01 COA & batch testing

What the COA for creatine monohydrate actually contains

Technical document 01 · Reviewed 2026-11-18 · Creatine Monohydrate

On this page
  1. Reading the certificate line by line
  2. What the HPLC assay is measured against
  3. The limit column and the lot result
  4. Where lots differ
  5. When to bring in a referee laboratory
  6. A side-by-side supplier check

A creatine monohydrate COA should let a reader audit the lot: appearance, IR identity, HPLC assay, creatinine, dicyandiamide, loss on drying, residue on ignition, mesh, heavy metals and microbiology. Each line needs the method and the release limit that applied, not a bare pass. The ranges below are the catalog specification; the batch COA is the record for the drum in front of you.

In one line

The certificate has to show the assay basis, the creatinine and dicyandiamide figures and the water of crystallisation, not only a purity number.

Reading the certificate line by line

Auditing a COA means reading every line for three things: the parameter, the method that produced the value, and the limit the lot had to meet. A sheet with only a parameter and a result is a summary; it cannot be checked.

Identity, assay and physical tests · creatine monohydrate
ParameterMethodRelease limit
AppearanceVisualWhite crystalline powder
IdentificationIR compared with a reference spectrumConforms
AssayHPLC, calculated on the dried basis99.0–101.0%
CreatinineHPLC or another specific method named on the certificate≤100 ppm
DicyandiamideHPLC or another specific method named on the certificate≤50 ppm
Loss on dryingGravimetric≤12.0%
Residue on ignitionGravimetric, ashing≤0.1%
MeshSieve analysis80 or 200 mesh, as ordered
Contaminant and microbiological limits
ParameterMethodRelease limit
LeadICP-MS or ICP-OES≤3 mg/kg
ArsenicICP-MS or ICP-OES≤2 mg/kg
CadmiumICP-MS or ICP-OES≤1 mg/kg
MercuryICP-MS or ICP-OES≤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 HPLC assay is measured against

The 99.0–101.0% figure is an assay on the dried basis, not a percentage of the powder as sold. The monohydrate carries roughly one molecule of water per creatine molecule, about 12% of its mass, so an as-is figure of 99% would leave no room for the water that defines the hydrate. The certificate should state the basis, and the calculation should use that lot's loss-on-drying result, not a fixed factor.

The method also needs enough detail to be repeatable: column, mobile phase, detection mode, standard and the integration rule. Creatine has a weak chromophore, so a low-wavelength UV method has to be validated for interference from creatinine and from the sample matrix. Peak purity or a diode-array check belongs on the method summary, because a single retention time can match more than one compound.

Creatinine and dicyandiamide are separate lines, not figures inferred from the assay. Creatinine is the cyclised degradation product that forms when creatine meets moisture and heat; dicyandiamide is a nitrogen-rich process impurity associated with the cyanamide route. Both can sit inside a passing assay, which is why the specification caps them independently.

The limit column and the lot result

The limit is the gate; the result is the measurement. They should sit in separate columns with the method alongside, so the reader can decide whether the lot passed rather than taking the word “pass” on trust. A result column that repeats the specification range is not a result at all.

Creatinine and dicyandiamide show why the distinction matters. A lot at 20 ppm creatinine and a lot at 90 ppm both pass a 100 ppm limit, but they are not equivalent starting points for a formula that will see heat or water. The same is true of residue on ignition: 0.02% and 0.09% both pass 0.1%, while the second leaves much less room for an inorganic contaminant.

Loss on drying is the line most often misread. The limit is high by design because the water of crystallisation is part of the material. A result far below 12.0% is not a cleaner lot; it may be an over-dried or partly anhydrous powder, and the hydrate identity should be checked before the result is accepted. Mesh is a grade descriptor rather than a purity limit, and the certificate should name the sieve cut and the test method rather than only “80 mesh” or “200 mesh”.

Where lots differ

Synthesis and crystallisation set the impurity baseline. The cyanamide route can leave dicyandiamide, while a route that starts from sarcosine has a different impurity map. Crystallisation and drying control the water of crystallisation, and drying is also where creatinine can climb if the temperature is pushed to reach a low moisture figure. Milling and sieving then set the mesh and the flow, without changing the chemistry.

Assay usually moves in a narrow band, which is why a single figure near the edge says less than a trend across lots. Creatinine drifting upward across three lots points to the drying step; dicyandiamide drifting upward points to the synthesis or the wash; residue on ignition drifting upward points to desalting. Colour and flow are less quantitative but can reveal a hot dry or a fines-heavy mill cut before the numbers do.

When to bring in a referee laboratory

Order an independent test for a new source, a first shipment into a regulated market, a disputed result, or any identity claim the supplier's certificate does not prove. The useful panel is not a full repeat of every line: IR for identity, HPLC for assay and peak purity, creatinine and dicyandiamide by a specific method, loss on drying, and ICP-MS or ICP-OES for the heavy-metal panel, all on one submitted sample.

Every screen should carry its detection limit. “Not detected” without a limit describes the method's sensitivity, not the sample. The assay report should also name the column, detector and calculation basis, because a percentage without those details cannot be reproduced.

A contested result is the other reason to go outside. Send the referee the retained sample, the supplier's COA and the agreed specification, and ask for the tests that distinguish the monohydrate from a dried or blended powder. Keep the chain of custody intact: the material tested must be the material you retained.

A side-by-side supplier check

Put six lines side by side: hydrate and loss on drying; assay and basis; creatinine; dicyandiamide; residue on ignition; mesh and flow. Two quotations that match on assay but diverge on the other five are not describing the same powder. Add the heavy-metal method and the microbiological panel, because “conforms” without a method is not comparable.

Traceability is the other half. The lot code must match across the COA, the drum and the retained sample, and the certificate should identify the manufacturer, grade and pack. A genuine certificate for the wrong drum is still the wrong document.

The catalog specification sits on the creatine monohydrate product page. The mesh and flow question is set out on the grade page; the full identity sequence is on the identity page; the release workflow is on the quality page. The assay method is described in the HPLC method note, and the moisture calculation is defined in the loss-on-drying glossary entry.

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