Corine Lormel, Ph.D.
Some customers are surprised to see a compound named ‘’Unknown’’ in the listing of their essential oils analysis reports. Why do chemists keep them in all reports? Are they just unidentified compounds? Why don’t chemists try to identify them?
In this paper, we will try to understand what is behind the famous unknown compounds.
Unknowns can be found in two places in your reports.
- The first place is explicitly indicated in the result table of the report (if a GC report is something completely new, I suggest, you first read this previous paper blog here). To understand this type of unknown we will use an analogy to a criminal investigation. On a crime scene, the police take fingerprints and DNA. Based on their finding they can state if the criminal is a child or an adult, a woman or a man. They have a unique map identifying one specific person. That is the same for molecules. Each of them has a special signature which is given in GC-FID by the retention index and/or in GC-MS by the mass spectrum. So, the analyst can often define the family of the molecule: a ketone, a monoterpene, etc.
Similarly to the police, who know when the DNA test provides a match wether the same person was present on two crime scenes, an unknown molecule that matched a saved DNA (retention indexes, mass spectrum) in our reference means we have met it before. That is why we keep track of all the unknows in our databases and the reports because they provide relevant information on the essential oils. For some samples, that can be even more useful, as a specific unknown can be a clue for species and if it is missing, the sample is considered as unusual.
Also, like the police which cannot deduce a person’s name or picture based only on his DNA (except if he has already been arrested) more clues are necessary, like ideally a video of the criminal during his forfeiture. In phytochemistry which is a relatively young science, several molecules haven’t been “arrested and pictured” yet. The extra analyses to give them a complete structure and a name can be a huge and long journey as related in these articles (St-Gelais, et al., 2018) (Tissandié, et al., 2018).
Another hindrance in the recognition is the sharing of information. If a police department completely identifies a criminal (picture, name, address, DNA) but keeps the information only in their own databases, the other departments all around the world will not get any access. Worse they will have to go through the whole process from the start of their own. That is what happens with the different databases of molecules currently available: if the newly identified unknowns are not integrated, they can remain unknown for a while to most laboratories.
To conclude with this type of unknown, we have their DNA, but we don’t know yet their names.
- The second place to find unknowns is not as obvious. If you can figure out it is linked to the percentage of compounds identified at the end of our reports (the famous % that will never be 100 % as explained here ) you are closed. What is lacking to reach the 100 % is not identified. To sum up, this unrevealed part came from information which level is close to the background noise (Image 1). For instance, it is like ignoring the clothes sewing threads individual colours in a people description because most of the time it is insignificant or irrelevant.
Then with a settled zoom level for all the analysis, we gathered these tiny pieces of information in a global unknown portion which is lacking to reach the 100%.
Image 1 : A typical GC chromatogram with the neglected peaks circled in red.
Now, we hope this blog post helping you to have a better understanding of why there are one or several unknowns in your essential oils reports.
- St-Gelais, A. et al., 2018. Aromas from Quebec. VI. Morella pensylvanica from the Magdalen Islands: a (-)-α-bisabolol-rich oil featuring a new bisabolane ether. Journal of Essential Oil Research, 30(5), pp. 319-329.
- Tissandié, L. et al., 2018. Towards a complete characterisation of guaiacwood oil. Phytochemistry, Volume 149, pp. 64-81.