Chromatographic Selectivity – It’s Technical, but It’s Important

Posted September 20 2016

I’m finally writing my first technical blog for Emerald Scientific – hooray! While I’m perfectly content droning on about esoteric topics, I promise that I won’t do that in the Emerald Scientific blog. Your time is important, so I won’t waste it introducing technical concepts that you’ll never need to know. The topic of this blog – column selectivity – is extremely important for residual solvents analysis, and it’s something that has been largely ignored by the cannabis analytical industry.

What is selectivity and why is it important? In a nutshell, selectivity is the measure of column’s ability to separate different compounds. Selectivity is denoted by the greek letter alpha (α) and is applicable to both GC and LC separations. Basically, the higher the value of α for two peaks, the more separation there is between them (shown below).

Figure 1: The higher the α value, the more separation there is between two peaks

“So what?” you ask. The fact is, no single GC or HPLC column can separate every compound in existence. Or even every compound in a given compound class. If you’re running cannabis residual solvents using one GC column and a flame ionization detector (FID), what happens if there’s a co-elution between one of your compounds of interest and another solvent that happens to be in the sample? The answer is that you’ll never know that there’s a co-elution, and you’ll end up reporting that compound at an erroneously high value. If the co-elution puts your compound over the regulatory limit, then the producer will have failed a batch that should have passed. Definitely not an ideal situation.

What can we do to mitigate this situation? First off, don’t panic. This problem is common in a lot of industries (e.g. environmental), and they’ve already come up with a solution – two solutions, actually.

The first solution is to use a different detector. Where FIDs only measure how much material is eluting at a given time, mass spectrometric detectors (MSDs) not only measure how much is eluting at a given time, they also collect information on the identity of the eluting compounds. This way, co-eluting compounds can usually be identified and quantified separately. In cases where they can’t one can usually see that there is a co-elution so they know that their reported value will be erroneously high. All of this sounds great, but MSDs have some drawbacks:

  • They’re much more expensive than FIDs
  • They require more expertise to operate
  • They require more frequent calibration
  • Their calibration range is more limited than FIDs

All of the above can be dealt with, and MS detection is definitely a viable way to confirm your positive results.

The second solution is to employ a confirmation column. A confirmation column is a second column with very different selectivity compared to your primary column (the technical term is ‘orthogonal selectivity’ if you want to impress your friends). The way a confirmation column works is that the differing selectivity pretty much guarantees that a co-elution on the primary column won’t result in a co-elution on the confirmation column. It’s like Facebook stalking an old high school flame: there are probably several Justin Whatshisnames in California, but there’s probably only one Justin Whatshisname who is from California and went to your high school during the years you attended. While searching name + state won’t give you an answer you’re confident in, name + state + high school will ensure that you’re not messaging a stranger.

By using a confirmation column, you’re able to have confidence in your results without having to purchase a headspace-GC/MS instrument. Additionally, because the confirmation column also uses an FID, it can be implemented with little additional training for lab technicians. While using a confirmation column will give you confidence in your FID results with a lower pricetag than GC/MS, there are a couple of drawbacks:

  • Confirmation columns require a separate FID, meaning you need a dual-FID system, but an FID can usually be added to a system at a fraction of the cost of a MS.
  • The initial installation of a confirmation column can be a little tricky – but if you contact, we can talk you through it.

Since this blog is all about selectivity, I’m going to go a little bit more in depth on solution #2. To demonstrate what the results from a confirmation would look like, here’s an example of differing column selectivities for cannabis residual solvents. The first chromatogram shown is a list of solvents run on an Rxi-624Sil MS (Emerald PN: SUGC01038) using the 8610C GC kindly loaned to me by SRI.

Figure 2: Solvents on Rxi-624Sil MS (Total Vaporization Technique)

The next chromatogram shows what happens when I analyze the same compounds under the same conditions on a column with a completely different selectivity – the Rtx-5.

 Figure 3: Solvents on Rtx-5 (Total Vaporization Technique)

We can see that the Rtx-5 generates a completely different chromatogram than the Rxi-624Sil MS (check out acetone and isopropanol!). This is because the Rtx-5 column has a very different selectivity for these compounds (especially the alcohols). However, just because a column generates a different chromatogram than our primary column (the Rxi-624Sil MS), it doesn’t mean it’s a good confirmation column. We can see that with this very abbreviated list, there are co-elutions on the Rtx-5. While the co-elutions wouldn’t necessarily preclude it from being used as a confirmation column for this solvent list, there is a better alternative. The Rtx-1 (Emerald PN: SUGC01039) does a much better job of separating the residual solvents of interest to the cannabis industry. I’m sure you’d like to see a chromatogram of that, but I’m going to leave you in suspense until my next blog, which will address setting up a dual-column system using the Rxi-624Sil MS and the Rtx-1, and a proposed workflow for those of you who don’t have two FIDs.

If you really can’t wait for the unveiling of the chromatogram or have any questions regarding residual solvents analysis, feel free to contact me.


Terms used in this blog:

Selectivity: The ability of a column to separate specific compounds under specific conditions. Measured as the ratio of retention times between two specific peaks.

Elute: to exit the column. Each compound elutes at a specific time, referred to as the retention time, but if you hear someone say ‘elution time’, then you’ll know that they’re just trying to sound smart by using an uncommon term for retention time.

Co-elution: two compounds eluting at the same time, resulting in one peak instead of two.

Orthogonal selectivity: A term for describing the selecitivity of one chromatographic column as compared to another, very different column. Not all columns have orthogonal selectivity. For example, a 1-type phase and a 5-type phase do not have orthogonal selectivity. However a 1-type phase and a 624-type phase do have orthogonal selectivity. Whether or not a column pair has orthogonal selectivity depends on the chemical makeup of each column’s stationary phase.