This part is a reference set for reading lot-by-lot GC-MS certificates. For 12 core essential-oil groups, the table below gives only the ranges of the main identifying constituents in the current ISO standards, together with the constituents used to distinguish species or detect adulteration. The full table of each standard contains more constituents and more origin columns; the ISO text or the equivalent TCVN must be consulted for formal comparison.
5.1 What a chemotype is, and why it matters
Chemotypes are groups of individuals of the same species that look alike morphologically but differ clearly in the main constituents of their essential oil, usually because of differences in terpene-synthesis genes. Three examples with data Measured:
- Thyme Thymus vulgaris. Cluster analysis of 85 profiles yielded 20 groups; within a single species there are samples with thymol 47.1%, linalool 76.2%, geraniol 59.8% and sabinene hydrate 30.8%. The enantiomeric ratio of linalool also reverses with chemotype: d/l 1:99 in the linalool type but 97:3 in the geraniol type (Satyal et al. 2016, doi:10.3390/foods5040070).
- Tea tree Melaleuca alternifolia. There are 6 natural chemotypes, the three main ones dominated by terpinen-4-ol, terpinolene and 1,8-cineole; only the terpinen-4-ol type is traded under ISO 4730 (Padovan et al. 2017, doi:10.1186/s12870-017-1107-2; abstract).
- Rosemary. At least 5 chemotypes across 78 profiles (Satyal et al. 2017, doi:10.3390/foods6030020). The verbenone type has no ISO range.
Some essential oils have no chemotypes in the population sense because they come from clonally propagated hybrids: peppermint, lavandin and most commercial vetiver. For these, variation comes from the clone, the harvest season and the growing region. In Spanish lavandin (215 samples, 2016–2022), the clone determined composition more than the environment did: terpinen-4-ol averaged 3.49% in the Grosso clone versus 0.21% in the Super clone Measured (Ortiz de Elguea-Culebras et al. 2024, doi:10.1016/j.jarmap.2024.100550).
5.2 Reference table of main identifying constituents
Figure 4 places these ranges on a single scale to show which specifications are narrow and which are wide.
| Essential oil | Standard | Main identifying constituents | Distinguishing signs or points to check |
|---|---|---|---|
| True lavender Lavandula angustifolia | ISO 3515:2002 (6 origin columns) | linalool 25–38 (wild French), 30–45 (Maillette clone), 22–34 (Bulgaria); linalyl acetate 25–45, 33–46, 30–42 | low camphor and cineole (camphor ≤ 0.5–1.5); raised camphor suggests blending with lavandin |
| Spike lavender L. latifolia | ISO 4719:2012 | 1,8-cineole 16–39; camphor 8–16; linalool 34–50 | almost no linalyl acetate (≤ 1.6) |
| Rosemary | ISO 1342:2012 | Tunisia/Morocco type: cineole 38–55, camphor 5–15; Spanish type: cineole 16–23, camphor 12.5–22 | verbenone only 0.7–2.5 in the Spanish type; the verbenone chemotype has no ISO |
| Thyme, thymol type | ISO 19817:2017 | thymol 35–55; p-cymene 14–28 | a single standard covers both T. vulgaris and T. zygis, so it cannot distinguish the two species |
| Tea tree | ISO 4730:2017 (the 2025 edition adds an enantiomeric criterion for α-terpineol) | terpinen-4-ol 35–48; cineole ≤ 10; γ-terpinene 14–28 | (+)-terpinen-4-ol 67–71% of total terpinen-4-ol; p-cymene rises as the oil ages |
| Cineole-rich eucalyptus | ISO 3065:2021 (Australian type, several species combined) | 1,8-cineole 80–95 | unrectified E. radiata oil has only about 62–72% cineole |
| Lemon eucalyptus Corymbia citriodora | ISO 3044:2020 | citronellal 75–85 | isopulegol 5–10 (cyclisation product of citronellal) |
| Peppermint | ISO 856:2006 | menthol 32–49; menthone 13–28 | menthofuran 1–8 must be present; pulegone 0.5–3; very low menthofuran suggests blending with cornmint |
| Rose Rosa × damascena | ISO 9842:2024 (3 columns) | citronellol 20–34 (Bulgaria), 28–47 (Turkey), 30–47 (Morocco); geraniol 14–22, 6–29, 6–23 | 2-phenylethanol only ≤ 2.5–3; several tens of per cent means an absolute or a compounded oil; methyl eugenol 0.8–6 |
| Bergamot | ISO 3520:2022 (Calabrian type) | linalyl acetate 22–36; linalool 3–15; limonene 32–47 | bergapten 0.18–0.38% (measured by HPLC); furocoumarin-free (FCF) oil no longer matches this criterion |
| Sweet orange | ISO 3140:2019 | limonene 93–96 | octanal, decanal, valencene and β-sinensal are more useful than limonene for detecting adulteration |
| Sandalwood S. album | ISO 3518:2022 (the 2025 edition revises the Z-lanceol range) | Z-α-santalol 41–55; Z-β-santalol 16–24 | E,E-farnesol ≤ 1; at several per cent, suspect blending with Australian sandalwood |
| Australian sandalwood S. spicatum | ISO 22769:2022 | Z-α-santalol 15–30; E,E-farnesol 2.5–18 | also contains epi-α-bisabolol and Z-nuciferol |
| Vetiver | ISO 4716:2013 (5 columns) | khusimol 12–18 (Bourbon), 6–11 (Indonesia), 9–15 (Haiti) | percentages calculated without integrating 1,8-cineole |
| Patchouli | ISO 3757:2002 | patchoulol 27–35; α-bulnesene 13–21; α-guaiene 11–16 | patchoulol above 35 lies outside ISO (refined or fortified oil); optical rotation −60° to −40° |
Data sources: the official preview or summary page of each standard. For ISO 856:2006, ISO 3757:2002 and ISO 4716:2013, figures were read from the preview; for some 2025 standards only the foreword could be read. For ylang-ylang (ISO 3063:2004), lemon (ISO 855:2003) and E. globulus (ISO 770:2002), the chromatographic profile tables could not be read.
