What Purity on a Certificate of Analysis Really Means

A certificate that reads 99.8 percent looks like a statement about the bottle. It is closer to a statement about one chromatogram, recorded with one detector, on one day, against one set of acceptance criteria. What purity on a certificate of analysis means is inseparable from the method that produced it. Strip the method away and the number stops being a measurement and becomes a slogan.

This page is the map for the purity and quality control category. It defines the working vocabulary — impurity, threshold, specification, degradation product, certified reference material — and shows where each term comes from and what it can support. Individual techniques and worked cases are handled in separate articles.

What purity on a certificate of analysis actually means

Purity is almost never measured directly. It is inferred from the absence of something else. A chromatographic purity figure is the fraction of the total detector response attributable to the main peak, which is a different quantity from the fraction of sample mass that is the intended compound.

The gap matters because a detector responds to a property, not to a substance. An ultraviolet detector responds to light absorption, so a co-eluting impurity without a chromophore contributes nothing to the total response and is silently excluded from the denominator. Its absence from the chromatogram is a property of the detector, not of the sample.

ICH Q2(R2), the harmonised guideline on validation of analytical procedures, addresses this directly. It states that specificity or selectivity can be demonstrated through absence of interference or by comparison of results to an orthogonal procedure, and that where one analytical procedure does not provide sufficient discrimination, a combination of procedures is appropriate. A single number from a single method is, in that framing, a partial answer.

Three questions therefore sit behind any purity claim. Which method produced it. What that method is blind to. And what else was run alongside it.

Why area percent and mass percent are not the same number

Area normalisation treats every peak as if it responded like the main compound. That assumption is convenient and usually wrong. Two substances at identical mass concentration can differ severalfold in detector response, so the relative response factor — the ratio between an impurity’s response and the main compound’s response at equal mass — decides how much the impurity appears to contribute.

ICH Q3A(R2) is explicit that organic impurity levels can be measured by techniques that compare an analytical response for an impurity to that of an appropriate reference standard or to the response of the drug substance itself, and it refers to impurity levels calculated using the response factor of the drug substance. The guideline treats the response factor as a variable to be handled, not a constant that can be ignored.

The practical consequence is that an area percent figure and a mass percent figure answer different questions.

Expression of purity	What it counts	Main blind spot
Area percent by area normalisation	Share of total detector response in the main peak	Anything the detector cannot see, plus unequal response factors
Assay against a reference standard	Mass of the compound relative to a characterised standard	Inherits the quality and the assigned value of that standard
Mass balance	Assay plus all quantified impurities, water, solvents and residue	Needs several orthogonal determinations, not one chromatogram

Mass balance is the honest version of the question, because it forces every component to be either quantified or declared missing. It is also the expensive version, since it needs water content, residual solvent content and inorganic residue alongside the chromatogram. Certificates differ enormously in how much of that work they show.

The three thresholds that decide how far an impurity is chased

Pharmaceutical impurity control is built on a threshold ladder rather than a single limit. ICH Q3A(R2) defines each rung in one line. A reporting threshold is a limit above which an impurity should be reported. An identification threshold is a limit above which an impurity should be identified. A qualification threshold is a limit above which an impurity should be qualified, meaning its biological safety at that level has to be supported by data.

The rungs are keyed to the maximum daily dose of the drug substance, because the quantity of impurity that reaches a patient depends on how much material is involved. Attachment 1 of the guideline sets them out.

Maximum daily dose	Reporting threshold	Identification threshold	Qualification threshold
Up to 2 g per day	0.05%	0.10%, or 1.0 mg per day intake, whichever is lower	0.15%, or 1.0 mg per day intake, whichever is lower
More than 2 g per day	0.03%	0.05%	0.05%

A worked example in the same attachment makes the ladder concrete. For a maximum daily dose of 0.5 g, the reporting threshold is 0.05 percent, the identification threshold is 0.10 percent and the qualification threshold is 0.15 percent. An impurity at 0.04 percent is not reported at all. The same impurity at 0.12 percent has to be identified. Higher still, it also has to be qualified.

