How to Read a Metal Powder Certificate of Analysis
A metal powder certificate of analysis is one of the few documents a buyer can read in ninety seconds and know whether a delivery is qualified. Most of them are technically valid and practically useless, because they report typical values instead of measured ones, or omit the method so the number cannot be compared to anything.
If a document could describe any shipment of this material, it is a datasheet. A certificate of analysis must be tied to one lot, and every number on it must be a measurement of that lot rather than a typical value. That single test disqualifies more documents than any other check.
Our own position is stated plainly: we will supply a lot certificate for the lot we ship, and if a field is not measured we will say so rather than print a typical value in its place.
On a metal powder certificate of analysis, which fields are gatekeepers?
Read the certificate twice: once for the fields that can make you reject a delivery, and once for the fields that are simply useful. Mixing them up is how people end up refusing a perfectly good lot over a rounding difference while missing a missing method.
| Field | Gatekeeper? | What has to be there |
|---|---|---|
| Lot / heat number | Yes | Unique, and traceable back to a production record. Without it the other fields are unattributable |
| Alloy designation | Yes | Full designation including the grade — for example Ti-6Al-4V Grade 5 versus Grade 23 ELI, which are different products |
| PSD method and D10/D50/D90 | Yes | A named method, not just a band. Sieving to ASTM B214 / ISO 4497 or laser diffraction to ISO 13320 — and the D-values, not only the range |
| Oxygen, with method | Yes | The measured figure and the limit the lot is held to. This is the field that separates grades in titanium |
| Nitrogen and hydrogen | Yes | Reported separately, not lumped into “interstitials” |
| Full chemistry | Yes | Against the alloy specification — ASTM F2924 for Ti-6Al-4V, ASTM F3318 for AlSi10Mg |
| Sphericity / morphology | Usually | A figure rather than an adjective, or SEM images. Sphericity of 0.9 or better is a common LPBF screening gate |
| Hall flow | Usually | Seconds per 50 g with the funnel size. Method to ISO 4490. Ni superalloys are specified far tighter than general grades |
| Apparent / tap density | Informational | g/cm³ for the specific alloy, to ISO 3923-1. Compare only within the same material |
| Production route | Informational | Gas atomised, VIGA, EIGA, PREP or plasma. Not a quality grade by itself, but it explains the cost |
| Sampling and date | Informational | How the sample was taken, to a sampling standard such as ASTM B215, and when. Old certificates are worth asking about |
| Reuse declaration | Informational | Virgin feedstock? Blending? Cycle count? This belongs to the powder maker's qualification, not to a standard |
Standards referenced are the ones a buyer normally writes onto a purchase order. Requirements vary by application and jurisdiction — check your own quality system. Nothing here is a quotation.
Scroll the image sideways to read it →
Three ways a certificate gets fudged without being wrong
Typical instead of measured
The most common problem, and it is rarely dishonest. A field reads “0.15 % max” or “0.9 min” because that is the specification, not because anyone measured your lot. Look for the words typical, nominal or target, and for limits written as “max” or “min” with no measured value beside them. A specification is a promise; a certificate should be a report.
A number with no method
A PSD reported without a method cannot be compared to another PSD reported without a method, because the two are not the same measurement. Sieve analysis and laser diffraction will not agree, and neither is wrong. This is why the method field is a gatekeeper and not a footnote — and the general vocabulary for all of it sits in ISO/ASTM 52900, which is worth reading once.
The datasheet in a certificate's clothing
A clean, glossy document with no lot number, no date and no signature is a datasheet that got attached to a delivery note. It is not a fraud and it is not worthless — it tells you the product — but it does not tell you anything about the powder in the container in front of you.
Two certificates for the same alloy, side by side
Two suppliers quote the same alloy in the same quantity. Both send a document called a certificate of analysis. One is usable and one is not, and the difference is visible in about ninety seconds.
The two documents below are worked examples built to show what each field should look like. They are not certificates we have issued and the values are not measurements of any lot we have shipped — they illustrate the difference in how a field is written.
| Field | Document A | Document B |
|---|---|---|
| Alloy designation | Ti-6Al-4V (Grade 5), powder for LPBF | “Titanium powder” |
| Lot number | Lot Ti-6Al-4V-2607-14, sampled 2026-07-14 | not stated |
| Oxygen | 0.13 wt%, measured, method named | “0.15 % max” — a specification |
| PSD | D10 / D50 / D90 by laser diffraction to ISO 13320 | “15–45 µm” |
| Sphericity | 0.94 mean, SEM images attached | “high sphericity” |
| Flow | 24 s/50 g to ISO 4490 | “good flow” |
| Traceability | Lot number resolves to a production record and a COA we can re-request | no lot to trace |
Document A is a certificate of analysis for a lot. Document B is a brochure that happens to be attached to an invoice. Both are technically printable, neither is fraudulent, and only one of them survives a supplier audit.
