How to Read a Certificate of Analysis: A Researcher's Guide

A Certificate of Analysis is either the single most honest document in the research peptide industry, or a piece of marketing collateral with a laboratory aesthetic. Learning to tell the difference is one of the most practical skills a researcher can develop.

Why COAs matter for research-grade compounds

A Certificate of Analysis (COA) is the document that answers a deceptively simple question: what is actually in the vial? For a researcher working with reference compounds, the identity, purity, and sterility of the material are not optional context — they are the preconditions for any observation drawn from the experiment being meaningful at all. A noisy result from a pure compound is a data point. A noisy result from an unknown mixture is nothing.

In a regulated pharmaceutical environment, COAs are governed by pharmacopoeias (official reference books that define how drug tests must be performed) — the European Pharmacopoeia (Ph. Eur.) in Europe, the United States Pharmacopeia (USP), and the Japanese Pharmacopoeia (JP), among others — which specify how tests are to be performed, reported, and signed off. The research peptide market does not fall under these frameworks, which is precisely why understanding what a legitimate COA looks like matters so much. There is nobody checking for you.

Who issues a COA, and why independence matters

There are two distinct species of Certificate of Analysis. The first is issued by the manufacturer of the compound — an internal quality control document produced by the same laboratory that synthesized the material. The second is issued by an independent third-party laboratory that received a sample of the batch from the manufacturer or vendor and ran its own tests on equipment the manufacturer does not own or control.

The distinction is not academic. A manufacturer's internal COA describes what the manufacturer says it found. An independent COA describes what someone else, with no commercial interest in the result, found when they looked. In every other corner of analytical chemistry — food testing, environmental monitoring, pharmaceutical release — the independent result is considered the more credible of the two, for obvious reasons.

The serious research peptide vendors — the ones who behave as if their customers are people who actually know how to read the documents — have largely migrated to independent third-party testing as the default. The rest have not.

Janoshik Analytical and the peptide research community

In practice, one name has become something close to a gold standard in the English-speaking peptide research community: Janoshik Analytical, an independent laboratory based in the Czech Republic. Janoshik reports are widely circulated, compared, and debated on research forums, and their format — clean HPLC chromatograms, mass spectrometry identity confirmation, signed and dated documents — has become the template that other independent labs are measured against.

There is nothing magical about Janoshik specifically; there are other competent independent labs. What matters is the principle: a COA from an independent laboratory with a public reputation is structurally more trustworthy than a COA produced by the same operation that is selling you the vial.

Reading an HPLC result

What High-Performance Liquid Chromatography is

HPLC is a separation technique. A small amount of the sample is dissolved in a solvent and pushed under high pressure through a column (a narrow tube) packed with a stationary phase — typically, for peptide work, a reverse-phase C18 material (a sorbent coated with 18-carbon chains that grip molecules by their hydrophobic parts). Different molecules in the sample interact with the stationary phase to different degrees, which causes them to emerge from the far end of the column at different times. A detector — usually a UV absorbance detector for peptides — records what comes off the column, when, and in what quantity. The result is a chromatogram: a trace showing peaks at various retention times.

How purity percentages are calculated

The number that appears on a COA as "purity" is almost always an area-percent calculation. The detector records a signal intensity over time; each peak encloses an area; the largest peak (assumed to be the main compound) is divided by the sum of all peaks to produce the purity figure. A result of "99.1% purity" means the main peak represents 99.1% of the total integrated UV-absorbing material on the chromatogram at the wavelength measured.

This is not the same as saying that 99.1% of the mass in the vial is the target compound. It is saying that 99.1% of the material detected by this specific technique, at this specific wavelength, is associated with the main peak. Salt, water, and any compound that does not absorb at the measured wavelength will not appear in the calculation at all.

What "99% purity" actually means

With that caveat in mind, 98–99% area-percent purity by reverse-phase HPLC with UV detection is a reasonable shorthand for a clean synthesis — meaning the peptide was assembled, cleaved, and purified without accumulating a large population of truncated or modified side-products. Numbers substantially below 95% suggest either a difficult sequence or sloppy purification. Numbers above 99% are common for short peptides and for sequences that purify easily.

Chromatogram interpretation

A useful COA reproduces the actual chromatogram, not just the summary number. When researchers read the chromatogram, they look for: a single dominant peak at the expected retention time; a clean, symmetrical peak shape; a flat baseline before and after; and small, discrete impurity peaks rather than a broad hump of unresolved material. A COA that reports a headline purity number but hides the underlying trace is conspicuously less useful than one that shows its work.

Mass spectrometry identity confirmation

Why identity matters

HPLC tells a researcher how pure the sample is. It does not tell them what the sample is. In principle, a vial could be 99% pure of the entirely wrong compound — a synthesis that produced a single clean molecule that happens not to be the one on the label. HPLC alone cannot detect that. Mass spectrometry is the technique that closes the gap.

