What the COA for fish oil actually contains
The certificate for a refined fish oil lot reports appearance, identification by the fatty-acid profile, EPA and DHA by GC, free fatty acid, peroxide value, p-anisidine value, moisture, heavy metals and microbiology. Each line needs its method, its release limit and a measured result in a separate column; the limits here are the catalog specification, not a batch result.
A fish oil certificate is read as a set: the EPA and DHA figures need their GC basis, the peroxide and anisidine values need each other, and the contaminant panel belongs to the same lot.
The certificate as a lot report
A certificate is auditable when the reader can see what was measured, how, and against which limit. For fish oil the identity line is a profile, not one peak: the oil is a triglyceride mixture.
The header should carry the lot code, species or mix, origin, production date and CAS 8002-50-4. That entry covers fish oil as a class, so it cannot replace the species line.
| Parameter | Method | Acceptance limit |
|---|---|---|
| Appearance | Visual | Clear to slightly hazy; pale yellow to golden |
| Identification | GC-FAME, full profile against a reference | Profile conforms |
| EPA (C20:5 n-3) | GC-FAME | 18.0–22.0% |
| DHA (C22:6 n-3) | GC-FAME | 12.0–16.0% |
| Free fatty acid, as oleic | Acid–base titration | ≤0.50% |
| Moisture | Karl Fischer titration | ≤0.1% |
EPA and DHA are separate lines because a total hides the split: 30% can be 20% EPA with 10% DHA, or 15% with 15%, and the ratio differs. DPA (C22:5 n-3) sits outside both figures.
The GC basis and the oil basis
The method transesterifies the oil to methyl esters, separates them on a capillary column and measures them with a flame-ionisation detector. Results are usually area per cent of the total fatty-acid peak area, sometimes weight per cent after response-factor correction; the certificate should state the basis, column and standard.
An area percentage is not milligrams per gram of oil. The glycerol backbone and the unsaponifiable fraction sit outside the fatty-acid total, so 20% EPA by area is not 200 mg of EPA in every gram without a conversion. A label dose in milligrams needs an internal standard or a validated factor, not the area per cent multiplied by the fill.
Free fatty acid is titrated as oleic acid. It measures hydrolysis, not EPA, and a low FFA does not prove freshness: an oil can be barely hydrolysed and still oxidised. Moisture by Karl Fischer is small, but it belongs on the certificate because water drives hydrolysis and a wet oil oxidises faster.
Release limit, result and the oxidation pair
The release limit is the boundary; the result is what the laboratory measured. Both belong on the page in separate columns, with the method beside them. A result column that repeats the specification range is a copy of the gate, not a measurement.
Oxidation is read as a pair. Peroxide value counts hydroperoxides, the first products; p-anisidine value counts the aldehydes that follow. PV can fall as hydroperoxides decompose while p-AV climbs, so a low PV on an older lot is not proof of freshness. TOTOX is twice PV plus p-AV; the catalog limits allow a TOTOX of 30, while the GOED voluntary monograph uses a ceiling of 26. A buyer who wants 26 writes it into the purchase specification; it is a benchmark, not a statutory limit.
| Parameter | Release limit | Evidence on the certificate |
|---|---|---|
| Free fatty acid | ≤0.50% as oleic | Measured percentage, with the titration method |
| Peroxide value | ≤5.0 meq O₂/kg | Measured value, method and test date |
| p-Anisidine value | ≤20.0 | Measured value; TOTOX if agreed |
| Moisture | ≤0.1% | Measured percentage by Karl Fischer |
| Heavy metals | Pb ≤3 · As ≤2 · Cd ≤1 · Hg ≤0.1 mg/kg | One measured value per metal, in mg/kg |
| Microbiology | Plate count and absence tests | The plate count and the absence result for the lot |
Report heavy metals per element; a single sum hides the element that matters. Microbiology needs a count plus an absence statement, not a tick. “Conforms” is a conclusion the reader should reach from the numbers.
What moves between lots
Species, fishing ground, season and refining move the profile. Anchovy, sardine and menhaden do not deliver the same EPA to DHA ratio, so a species change without a specification change is a material change. Refining moves free fatty acid, colour, moisture and oxidation more than it moves the fatty-acid ratio, but each step adds heat and oxygen contact.
An 18/12 designation is a commercial target, not a species. Two conforming lots can sit at 18/12 and 22/16, about 300 mg and about 380 mg of EPA+DHA per gram. That difference is visible to the formulator and invisible to a label that says only “fish oil”.
Colour runs from pale yellow to golden; a greenish or brown cast can point to oxidation or a different source. Haze at a cool temperature can be a wax fraction, not an adulteration signal. Odour is the least quantitative release check and one of the most useful: a fresh oil smells mild, an oxidised one is fishy or paint-like.
When an independent test earns its place
An outside laboratory is worth using when the source is new, a market is being entered for the first time, an oxidation result is disputed, or the certificate does not prove a form or species claim. The panel should close the gaps rather than repeat the certificate: GC-FAME with the chromatogram and EPA, DHA and DPA separately; PV, p-AV and TOTOX by named methods; free fatty acid; moisture; ICP-MS for the metals; dioxins and PCBs by GC-HRMS where the market requires them; intact-lipid analysis for triglyceride against ethyl ester; and sterols or cholesterol where a vegetable-oil dilution is suspected.
Every screen should come with its limit of quantification: “not detected” without a limit describes the method, not the oil. Keep the chain of custody intact, test the retained sample, and put the lot code on the report. In a dispute, send the referee the retained sample, the supplier certificate and the agreed specification, and ask for the discriminating test rather than another run of the same assay.
Supplier comparison
Compare six lines: species and source; form and concentration; EPA and DHA with their basis; FFA, PV, p-AV and TOTOX; the contaminant panel; and pack size with the lot code. Two offers that agree on “30% omega-3” but differ on species, form or oxidation describe different oils.
Traceability: who refined the lot, which steps were applied, whether an antioxidant was added and whether the pack was nitrogen-blanked. The catalog specification sits on the fish oil product page. The identity tests are set out on the identity page; the oxidation story is on the stability page; the separation method is indexed under methods; the release workflow 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