What an Activated Carbon Certificate of Analysis Must Contain

A certificate of analysis that says “conforms” is not a certificate of analysis. ICH Q7 §11.4 — published by the FDA as good manufacturing practice guidance for active pharmaceutical ingredients — requires that a certificate show “each test performed,” including the acceptance limits and the numerical results obtained. Both. The limit and the number. It also requires the batch number, the date of release, a date and signature from authorised quality personnel, and the name, address and telephone number of the original manufacturer. That is eight distinct requirements in one short clause, and most certificates travelling with bulk activated carbon satisfy fewer than half of them.

This matters more for carbon than for most commodities, because as covered in how to read an activated carbon datasheet, the datasheet is a selection document that publishes floors and ceilings. The COA is the only document that tells you what you actually received.

The short version

ICH Q7 §11.4 requires the acceptance limit and the numerical result for each test. A certificate showing only “pass,” “conforms” or “complies” does not meet that bar.
ISO/IEC 17025:2017 clause 7.8.2.1 lists the elements of a test report — among them identification of the method used, results with units of measurement, any deviations from the method, and identification of the person authorising the report.
Neither document was written for bulk adsorbent media. ICH Q7 governs pharmaceutical active ingredients; ISO/IEC 17025 governs accredited laboratories. No ASTM practice governing COA content for bulk media could be verified. Together these two are the only citable benchmarks that exist.
A COA certifies one lot, not a process. It says nothing about the lot you buy next quarter.
ISO 9001 certification attests to a management system, not to product quality. ISO states plainly that it “does not perform certification or issue certificates,” and that individuals are not certified to ISO 9001.

What a certificate of analysis is for

It is the evidence that a specific, identified batch was tested and what the tests returned. A datasheet describes a grade; a certificate describes a lot. The two documents do different jobs and one cannot substitute for the other.

The distinction has a third neighbour that gets confused with both: the affidavit of compliance. ANSI/AWWA B604, the US standard for granular activated carbon in water treatment, requires at Sec. 6.3 that the supplier provide an affidavit indicating the product proposed for use complies with the standard and the purchase documents with no exceptions. That is a statement of conformance. It is a useful document, and it is not a certificate of analysis, because it carries no results.

What must a certificate of analysis contain?

The most specific, citable requirement list in public circulation is ICH Q7 §11.4, as published by the FDA. It was written for active pharmaceutical ingredients, but its content requirements are generic to any tested, batch-released material.

Requirement (ICH Q7 §11.4)	What it means on a carbon COA
Authentic certificates issued for each batch	One certificate per production lot, on request — not a generic sheet reused across shipments
Name of the material, including grade where appropriate	“SURSORB PAC-1050, 4 mm pellet,” not “activated carbon”
Batch number	The production lot identifier, traceable back to the manufacturing record
Date of release	When quality released the lot, distinct from the date the certificate was printed
Each test performed, with acceptance limits and the numerical results obtained	Iodine number, ASTM D4607, limit min 1050 mg/g, result 1104 mg/g
Dated and signed by authorised personnel of the quality unit	A named individual with quality authority, not an unsigned PDF
Name, address and telephone number of the original manufacturer	The plant that made the carbon, identifiable and contactable
For repacked material: repackager details plus a reference to the original manufacturer’s certificate	The chain back to the production lot, unbroken

Every element on this specimen traces to a clause in ICH Q7 §11.4 or ISO/IEC 17025 clause 7.8.2.1. The line that most often goes missing is the numerical result.

Why “conforms” is not a result

Because it destroys the information you bought the test for. Consider two lots of the same grade against an iodine number specification of minimum 1050 mg/g. Lot A returns 1180 mg/g. Lot B returns 1052 mg/g. Both conform. On a certificate that prints only “conforms,” they are identical documents.

They are not identical carbons. Lot B is sitting on the limit, which tells you something about where the production window is running, and it is the lot that will generate a dispute if the receiving laboratory’s result differs by 1 %. That difference is exactly the situation ASTM E29-22 governs — the absolute conformance method and the rounding method can return opposite verdicts on a borderline number, and E29 binds only if the specification cites it. A “conforms” certificate makes that argument impossible to have, because the number is not on the page.

