Reference library

How to Read a Peptide COA: A 4-Pass Guide

Check identity, purity, quantity, and provenance on a peptide certificate of analysis — and why the 98% HPLC purity number tells you the least.

A peptide certificate of analysis is read in four passes, in this order: identity (does the mass spectrum match the theoretical mass?), purity (what percentage of the chromatogram area is the target peak, and at what wavelength?), quantity (how much of the vial's labeled mass is actually peptide, after counterion and water?), and provenance (does the lot number on the document match the vial, and who ran the test?). Most people stop at the purity number on the summary page. That number — usually 98.0% or 99.0% — is the least informative figure on the document, because it is an area-normalized UV measurement that describes the ratio of things that absorb light at 214 nm, not the composition of the powder in the vial.

Pass One: Identity Before Purity

Purity is meaningless if the compound isn't what the label says. Identity comes from mass spectrometry, almost always ESI-MS, and the COA should show three things: the theoretical mass, the observed mass, and the spectrum itself.

Check which mass is quoted. Monoisotopic and average mass diverge as chains get longer — for a 40-residue peptide around 4,100 Da, the gap between monoisotopic and average mass runs roughly 2.5–3 Da, which is larger than the ±0.5 Da tolerance many COAs claim. A document quoting a match "within 0.3 Da" while comparing a monoisotopic theoretical to an average observed value is arithmetically confused.

Look at the charge states. Peptides above ~1,500 Da typically ionize as an envelope of multiply charged species: [M+2H]²⁺, [M+3H]³⁺, [M+4H]⁴⁺. A COA showing a single clean peak at exactly the neutral mass, with no charge envelope, is showing a deconvoluted spectrum at best and a drawn figure at worst. Ask for the raw envelope.

Two mass shifts matter more than any others and are visible if the spectrum is included at reasonable resolution:

  • +16 Da — methionine or tryptophan oxidation. Common in material stored warm or handled in air.
  • +0.98 Da — asparagine or glutamine deamidation. On a unit-resolution quadrupole analyzing a 4,000 Da species, this is effectively invisible; it requires a high-resolution instrument (Orbitrap or TOF) or an orthogonal method to detect at all.
  • −18 Da — dehydration, or a cyclization artifact.
  • +22 / +38 Da — sodium and potassium adducts, which indicate desalting was incomplete rather than a synthesis problem.

Pass Two: Read the Chromatogram, Not the Percentage

The HPLC trace is where a real COA distinguishes itself from a graphic. A usable trace names the column (commonly a C18, 4.6 × 250 mm, 5 µm), the mobile phases (typically 0.1% TFA in water and 0.1% TFA in acetonitrile), the gradient, the flow rate (1.0 mL/min is standard for that column geometry), the detection wavelength, and the injection volume.

Specific things to check:

  • The wavelength. 214 nm detects the peptide bond and sees nearly everything peptidic. 280 nm sees only Trp, Tyr, and Cys. A purity number generated at 280 nm on a peptide with one tyrosine is a nearly worthless figure and will read artificially high.
  • The run time versus the last peak. If the gradient runs 30 minutes and the target elutes at 12 minutes, fine. If the target elutes at 27 minutes on a 30-minute run, hydrophobic late-eluting impurities may simply never have come off the column before the run stopped. Truncated runs inflate purity.
  • The baseline. Look at where the y-axis starts. A trace cropped so the axis begins at 40 mAU hides small impurity peaks entirely. Integration parameters and the noise floor should be stated.
  • The integration table. A genuine table lists every integrated peak with retention time and area percent, including the ones at 0.3% and 0.4%. A COA reporting only "main peak: 99.2%" has discarded the information you'd use to compare lots.

The practical difference between purity grades is worth stating in mass: at 98.0% purity, one gram of material contains 20 mg of related substances. At 99.5%, it contains 5 mg. Whether that 15 mg delta matters depends entirely on what the material is for — a point the section below returns to.

Pass Three: Net Peptide Content — The Number Most COAs Omit

This is the single largest gap between what a COA appears to say and what a vial contains.

HPLC area-normalized purity says nothing about mass. Lyophilized peptides purified by RP-HPLC in TFA-containing buffers carry TFA counterions, typically 10–20% of the total mass depending on the number of basic residues. Lyophilized powders also retain water, commonly 4–8% by mass measured by Karl Fischer titration.

Stack those: a vial labeled 10 mg gross, at 99% HPLC purity, may contain roughly 7.5–8.5 mg of actual peptide. The COA is not lying. It is answering a different question than the one most buyers think they asked.

The tests that close this gap are amino acid analysis (AAA) or nitrogen determination for net peptide content, and Karl Fischer for water. Most research-use-only COAs include neither. If lot-to-lot mass consistency matters to your work, request them explicitly or accept a ±15% mass uncertainty between lots that both read "99% pure."

