A COA is three narrow answers about one sample on one date — not a verdict on the vial in your hand. Here is what each test measures, and what each one cannot see.
A certificate of analysis looks like a verdict. It is actually a set of answers to very specific questions, each tied to a method, about one sample, taken from one batch, on one date. For a peptide, a typical COA answers at most three:
Every word in those questions is load-bearing. "Something in the sample" is not "every molecule." "Fraction of UV signal" is not "fraction of molecules." "One sample" is not "your vial." The sections below take each test apart — what it measures, and the specific, published ways it fails — because the failures are not hypothetical. Each one is documented in the analytical literature with named examples.
The headline number on almost every peptide COA — "99.1% purity" — comes from reversed-phase HPLC. The sample is pushed through a column that separates molecules by how water-shy they are; a UV detector, usually watching at 214 nm where peptide bonds absorb, draws a peak for each component; and purity is reported as the main peak's share of the total peak area.
That number is real, and it is also narrower than it reads, for three documented reasons.
Co-elution: two molecules, one peak. HPLC separates by one property. An impurity similar enough to the target rides underneath the main peak and is counted as product. This is not a theoretical risk: a 2025 study documented des-Gly4 and des-Thr5 deletion impurities of glucagon — molecules missing a single amino acid — co-eluting with the parent peak on standard RP-HPLC (PMID 40857144). The fix used by pharmaceutical analysts is a second, independent method: the FDA's own laboratory has published LC–high-resolution-MS workflows precisely because one chromatogram cannot carry the whole identity-and-purity argument (PMID 25716148).
Area percent is not molecule percent. The detector measures UV absorbance, and components differ in how strongly they absorb. A peak area share is a signal share. Non-absorbing material — salts, water, many small contaminants — contributes no peak at all and is invisible to the calculation, while a strongly absorbing minor impurity is overweighted. "99% by HPLC" means 99% of what the detector saw.
The impurity family is mostly near-relatives. The canonical taxonomy of peptide impurities (D'Hondt et al. 2014, PMID 25044089) is dominated by molecules a single step removed from the target: deletion and insertion sequences, oxidised forms, deamidated forms, and stereoisomers. Those are exactly the species most likely to co-elute or to hide at the edges of the main peak — the method is weakest precisely where the contamination is densest.
Mass spectrometry is the COA's identity check: it weighs the molecule. If the measured mass matches the theoretical mass of the intended peptide, the right molecule is present in the sample. That is a strong and necessary result. It is also, on its own, blind in specific places:
The pharmaceutical standard reflects this: value and identity for peptide reference standards are assigned by mass balance — several orthogonal methods combined and reconciled — the approach the US Pharmacopeia itself uses (PMID 36949371). One chromatogram plus one mass is the floor of the argument, not the ceiling.
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeHere is the least understood gap on a peptide certificate. Purity describes the peptide fraction: of the peptide material present, how much is the right peptide. Net peptide content describes the powder: of the mass in the vial, how much is peptide at all — the rest being water and the counter-ions (usually trifluoroacetate, TFA) that the peptide carries as a salt. A COA can honestly report 99% purity for a powder that is 80% peptide by weight.
The counter-ion arithmetic is unforgiving for small peptides. Take a worked example — this is arithmetic, not a measurement: a peptide of about 1,400 Da carrying two TFA counter-ions (about 114 Da each) is roughly 86% peptide before any water is counted. Now follow a hypothetical "5 mg" vial through both numbers: at 82% net content and 98% purity, the vial holds about 4.0 mg of the intended peptide. Neither figure is a lie; multiplying them is the buyer's job, and most certificates make it impossible by omitting content altogether. The quantitative anchor method pharmacopeias use for content is amino acid analysis (PMID 26424264) — if a certificate reports content, that is usually the method behind it.
The salt form matters beyond arithmetic. TFA is biologically active in its own right: it inhibited cell proliferation at concentrations of 10⁻⁸ to 10⁻⁷ molar in a classic study (Cornish et al. 1999, PMID 10567002), and it modulates glycine receptors at higher, micromolar-to-millimolar concentrations (PMID 22548713) — two different endpoints at two different concentration ranges, often conflated in marketing. This is one reason pharmaceutical peptides are usually supplied as acetate salts, and why a certificate that names its salt form is telling you something real about the manufacturer's process. (The companion guide on how peptides are made explains where the TFA comes from.)
Sterile is not the same as endotoxin-free. Sterility testing asks whether living microbes are present. Endotoxin — lipopolysaccharide, LPS, shed from the walls of gram-negative bacteria — is not alive, sails through sterilising filters, and provokes fever and inflammatory responses on its own. A product can pass every microbial test and still carry endotoxin.
This is not a pedantic distinction; it has a documented falsifying case. A 2024 analysis of semaglutide products bought online (Ashraf et al., PMID 39509151) measured actual purities of 7.7 to 14.37 percent against claims of 99 percent — and found every sample endotoxin-positive while containing zero viable microbes. The products were, in the narrow sense, sterile. They were not clean.
