keep GC×GC data acquisition simple and attract more people to the club