A purity percentage tells you the ratio of UV-absorbing species. Net peptide content tells you how much peptide is in the vial. They are unrelated numbers, and only one of them is on most certificates.

Pass Four: Provenance, Dates, and Document Forensics

  • Lot match. The lot on the COA must match the vial label character for character. A COA for lot BPC-2411A supplied against a vial marked BPC-2503B is a different batch of material.
  • Date of analysis versus date of manufacture. A gap of 18 months means the chromatogram describes the material as it was, not as it is.
  • Who ran it. In-house QC is legitimate but is a self-report. A third-party COA names the contract lab, its address, the instrument, and an analyst or QA signature. Search the lab name; contract analytical labs have websites and accreditations.
  • PDF metadata. Open the file properties. Instrument-exported reports carry the software in the producer field (Chromeleon, Empower, Analyst). A COA whose producer field reads as a design tool was assembled by hand, which does not prove fabrication but does mean no chain of custody exists between the instrument and the document.
  • Storage and stability. Lyophilized peptide is typically specified at −20 °C for long-term storage (24–36 months), with 2–8 °C acceptable for shorter holding. A shipping excursion of 3–5 days at ambient temperature is generally tolerated by lyophilized material; the same excursion is a different question entirely for material in solution. Our shipping and quality pages describe how documentation travels with a lot.

When a COA Isn't Worth Paying For

This section costs us sales, and it should.

A COA proves a sample met a spec on a date. It does not describe the vial in your hand. Fill lots run to thousands of vials from a single bulk. One or a few vials are tested. Fill uniformity, vial-to-vial residual moisture, and stopper integrity are all untested variables. Anyone who tells you a COA guarantees the contents of a specific vial is overselling the document.

Independent verification often costs more than the material. Third-party HPLC-MS on a submitted sample typically runs $75–$250; a fuller panel adding Karl Fischer, residual solvents by USP <467>, and amino acid analysis runs $400–$900, with 5–10 business days turnaround. If a vial costs $28, sacrificing one vial and paying $150 to test it adds well over 100% to the effective cost of that lot. For an exploratory screen where identity is the only real concern, that is a poor allocation. Verify when you are locking in a multi-lot study, not when you are running a single pilot.

COAs cannot detect what the method doesn't look for. Mannitol, sucrose, glycine, and similar bulking agents are essentially transparent above 210 nm. A vial that is 30% mannitol by mass can still report 99% HPLC purity at 214 nm, truthfully, because the mannitol never appears in the chromatogram. Only a mass-based method (AAA, quantitative NMR, or a differential weight measurement) catches this. If cut material is your actual concern, the purity number on a COA is not the test that answers it.

Chasing precision your assay can't resolve is wasted budget. If your downstream measurement carries 10–15% coefficient of variation, the difference between 98.5% and 99.6% material is buried in noise. Paying a 40% price premium for the higher grade buys a number for the file, not a better experiment.

A research-use-only COA is not a GMP release certificate. It carries no validated method package, no stability program, no batch record. Anyone whose work requires the latter should be sourcing from a GMP supplier and should expect per-gram costs one to two orders of magnitude higher. See research use only for what that designation does and doesn't cover.

Some buyers should not be buying documented reference material at all. If you have no HPLC access, no comparator lot, and no way to act on a discrepancy, a COA is a document you can file but cannot use. The honest version of supplier qualification for that situation is: buy the smallest quantity, from a supplier who publishes third-party lot documentation, and accept that you are trusting the supplier rather than verifying them.

Building a Lot-Comparison Habit

The highest-value use of a COA is comparative, not absolute. Keep a simple record per lot: lot number, analysis date, retention time of the main peak, purity at the stated wavelength, observed mass, and the two largest named impurities with their area percentages. Across four or five lots from the same supplier, retention time should hold within roughly ±0.2 minutes on the same method, and the impurity profile should look like the same fingerprint. A lot whose impurity pattern suddenly changes shape — new peaks, different ratios — signals a changed synthesis route or a changed manufacturer, even when the headline purity number is identical.

That drift is invisible if you only ever read the summary line. It is obvious within thirty seconds if you keep the chromatograms. Suppliers who resist sending raw traces, or who send the same chromatogram image for three consecutive lots with only the header text changed, have told you something useful about their documentation practices. Our writeup on high purity peptides review covers how to structure that comparison across vendors, and contact reaches us for lot-specific documentation requests.


Supplied strictly for in-vitro laboratory research by qualified researchers and institutions. Not a drug, food, cosmetic or dietary supplement. Not for human or veterinary consumption. Not evaluated by the FDA.

Every batch ships with its own certificate.

Purity, identity and lot number, documented for the exact vial you receive.

How we test