A food-grade CMC certificate of analysis can look reassuring while still leaving the buyer with unanswered questions. A line may say “viscosity: pass,” but omit the solution concentration. A purity result may be reported on a dried basis while the shipment is received as-is. A batch may meet a broad regulatory identity specification yet still fall outside the tighter range approved for an acidic beverage or a filling line.
This guide explains how distributors, food factories, procurement teams, and quality teams can read a food-grade CMC COA as a batch-release document—not as a marketing sheet. It focuses on the connection between the specification, the actual result, the test method, and the buyer’s approved application.
Quick answer: Read a food-grade CMC COA in four passes. First, confirm that the document identifies the correct manufacturer, grade, and batch. Second, compare the specification column with the specification approved in your purchase contract. Third, check the actual result, unit, basis, and test method—not only “Pass.” Fourth, review the application-critical items, especially viscosity, degree of substitution (DS), purity, loss on drying, pH, and any market- or customer-specific contaminant requirements. A COA can support batch release, but it cannot replace incoming inspection or an application trial.
What a CMC COA can—and cannot—prove
A certificate of analysis records quality results for a defined material or batch. It should help the buyer answer three practical questions:
- Is this the material and batch that we ordered?
- Was it tested against the specification we approved?
- Do the reported results support release for our intended use?
A COA is not the same as a technical data sheet (TDS), safety data sheet (SDS), food-safety certificate, allergen statement, or regulatory declaration. A TDS usually describes a product range or typical properties. A COA should report results for a specific batch. An ISO, BRCGS, Halal, or Kosher certificate addresses a management system, site, process, or religious requirement; it does not replace batch-specific analytical results.
A COA also does not prove that CMC will stabilize a particular beverage, retain water in a bakery filling, or hydrate correctly in the buyer’s process. Those decisions require a representative sample and a controlled application trial.
Start with identity and traceability
Before reviewing any number, confirm that the COA belongs to the material in front of you. At minimum, the document and package information should allow the buyer to connect the following:
- product name and commercial grade;
- manufacturer or responsible releasing entity;
- batch or lot number;
- production or release date, where applicable;
- referenced specification or document revision;
- package or shipment identification;
- authorized quality approval.
The batch number on the COA should match the lot marking on the bags and the commercial documents. If a distributor repacks the material, the original manufacturing lot and the new packing lot should remain traceable. A COA with no lot number, no grade, or only a copied “typical” table should not be treated as batch evidence.
Read all four parts of every result
Most COA rows contain four pieces of information. A result is only comparable when all four are understood.
| COA field | What it tells the buyer | Review question |
|---|---|---|
| Test item | The property being evaluated | Is this the same parameter named in our approved specification? |
| Specification | The allowed range, minimum, maximum, or acceptance statement | Does it match the contract and destination-market requirement? |
| Result | The measured value or reported outcome for this batch | Is it an actual value, “Pass,” “ND,” or a less-than result? |
| Method and unit | How the result was produced and expressed | Are the method, concentration, basis, temperature, and unit comparable? |
When the method is printed only as an internal code, ask for the method title or a cross-reference to the agreed standard. The buyer does not always need the full laboratory procedure, but the conditions that materially affect comparison must be visible or available.
How to interpret the main food-grade CMC parameters
Appearance and identification
Appearance is a first screening check, not proof of chemical identity. Food-grade sodium carboxymethyl cellulose is generally described in official specifications as a white to slightly yellowish, hygroscopic powder, granule, or fibrous material. The accepted description should match the contracted form.
If the product is ordered as fine powder, granular CMC, or an instant-dispersing form, make that distinction clear in the product grade or particle-size requirement. “Appearance: conforms” does not tell the receiving team whether the physical form is correct for the dosing system.
Viscosity
Viscosity is usually the most misunderstood line on a CMC COA. The number alone is not a grade definition. At minimum, the comparison needs:
- CMC concentration, such as 1%, 2%, 4%, or another stated level;
- whether the concentration is weight/weight or weight/volume;
- water quality and preparation sequence;
- mixing and hydration time;
- measurement temperature;
- instrument type and, where relevant, spindle and speed.
A result measured in a 1% dispersion is not directly interchangeable with a result measured at 2%. Changing concentration changes polymer interaction and apparent viscosity, often by far more than a simple two-times factor. The U.S. provision in 21 CFR 182.1745, for example, refers to a 2% by-weight aqueous preparation at 25 °C, while commercial grades may be sold using other agreed methods. This is why the purchase specification must name the exact method.
SINOCMC’s food-grade CMC viscosity testing method provides additional laboratory context. Use it as a method reference, while keeping the full buyer-approved conditions on the specification.
