HPLC Purity & Peptide Certificates of Analysis

The number that ships the lot

A receiving lab is looking at two candidate lots of the same sequence. One certificate reports 98.7% purity, the other 99.4%. The purchasing decision often stops there, as though the higher figure settles the matter. It rarely does. The two results may have come from different gradients, different detection wavelengths, or different integration choices, which means the percentages are not strictly comparable and neither one describes how much of the vial is actually the intended molecule.

A certificate of analysis is a summary of decisions made during release testing. Reading it well means understanding what each line measures, what method produced it, and where the silent gaps are. The purity figure is the headline, but identity, content, appearance, and the method context around them are what let you decide whether the material fits your experiment.

It helps to treat the document as an argument rather than a verdict. Each result is a claim, and each claim is only as strong as the method and acceptance criteria behind it. A confident reviewer does not ask whether the numbers look good; they ask what question each number was designed to answer, and whether that question is the one their own work actually depends on.

Area purity is a ratio, not a mass balance

In reverse-phase HPLC, a dissolved sample travels through a column while the mobile-phase composition changes. Components partition between the stationary and mobile phases at different rates and elute at different times, producing peaks on a chromatogram. Software integrates the area under each peak, and area purity is the principal peak expressed as a percentage of the total integrated area under a defined method.

That definition carries an important limitation. Area purity is a relative measure among species the detector can see. A UV detector at, say, 214 nm responds to peptide bonds, so it reports peptide-related impurities well but is largely blind to counter-ions, residual water, and non-absorbing salts. A vial can be 99% pure by peak area and still contain a meaningful mass fraction of trifluoroacetate or bound water. Purity answers "how clean is the chromatographic profile," not "what fraction of this powder is active peptide."

Integration parameters also shape the result. Baseline placement, peak-width settings, and whether small shoulders are resolved or merged all move the number. This is why an area purity figure is only interpretable alongside the method that generated it, and why two honest laboratories can report slightly different values for the same material.

The detection wavelength deserves particular attention. Many peptide methods monitor at low ultraviolet wavelengths where the peptide bond absorbs strongly, but an impurity that lacks a strong chromophore at that wavelength will under-report or vanish entirely. Switching to a different wavelength, or to a detector with broader response, can change which impurities appear and therefore change the purity figure without any change to the material itself. A percentage tied to a single wavelength is a view of the sample, not a complete inventory of it.

Identity answers a different question

Purity tells you the profile is clean; it does not tell you the dominant peak is the molecule you ordered. A deletion variant missing one residue, or a sequence with an unintended modification, can elute close to the target and inflate the apparent purity of the wrong compound. Retention time is a weak identity argument on its own because it is method-dependent and shared by many species.

Mass spectrometry supplies the identity evidence. By measuring the molecular mass of the principal component and comparing observed to expected values, LC-MS confirms that the clean peak corresponds to the intended sequence and salt-adjusted mass. A credible certificate reports both the expected and observed mass so a reviewer can check the match rather than trust a pass/fail stamp.

Purity and identity are complementary, not interchangeable. A lot can pass one and fail the other. Requiring both closes the most common gap in casual documentation, where a high purity figure implies a confidence about identity that the purity method never actually tested.

There is a further subtlety worth naming. Mass confirmation establishes that the observed mass matches the expected mass, but two different sequences can share the same nominal mass, and some modifications shift mass by only a small amount that a low-resolution instrument may not resolve. For most research purposes an accurate mass match plus a clean purity profile is a strong combination, but for sequences with known isobaric risks, higher-resolution mass measurement or orthogonal sequence confirmation is what turns a plausible match into a confident one.

Content is what the assay column often omits

Peptide content, sometimes called net peptide content or assay, is the fraction of the powder that is peptide backbone rather than water, counter-ion, or salt. It is measured by orthogonal methods such as quantitative amino acid analysis, nitrogen determination, or a calibrated quantitative HPLC assay against a reference standard. Content and purity are frequently confused, but they measure different axes: purity is a ratio among peptide-related species, content is an absolute mass fraction of the whole.

