What the COA for borage oil actually contains
The certificate for a borage oil lot reports appearance, identification by gas chromatography of the fatty acid methyl esters, the GLA, linoleic and oleic figures, free fatty acid, peroxide value, moisture, the heavy-metal and microbiology panels and a pyrrolizidine-alkaloid result. Every row needs the method, the acceptance limit and the measured value in separate columns. The catalog limits are release targets; the lot certificate records the actual material in the pack.
A borage oil certificate has to show the fatty-acid profile, the oxidation state and the alkaloid result for the same lot, because the GLA figure alone does not describe the oil.
The certificate as a lot report
The catalog GLA range is wide enough that two conforming lots can sit six percentage points apart. The certificate has to show where in the range the lot falls, which acids accompany the GLA and what the oxidation values were at release, not only that the lot is borage oil. A row that says only “conforms” asks the buyer to trust a conclusion without the evidence.
The identity row and the GLA row are separate decisions. Identity asks whether the fatty-acid profile is consistent with Borago officinalis seed oil; the GLA figure asks how much of one acid is present.
| Parameter | Method | Release limit |
|---|---|---|
| Appearance | Visual | Clear to slightly hazy, pale yellow to golden |
| Identification | GC of fatty acid methyl esters against a reference profile | Profile conforms |
| Gamma-linolenic acid (C18:3 n-6) | GC-FAME | 18.0–25.0% |
| Linoleic acid (C18:2 n-6) | GC-FAME | 30.0–40.0% |
| Oleic acid (C18:1 n-9) | GC-FAME | 15.0–25.0% |
The profile is the part a buyer can use. A certificate that reports GLA but omits linoleic and oleic leaves the identity question open, because a blend can be built to hit a GLA figure while the rest of the profile drifts.
The GC basis: area per cent, weight per cent and the oil basis
Fatty-acid results come from transesterification to methyl esters and capillary GC with flame-ionisation detection. Percentages can be reported as area per cent or as weight per cent after response-factor correction, and the two are not identical. The certificate should state the basis, the column and the standard, and include the chromatogram when identity or the GLA figure is contested.
Resolution matters at the C18:3 position. Gamma-linolenic acid (C18:3 n-6) and alpha-linolenic acid (C18:3 n-3) have the same carbon count and can co-elute on a short or non-polar column. A method that does not separate them can report a combined C18:3 figure and overstate the GLA. The method note should show the two are resolved, with the integrated peak on the chromatogram.
The oil basis is a second conversion. A figure of 20.0% GLA on the oil means 200 mg of GLA in a gram of oil, before any carrier. If the oil is carried on a powder, the figure on the powder is the oil figure multiplied by the oil load, and the certificate has to say which basis it is reporting. The label arithmetic on the grades page uses the same conversion.
Release limit, result and the oxidation pair
Free fatty acid and peroxide value are release limits and the two numbers most likely to move after release. Peroxide value measures primary oxidation products and rises early; the p-anisidine value measures the secondary aldehydes that appear as the oil ages. A lot can pass the peroxide limit and still develop an off-odour in a softgel within months, which is why the anisidine value is worth agreeing even though the catalog does not fix it.
| Parameter | Method | Release limit |
|---|---|---|
| Free fatty acid (as oleic) | Titration | ≤0.50% |
| Moisture | Karl Fischer | ≤0.1% |
| Peroxide value | Iodometric titration | ≤5.0 meq O₂/kg |
| Heavy metals | ICP-MS | Pb ≤3 · As ≤2 · Cd ≤1 · Hg ≤0.1 mg/kg |
| Microbiology | Plate count and absence tests | APC ≤1,000 cfu/g; yeast and mould ≤100 cfu/g; E. coli and Salmonella absent |
| Pyrrolizidine alkaloids | LC-MS/MS, sum of the regulated list | Reported against the destination limit |
| p-Anisidine value | Spectrophotometric | Agreed at order |
The result column needs a measured figure for every line: a number per element for heavy metals, not a single total; the count and the absence statement for microbiology; the sum, individual analytes and limit of quantification for pyrrolizidine alkaloids. A “not detected” entry without a limit describes the method, not the oil.
What moves between lots
The seed lot, the growing season and the pressing temperature move the GLA within the catalog range. Refining moves the free fatty acid, the colour and the alkaloid load much more than the fatty-acid profile. A cold-pressed oil can sit at 0.3% free fatty acid and a refined oil from the same seed at 0.05%, both reporting the same GLA; the grade explains the difference.
Peroxide value is the most variable line because it depends on the seed condition, the pressing temperature, the headspace and the time since production. A result close to the 5.0 limit leaves little room for a warm warehouse. Colour is the least quantitative line: pale yellow to golden covers a fresh cold-pressed oil and a lightly refined one, and a darker shade prompts a storage question rather than an identity failure.
A single result near an edge is less informative than a trend: GLA drifting low, peroxide climbing, or free fatty acid rising in a refined grade. The lot code on the certificate should match the bottle, the drum and the retained sample, or the trend belongs to a different material.
When to order an outside test
Order an independent test for a new source, a first shipment into a regulated market, a disputed GLA figure, a suspected blend or an unanswered PA question. The panel is narrow: GC-FAME with the chromatogram and full profile; peroxide and p-anisidine values; free fatty acid; a sterol profile if the source is disputed; a PA screen by LC-MS/MS; heavy metals and microbiology.
Ask for the limit of quantification for each screen, and choose a laboratory working to ISO/IEC 17025 for the tests in scope. Chain of custody matters: the sample tested should be the sample retained, the report should identify the lot, and the referee should receive the certificate and specification with the sample.
What to compare between suppliers
Put six lines side by side: species and process; GLA with its basis; the full fatty-acid profile; free fatty acid, peroxide and p-anisidine values; the PA statement; and pack size with lot traceability. Two certificates that agree on GLA but disagree on linoleic or oleic are not the same oil, and one without a PA statement leaves the borage-specific question unanswered.
The catalog limits are on the borage oil product page. Test terms are defined in the COA glossary entry, the method library is indexed under analytical methods, the borage-versus-evening-primrose comparison is on the evening primrose oil powder page, and lot release is described on the quality page.
Related documents
Send the specification you intend to use — grade, quantity and destination market — and we will reply with the batch COA, the technical data sheet and a quotation.
Request a quotation