Alexis St-Gelais, chimiste
Essential oils are worth quite a lot of money. It so happens that, in order to boost profits, unscrupulous people alter oils in various ways. This “Adulterants and you” series is there to introduce you to some of the adulterations we encounter. This is because not all of them are necessarily obvious, nor bear the same level of risk for the final consumer.
Methyl salicylate-rich oils (wintergreen and birch) are a challenge for people controlling the quality of essential oils. Unlike most other essential oils, they are made of mostly a single constituent, namely methyl salicylate, which is also available as a very cheap synthetic ingredient. Adulteration of these oils with their synthetic equivalent is common.
Today’s post has a particular structure – it is an hybrid between the series and a research note, since the adulterant in question was identified through our systematic investigation. The “How do we detect it?” section has thus been enhanced to present our scientific investigation, followed by the others usual sections of the “Adulterants and you” series.
What is this adulteration about?
Wintergreen essential oil (from either Gaultheria yunnanensis, G. procumbens or G. fragrantissima) or birch oils (Betula lenta) are mostly constituted of methyl salicylate, which can easily be replaced or diluted with cheaper synthetic methyl salicylate. The industrial production of methyl salicylate relies on the Kolbe-Schmitt reaction, a classic basic organic chemistry reaction shown on Figure 1 . The process is conducted at high pressure and temperature, which ensures high yields of quite pure salicylic acid. Salicylic acid can be methylated using e.g. methanol to yield methyl salicylate.
Synthetic methyl salicylate, just as its natural counterpart, shows as a colorless liquid with a characteristic scent of wintergreen. It is very cheap as a bulk ingredient that is produced in huge amounts in China, usually sold at a purity of >99% (which is the regular purity of a decently well ran Kolbe-Schmitt reaction).
What does synthetic methyl salicylate do as an adulterant?
Synthetic methyl salicylate is used to dilute or mimic the costly natural birch and wintergreen oils, and then make huge profits. From a chemical point of view, synthetic methyl salicylate is exactly the same thing as the natural ingredient: same smell, same structure, same biological properties.
Is it dangerous?
Methyl salicylate itself has just the same properties, wether it comes from a plant or an industrial facility. The main difference stands in minor constituents. Trace impurities of synthetic methyl salicylate can be toxic – but at such low levels, this should not really be a concern, unless one starts drinking his essential oils. Overall, it is unlikely that there are serious health concerns over this particular adulteration.
How do we detect this adulteration: Isolation and Characterization of Dimethyl 4-Hydroxyisophthalate as a Marker of Synthetic Methyl Salicylate Addition
Genuine oils should contain small amounts (often <0.05%) of other natural compounds, such as linalool, pinenes and ethyl salicylate. Lack of such compounds is sufficient to suspect that an oil in fact is entirely synthetic methyl salicylate. However, clever people will rather dilute a true oil with the synthetic equivalent. In such cases, small amounts of natural compounds can still be detected, and it becomes much harder to tell if the oil has been altered or not from a routine analysis.
First, if any trace of phenol is detected, synthetic methyl salicylate can be suspected. Phenol is the starting material of the Kolbe-Schmitt reaction. However, it usually is removed fairly easily from the manufacturing process using some high temperature vacuum, since it has a much, much lower boiling point than salicylic acid.
Furthermore, unless some further purification is added to the process, a small quantity (it can be as low as 0.05%) of 4-hydroxyisophthalic acid (4) is generated during the Kolbe-Schmitt reaction  (see again Figure 1). This byproduct arises from the double addition of carbon dioxide to the phenol, a coreaction that can hardly be avoided at all during the manufacturing process owing to universal chemical mechanisms. Any quantity of compound 4 remaining in unrefined technical grade salicylic acid will be methylated as well in the course of methyl salicylate synthesis, generating compound 5. Compound 4 can be removed from salicylic acid, but this likely is kept for pharmaceutical-grade products, which are much costlier and make the adulteration less attractive. As such, compound 5 represents an excellent marker of synthetic methyl salicylate addition.
Technical-grade synthetic methyl salicylate featured a late-eluting unknown constituent in the 0.1-0.6% concentration range (Figure 2). Back when we did this study, a few years ago, we had older databases that did not include this constituent. As such, we had to find our own mass spectral reference. So, we decided to check what it was by systematically isolating and characterizing it.
We started from 140 mL of the mixture. The first challenge was to obtain a fraction that was enriched with this molecule, with less methyl salicylate to get rid of. Since both molecules were quite far appart on the GC chromatogram (Figure 2), we could conclude that their boiling points were also very different (for more details on this inference, take a look back at our post on polarity). We thus used a Vigreux column (Figure 3) to selectively distill the methyl salicylate.
