What a peptide certificate of analysis does not test for

Reference page · published 2026-09-20

A peptide certificate of analysis answers two questions: is this the right molecule (identity, usually by mass spectrometry) and how much of what the detector saw was the main peak (purity, usually an HPLC area percentage). Both are real measurements. Neither one looks for metal, and neither one looks for leftover solvent. Those are separate tests, on separate instruments, against separate international reference limits — and they are almost never on the document you are handed.

This page is about that gap. It is not an accusation against any seller, including us. It is a description of what two well-documented classes of contaminant are, why they exist in synthetic peptides at all, what happened the one time a government laboratory systematically went looking for them in peptides bought from the internet, and — just as important — the several reasons that finding cannot be stretched into a claim about the market as a whole.

The two classes a certificate usually skips

Making a peptide is chemistry, and chemistry leaves residue. Two categories of it have nothing to do with the peptide's sequence, which is exactly why a sequence-and-purity certificate does not surface them.

Elemental impurities are metals and metalloids — arsenic, lead, cadmium, mercury and others. They can arrive from catalysts, from reagents, from water, from equipment, or from the container. They are measured by a different instrument than the one that weighs a peptide: typically inductively coupled plasma mass spectrometry (ICP-MS), which vaporises a sample and counts elements. A peptide identity scan is looking for one molecule's mass. It is not counting lead atoms, and it would not report them if they were there.

Residual solvents are the organic liquids used to build and clean the molecule and never fully removed. Solid-phase peptide synthesis is a wet process from start to finish. Whatever fraction of those liquids stays behind in the dried powder is, by definition, part of what is in the vial — and it is invisible to both of the tests a certificate normally reports.

Each class has its own international reference framework, and it is worth reading what those documents say they are for, because their scope is narrower than people assume.

What the two frameworks actually say

The International Council for Harmonisation publishes both. The European Medicines Agency's scientific-guideline page for ICH Q3D describes the elemental-impurities guideline this way:

This document presents a process to assess and control elemental impurities in the drug product using the principles of risk management as described in ICH Q9. This process provides a platform for developing a risk-based control strategy to limit elemental impurities in the drug product.

— European Medicines Agency, ICH Q3D Elemental impurities — Scientific guideline. Revision 2 is recorded there as effective from 24 September 2022.

The companion page for ICH Q3C, on residual solvents, is equally plain about its purpose:

This guideline recommends acceptable amounts for residual solvents in pharmaceuticals for the safety of the patient. It recommends use of less toxic solvents and describes levels considered to be toxicologically acceptable for some residual solvents.

— European Medicines Agency, ICH Q3C (R9) Residual solvents — Scientific guideline, reference EMA/CHMP/ICH/82260/2006, recorded with a legal effective date of 29 April 2024. The page's own keyword list includes permitted daily exposure (PDE), which is the unit these limits are expressed in.

Read those two sentences closely and a boundary appears that matters for this whole subject. Both frameworks are written for medicines for human use. Q3C says so in as many words — for the safety of the patient. A research reagent is not a drug product and is not within the legal scope of either guideline. That cuts in an uncomfortable direction, not a reassuring one: the limits are not rules that bind a research-chemical seller, so where they appear in this field at all, they appear as reference points a laboratory chose to measure against, not as a standard anybody is obliged to meet.

What was found when someone systematically looked

In 2018, a team drawn from Belgium's federal public-health institute, its national veterinary and agrochemical laboratory, and the University of Ghent published the study that anchors this page. Its stated starting point was that the work had not been done before:

Although falsified samples of biotherapeutics were already analysed, looking at a specific aspect of their quality or identity, no systematic studies have been performed regarding the presence of different impurities or possible contaminations.

— Janvier S, Cheyns K, Canfyn M, et al. Impurity profiling of the most frequently encountered falsified polypeptide drugs on the Belgian market. Talanta. 2018 (PMID 30029448).

They took the ten most frequently encountered falsified peptide drugs on the Belgian market, acquired from three different suspected illegal internet pharmacies, and screened each for five things, not two: the active ingredient, impurities related to that ingredient, small-molecule contaminants, elemental impurities, and residual solvents. That five-part scope is the whole point. It is a superset of what a certificate reports, and the extra categories are where the findings came from.

On purity and content, the result was wide:

This comprehensive study showed that these type of falsified drugs not only have a high variation in amount of drugs per unit and a low purity (ranging between 5% and 75% for cysteine containing peptides), but also contained the known toxic class one elemental impurities arsenic (As) and lead (Pb).

And on the metals specifically:

One sample was contaminated with Pb while multiple samples were found with concentrations up to ten times the ICH toxicity limit for parenteral drugs.

