Terpenes in Cannabis – to MS or not to MS?

In my last blog, I showed how FID is a more suitable detection method for cannabis residual solvent analysis than MS. But what about terpene analysis? Can our old friend the FID hold its ground against the mighty mass spectrometer for this application? Actually, it can!

Terpenes are much larger molecules than residual solvents, so terpene analysis via MS doesn’t suffer from low molecular weight interferences like we see for residual solvents. However, terpenes fall into another weakness of MS: mass spectrometers are wonderful at separating and quantifying ions of different masses, but if a MS encounters two co-eluting compounds with the same mass and/or mass fragments, there is no way for the MS to differentiate between those compounds, which means that they can’t be quantified/identified separately. Let’s take a look at a table of the molecular weights and major ions for a bunch of terpenes of potential interest to the cannabis industry:

In this list of 37 terpenes, we actually only have 11 unique molecular weights and 12 unique major fragment ions. If you look through the list, you can see that these compounds share a lot of ions. To show a graphical example, we can look at the mass spectra for alpha-phellandrene and delta-3-carene, which elute very close to each other on Restek’s recommended column (by the way, you’ll always have groups of closely-eluting terpenes on any column you use – there are just too many of them to separate them all to baseline):

Mass Spectrum for Alpha-Phellandrene from NIST Database*

*http://webbook.nist.gov/chemistry/

Mass Spectrum for 3-Carene from NIST Database*

*http://webbook.nist.gov/chemistry/

These spectra are almost identical, and will be very difficult to deconvolute for quantitative purposes. There are many more cases like this amongst all the terpenes since they’re all part of the same structural class. So basically, you can definitely analyze terpenes using MS, but you lose much of the advantage MS gives you from a spectral deconvolution standpoint, since the MS can’t really tell the difference between many of these terpenes anyway. You might achieve a slight increase in sensitivity using MS versus FID, but since we’re interested in terpenes in the percent level (even fractions of a percent), FID has plenty of sensitivity at a fraction of the cost.

So once again, the humble (and affordable) GC-FID is the best choice, even over MS. I hope this helps explain why.