How to read a Certificate of Analysis
A practical walkthrough of what a peptide COA contains, how to interpret HPLC and mass-spec numbers, and which sections most often hide red flags.
Updated May 7, 2026 · 5 min read
A Certificate of Analysis is the single most important document in research-chem peptide sourcing. A real peptide COA tells you what's actually in the vial; a fake or incomplete one tells you that the vendor either doesn't know or doesn't want you to know. Learning to read one in detail is a one-time investment that pays back across every order you'll ever place.
This guide walks through the sections of a typical COA, what each number means, and where the most common red flags hide.
What a COA actually is
A Certificate of Analysis is a third-party laboratory report linking a specific batch of product to a set of test results. The document is generated by an analytical lab — not the vendor — and is dated and signed (often electronically) by the lab. The vendor passes the COA along to the customer.
A complete peptide COA covers four core categories:
| Category | Test method | What it answers |
|---|---|---|
| Identity | Mass spectrometry (MS) | Is this molecule the labeled peptide? |
| Purity | HPLC (usually reverse-phase) | What percent is the labeled compound vs. impurities? |
| Endotoxin | LAL assay | Is bacterial endotoxin below threshold? |
| Sterility | Microbial culture | Is the product free of viable organisms? |
For lyophilized research-chem peptides, identity, purity, and endotoxin are the three you'll see most often. Sterility appears on injectable-grade material and on products reconstituted by the vendor.
Reading the identity section
Mass spectrometry confirms the molecular weight of the peptide. The COA will list:
- Theoretical mass — the calculated molecular weight of the labeled peptide
- Observed mass — what the instrument measured
- Difference / tolerance — how close the observed value is to theoretical
For most strength peptides, the observed and theoretical should match to within roughly 1 Da. Larger discrepancies mean either the molecule isn't what's labeled, the peptide is degraded, or there are significant byproducts in the mix.
If the COA shows mass-spec data, look at the actual numbers. "MS confirmed" with no values is a yellow flag. "Identity confirmed" with no method listed is a red flag.
Reading the purity section
HPLC purity is reported as a percentage of total peak area at a given wavelength (typically 214 nm or 220 nm). The COA usually shows:
- Purity percentage — main peak as a percent of all peaks
- Chromatogram — visual trace of the run, with the main peak and any impurity peaks
- Method parameters — column, gradient, flow rate, detection wavelength
How to interpret the number:
| HPLC purity | Interpretation |
|---|---|
| 99%+ | Pharmaceutical-grade target |
| 98 to 99% | Reputable research-chem |
| 95 to 98% | Marginal; impurities may drive side effects |
| Below 95% | Problematic |
A chromatogram with one tall peak and a flat baseline is what you want. Multiple peaks of comparable size mean meaningful impurities. For more on what those impurities tend to be, see peptide purity explained.
Reading the endotoxin section
The LAL (Limulus Amebocyte Lysate) assay measures bacterial endotoxin in EU/mg (endotoxin units per milligram). The COA will list:
- Result — measured value, e.g. "0.8 EU/mg"
- Method — gel-clot, kinetic chromogenic, or kinetic turbidimetric
- Limit applied — what threshold the lab compared to
Reasonable thresholds:
| Use case | Endotoxin limit |
|---|---|
| Pharmaceutical injectable | Below 0.5 EU/mg |
| Research-chem injectable | Below 5 EU/mg |
| Topical | More lenient |
Endotoxin is the single most common cause of "the peptide gave me flu-like side effects" reports. A COA without endotoxin testing on an injectable product is incomplete. See endotoxin testing explained for the deeper background.
Batch number, test date, and lab identification
The administrative section of the COA is where many fake or recycled COAs fall apart:
- Batch / lot number — must match the number printed on the vial you're shipped. If it doesn't match, the COA is for a different production run.
- Test date — should be recent and predate the ship date. A COA from 18 months ago tells you nothing about a fresh batch.
- Manufacturing date — when the peptide was produced
- Expiry / retest date — when the lab recommends retesting
- Laboratory name and address — the analytical lab, not the vendor
Verify the lab is real. A web search for the lab's name should return a website with services, contact information, and a physical address. Labs that exist only as a logo on COAs are not third-party.
Common COA red flags
| Red flag | What it indicates |
|---|---|
| No batch number | Generic or recycled COA |
| Batch number doesn't match the vial | COA is for different material |
| Only one test category | Likely identity-only with no purity or endotoxin data |
| "In-house testing" | Not third-party |
| No test date | Could be years old |
| No chromatogram or mass-spec trace | Numbers without backup |
| Lab not findable online | Possibly fabricated |
| All values rounded to clean numbers | Possibly fabricated |
| Same COA across multiple batches | Recycled document |
When in doubt, ask the vendor for the raw chromatogram and mass-spec trace as image files. Reputable vendors will share them.
What to do with a COA after you receive it
Save it. Keep a folder by vendor and batch number, with the COA filed alongside the vial photo and order receipt. If a side effect shows up later, that paper trail is what makes investigation possible. Without it, you have no way to distinguish a bad vial from a bad protocol.