Two details in Q3A(R2) are easy to miss and they change how a certificate reads. The quantitation limit of the analytical procedure should be not more than the reporting threshold, so a method that cannot see down to the reporting level cannot support the claim built on it. And quantitative results should be presented numerically, not in general terms such as complies or meets limit. A certificate that says only conforms has declined to answer the question.

One caution about borrowing these numbers. The ICH thresholds are written for new drug substances and drug products, where a maximum daily dose exists and patient exposure is the governing concern. An analytical reference standard sitting in a laboratory has no daily dose, so the threshold ladder transfers as a way of thinking about proportionality, not as a set of limits that automatically apply.

Where impurities come from and why a compound degrades on the shelf

Impurities are not contamination in the everyday sense. Most of them are the predictable residue of the route that made the compound, and each family has its own guideline.

Process-related impurities arrive from starting materials, intermediates, reagents and by-products, and Q3A(R2) separately flags carry-over of catalysts into the drug substance as an inorganic concern. Residual solvents have a guideline of their own. ICH Q3C(R8) sorts solvents into three classes by possible risk to human health, and the direction of that scale is worth stating plainly, because it is frequently reported backwards.

Q3C class	Definition in the guideline	How it is controlled
Class 1	Solvents to be avoided: known human carcinogens, strongly suspected human carcinogens, and environmental hazards	Individual concentration limits in parts per million
Class 2	Solvents to be limited: non-genotoxic animal carcinogens, or possible causative agents of other irreversible toxicity such as neurotoxicity or teratogenicity	Permitted daily exposure values per solvent
Class 3	Solvents with low toxic potential to man, for which no health-based exposure limit is needed	A general limit rather than substance-specific values

Class 1 is the dangerous end of that scale, not the safe one. Q3C(R8) states that Class 1 solvents should not be employed in the manufacture of drug substances, excipients and drug products because of their unacceptable toxicity or their deleterious environmental effect. Where their presence is unavoidable, the guideline fixes limits in parts per million, with benzene at 2 ppm and carbon tetrachloride at 4 ppm. 1,1,1-Trichloroethane sits in the same table at 1500 ppm, and the guideline is explicit that it is listed there as an environmental hazard rather than for the same toxicological reason.

Elemental impurities have their own document again. ICH Q3D(R2) places arsenic, cadmium, mercury and lead in Class 1, as human toxicants that require evaluation across all potential sources and routes of administration. Cobalt, nickel and vanadium fall into Class 2A, elements with a relatively high probability of occurrence in the product. Class 2B collects silver, gold, iridium, osmium, palladium, platinum, rhodium, ruthenium, selenium and thallium, which may be excluded from the risk assessment unless they are intentionally added.

Impurities carrying a structural alert for DNA reactivity are handled separately. ICH M7(R1) sorts them into five classes and controls the unstudied ones through the threshold of toxicological concern, a concept developed to define an acceptable intake for an unstudied chemical that poses a negligible risk of carcinogenicity. An impurity with an alerting structure unrelated to the drug substance and no mutagenicity data can be controlled at that level, or put through a bacterial mutagenicity assay and reclassified on the result.

Degradation is the category that changes after the certificate is printed. ICH Q3B(R2) defines a degradation product as an impurity resulting from a chemical change in the drug substance brought about during manufacture or storage by the effect of, for example, light, temperature, pH, water, or by reaction with an excipient or the immediate container closure system. That list doubles as a list of what a storage decision controls. A purity figure is a time stamp, not a permanent property.

Why a specification and a certificate answer different questions

A specification is a set of tests with acceptance criteria that material has to meet to be acceptable for its intended use. A certificate of analysis is the record of what one batch returned when those tests were run. The first is a contract, the second an observation, and conflating them is the most common misreading in the whole area.