What to do when a field is missing
Ask, in writing, and be specific about which field. “Is this powder certified?” gets a cheerful yes; “what is the measured oxygen content for lot XXXX, and by which method?” gets a usable answer or a useful silence.
If a supplier cannot answer a gatekeeper field, you have three honest options: accept the gap knowingly and record it in your own quality file, source the material elsewhere, or pay for the measurement. All three are legitimate. What is not legitimate is filling the gap yourself with a typical value and then treating the purchase as qualified.
One line on your certificate is also an export-control question
The chemistry table on a CoA decides more than whether your part meets specification. For some alloys it decides whether the shipment can leave the country at all.
MOFCOM and the General Administration of Customs Announcement No. 18 of 2025 puts
seven medium and heavy rare earths under export control, including
scandium. The scandium entries sit at 1C907, and
1C907.a.2.a covers scandium-aluminium alloys. Note 1 states that the
controlled alloys include ingots, bars, wire, sheet, plate, tube, granules and
powders — powder is named explicitly, not implied.
Compare two otherwise familiar certificates. An AlSi10Mg CoA lists aluminium, silicon and magnesium, no rare earth appears, and the alloy matches none of those entries. An AlMgSc CoA carries a scandium figure, and that figure is what triggers the question. Nothing differs in kind between them: same format, same methods, same measured values.
A scandium figure is a trigger to check, not a verdict. Whether a given scandium-bearing alloy meets the controlled definition is a question for MOFCOM. We do not answer it by inference; we ask, and we tell you the answer we got. Anyone who says scandium content by itself settles the question is guessing.
The stated procedure applies to every shipment: declare in the customs remarks whether the item is controlled, and if it is, give the dual-use export control code. Customs may query the declaration, and goods are not released while a query is open. The HS codes in the announcement are explicitly for reference only — the current tariff and control list govern. If your program needs this settled before you commit, ask before you order; our powder export eligibility by country page covers the process.
Control position stated as of 2026-10-13, from MOFCOM and GAC Announcement No. 18 of 2025 (issued 2025-04-04). Verify against the original text before relying on it, and treat this as information rather than legal advice.
How this connects to the rest of your supply
A certificate is one document in a chain, and each link has an owner:
- Before the order — the specification is agreed in writing, ideally against a named standard. Our metal powder buyer's guide covers what that specification should contain and why.
- On arrival — check the container, the seal and the lot number against the paperwork before anything is opened. If they disagree, stop.
- After opening — record the lot against your build record. This is what makes a certificate traceable rather than decorative.
- Between builds — reuse is a separate decision with its own limits, covered in our separate guide to reusing 3D printing powder, which goes into the cycle limits in full.
If the order also needs to cross a border, check powder export eligibility by country — that is a different question from whether the material is any good, and it is worth answering before you commit to a date.
Where this leaves you
The habit worth forming is small and unglamorous: before accepting any powder delivery, find the lot number and check that at least one gatekeeper field carries a measured value and a named method. If it does, you have a certificate. If it does not, you have a brochure with a barcode.
We supply metal powder on its own or against printed parts — see metal and nylon powder supply for the grades. Two grades we are asked about more than any other are Titanium Ti-6Al-4V (Grade 5) and Stainless Steel 316L — both have a CoA field worth reading closely, because oxygen and nitrogen are reported separately and are the two numbers that decide whether a lot is in spec for your application. Our material pages list what we work to for each one. If you want a written price against a specification you already have, send it through the quote form and we will tell you which of these fields we can report and which we cannot.
The yield number behind your particle size distribution
Part of PSD is upstream of the powder: fine-powder yield, the share of a melt or wire that can be cut into the band a machine uses, typically 15–45 µm for laser powder bed fusion. The rest becomes coarser fractions or recycles as feedstock. Yield is the largest single lever on the cost of a spherical lot, and one reason the same alloy can be quoted far apart.
Fraunhofer ILT reported in Powder Technology on 2026-10-10 that a hybrid plasma-gas atomisation route reached a fine-powder yield of 73.22 % in the 10–45 µm band. The route is crucible-free and induction-free, with one direct-current plasma torch and gas atomisation tightly coupled into the same system.
Read the benchmark material before drawing a conclusion from that figure. It was measured on commercial 308L stainless steel wire. No published equivalents exist for titanium or aluminium yet, so the lower titanium and aluminium powder prices promised from this line of work are not currently purchasable.
Two questions follow for any supplier: what does your fine cut yield, and is the band you quote the fine cut or the total distribution. The first says where the cost sits; the second says whether you are being quoted the part you need or the whole thing it was cut from.
Yield figure as reported by Fraunhofer ILT in Powder Technology, 2026-10-10, read via an industry weekly that cites it; the journal text itself sits behind a paywall. Figures as of 2026-10. Nothing here is a specification we offer or a price we quote.