How to Read a Fragrance Oil COA and SDS: A Worked Example
A certificate of analysis that lists three physical constants is a consistency check, not proof of what is in the bottle. We say that about our own: the published COA for the Lilac Blossom Oil gives appearance, smell, relative density, refractive index and optical rotation, and that is all it gives. This post reads it line by line, then does the same for the safety data sheet (SDS) it is not a substitute for.
Our own COA, line by line
The document is headed "Lilac Oil (Syringa vulgaris) - Analytical Data," for batch LILIBCUNINC001A. Here is each result, the window it is judged against, and what the test physically measures.
| Line on the COA | Result (standard) | What it measures | What it cannot tell you |
|---|---|---|---|
| Appearance | complies (yellow liquid) | Color and clarity as judged by eye | How strong the color is; no color scale is given |
| Smell | complies (floral, sweet, fresh) | A trained or untrained nose against a description | Anything repeatable between two people |
| Relative density at 20 C | 0.973 (0.950 to 0.990) | Mass of a volume of oil compared with the same volume of water | Which ingredients make up the mass |
| Refractive index at 20 C | 1.501 (1.480 to 1.520) | How much the liquid bends light | Which ingredients bend it |
| Optical rotation at 20 C | -0.72 degrees (0 to -10 degrees) | How far the liquid turns polarized light | Whether any optically active ingredient is present |
| Best by | 05/31/2027 | A date the supplier assigns | Storage history after it leaves the supplier |
| Extraction method / parts used | "Blended" / "N/A" | A label, not a measurement | Origin, source material, process |
The relative density line is worth doing arithmetic on, because it is the one you can check yourself. At 0.973, 20 ml of oil weighs about 19.46 g. The window of 0.950 to 0.990 allows anything from 19.00 g to 19.80 g, so a second lot can differ by roughly 4% of its weight and still sit inside the stated standard. The refractive index window is 1.480 to 1.520, about plus or minus 1.3% around the middle. Both windows are wide enough that they confirm "a liquid in the expected range," not "this exact composition."
The optical rotation window has a property that is easy to miss. It runs from 0 to -10 degrees, which means a reading of exactly 0 would also pass. Reasoning from the window itself, this test cannot separate an oil containing optically active molecules from one containing none, and our batch's -0.72 is a small number near the edge of the range. The COA also names "Pharmacopoeia" as its criterion without saying which one, and its conclusion line says the tested parameters comply with pharmacopoeia standards for the oil it names. Standards bodies do publish methods for these tests: ISO's catalogue lists reference methods for essential oils, such as ISO 279:1998 for relative density at 20 C and ISO 592:1998 for optical rotation, and ISO product standards for single essential oils such as cardamom (ISO 4733) and lemongrass (ISO 4718) list relative density, refractive index and optical rotation among their characteristics. Our COA does not state which method was used, and our oil is a composition, not a single essential oil, so those standards are context for the tests, not a statement about ours.
What is not on it
No GC-MS (a chromatogram or constituent list), no percentages of ingredients, no flash point, no allergen data, no origin, no solvent or residue data, no IFRA data. That absence is the position: the COA shows that the batch met three physical windows and two sensory descriptions. It is also the COA of one batch, and we say "the batch on the published COA" for that reason. It does not promise that a later lot matches it.
It helps to know what the missing test would add. In GC-MS, a column separates a mixture so each component exits at its own retention time, and a mass spectrum for each peak is matched against reference libraries. Fiehn's 2016 protocol describes pairing spectral matching with retention indices as two independent parameters, against a library of 242,477 compounds, and warns that some detected compounds are artifacts of the hot injection and that fragment ions overlap between compounds (Fiehn, 2016). That paper is about metabolomics, not fragrance, so read it as a description of the technique and its limits, not a standard for our oil. The practical point: GC-MS can show far more of what is there, with limits; three constants cannot.
