A Certificate of Analysis (COA) is the one document that tells you, batch by batch, whether what’s in the container matches what’s on the contract. Most buyers check that the numbers are “within spec” and move on. But a handful of these parameters explain far more about the batch than a pass/fail glance suggests — and knowing what to look for can save you a rejected shipment or a formulation problem three months down the line.
We’ll use sodium tripolyphosphate (STPP) as the working example, since it’s one of the most widely traded food phosphates, but the same six parameters apply — with different limits — across the whole phosphate family: TSPP, SAPP, DSP, and the potassium and calcium salts alike.
1. Main content / assay
This is the headline number — the percentage of the actual active compound in the powder, everything else being moisture, processing residue, or related phosphate species. For food grade STPP under GB 1886.335-2021, the minimum is 85.0%. Industrial grades run higher: 90–96% depending on class.
A batch sitting right at the minimum isn’t necessarily a problem, but it’s worth watching over several shipments — if a supplier’s main content keeps drifting toward the floor of the spec range, that’s usually the first sign of a process control issue before it shows up anywhere else.
2. Total phosphate (as P₂O₅)
This measures total phosphorus content, expressed as P₂O₅ equivalent, and it’s arguably more informative than the main content figure alone. Main content tells you how much of the target compound is present; total phosphate tells you the overall phosphorus loading — which is what actually drives functional performance in your application (water retention, leavening power, buffering capacity).
Why this matters in practice
Two batches can both pass “main content ≥ 85%” but have meaningfully different P₂O₅ readings if one carries more related phosphate impurities. For dosage-sensitive applications like bakery leavening, ask your supplier for both numbers, not just the headline assay.
3. Water insolubles
The fraction of the product that won’t dissolve — typically silica, iron compounds, or incompletely reacted raw material carried over from production. Food grade limits are tight (0.1–0.2% is typical across the phosphate family) because insolubles show up as visible sediment or haze in finished products, particularly in clear beverages and brines.
This number rarely fails outright, but a rising trend across shipments from the same supplier is worth a conversation — it often points to raw material quality drifting before other parameters move.
4. pH value
Measured on a standard solution (commonly 10 g/L, sometimes 1%), pH confirms the product is the compound it claims to be — sodium phosphates each have a characteristic pH range, and a reading outside that band can indicate cross-contamination between grades during packing, or an off-spec production run.
pH also matters functionally: it’s often the single biggest factor in how a phosphate performs in your process, from meat brine penetration to dough leavening rate.
5. Heavy metals, arsenic & lead
These are the parameters regulators focus on for good reason, and they’re the ones we’d encourage buyers to actually read rather than skim. Food grade limits are typically expressed in mg/kg (ppm): lead around 2–4 mg/kg, arsenic around 3 mg/kg, and a combined “heavy metals as Pb” figure around 10 mg/kg, depending on the specific phosphate and standard.
Because these limits are set close to what’s achievable with clean raw material and good process control, a batch trending upward — even while staying technically within spec — is worth flagging to your supplier before it drifts further.
6. Fluoride
Fluoride is a natural companion element in phosphate rock, and every phosphate manufacturer has to control it through the production process. Limits vary noticeably by grade: food grade STPP is capped around 20–50 mg/kg, while feed grade allowances can run into the hundreds — the difference reflects how each end use tolerates fluoride exposure over time.
If you’re buying the same nominal product for two different applications (say, a food line and a feed line), check that you’re actually receiving the grade with the fluoride limit your application needs — it’s an easy mix-up when purchasing teams order by product name alone.
Reading them together
No single number on a COA tells the whole story. A batch with a strong main content assay but insolubles creeping toward the limit is a different quality signal than a batch with borderline heavy metals but everything else comfortable mid-range. The most useful thing you can do as a buyer is track these numbers shipment over shipment, not just batch by batch in isolation — patterns show up in trends long before they show up as a failed spec.
Frequently asked questions
Is a COA the same as a lab test I can independently verify?
Not automatically — a COA reflects the manufacturer’s own quality control testing for that specific batch. For shipments where independent verification matters, request third-party pre-shipment inspection (SGS, BV, or similar) alongside the COA.
What should I do if a COA value is right at the spec limit?
A value at the limit still passes, but it’s worth asking your supplier whether that’s typical for the batch or an outlier. Reputable manufacturers can usually show you a trend across recent shipments on request.
Do industrial grade and food grade COAs test the same parameters?
Mostly yes, but food grade adds tighter controls on heavy metals, arsenic, and fluoride, and typically includes a pH specification. Industrial grade COAs more often report whiteness and iron content instead, which matter for applications like detergents and ceramics.