01 COA & batch testing

What the COA for biotin actually contains

Technical document 01 · Reviewed 2026-11-18 · Biotin

On this page
  1. The certificate as a line-by-line report
  2. What the assay, the rotation and the drying step measure
  3. Release limit versus batch result
  4. What moves between batches
  5. When to order an outside test
  6. What to compare between suppliers

A biotin COA is short, and most of its value sits in the columns that explain how each figure was obtained. It should identify the material by IR, quantify it by HPLC, fix the chiral form by specific rotation, and place water, ash, particle size, metals and micro results beside the limits they were released against. The values on this page are catalog limits; the lot-specific numbers belong on the certificate that travels with the pack.

In one line

The assay says how much biotin is in the drum; the rotation and the IR say what it is, and the drying basis says what the number means.

The certificate as a line-by-line report

Start with the identity block, because a lot that fails identity cannot be rescued by a good assay. The first table covers appearance, the IR match, the HPLC assay, the rotation and the two gravimetric lines.

Identity, assay and moisture · biotin
ParameterMethodRelease limit
AppearanceVisualWhite crystalline powder
IdentificationIR, compared with a reference spectrumConforms
Assay (C10H16N2O3S)HPLC97.5–102.0%
Specific rotationPolarimetry, 2% in 0.1 M NaOH+89° to +93°
Loss on dryingGravimetric≤1.0%
Residue on ignitionGravimetric≤0.1%

The second table covers the physical and contamination block. Sieve analysis describes the powder, ICP-MS describes the inorganic contaminants, and the plate count and absence tests describe the microbiological state. These are separate measurements and should not be merged into one purity figure.

Physical, contaminant and microbiological limits
ParameterMethodRelease limit
Particle sizeSieve analysis≥95% through 80 mesh
Heavy metalsICP-MSPb ≤3 · As ≤2 · Cd ≤1 · Hg ≤0.1 mg/kg
MicrobiologyPlate count and absence testsAPC ≤1,000 cfu/g; yeast and mould ≤100 cfu/g; E. coli and Salmonella absent

What the assay, the rotation and the drying step measure

HPLC answers the amount question. It does not answer the identity question, because a retention time can be shared by a related substance; the certificate needs an IR match and a rotation result alongside the assay. The method write-up should name the detector and the reference standard, since biotin's ultraviolet absorbance is weak and the detector conditions affect the result.

Rotation is the chiral measurement. The catalog window is +89° to +93°, run at 2% in 0.1 M NaOH. The observed angle depends on concentration, solvent, path length and temperature, so a certificate that omits those conditions gives the reader a number that cannot be compared with another laboratory's. The basis also matters: the figure may be calculated on the dried material or on the material as received.

Loss on drying measures water and volatiles under the stated conditions, with a ceiling of 1.0%. That ceiling is wide enough to shift an assay by about one percentage point when the basis changes. For example, 97.6% as-is with 0.9% loss on drying would be about 98.5% on the dried basis. The certificate should say which basis the 97.5–102.0% window uses, or two readers can disagree while both are correct.

Residue on ignition stands for the inorganic fraction left after ignition. It is not an ICP-MS result and should not be read as one. The same separation applies to the micro panel: a total plate count and an absence test describe different things, and both belong on the page.

Release limit versus batch result

A limit is a decision rule; a result is a measurement. The certificate needs both, in separate columns, with the method close enough that the reader can see how the number was produced. When the result column simply repeats the specification, the reader has no evidence that the lot was tested at all.

Some parameters are reported as conformity. IR is the usual case: the certificate may say “conforms” when the sample spectrum matches the reference, but the spectrum should be retained for inspection. For rotation, a bare “conforms” is less useful than the observed angle, because +89.2° and +92.8° both pass and sit at opposite ends of the window.

Absence tests also need a test portion. “Salmonella absent” is a statement about the portion that was tested, not about the whole drum, and the certificate should name that portion. A conclusion of “passes specification” should be reachable from the page, not assumed from a tick box.

What moves between batches

Assay, water content and rotation follow the synthesis, crystallisation and drying steps; milling and sieving change the particle-size result. Colour and flow are less quantitative: a slightly off-white lot can pass every release test while indicating a different drying history or a trace of process residue.

Look at the direction of travel, not only the latest result. Rotation moving toward +89°, water moving toward 1.0% or the 80-mesh fraction falling toward 95% all show a process changing while each individual lot is still released. For a formula with a tight fill weight or a low-dose blend, the previous four certificates are more informative than one.

When to order an outside test

Use an independent laboratory when the source is new, when the lot is the first into a regulated market, when a result is disputed, or when the supplier's own certificate does not settle an identity question. A focused panel is enough: IR with a reference spectrum, HPLC for the assay, polarimetry for the rotation, ICP-MS for the metals and the micro tests named in the specification.

The report should give the detection limit for each contaminant and include the raw IR spectrum and chromatogram. “Not detected” without a limit tells the reader about the method, not the sample. Keep the chain of custody intact: the sample that was tested should be the sample that was retained, and the lot code should match the certificate and the pack.

What to compare between suppliers

Compare six columns: grade and carrier; assay method and basis; rotation with its conditions; water content; particle-size result; and the contaminant and micro panel. If two offers agree on assay but diverge on the other five, they are not the same material.

Traceability is the next layer. The lot code on the certificate, the pack and the retained sample should be the same, and the manufacturing site should be named on it. A dilution needs its biotin content in mg/g and its carrier on the same page as the pure-material assay; the 97.5–102.0% window is not the right limit for a 1% blend.

The catalog limits are on the biotin product page. The grade comparison is on the grade page, the identity tests on the identity page, and the release workflow on the quality page. The HPLC method is described in the HPLC method note.

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