Q3A(R2) draws a second distinction that certificates often blur. A specified impurity is individually listed and limited with its own acceptance criterion. An unspecified impurity is limited only by a general criterion. On a separate axis, an identified impurity is one for which a structural characterisation has been achieved, while an unidentified impurity is defined solely by a qualitative analytical property such as chromatographic retention time. The two axes are independent, so a specified impurity can still be unidentified: listed, limited, and structurally unknown.

This is why two batches quoted at the same purity are not necessarily equivalent. If one certificate lists three named related substances with individual criteria and the other reports a single largest unspecified peak, the two documents carry very different amounts of information even when the headline figure matches. The number is identical; the knowledge behind it is not.

What a certified reference material adds that a purity figure does not

Metrology has a formal definition for the object that a purity claim ultimately leans on. The International Vocabulary of Metrology, published as JCGM 200:2012, defines a certified reference material as a reference material accompanied by documentation issued by an authoritative body and providing one or more specified property values with associated uncertainties and traceabilities, using valid procedures.

Two words in that definition do most of the work. Uncertainty means the document states how well the value is known, not only what the value is. Traceability means the value is linked through a chain of comparisons to a stated reference. A bare purity percentage carries neither, which is why it cannot on its own support a quantitative result somewhere else. The vocabulary points to ISO Guide 35 for how such materials are produced and characterised.

The practical reading is that a number is only as transferable as the documentation around it. That is also why supply terms and intended-use statements are part of the technical record rather than legal decoration; ours are set out in the terms of supply.

Frequently asked questions

What does 99.5 percent purity by HPLC actually mean?

It normally means that 99.5 percent of the total detector response in the chromatogram came from the main peak. That is a share of response, not a share of mass. Components that do not absorb at the monitored wavelength, or that never elute from the column, are missing from both the numerator and the denominator of that calculation.

Is area percent the same as mass percent?

No. Area normalisation assumes every component responds like the main compound, which is rarely true. Relative response factors can differ severalfold between an impurity and the parent structure, so an impurity may be over- or under-represented. Converting between the two requires either measured response factors or an orthogonal method that determines mass directly.

What is the difference between a reporting threshold and an identification threshold?

ICH Q3A(R2) defines a reporting threshold as the limit above which an impurity should be reported, and an identification threshold as the limit above which its structure should be established. Reporting is disclosure; identification is structural work. Both are keyed to the maximum daily dose of the drug substance rather than being fixed percentages that apply everywhere.

Why do two batches at the same purity behave differently?

Because an identical headline figure can conceal different impurity profiles. One batch may carry a single well-characterised related substance, another a spread of unidentified peaks plus residual solvent. Purity states how much of the material is not the main compound. It does not state what that remainder is, and the remainder drives reactivity and interference.

Are ICH impurity thresholds binding on analytical reference standards?

No. The ICH thresholds are written for new drug substances and drug products, where a maximum daily dose defines patient exposure. A reference standard used in a laboratory has no such dose. The threshold logic remains useful as a proportionality argument, but quoting the percentages as if they were laboratory limits misstates their scope.

Which residual solvents are the ones to avoid entirely?

ICH Q3C(R8) collects them in Class 1, described in the guideline as solvents to be avoided: known human carcinogens, strongly suspected human carcinogens and environmental hazards. Benzene and carbon tetrachloride sit in that class. Class 3, at the opposite end of the scale, covers solvents with low toxic potential for which no health-based exposure limit is needed.

Does a certificate of analysis expire?

The measurement does not expire, but its relevance does. ICH Q3B(R2) defines degradation products as impurities formed during manufacture or storage through light, temperature, pH, water or reaction with the container closure system. A certificate describes the material on the test date, and re-test intervals exist precisely because that description drifts.

References

Research use only. Nonsensia Lab supplies analytical reference standards for laboratory and research applications. This article is published for scientific and educational purposes. It is not medical advice, it does not describe any use in humans, and nothing in it should be read as a recommendation to administer any substance to a person or animal.

Filed under: Purity & Quality Control

Nonsensia Lab supplies the compounds discussed in this guide as analytical reference standards for laboratory and research use.

What Purity Actually Means: Impurities, Certificates and Specifications