Three practical uses of the numerical result that “conforms” removes:

Trend detection. Ten certificates with numbers show whether a supplier’s process is drifting toward a limit. Ten certificates saying “conforms” show nothing.
Dispute resolution. If your laboratory and the supplier’s disagree, the comparison starts from two numbers. Without the supplier’s number there is nothing to compare.
Design margin. If you sized a bed on a minimum and the delivered lot is well above it, that is real margin you can account for. You cannot account for margin you cannot see.

What ISO/IEC 17025 adds

ISO/IEC 17025:2017 is the international standard for the competence of testing and calibration laboratories, and its clause 7.8.2.1 enumerates what a test report must contain — sub-clauses (a) through (p). The list is longer than ICH Q7’s and covers the laboratory side of the same problem. The elements that matter most when reading a carbon certificate:

Identification of the method used. Not “iodine number” but ASTM D4607, with its revision. Without it, you cannot tell whether two laboratories ran the same test.
Unambiguous identification of the item tested, and the dates of receipt, sampling and performance of the test.
Results with the units of measurement. As set out in the datasheet article, 0.5 g/cc and 500 g/L are the same number, and a certificate that omits the unit is not reportable data.
Any additions to, deviations from, or exclusions from the method. This is the clause most often quietly ignored. If a laboratory used a shortened contact time or a variant procedure, the report must say so.
Identification of the person or persons authorising the report.
A statement that the results relate only to the items tested, plus identification of any external providers used.

A carbon certificate is not a 17025 accredited test report unless it is issued by an accredited laboratory within its scope, and most manufacturer certificates are not. That does not make 17025 irrelevant. It makes it the reference for what complete reporting looks like, and it gives you a defensible list of questions.

Being straight about the two benchmarks

Neither ICH Q7 nor ISO/IEC 17025 was written for activated carbon, and it would be dishonest to imply otherwise.

ICH Q7 is Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients. Its subject is APIs and API intermediates. A pelletised carbon for biogas desulphurisation is not within its scope, and no regulator will hold a carbon plant to it. ISO/IEC 17025 governs the competence of testing laboratories. It applies to the laboratory issuing a report, not to a manufacturer issuing a shipment certificate.

We also could not verify any ASTM practice governing certificate-of-analysis content for bulk chemicals or adsorbent media. If one exists we have not found it in public material, and we are not going to invent a citation.

So the honest framing is this: these two documents are the only citable benchmarks that exist for COA content, and they are borrowed. They are borrowed from two industries that took the question seriously and wrote it down. Applying them to bulk carbon is a choice, not an obligation. It is a defensible choice, because the elements they require — batch identity, method identity, limit, result, units, date, signature, manufacturer identity — are the elements any buyer needs regardless of industry.

Repacked, re-bagged and traded carbon

This is where the ICH Q7 requirements earn their place in a carbon article. A large fraction of activated carbon reaching end users has passed through a trader, a repacker or a blender. Material may be produced in one country, shipped in bulk, re-bagged into 25 kg sacks under a different brand, and sold on.

ICH Q7 handles that case explicitly: a certificate issued by a repacker or reprocessor must show the repacker’s name, address and telephone number, and include a reference to the original manufacturer’s certificate. Applied to carbon, that means the certificate in your hand should let you get back to the plant that made the material and to the certificate that plant issued for the production lot.

Questions worth asking when a certificate arrives on a trader’s letterhead:

Which plant produced this material, and where is it?
What is the original manufacturer’s lot number, and does the original certificate exist?
Were the reported results measured on this lot, or copied from the producer’s certificate?
If the material was re-bagged, was anything re-tested after handling? Moisture is the obvious candidate — ASTM D2867 notes that moisture in carbon packed in typical shipping containers usually increases during transport and storage.

None of that is an accusation. Repacking is a legitimate and necessary part of the supply chain. It is simply a break in the documentary chain that a properly written certificate repairs.

What a certificate of analysis does not tell you

It certifies one lot, not a process. A certificate is a snapshot of material that has already been made. It carries no commitment about the next lot, and it is not evidence of process capability. If lot-to-lot consistency matters to you, ask for certificates from several consecutive lots and read the spread of the numbers — which you can only do if the numbers are printed.

It does not predict service performance. Every capacity-type ASTM method in this family states that its result is a quality-control or comparability index. ASTM D3467 says CTC activity does not necessarily measure effectiveness on other adsorbates or at other conditions. ASTM D3802 says ball-pan hardness does not measure in-service resistance to degradation, only the comparability of lots (the method is set out here). ASTM D6646 says its accelerated H2S method does not duplicate the conditions an adsorber meets in practical service. A COA full of conforming results is not a prediction of bed life.