Then the team did something a total-metals number would not have done on its own. Arsenic exists in organic and inorganic forms with very different toxicity, so they went back and separated them:

Subsequent speciation of As confirmed the elevated concentrations for As and demonstrated that all As was present in the more toxic inorganic form.

That last step is the methodological lesson buried in the paper. A single number for “arsenic” is not one fact but two questions, and the second one — which arsenic — required a further analysis that most testing never reaches. Anyone reading an elemental result anywhere should know whether it is a total or a speciated figure.

The authors' own framing of why any of this was possible is worth quoting, because it names a mechanism and not a villain:

Since these falsified preparations are produced outside the legally required quality systems, end-users have no guarantee regarding the efficacy and safety of these products.

Two other datasets on what arrives versus what is claimed

The Belgian study measured contaminants. Two other published datasets measured a simpler thing — whether the contents matched the label — and both are useful here because that question is also outside a purity percentage.

A French forensic-toxicology group analyzed products seized from the black market among bodybuilders between January 2016 and December 2019, screening by gas chromatography–mass spectrometry and liquid chromatography–high-resolution mass spectrometry. Of 110 products submitted, 75 were pharmaceuticals, and the breakdown of those 75 was:

Category	Count	Share	The paper's definition
Substandard	25 / 75	33%	dosage not on the acceptable range defined for original products
Counterfeit	24 / 75	32%	qualitative formulation does not match the label
Original	14 / 75	19%	qualitative formulation and levels of active ingredients fully matches the declared formulation

Fabresse N, Gheddar L, Kintz P, et al. Analysis of pharmaceutical products and dietary supplements seized from the black market among bodybuilders. Forensic Sci Int. 2021 (PMID 33838562). The three categories above account for 63 of the 75; the paper's abstract does not classify the remainder.

Note what separates the first two rows. A substandard product is the right compound in the wrong amount — a quantity failure. A counterfeit one is qualitative formulation does not match the label — an identity failure. They are different defects, they are caught by different tests, and a purity percentage on its own reports neither.

The quantity question has a cleaner illustration. A Danish and British team bought melanotan II from three online shops and measured both identity and content by liquid chromatography with ultraviolet and tandem mass spectrometric detection:

Vials from two shops contained unknown impurities ranging from 4.1 to 5.9%; impurities from one shop were below the quantification limit. The total amount of melanotan II in vials ranged between 4.32 and 8.84 mg, although each shop claimed that vials contained 10 mg melanotan II.

— Breindahl T, Evans-Brown M, Hindersson P, et al. Identification and characterization by LC-UV-MS/MS of melanotan II skin-tanning products sold illegally on the Internet. Drug Test Anal. 2015 (PMID 24771717).

Every vial in that set was the right molecule. The identity test would have passed. What varied was how much of it was there — from 4.32 mg to 8.84 mg against a uniform 10 mg claim — and that is a third measurement again, distinct from both identity and purity.

Disclosure: Artemis Labs, which publishes this site, sells melanotan 2 as a research compound. We cite a study that found under-filled vials of it anyway, because a finding is not less true when it is inconvenient, and a reference site that quietly skipped the compounds its publisher stocks would be worth nothing to anybody. The study examined products from shops the authors describe as selling illegally; it did not examine ours, and we are not claiming it did.

What does the evidence not show?

This is the section that decides whether the page above is useful or misleading, so it is the longest one.

None of these studies sampled the research-chemical market. The Belgian work is explicitly about falsified peptide drugs from suspected illegal internet pharmacies. The French work analyzed products seized from the black market. The melanotan II work bought from shops the authors describe as selling illegally. A falsified copy of a prescription medicine and a research reagent sold as a research reagent are different objects reaching a buyer by different routes. Nothing in this literature measures what is in any particular seller's vial, including ours, and it should not be read as though it does.

The sample sizes are small and the selection is not random. Ten products from three sources; three online shops; a seizure set that arrived at a laboratory because law enforcement sent it there. A seizure sample is drawn from material already suspected of being illicit, which is close to the opposite of a representative sample. Percentages calculated on 75 items selected that way describe those 75 items.

The data are old enough to matter. The melanotan II analysis was published in 2015, the Belgian impurity profiling in 2018, the French seizure analysis covers 2016–2019. Supply chains in this field have changed repeatedly since. These are the best systematic measurements that exist; they are not current measurements.

And the contrary result is real. A 2026 study assessed 15 elemental impurities in 38 active pharmaceutical ingredient samples from Jordanian pharmaceutical manufacturers by ICP-MS, and reported that concentrations were within or below the permissible limits except for 208Pb as recommended by the United States Pharmacopeia (USP - NF 2022) and the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use ICH Q3D (R2), with a hazard index below 1 in every sample except amoxicillin (PMID 42461535). Read carefully, that is two findings at once: material made inside a quality system mostly meets the limits, and one element still exceeded them. It argues against panic and against complacency in the same sentence.

