How to Read a Nanomaterial Certificate of Analysis

What a certificate of analysis is, and what it is not

A certificate of analysis is a supplier's statement about a specific lot of material. That is the whole of its authority. It is not a specification, not a product description, and not a guarantee that the next lot will behave the same way. When it is doing its job it tells you what was measured on the batch in the box, by what method, and against what acceptance limits.

The trouble is that in nanomaterials the document has drifted. A large fraction of what circulates as a CoA is a product datasheet with a lot number stamped on it: nominal values, techniques unnamed, and no distinction between a number somebody measured on your batch and a number copied from the catalogue. That is not fraud, and often nobody involved intends to mislead. It is simply a document that answers a different question from the one you are asking.

This guide is the conversion page for everything else in this cluster: every technique guide on this site ends with how the result appears on a CoA, and this is where they all land.

Measured, typical, nominal: three words that are not synonyms

Before reading any value, find out which of three categories it belongs to. Good certificates label this explicitly; most do not, and asking is free.

• Measured — this lot, this instrument, this date. The only class of number that says anything about what is in the box.
• Typical — the historical average across production. Useful, honest when labelled, and silent about your batch.
• Nominal or target — the design intent. A "20 nm" gold nanoparticle product is a nominal 20 nm; the measured mean of your lot is a different number and the certificate should carry both.

The single most useful habit when reading a CoA is to go through it once and mark each line M, T or N. Most of the ambiguity in the document disappears in that pass, and what is left is a short list of things to ask about.

Reading it line by line

Identity: name, formula, CAS number

Chemical identity is the easy part and the part that misleads most. A CAS registry number identifies a substance, not a nanostructure. Nanoscale forms of a material frequently share a registry number with the bulk allotrope, which means the CAS number on a nanomaterial certificate confirms the chemistry and tells you nothing whatsoever about size, morphology, layer count or surface chemistry — the properties you are buying the material for. Treat it as a necessary check, never a sufficient one.

The naming is worse. "Graphene" is used commercially for materials ranging from genuine monolayer sheets to graphite nanoplatelets tens of layers thick, which is why ISO/TS 80004-13 exists to define the vocabulary and ISO/TS 21356-1 exists to specify how the claim should be substantiated. If a certificate says "graphene" and does not say how many layers, by what measurement, the identity line is incomplete.

Purity, and what the balance is

"Purity: 95%" is the most information-poor line on a typical certificate, because the useful question is not the 95 but the 5. Five per cent residual iron catalyst in a carbon nanotube sample is a completely different material from five per cent amorphous carbon: the first is magnetic, catalytically active, cytotoxic in some assays, and removable by acid; the second is none of those, and removing it costs you nanotubes. A certificate that quotes purity without composition of the balance has told you the least useful half.

Ask which method produced it. Carbon purity is normally a thermogravimetric number, and TGA in air reports a residue that is metal oxide, not metal — a conversion suppliers sometimes forget, which inflates the apparent catalyst content. The TGA guide works that arithmetic through. Metal purity is usually ICP-OES or ICP-MS, and there the question is whether the figure is on a metals basis or total mass, which can differ substantially for a coated particle.

Particle size — by which technique?

This is where certificates most often become internally inconsistent, because the three common sizing techniques legitimately give three different numbers for the same sample and the certificate names none of them.

• TEM or SEM gives a number-weighted geometric size of the dried, imaged particle.
• DLS gives an intensity-weighted hydrodynamic diameter in suspension, which includes the solvation layer and any coating, and is biased upward by the largest objects present.
• XRD gives a crystallite domain size, which is smaller than the particle whenever the particle is polycrystalline.

A certificate reporting "size: 30 nm" without the technique is unusable for anything quantitative. One reporting 25 nm by TEM and 60 nm by DLS is not contradicting itself — it is telling you the particles are coated, solvated, or partly agglomerated, which is exactly the information you wanted. The comparison of DLS, NTA and TEM is the long version.

Look also for the distribution, not just the mean. A mean with no width, no standard deviation, no polydispersity index and no histogram is a single statistic standing in for a distribution that may be bimodal.

Specific surface area

Usually BET nitrogen adsorption, and it is one of the more trustworthy lines on a certificate because the method is standardised in ISO 9277 and hard to fudge. Two things to check: the degas conditions, because an under-degassed sample reports a low area and the temperature has to be high enough to clear adsorbed water without sintering the powder; and whether anybody has converted the area to an "equivalent particle size" for you. That conversion assumes smooth, dense, monodisperse, non-porous, unaggregated spheres, and the BET guide shows what it does when they are aggregated, which they always are.