Expected mass vs. observed mass

A peptide's expected mass is calculated by summing the monoisotopic masses (the exact weights of the most common isotope of each atom) of its constituent amino acid residues and adjusting for the water lost at each peptide bond and any terminal modifications. This gives a theoretical molecular mass — a specific number, usually reported in daltons (the standard unit for molecular weight). A mass spectrometer ionizes the sample (gives molecules an electrical charge so they can be manipulated) and measures mass-to-charge ratio for the resulting ions; a match between the observed m/z (accounting for charge state) and the calculated expected mass is the confirmation that the compound in the vial is the compound on the label.

Common identity confirmation formats

A standard independent COA will quote both the calculated and observed masses, often with a spectrum attached showing the relevant m/z peak. Matches to within a fraction of a dalton are expected for modern electrospray-ionization (ESI) instruments. A COA that reports purity but declines to report identity is, for a researcher, missing its single most important piece of information.

Endotoxin testing

Endotoxins are lipopolysaccharide fragments (pieces of the outer "skin") from the outer membrane of Gram-negative bacteria. They are not the bacteria themselves — sterilizing a sample by filtration or heat does not remove them — and they are biologically active at extremely low concentrations. In injectable research contexts, endotoxin contamination is a serious confound: it can itself produce inflammatory, febrile (fever-inducing), and behavioral responses in animal models, which means an endotoxin-contaminated sample can look biologically active without containing anything of interest.

The standard test is the Limulus Amebocyte Lysate (LAL) assay, which exploits a clotting reaction in horseshoe crab blood cells that is exquisitely sensitive to endotoxin. Results are reported in endotoxin units per milligram (EU/mg) or per milliliter. The European Pharmacopoeia and USP both set threshold values for injectable pharmaceuticals; research-grade material from reputable sources is generally tested against comparable thresholds.

Heavy metals testing

Synthetic peptide production introduces the theoretical possibility of heavy metal contamination from reagents, catalysts, and equipment. The four metals routinely tested for are lead, cadmium, arsenic, and mercury. Detection limits are typically reported in parts per million (ppm) or parts per billion (ppb), with acceptable thresholds defined by the relevant pharmacopoeia for the closest regulatory analog. A reputable COA will either report each metal with a numeric value or state that the compound was tested against a specific standard and passed.

Bioburden testing

Bioburden refers to the viable microbial load in a sample — bacteria, yeasts, and moulds that can grow if given the opportunity. Bioburden is distinct from endotoxin (which is a chemical entity) and from sterility (which is an absolute claim rather than a measurement). For research-grade peptides, the expectation is a low bioburden count under defined conditions, reported as colony-forming units (CFU, the count of microbes able to grow into visible colonies) per gram or per milliliter. The relevant Ph. Eur. chapters on microbiological examination of non-sterile products provide the standard framework.

How to spot a FAKE or manipulated COA

The research peptide community has, over the years, developed a practical catalog of the ways COAs are faked, reused, or manipulated. Researchers learning to evaluate vendors should be alert to all of them.

• Mismatched dates. The test date predates the batch manufacturing date, or postdates it by a period that makes no logistical sense.
• Inconsistent formatting. Fonts change between sections, alignment drifts, or the same report format is used for both "manufacturer" and "independent lab" results.
• No independent lab signature. The document carries only the vendor's branding. No signature, no analyst name, no laboratory contact details.
• Photoshopped documents. Visible compression artefacts around numbers, faint traces of previous values beneath new ones, or watermarks that do not match the claimed issuing lab.
• Missing batch numbers. A COA with no batch identifier cannot be tied back to a specific vial; this makes it structurally impossible to verify.
• "Generic" COAs reused for multiple products. The same document, with the same dates and the same chromatogram, circulates under different product names. This is a common fraud and is easier to catch than one might expect.
• No raw traces. The report presents a headline purity number but omits the chromatogram and the mass spectrum. A legitimate independent COA shows the underlying data.
• Suspiciously identical numbers across batches. Genuine analytical measurements have variance; five consecutive batches reporting 99.00% purity to two decimal places is not plausible.

Questions to ask a vendor about their COAs

A researcher evaluating a new supplier can cover most of the important ground with a short list of questions. The answers — or the reluctance to provide them — reveal a great deal.

• Which laboratory issued this COA? An independent third-party name, publicly identifiable, is the only acceptable answer.
• Is every batch tested, or is testing spot-checked? The serious answer is every batch.
• Can the batch number on the COA be matched to the batch number on the physical vial? If not, the COA is decorative.
• Are the raw HPLC chromatogram and mass spectrum included? Summary numbers without underlying traces are substantially less informative.
• What is tested beyond purity? Identity, endotoxins, heavy metals, and bioburden should all be covered for serious research use.
• How long has the vendor been using this lab, and can they produce historical COAs for older batches? A vendor who can show continuity is harder to fake than one who cannot.

The Lipa standard

Lipa's position on this is straightforward and, to be honest, unremarkable by the standards of how analytical work should be done. Every batch of every compound is tested by an independent third-party laboratory. Every COA is tied to a specific batch number that appears on the physical vial. Every COA includes the underlying HPLC chromatogram and mass spectrometry identity confirmation. Purity, identity, endotoxins, heavy metals, and bioburden are all covered. The documents are published, not hidden.

This is not a marketing point. It is what the minimum reasonable answer to "what is in the vial?" looks like in the research peptide space, and the only thing that makes it worth writing down at all is the number of competitors who do not meet it.