And a management-system certificate is not a product certificate. ISO 9001:2015 defines requirements for a quality management system, and in ISO’s own words it “does not prescribe how an organization must operate.” ISO is equally direct about what certification is and who does it: ISO “does not perform certification or issue certificates,” which is done by independent certification bodies that may themselves be accredited by national accreditation bodies. ISO also states that individuals are not certified to ISO 9001.

Put those together and the position is clear without needing to overclaim: an ISO 9001 certificate is third-party attestation that an organisation’s management system conforms to the standard. It says nothing about any individual lot of carbon. It is a reasonable thing to ask a supplier for, and it is not a substitute for a certificate of analysis. (ISO 9001:2015 is currently under revision, with a replacement expected around September 2026.)

Where Suracsh stands

Suracsh Filters Pvt Ltd holds ISO 9001:2015, certificate number Q-707024081304, issued by MyCert on 13 August 2024 and valid to 12 August 2027, for the scope “manufacturing of NBC filters, activated carbon & filtration media.” The certificate is a management-system attestation, which is exactly the point made above: it says nothing about any individual lot of carbon, and it is not a substitute for the certificate of analysis that ships with your container. The in-house R&D unit is separately recognised by the Government of India’s Department of Scientific and Industrial Research, F.No. TU/IV-RD/5079/2025, to 31 March 2027.

Our published datasheets currently carry defects, set out in full in how to read an activated carbon datasheet — including a hydrogen sulfide capacity printed without a test method and a methyl iodide efficiency printed without a protocol. Those are being corrected at source. They are also a demonstration of why the certificate matters more than the sheet: a properly written certificate names the method, so it cannot carry that particular defect.

A redacted specimen certificate of analysis for a real production lot will be published on this page once the factory has cleared it — with the customer, price and consignment details removed and the test data intact. Publishing the template is easy; publishing a real one is the part that counts.

The checklist to send with a purchase order

Ask for these, in writing, before the first shipment rather than after a dispute:

A certificate of analysis for each production lot, referenced to the lot number on the bags.
Every test named by designation and revision — ASTM D4607-14(2021), not “iodine.”
The acceptance limit and the measured numerical result, with units, for every test.
The date of release and the name and signature of the authorising quality person.
The name, address and telephone number of the plant that produced the material.
Where the material has been repacked, a reference to the original manufacturer’s certificate.
Which ASTM E29 conformance method applies to borderline results, and at what rounding interval.
Which laboratory performed each test, and whether it is accredited for that method.

A supplier that publishes its specifications openly should be able to answer all eight without a meeting. The full specification for every grade named here is published in HTML at suracsh.us, and every datasheet downloads without a form or an NDA; the certificate is the document that tells you what arrived on the truck.

Sources

Questions

What should an activated carbon certificate of analysis contain?

Per ICH Q7 section 11.4: the product name and grade, the batch number, the date of release, each test performed with its acceptance limits and the numerical result obtained, a date and signature from authorised quality personnel, and the name, address and telephone number of the original manufacturer.

Is a certificate of analysis that says conforms acceptable?

No. ICH Q7 requires the acceptance limits and the numerical results obtained for each test. The word conforms hides where in the range the lot fell, which is exactly the information you need to spot drift between lots or a result sitting hard against a limit.

Does ISO 9001 certification mean the carbon meets specification?

No. ISO 9001 attests that a management system conforms to the standard's requirements. ISO states that it does not perform certification or issue certificates, and that individuals are not certified to ISO 9001. Product conformity is evidenced by test data on a certificate of analysis, not by a management system certificate.

What should a COA show for re-bagged or traded activated carbon?

ICH Q7 requires that a certificate issued by a repacker or reprocessor shows the repacker's name, address and telephone number, and includes a reference to the original manufacturer's certificate. Without that reference, the chain back to the production lot cannot be established.

Is there an ASTM standard for certificate of analysis content?

No ASTM practice governing certificate of analysis content for bulk chemicals or adsorbent media could be verified. The citable benchmarks are ICH Q7 section 11.4, written for pharmaceutical ingredients, and ISO/IEC 17025 clause 7.8, written for accredited testing laboratories.

Check this against your own duty Everything here is general engineering and published data, not an assurance of performance in your service. Send the contaminant, concentration, flow, temperature and humidity and we will tell you what we would trial — trial batches run from 1 to 10 tonnes.