Surface chemistry and functionalisation

For any functionalised product this is the line you are actually buying, and it is the least well supported. "COOH-functionalised" is a claim; the supporting measurements are XPS for elemental surface composition and oxidation state, FTIR for functional-group identification, TGA for organic loading, and a titration or zeta-potential-versus-pH curve for accessible acid groups. A certificate that asserts functionalisation without any of these is asserting the process, not the product.

If XPS is quoted, note that its sampling depth is on the order of several nanometres. On a 20 nm particle that is not a surface measurement in the way the word implies — it is a substantial fraction of the whole particle, and the geometry has to be reasoned about. The XPS guide is about exactly this misreading.

Zeta potential

Meaningless without pH and ionic strength, and reported without them more often than not. A zeta potential is a property of a particle in a specified medium, not a property of the particle. See what the zeta number means and when it lies.

Concentration, for dispersions

Check the basis. Milligrams per millilitre of what — total solids, the nanomaterial only, or the metal content? For a coated particle these differ by the mass of the coating, which can be a large fraction. Check also whether a particle-number concentration is given, and if so how it was obtained; converting mass to number requires an assumed density and size distribution, and small errors in the assumed size are cubed.

Residual solvent, moisture, and endotoxin

Routinely omitted and routinely important. Residual synthesis solvent affects dispersion behaviour and downstream chemistry. Moisture affects mass-basis calculations for hygroscopic powders. Endotoxin matters for any biological application, and "low endotoxin" without a number and a method (typically LAL) is not a specification.

Lot number, dates, and storage

A lot number that does not appear anywhere else in your records is administratively useless. Look for the manufacture date, the analysis date — which is not the same date, and if the analysis predates the manufacture the document is a template — and a retest or expiry date with the storage conditions it assumes. Storage conditions are a specification, not advice.

Which technique produced which line

CoA line	Usual method	What to check
Particle size, geometric	TEM / SEM	How many particles counted; is a histogram attached
Particle size, in suspension	DLS or NTA	Medium, concentration, weighting, PDI
Crystallite size and phase	XRD	Instrumental broadening subtracted; phase assignment
Specific surface area	BET nitrogen adsorption	Degas temperature and time; multi- or single-point
Purity / residual catalyst	TGA, ICP-OES, ICP-MS	Oxide-to-metal conversion; what the balance is
Surface composition, oxidation state	XPS	Charge referencing; sampling depth vs particle size
Functional groups	FTIR, titration, TGA	Whether loading is quantified or only detected
Defect density, layer count (carbon)	Raman	Laser wavelength; height or area ratio; points sampled
Colloidal charge	Electrophoretic light scattering	pH and ionic strength stated
Optical properties	UV-Vis, PL	Solvent, concentration, path length

The fields that are routinely missing

Absence is information. The following are omitted often enough that their absence is worth a question rather than an assumption:

• The method behind every numeric line.
• The date of analysis, distinct from the manufacture date.
• The distribution width behind any size or mass mean.
• The composition of the impurity balance.
• The medium and pH for any suspension property.
• Whether the analysis was performed in-house or by a third party.
• Raw data — spectra, micrographs, thermograms — as opposed to derived single numbers.

Five questions to send back before you order

1. Which values on this certificate were measured on this lot, and which are typical or nominal? The most valuable question on the list, and the cheapest to answer.
2. What technique produced the size figure, and can I have the distribution rather than the mean?
3. What is the impurity balance made of? For carbon materials, metal or amorphous carbon; for metal particles, which metals and at what level.
4. Can I have the underlying data — the micrographs, the Raman spectrum, the thermogram — rather than the derived numbers? A supplier who has them will usually send them.
5. What are the storage conditions and the retest date, and what changes if they are not met?

None of these are adversarial questions. Suppliers who characterise their material properly are generally pleased to be asked, because it is the part of their work that never shows up in the price comparison.

Why this marketplace publishes fields rather than prose

nanoMani lists specifications as structured fields rather than a paragraph of description, and how it works sets out why. A field can be empty, and an empty field is visible. A paragraph can omit the same information and read as complete. Everything above is the argument for that design from the buyer's side: the failures in a certificate of analysis are almost never wrong numbers, they are missing context — and a form that has somewhere to put the context is the cheapest fix available.

Related guides

• The techniques behind every line above: the characterisation workflow.
• What to do with the material once it arrives: storing and handling nanomaterials safely.
• What drives the price behind the certificate: the nanomaterial price guide.