The Vigreux column is set over a boiling flask containing the oil, the whole apparatus being under slight vacuum to bring down boiling points and allow the distillation to occur at reasonable temperatures. The column contains many glass dents, creating a step with a large surface in contact with the air. At each step, the gaseous compounds come in contact with a slightly cooler environment from the ambiant air, and condensate. They are then heated back by the upcoming vapor. Since the unknown compound boils at a much higher temperature than methyl salicylate, the latter is more readily turned back to vapor, and proceeds faster to the next step. Consequently, at every step, the relative methyl salicylate content of the vapor phase increases. As a result, at the exit of the column, mostly pure methyl salicylate is condensed, while the unknown falls back into the boiling flask where it becomes more and more concentrated as the methyl salicylate is evaporated. We thus obtained a “light” fraction, which was almost 100% methyl salicylate by GC, and about 4 mL of a brownish “heavy” liquid fraction remaining in the original flask. GC analysis showed that it now contained 6% of the unknown.
The heavy fraction was then submitted to column flash chromatography over 75g of silica, eluting isocratically with a mixture of hexanes/acetone 50:1. Flash chromatography is a lab-scale variant of chromatography used to separate mixtures of compounds in order to isolate some of them. In this case, previous testing using thin layer chromatography indicated that the conditions used would provide suitable separation of the compounds. The methyl salicylate would come out of the column first, followed by the unknown constituent, while the yellow-brownish impurities would mostly remain adsorbed on the silica. The predicted conditions worked out well, as can be seen on Figure 4.
Fractions of 50 mL were collected, and checked on thin layer chromatography using hexane/acetone 15:1 as an eluent and silica plates. Similarly to flash chromatography, thin layer chromatography can separate the compounds of a mixture, but on a much smaller scale. It is mostly used as the “eyes” of the bench chromatographer, allowing him to check how the separation went. The plate is coated with silica containing fluorescent material, which glow green under UV light at 254 nm. Many organic compounds, and especially those containing aromatic rings (such as methyl salicylate and most likely the unknown compound), quench this light and thus appear as dark spots on the plate, negating the background fluorescence. Figure 5 shows how the collected fractions fared.
Fraction 9 contained both compounds and was discarded, and fractions 10-15 were pooled and gently evaporated under vacuum to yield powdery white cristals, presumed to be the relatively pure unknown compound (figure 6).
The final step was to characterize the obtained unknown constituent, which was determined to be dimethyl 4-hydroxyisophthalate (5). This finding was supported by nuclear magnetic resonance experiments, which allow structural determination of molecules – I will keep a more popularized description of this technique for a later post, and only present the results for our chemists readers and the sake of transparency. The MS spectrum of 5 was also recorded (Figure 7), and we share its retention indexes on two capillary columns in GC-FID. The following information are given as reference for anyone attempting to identify this compound in suspect wintergreen oils.
Dimethyl 4-hydroxyisophthalate (5). White crystals, 1H NMR (400 MHz, DMSO-d6/CDCl3 1:1) δ: 11.07 (s, 4-OH), 8.37 (d, J = 2.1, H-2), 8.02 (dd, J = 8.8, 2.1, H-6), 6.99 (d, J = 8.8, H-5), 3.96 (s, H-2’), 3.86 (s, H-1’); 13C NMR (100 MHz, DMSO-d6/CDCl3 1:1) δ: 168.9 (C-8), 164.8 (C-7), 164.0 (C-4), 135.8 (C-6), 131.7 (C-2), 120.7 (C-1), 117.3 (C-5), 112.0 (C-3), 52.3 (C-2’), 51.5 (C-1’); EI MS m/z (rel. int.): 178 (100), 147 (65), 210 (45), 179 (35), 119 (34), 63 (13), 91 (11), 120 (9), 136 (8); retention indexes (normal alkanes RI, H2, FAST GC): 1615 (BPX-5, SGE Analytical Science), 2313 (Solgel-Wax, SGE Analytical Science).
Compound 5 is the methylated product of the documented byproduct 4 of the Kolbe-Schmitt reaction . It has also been mentioned in a Chinese journal as being present in waste material from methyl salicylate production . Our investigation bridged the gap between these information and the routine detection of synthetic methyl salicylate adulteration: any occurence of compound 5 in a methyl salicylate-rich oil indicates adulteration with synthetic material.
Bottom of the line
Synthetic methyl salicylate can easily be used to diluted or mimic natural oils of wintergreen and birch, being for the most identical to the natural molecule. Although not dangerous, this adulteration results in a strongly biased price for a not-anymore natural oil. It can be detected either by absence of natural minor constituents, or the presence of dimethyl 4-hydroxyisophthalate (5), an impurity arising from the Kolbe-Schmitt reaction, which has been isolated and has been duly characterized by NMR and MS. The data provided in this blog post should allow any testing laboratory to identify the adulteration.
 Boullard, O, Leblanc, H, Besson, B. Salicylic Acid. In Ullmann’s Encyclopedia of Industrial Chemistry, 7th ed., vol. 32, p. 127-134, [Online], URL: http://onlinelibrary.wiley.com/doi/10.1002/14356007.a23_477/abstract
 Yumei, L. Preliminary characterization of residue from methyl salicylate production, Shanghai Huanjing Kexue, 1998, 17 (2), 40-41 (in Chinese).