5.3 Three special groups
Ylang-ylang. The oil is fractionally distilled over time into the extra, I, II and III grades. According to ISO 3063:1983 (an old edition, read from an OCR scan, so the pairing of min/max values may be wrong), the ester value of Madagascan oil falls from 125–160 in the extra grade to 38–58 in grade III, and the relative density from 0.950–0.965 to 0.906–0.921 Standard. The extra grade is rich in linalool, p-cresyl methyl ether, methyl benzoate and benzyl acetate; later grades are rich in sesquiterpene hydrocarbons such as β-caryophyllene and germacrene D Industry (Tan et al. 2015, doi:10.1155/2015/896314). A sample of C. odorata var. fruticosa from Ha Noi (n = 1) had β-caryophyllene 26.8% and linalool 8.7%, closer to the grade III or complete-oil profile than to the extra grade Inference.
Basil. There are at least four chemotypes: methyl chavicol (estragole), linalool, methyl eugenol and methyl cinnamate (Viña & Murillo 2003, cited by Lawrence). Two of them carry compounds restricted for genotoxicity. The current ISO number for basil essential oil could not be verified (NR).
Bergamot and furocoumarins. IFRA limits total bergapten to 15 ppm (0.0015%) in finished products applied to sun-exposed skin; rinse-off products are exempt Standard. For a cold-pressed oil containing 0.18–0.38% bergapten, the maximum oil concentration in a leave-on product is about 0.39–0.83% Inference (15 ppm divided by 1,800–3,800 ppm). Vacuum-distilled oil can be almost free of bergapten while its volatile fraction stays close to that of cold-pressed oil (Belsito et al. 2007; abstract).
5.4 The limits of “meets ISO”
ISO is a commercial quality specification, intended “to facilitate the assessment of quality”. It is not a test of naturalness. The data show errors in both directions:
| Case | Data | Source |
|---|---|---|
| Lavender blended with 27% synthetic linalool and linalyl acetate | the percentage profile can still fall within ISO; only chiral analysis reveals it: (S)-linalool rises from 6.2% to 33.6%, (S)-linalyl acetate from 0.6% to 19.6% | Capetti et al. 2021, doi:10.3390/molecules26185610 |
| Australian tea tree blended with 50% Chinese oil | (+)-terpinen-4-ol falls to 54.5%, outside the 67–71% range | Capetti et al. 2021 |
| Romanian lavender, 64 genuine lots | only 3/64 lots fully met ISO 3515; 12 lots had high cineole, 11 lots low linalyl acetate | Floare-Avram et al. 2026, doi:10.3390/chemosensors14070167 |
| Spanish lavandin Grosso, 112 samples | only 18% met all 11 constituents against ISO 8902:2009 | Ortiz de Elguea-Culebras et al. 2024 |
| Commercial sandalwood in the EU, 6 samples | 0/6 met ISO 3518 | Kucharska et al. 2021 |
There are also four technical limitations Inference, drawn from the standards themselves:
- One standard may cover several species (thyme, eucalyptus), so “meets ISO” does not prove the species.
- Many chemotypes have no ISO: verbenone rosemary; linalool, geraniol and thujanol thyme; cineole tea tree. These can be compared only with the literature.
- ISO ranges change between editions: ISO 3520 raised the limonene ceiling from 45 to 47% (1998 → 2022); ISO 9842 changed the measurement temperature from 20 to 25 °C and dropped the “Turkey peasant” column (2003 → 2024). A certificate must state the standard number together with its year.
- Percentages depend on the chromatographic column: 1,8-cineole co-elutes with β-phellandrene or limonene on some columns; ISO 3518 advises against non-polar columns. Comparing a certificate with ISO is meaningful only with the same column type and the same identification method.
5.5 Enantiomeric benchmarks for authenticity testing
| Essential oil | Benchmark | Source |
|---|---|---|
| True lavender | (R)-linalool 93.8%; (R)-linalyl acetate 99.4% | Capetti et al. 2021 |
| Bergamot | (R)-linalool 98.6%; (R)-linalyl acetate 98.7% | Capetti et al. 2021 |
| Tea tree | (+)-terpinen-4-ol 67–71% | ISO 4730:2017 |
| Rose | S-(−)-citronellol above 99% | Ph. Eur., via Dobreva & Nedeltcheva-Antonova 2023 |
| Rosemary | (+)-α-pinene 97–99%; (−)-borneol 95–98% | Satyal et al. 2017 (5 samples) |
| Thyme | linalool d/l 1:99 (linalool and thymol types) but 97:3 (geraniol type) | Satyal et al. 2016 |
The thyme example shows that a single enantiomeric threshold cannot be applied to every lot sold under the same name: the benchmark must be tied to the chemotype.
A monograph compiled from 718 sourced facts collected under a common protocol and cross-checked independently. Items labelled Inference are the compilers’ own judgements. Diagrams are newly drawn from published data. Scientific reference material — not legal or medical advice, and no substitute for the text of the standards.