Degree of substitution (DS)
DS is the average number of carboxymethyl groups attached per anhydroglucose unit in the cellulose chain. It helps define chemical identity and can influence hydration, solubility, salt response, and performance in acidic systems. It is not a substitute for viscosity.
Official identity specifications do not all use the same boundary. FAO/JECFA Monograph 11 describes a DS range of 0.20 to 1.50, while the current U.S. provision includes a maximum substitution of 0.95. These are regulatory identity references, not a recommendation that every value within the broad range will perform the same way in a buyer’s formulation.
For a sensitive application, the contractual range may need to be narrower than a regulatory identity range. The agreed method also matters. See SINOCMC’s CMC degree-of-substitution determination method for the measurement principle used in an existing technical article.
Purity or assay
Purity should always be read together with its calculation basis and method. “99.5% minimum” may be reported on a dried or anhydrous basis. It is therefore not automatically the percentage of active material in an as-received bag.
In the FAO/JECFA monograph, the assay is stated on a dried basis and is calculated with reference to defined process-related components. A supplier or customer method may use different wording. Ask these questions:
- Is the result on a dried basis or as received?
- Which components are included or subtracted in the calculation?
- Does “purity” refer to assay, active polymer, or another internal definition?
- Is the specification the one required by the destination market and customer?
SINOCMC’s existing food-grade CMC purity testing article explains one laboratory approach. The COA should still state or reference the method used for the shipped batch.
Loss on drying
CMC is hygroscopic, so loss on drying is important for storage, handling, flow, and dry-basis calculations. It is a method-defined result: the sample mass, temperature, time, and endpoint affect the value. Do not compare a “moisture” value and a “loss on drying” value as if they are automatically identical.
A compliant loss-on-drying result does not rule out poor handling after release. Damaged liners, humid storage, or a wet container can change the condition of the product before arrival. Incoming inspection should therefore include package condition and, where risk justifies it, a receiving test.
pH
The pH line describes a defined CMC preparation, not the pH that the ingredient will create in the final food. Confirm the concentration and preparation specified by the method. FAO/JECFA, U.S., EU, customer, and supplier specifications may not use identical conditions or limits.
For acidic beverages, a neutral-water COA pH is not an acid-stability test. The buyer still needs an application trial under the real acid addition sequence, protein level, salts, heat treatment, and shear conditions.
These parameters can help control residual salts and process-related materials. The exact list and terminology vary by specification. “Ash,” “sulfated ash,” “sodium chloride,” “free glycolate,” and “total glycolate” are not interchangeable labels.
Review the named method and basis before comparing two suppliers. If the buyer’s specification combines components while the supplier reports them separately—or the reverse—ask for a calculation bridge rather than assuming equivalence.
Heavy metals and microbiological criteria
Required items and limits depend on the destination market, customer standard, food application, risk assessment, and current regulation. Do not copy a contaminant panel from an old COA and assume it is globally sufficient. The European Union amended specifications concerning E466 in 2025, illustrating why regulatory references should be checked for currency.
For any result reported as “less than” a value or “not detected,” confirm the unit, test method, and reporting or quantification limit. “ND” does not mean absolute zero. It means the analyte was not detected under the stated method and detection capability.
If some microbiological or contaminant tests are conducted periodically rather than on every batch, the testing frequency and release rule should be agreed in writing. A periodic third-party report should not be presented as though it were a batch-specific measurement.
Particle size and bulk density
Particle size is not always part of a regulatory identity specification, but it can be critical to dust control, feeding, wetting, dispersion speed, and lump formation. Bulk density can affect bag volume and dosing equipment. If either parameter matters to the plant, add it to the purchase specification and COA rather than relying on an informal product description.
Understand “Pass,” “ND,” “<,” and actual values
An actual value gives the buyer more information than “Pass.” It supports lot-to-lot trending and helps identify drift before a batch becomes out of specification.
- Pass or conforms: The result met the stated acceptance rule, but the measured value may be hidden.
- ND: Not detected under the stated method. Check the detection limit.
- < value: The measured amount was below a reporting or quantification threshold. Check which threshold the laboratory uses.
- Typical value: A descriptive target, not necessarily a batch result or release limit.
- Actual value: A batch-specific reported result that can be compared with the limit and historical trend.
For application-critical items such as viscosity and DS, request actual results whenever practical. A batch can be within specification but still be unusual compared with the supplier’s recent history. This is called an out-of-trend signal; it is not automatically a failure, but it may justify review before the material enters a sensitive process.
Separate three levels of specification
Many COA disputes begin because the parties are using different definitions of “specification.” Separate these three levels before approval:
- Regulatory identity and purity requirements: The applicable legal or recognized reference for the destination market.