The pharmacopeial endotoxin framework (USP chapter <85>) sets limits as K/M — a threshold constant (K = 5 endotoxin units per kilogram for most parenteral routes) divided by the maximum dose — which means a defensible endotoxin limit is dose-dependent, and a bare "endotoxin: pass" or a per-milligram figure quoted without a dose context is a vendor convention, not a standard. Endotoxin also corrupts research itself: LPS contamination is a recognised confounder that can manufacture apparent "immune effects" in cell experiments (PMID 28355240) — a contaminated batch does not just risk the user, it falsifies the data.
The wider falsified-products literature makes the same point at market scale: a 2018 analysis of seized falsified polypeptide products found purities ranging from 5 to 75 percent, with arsenic and lead contamination up to ten times the ICH parenteral limits in some samples (Janvier et al., PMID 30029448). And even a genuine product drifts: peptide stability work (PMID 36986796) is the reminder that a certificate describes the batch on the day it was tested, not after a year in a warm warehouse.
Dosing cheat sheet, reconstitution reference, and cycle planning — delivered to your inbox.
It is tempting to assume a framework of official thresholds stands behind research-peptide certificates. Mostly, it does not — and knowing exactly where the real rules stop is the most useful thing this guide can give you.
Then the chain-of-custody problem, which no method fixes. A certificate is a document about a sample. Nothing in the document itself binds it to the vial being sold — the binding is the seller's batch numbering and honesty. The documented, quantified problem in the literature is falsified products with false label claims (PMIDs 30029448, 39509151). Claims that a specific share of certificates in circulation are forged documents circulate widely but could not be verified against a primary source for this guide, so they are not printed here — the honest version is stronger anyway: even a genuine COA, genuinely earned, certifies a sample, and the distance from that sample to your vial is trust, not chemistry.
A short checklist, in the order an analyst would ask the questions:
None of this makes a verified COA worthless — it makes it worth exactly what it is: method-bound evidence about a sample, strongest when several orthogonal methods agree, silent about everything it did not measure.
Nothing above tells you where to buy, because sourcing is the part that varies most. Whoever you use, require the same three things:
The vendor we recommend — full panel published per batch. A reader discount use it at checkout.
For research use only. Not for human consumption. 21+.
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeIt means that in one HPLC run, the main peak accounted for about 99% of the total UV peak area — usually measured at 214 nm. It does not mean 99% of the molecules are the target: UV area is a signal share, not a molecule count; impurities similar to the target can hide under the main peak (documented for glucagon deletion impurities, PMID 40857144); and material that does not absorb UV, including water and salts, is not counted at all.
Purity describes the peptide fraction — of the peptide present, how much is the correct peptide. Content describes the whole powder — of the vial's mass, how much is peptide at all, versus water and counter-ions such as TFA. A small peptide carrying two TFA counter-ions is already down to roughly 86% peptide by weight on arithmetic alone. A COA that reports only purity leaves the content question — how much peptide you actually have — unanswered.
No, in two distinct ways. Analytically, it certifies one sample from one batch on one date, and only for the properties its methods can see — mass spec cannot distinguish leucine from isoleucine, and standard HPLC can miss co-eluting impurities. Logistically, nothing inside the document binds it to your vial; that binding is the seller's batch numbering. The documented market problem is falsified products with false claims — purities of 5 to 75% in seized products (PMID 30029448) and 7.7 to 14.37% in online semaglutide sold as 99% pure (PMID 39509151).
Not necessarily — they are different tests for different things. Sterility means no living microbes. Endotoxin is a non-living bacterial fragment (lipopolysaccharide) that survives sterilisation and still provokes fever and inflammation. In the 2024 online-semaglutide analysis, every sample was endotoxin-positive while containing zero viable microbes (PMID 39509151). A certificate without an endotoxin line has not addressed the question.
Largely, no. The ICH impurity guidelines explicitly exclude peptides. The EMA's peptide guideline — 0.1/0.5/1.0% reporting, identification and qualification thresholds, two orthogonal identity methods — was verified for this guide as a 2023 draft (EMA/CHMP/CVMP/QWP/387541/2023). The FDA's synthetic-peptide ANDA guidance covers five named drug products only. USP chapters define methods for pharmaceutical peptides. Research-chemical certificates borrow the vocabulary of this framework without being governed by it, which is why reading the method behind the number matters more than the number.
Disclaimer: This guide is for educational and informational purposes only and is not intended as medical advice, diagnosis, or treatment. The compounds discussed are not FDA approved for human use. Always consult a qualified healthcare provider before starting any new supplement or peptide protocol. StackTrax does not sell peptides or supplements directly — purchase links go to third-party vendors. StackTrax is not responsible for the products, quality, or business practices of any third-party vendor.
© 2026 StackTrax, LLC. All rights reserved.
Privacy · Terms · Do Not Sell or Share My Personal Information
StackTrax guides cover peptides and compounds that are not FDA-approved for the uses discussed. The dosing, reconstitution, and safety information is compiled from published research and community protocols for educational purposes only.
Before using any compound mentioned here, consult a qualified healthcare provider. StackTrax does not sell, prescribe, or recommend these substances for personal use.