FAMEs blog part 3: If I Use a Rt-2560 with my GC-MS will it explode?

This blog a part of a series: part 1, part 2, and part 4

Disclaimer: Do not use non-bonded columns, such as Rt-2560, in GC-MS for routine analysis! If you have the need to run some samples, here is what to expect (i.e. quite a bit of bleed)

In my last blog, I talked about how exciting it was to finally test our columns with a sample containing all kinds of trans-fatty acids. The Rt-2560 provided the best separation of C18:1 isomers.  At first, I was having doubts about the identity of the peaks that didn’t align with the cis/trans standard, marked by a purple star in the chromatogram (Fig. 1).

Fig 1: Separation of C18:1 isomers in frosting using GC-FID on Rt-2560. The red trace is the frosting sample, blue trace is the cis/trans standard. Purple stars mark unidentified peaks.

The obvious course of action is to run the sample with GC-MS, however, the Rt-2560 is not bonded and using other columns is out of the question since the elution order is completely different.  As I mentioned above, Rt-2560 is not recommended for use with MS detector due to its high bleed, but on the other hand, it is a fast way to tentative identification.

On the first glance, the chromatogram doesn’t show too much of the bleed:

Fig 2: Total ion chromatogram of frosting FAMEs on Rt-2560

But when we look at the spectra, the background ion m/z 472 is present in all of the peaks:

Fig 3: Spectra of an unknown peak, m/z range 35-550

Extracting the m/z 472 showed uniform abundance across the run with a small dip around the time of solvent elution:

Fig 4: Extracted ion chromatogram of m/z 472

Such a huge background can be subtracted, but it can still affect the identification and sensitivity. The easy solution is to cut-off m/z higher than 450 amu in order to obtain a cleaner mass spectrum:

Figure 5: Spectra of an unknown peak, m/z range 35-450

While the main interference is gone, there is still a background present – the m/z 378 is also part of column bleed. This peak is much smaller than the previous 472 and doesn’t hinder identification as much. However, this ion can also be cut-off as long as the analytes of interest have lower molecular weights.

Figure 6: Spectra of an unknown peak, m/z range 35-350

So after removing all the interferences, I was finally able to move on with the identification. Unfortunately, here lies another caveat. The difference in EI spectra between individual FAME isomers is so minuscule that we can only determine how many carbons and double bonds we have, but not where the double bond is. Figure 7 shows a typical example of a library search for one of the isomer peaks. But even though the identification is very limited, it does help us to rule out whether the peak is an analyte or a matrix background.

Figure 7: Library search example

In conclusion, while tentative identification using GC-MS is possible with Rt-2560, the high bleed will contaminate the ion source. We do not recommend this setup for routine analysis or screening.

More about TFAs and Rt-2560 from Hansjoerg and rest of the blog series:

Part 1: Is the industry ready for a ban of partially hydrogenated oils?

Part 2: Let’s look at actual samples with incurred TFAs

Part 4: Using hydrogen as a carrier gas for FAMEs analysis


Leave a Reply