- Commercial purchase specification: The limits, methods, units, and document revision agreed between buyer and seller.
- Application control range: The tighter internal range that the buyer may need for consistent processing or finished-product performance.
A material may satisfy a broad regulatory identity specification but miss the buyer’s contractual viscosity range. It may meet the purchase specification yet perform poorly because the approved sample was tested under different process conditions. The COA review should therefore connect regulatory compliance, the purchase contract, and the application trial without treating them as the same test.
A practical COA approval workflow
- Before sampling: Define the destination market, application, critical parameters, test methods, units, and acceptable ranges.
- At sample approval: Record the sample lot, specification revision, supplier COA, your test conditions, addition level, and application result.
- Before the purchase order: Attach or reference the approved specification and state which changes require written approval.
- Before shipment: Review the commercial-batch COA against the same specification and confirm that critical results are actual values where required.
- At receipt: Match the lot numbers, inspect packaging, follow the agreed sampling plan, and keep the material controlled until the required checks are complete.
- After release: Trend important results and record any processing or finished-product deviation by lot.
SINOCMC’s CMC acceptance and sampling article gives additional context on receiving inspection. For broader supplier qualification, use the food-grade CMC manufacturer checklist.
COA red flags for buyers
- The COA has no batch number or the batch does not match the bags.
- The document repeats a TDS table and reports no actual batch results.
- Viscosity is listed without concentration, temperature, or method reference.
- Units or dry/as-received basis change between the specification and result.
- The specification differs from the approved quotation or purchase order.
- Every lot shows exactly the same actual values without an explanation of reporting precision.
- “ND” is reported without a method or detection capability for a critical contaminant.
- A third-party or periodic report is presented as a test of the shipped lot.
- The manufacturing source or test method changes without buyer approval.
- The COA is used to dismiss a failed application trial without investigating process and sample differences.
What to send with a food-grade CMC COA request
A clear request reduces email exchanges and prevents a quotation based on the wrong grade. Send the supplier:
- destination country and customer specification;
- food application and required function;
- target viscosity range with full measurement method;
- required DS, purity, loss on drying, pH, and other critical parameters;
- required contaminant and microbiological panel, methods, and testing frequency;
- particle form or size requirement;
- sample quantity, trial conditions, annual demand, and packing format;
- the document revision that should appear on the quotation, sample COA, and commercial-batch COA.
For food-grade product and trial information, visit the SINOCMC food-grade CMC page.
Frequently asked questions
What is the difference between a CMC COA and a specification sheet?
A specification sheet defines the allowed limits and methods for a grade. A COA should identify a specific batch and report its results against that specification. A technical data sheet may show typical properties but is not necessarily a batch-release record.
Does a food-grade CMC COA prove the product is suitable for my formulation?
No. It supports identity and batch-quality review under the stated tests. Suitability still depends on your formula, pH, salts, proteins, process order, shear, temperature, dosage, and finished-product targets. Use a representative sample and application trial.
Why can two CMC COAs show the same viscosity but represent different grades?
The concentration, preparation, hydration time, temperature, instrument, spindle, and speed may differ. A 1% and a 2% viscosity result are not directly comparable. DS, particle form, and substitution uniformity can also affect application behavior.
Should every COA show actual values?
Actual values are most useful for critical parameters and trend review. Some identity tests may reasonably be reported as “Pass,” and results below a method threshold may be reported as “<” or “ND.” The reporting format should be agreed according to risk and customer requirements.
Can I use one global COA specification for every market?
Not safely. Official specifications and permitted uses can differ and can be amended. Confirm the current destination-market requirement, then add the commercial and application limits needed by your customer and process.
What should I do if the COA meets specification but the batch fails in production?
Control the material and compare the shipped lot, retained sample, incoming sample, process conditions, water quality, addition order, and test methods. A passing COA does not eliminate the possibility of a handling, sampling, testing, or application mismatch. Follow the agreed complaint and investigation process before assigning cause.
Final recommendation
Treat the food-grade CMC COA as one link in a controlled approval chain: target market → agreed specification → representative sample → application trial → commercial-batch result → incoming release. The strongest COA is not the one with the most rows. It is the one that clearly connects the correct batch to the correct specification, method, unit, and buyer decision.
If you are preparing a food-grade CMC inquiry, send SINOCMC your application, destination market, required COA items, viscosity method, target ranges, trial quantity, and packing requirement. The team can discuss a suitable grade and documentation plan before quotation. Contact SINOCMC.
Regulatory note: This article is a procurement and quality-review guide, not legal advice. Buyers are responsible for confirming current rules and customer requirements in the destination market.