Absence of a test is not evidence of contamination. The honest statement of what this page establishes is narrow: these two contaminant classes exist, they have defined reference limits, standard peptide certificates do not report them, and in the one systematic study of falsified internet-sourced peptides they were found above those limits. Every step of that is documented. The step that is not documented — and that nobody should take — is inferring what is in an untested vial from the fact that it is untested.

Affiliations and conflicts, as far as they can be read

We report these on every page that leans on a study, because who paid for a measurement is part of the measurement.

Talanta 2018 — authors are affiliated with the Scientific Institute of Public Health (WIV-ISP), Brussels; the Veterinary and Agrochemical Research Centre (CODA-CERVA), Tervuren; and the Drug Quality & Registration laboratory at the University of Ghent. Government and university laboratories; no commercial sponsor is apparent from the affiliations.
Forensic Sci Int 2021 — hospital and university toxicology laboratories in France, including Université Paris-Saclay and the Institut de médecine légale, Strasbourg. One author is also listed with X-Pertise Consulting, a private consultancy. We note it because it is on the paper, not because we have any reason to think it shaped the result.
Drug Test Anal 2015 — a Danish hospital clinical-biochemistry department with Aalborg University, working with public-health and university researchers in the United Kingdom.
Biol Trace Elem Res 2026 — Jordanian university departments; the samples came from Jordanian pharmaceutical manufacturers, which is worth knowing when reading a compliance rate.

Reading a certificate with this in mind

None of this makes a certificate worthless. It makes it a document with a defined scope, which is what it always was. Four questions follow from the record above, and all four are answerable by looking at the paper in front of you, without having to trust anybody:

Which tests are actually listed? If the document reports identity and an HPLC purity figure and nothing else, then elemental impurities and residual solvents were not measured. Not “were measured and passed” — not measured.
Is a metals result a total or a speciated one? The Belgian team's arsenic finding only became interpretable after speciation. A total figure and an inorganic-fraction figure are different facts.
Does anything on the document address how much peptide is in the vial? Purity is a share of a signal. Content is a weight. The melanotan II set passed identity and still ranged from 4.32 to 8.84 mg against a 10 mg claim.
Whose laboratory produced it, and for which lot? That question is large enough to have its own page, linked below.

Where we stand on this

It would be an obvious move to end a page like this by pointing at our own testing. We are not going to, because it would not be true. Artemis Labs does not test for elemental impurities or residual solvents, and no certificate associated with our products reports them. The gap this page describes is a gap that includes us.

We publish it anyway, for a reason that is the whole basis of this site: a researcher who understands the scope of a document is better served than one who has been reassured about it. If that position ever changes, it will change because a test was actually run and can be shown, and this paragraph will say so on the date it happens.

Sources

Janvier S, Cheyns K, Canfyn M, Goscinny S, De Spiegeleer B, Vanhee C, Deconinck E. Impurity profiling of the most frequently encountered falsified polypeptide drugs on the Belgian market. Talanta. 2018 Oct 1;188:795–807. doi:10.1016/j.talanta.2018.06.023 · PMID 30029448
Fabresse N, Gheddar L, Kintz P, Knapp A, Larabi IA, Alvarez JC. Analysis of pharmaceutical products and dietary supplements seized from the black market among bodybuilders. Forensic Sci Int. 2021 May;322:110771. doi:10.1016/j.forsciint.2021.110771 · PMID 33838562
Breindahl T, Evans-Brown M, Hindersson P, McVeigh J, Bellis M, Stensballe A, Kimergård A. Identification and characterization by LC-UV-MS/MS of melanotan II skin-tanning products sold illegally on the Internet. Drug Test Anal. 2015 Feb;7(2):164–72. doi:10.1002/dta.1655 · PMID 24771717
Massadeh AM, Diak WEA, Tahboub YR, Smadi MM. Assessment of Elemental Impurities and Associated Health Risks in Common Active Pharmaceutical Ingredients Using ICP-MS. Biol Trace Elem Res. 2026 Jul 16. Online ahead of print. doi:10.1007/s12011-026-05240-0 · PMID 42461535
European Medicines Agency. ICH Q3D Elemental impurities — Scientific guideline. ema.europa.eu/en/ich-q3d-elemental-impurities-scientific-guideline
European Medicines Agency. ICH Q3C (R9) Residual solvents — Scientific guideline (EMA/CHMP/ICH/82260/2006). ema.europa.eu/en/ich-q3c-r9-residual-solvents-scientific-guideline