01 COA & batch testing

What the COA for micellar casein actually contains

Technical document 01 · Reviewed 2026-11-19 · Casein

On this page
  1. What the sheet has to list
  2. Kjeldahl, the factor and the dry basis
  3. Catalogue range, release limit and lot result
  4. What moves between lots
  5. When an outside laboratory earns its fee
  6. What to compare between suppliers

A certificate for micellar casein answers a narrow question: what did the lot in the drum measure, by which method, and against which limits. Four figures carry most of the answer — protein with its nitrogen factor and basis, ash with the mineral load behind it, lactose as the sugar ceiling, and moisture, which converts every dry-basis figure into the value a customer handles. The remaining lines matter, but they rarely decide the lot on their own.

In one line

A micellar casein certificate is read from the protein figure with its nitrogen factor and basis, the ash that carries the mineral load, the lactose ceiling and the moisture that sets the as-is value.

What the sheet has to list

The parameter list follows the specification, not the certificate template. It divides into composition and physical parameters on one side, and contaminants and microbiology on the other, because different people read them and different laboratories run them. Both tables are catalogue limits — the window the grade is sold inside. Particle size is a sieve pass, not a mean diameter, and pH is read on a 5% suspension. A lot certificate adds the measured result, and only the result belongs to that lot.

Composition and physical parameters · micellar casein, catalogue limits
ParameterMethodRelease limit
AppearanceVisualWhite to off-white, free-flowing powder
Protein (dry basis)Kjeldahl, N × 6.38≥90.0%
MoistureLoss on drying≤10.0%
AshGravimetric, furnace≤2.0%
FatSolvent extraction≤1.5%
LactoseHPLC≤1.0%
pH (5% suspension)Potentiometry6.5–7.5
Particle sizeSieve≥95% through 80 mesh
Contaminants and microbiology · micellar casein, catalogue limits
ParameterMethodRelease limit
Heavy metalsICP-MSPb ≤3 · As ≤2 · Cd ≤1 · Hg ≤0.1 mg/kg
Total plate countPlate count≤10,000 cfu/g
Yeast and mouldPlate count≤100 cfu/g
E. coliEnrichmentAbsent
SalmonellaEnrichmentAbsent

The method column is not decoration. A lead figure without ICP-MS beside it cannot be compared with one from another technique, and a plate count depends on the medium, the temperature and the time. Results without methods are a summary, and a summary is not what an auditor asks for.

Kjeldahl, the factor and the dry basis

Kjeldahl measures nitrogen and converts it to protein with a factor. For milk proteins the factor is 6.38, which assumes nitrogen at about 15.7% of the protein mass; the general factor of 6.25 assumes 16%. The gap is small but real: a lot that reads 90.0% with the dairy factor would read about 88.2% if the general factor were applied to the same nitrogen result. Neither number is wrong, but the factor has to travel with the figure.

The basis does more damage than the factor. This specification is on a dry basis, so the 10.0% moisture ceiling is excluded from the protein calculation. At that ceiling a 90.0% dry-basis result corresponds to about 81.0% as-is, and a purchase order that drops the basis is out by nine percentage points before any analytical variation is counted. The certificate should print “dry basis” beside the figure, and the specification should say it in both places.

Not all nitrogen is protein nitrogen. Casein carries little non-protein nitrogen, but urea or a nitrogen-rich adulterant would lift the Kjeldahl figure without lifting the protein. The factor and the basis explain the arithmetic, not the source of the nitrogen; that question belongs to the identity page.

Catalogue range, release limit and lot result

Three statements get confused on certificates. The catalogue range is the offer, the release limit is the narrower window agreed for a market or a customer, and the lot result is the measurement. The first two are commitments; only the third is evidence. A certificate that says “conforms to specification” without printing the result offers a conclusion and withholds the measurement.

Some parameters are limited and others are reported. Moisture, ash, fat and lactose have ceilings. Appearance is a description, and a colour that stays “off-white” can drift between lots. Particle size can be a sieve pass and still be worth reporting as a distribution, because a five-point change in fines is invisible in a pass-fail line and obvious on a filling line.

Everything in the tables above is a catalogue limit. This site publishes the window; it does not publish lot results, lot numbers or customer names.

What moves between lots

Micellar casein starts as skim milk, and skim milk is seasonal. Feed, stage of lactation and herd management move the protein and lactose in the raw milk, and microfiltration can only concentrate what arrives. A higher lactose load puts more work on diafiltration; if the wash is shortened, lactose and ash rise together while the protein still passes on a dry basis.

The dryer moves moisture, colour and dispersibility. A hot inlet gives a drier powder that flows well and wets slowly; a gentler dry disperses more easily and holds more moisture. Neither is a defect, but they are not interchangeable in a filling line.

Fines migrate to the top of a drum in transit, so the first scoop and the last scoop of one lot can behave differently. A properly taken sample and a retained sample matter more than a grab from the surface.

When an outside laboratory earns its fee

Independent testing is worth it at three moments: a new source, the first shipment into a regulated market, and a dispute where the two laboratories disagree. The panel should answer the open question rather than repeat the certificate: protein by Kjeldahl at a second laboratory, the mineral profile by ICP-OES or ICP-MS, an amino acid profile, SDS-PAGE for the protein pattern, a melamine and cyanuric acid screen, and the microbiological set on the same sample.

Two limits are worth stating before the sample is sent. A milk-protein ELISA confirms the material is milk-derived; it does not separate micellar casein from caseinate, because both are milk protein. A nitrogen result, however careful, does not identify the nitrogen source. The ICP-MS method page and the quality page describe how results are kept with the lot.

What to compare between suppliers

Put six lines side by side: protein with its factor and basis; ash with the calcium and sodium figures behind it; lactose; moisture; particle size and a dispersion result; and the allergen statement with site and lot traceability. Two offers that agree on 90% protein and disagree on the other five are not the same material, even when both are called micellar casein.

The second layer is the record: which site made the lot, which milk batch went in, which membrane and dryer were used, and what retained sample exists. A certificate can be accurate and still describe a different lot from the one that arrives. The grade comparison explains why ash is the first line to check when the form is in doubt, and the COA glossary entry defines the document. The specification behind the certificate is summarised on the micellar casein product page.

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