The SDS: sixteen headings and where the answers sit
Under the U.S. Hazard Communication Standard, chemical manufacturers and importers must obtain or develop an SDS for each hazardous chemical they produce or import, and the SDS must include at least sixteen sections in a fixed order: identification; hazard identification; composition/information on ingredients; first-aid measures; fire-fighting measures; accidental release measures; handling and storage; exposure controls/personal protection; physical and chemical properties; stability and reactivity; toxicological information; ecological information; disposal considerations; transport information; regulatory information; and other information, including the date of preparation or last revision (29 CFR 1910.1200(g), via Cornell LII). The rule's own note says OSHA does not enforce the information requirements in sections 12 through 15, because those fall outside its jurisdiction.
For a candle maker the useful sections are few. Flash point sits in section 9, and the United Nations GHS guidance for SDS preparation lists flash point and density or relative density among the section 9 properties, asks that the method be named where it matters (open cup or closed cup), and says an SDS should state when a property is not available (UNECE, GHS Rev. 7, Annex 4, 2017). Section 5 covers fire-fighting, section 7 handling and storage, section 3 is headed composition and information on ingredients, which is where an ingredient statement would appear if the sheet gives one. Notice that section 9 repeats relative density, the one number our COA also reports. An SDS and a COA can therefore be cross-checked against each other.
Our site publishes no SDS, IFRA certificate or GC-MS report. A buyer who needs one should ask in writing before ordering and get the answer in writing. We do not promise a specific document, and we are not the right authority on whether a given material needs an SDS; a regulatory consultant or product-safety attorney is.
The IFRA certificate is a different document again
IFRA describes the certificate of conformity as "a document established by companies creating fragrance mixtures," in which the supplier tells its customer the product complies with the IFRA Standards for an intended use. IFRA says it does not create the certificates, third parties can certify them, they apply only to mixtures meant for direct inclusion in a finished consumer product, and the certificate "does not replace a safety assessment" (IFRA, certification of IFRA Standards). So an IFRA certificate answers "may this go in this product category at this level," not "what is it" and not "will it burn safely."
What to ask, in writing, before you order
- Is there a flash point for this material, by which method (open or closed cup), and on which lot?
- Is there an SDS? If not, say so in writing.
- Is there an IFRA certificate of conformity, and for which product categories?
- Is there an allergen declaration or ingredient statement you can use for labeling?
- Will the COA I receive carry the lot number printed on my bottle, and what is that lot's best-by date?
Judge any answer against the sections above: a flash point without a method, or a COA without your lot number, is an incomplete answer. Our flash point post explains why question one matters for candles, and testing before bulk covers how to keep lot records. For label wording see reading a lilac oil label, and for what the oil is and is not, what's in the bottle.
A COA is worth having and worth reading precisely. Ours says the batch met its windows, and it says nothing more, which is the right amount of confidence to take from it.
Lilac Blossom Oil
Lilac Blossom Oil from our own shop, in sizes from 0.5 oz to bulk.
From $15.99
Shop Now →Sources & Further Reading
- U.S. OSHA, 29 CFR 1910.1200, Hazard communication, paragraph (g). law.cornell.edu/cfr/text/29/1910.1200
- UNECE, 2017. GHS (Rev. 7), Annex 4: Guidance on the preparation of safety data sheets. unece.org/ghs_rev07/07e_annex4.pdf
- IFRA. Certification of IFRA Standards. ifrafragrance.org/certification-of-ifra-standards
- Fiehn O., 2016. Metabolomics by gas chromatography-mass spectrometry: the combination of targeted and untargeted profiling. Current Protocols in Molecular Biology 114:30.4.1-30.4.32. pmc.ncbi.nlm.nih.gov/PMC4829120
- ISO. ISO 279:1998 (relative density at 20 C) and ISO 592:1998 (optical rotation), essential oils; ISO 4733 and ISO 4718 (catalogue entries). iso.org/standard/25308