The practical consequence appears when you weigh material for a stock solution. If a vial is 99% pure but only 80% peptide content, gravimetric dosing based on gross mass overstates the amount of active molecule by roughly a fifth. For qualitative screening this may not matter. For quantitative pharmacology, structure-activity work, or anything downstream that depends on concentration, content is the number that actually governs your calculation.

Content also interacts with the salt form and with residual water. A hygroscopic salt form can pick up moisture between manufacture and use, quietly lowering the effective peptide fraction over time, and a counter-ion that contributes significant mass will pull content down even when the peptide itself is pristine. This is why a certificate that reports salt form, water content, and counter-ion alongside content is far more useful than one that reports content as a single isolated figure. Each of those adjacent measurements explains where the non-peptide mass lives.

Method context is part of the result

A result without its method is difficult to trust and impossible to reproduce. A useful certificate names the column chemistry, the mobile-phase system, the gradient, the detection wavelength, and the acceptance criteria the lot was released against. That context lets an independent reviewer judge whether the method could resolve the impurities that matter for a given sequence.

Method validation is the discipline behind that trust. Regulatory guidance describes the characteristics an analytical procedure should demonstrate, such as specificity, linearity, accuracy, precision, and appropriate detection and quantitation limits, so that a reported value means the same thing across analysts and days. When a supplier can point to a validated or at least well-characterized method, the purity figure stops being an isolated claim and becomes a repeatable measurement.

• Column and mobile-phase system used for the release assay
• Gradient profile and detection wavelength
• Acceptance criteria the lot was tested against
• Whether the method is validated or qualified, and to what standard

Raw data versus the summary line

A certificate is a summary; the chromatogram and mass spectrum are the evidence. For routine catalog quantities, the summary may be sufficient. For high-value orders, API-scale purchases, or any material entering a regulated workflow, request the underlying raw data so your own analysts can review baseline quality, peak shape, and how impurities were integrated.

Raw data also lets you catch internal contradictions. Batch numbers and test dates should agree across the purity report, the mass spectrum, and the certificate itself. A mismatch does not always signal a problem, but it warrants a question before the material is committed to work you cannot easily repeat.

Reviewing raw traces is a skill worth cultivating even for those who do not run the instruments. A flat, stable baseline suggests a well-behaved separation; a drifting or noisy baseline can hide small impurities or inflate the apparent principal peak. Peaks that are sharp and symmetric integrate more reliably than broad or tailing peaks, where the boundary between the main component and a co-eluting impurity becomes a matter of judgment. None of this requires reprocessing the data yourself; it simply means the trace, not the summary line, is where the confidence actually lives.

When the same number means different things

Consider three vials that all read 98% pure. In the first, the remaining 2% is a single, well-characterized, closely related impurity that is irrelevant to the intended assay. In the second, that 2% is spread across a dozen small unidentified peaks. In the third, the 2% includes a known oxidation product of a residue that happens to sit in the functionally important region of the sequence. The headline number is identical, but the three lots carry very different risk for a given experiment.

This is why an impurity profile is often more informative than a purity total. A specification that names the impurities of concern, sets individual and total limits, and requires that any impurity above a threshold be identified tells you far more than a lone percentage. It shifts the question from how clean the material is in the aggregate to whether the specific things that would compromise your work are controlled.

Writing a fit-for-purpose specification

The right analytical panel depends on the intended use, not on maximizing the number of tests. Exploratory in-vitro screening may be well served by purity, identity, and net content. Formulation development or regulated substance work can additionally require water content, residual solvents, counter-ion determination, bioburden, and endotoxin testing, each governed by recognized standards for how the test is run and interpreted.

Every added test carries cost and lead time, so specification writing is an exercise in matching evidence to risk. The expensive mistake is not paying for one test too many; it is omitting a method that materially affects the experiment and discovering the gap after the material is consumed. Define the application first, then decide which lines on the certificate you genuinely need to see, and set acceptance criteria in writing before you request a quote.

A useful discipline is to write the specification as a set of decisions rather than a wish list. For each proposed test, ask what result would cause you to reject the lot, and what you would do with the number if it passed. A test whose outcome would not change any decision is a test you probably do not need; a decision you cannot make without a given number is a test you cannot afford to skip. Held to that standard, a specification tends to become both shorter and sharper, because it reflects the experiment rather than an abstract idea of quality.