Endotoxins, heavy metals and TFA: what a COA rarely shows

What '99% purity' does and does not mean

The number on most peptide certificates of analysis is HPLC purity: the share of the material that elutes as the target peptide rather than as related peptides, such as truncated or modified sequences. It is a useful number, and it answers a narrow question. It says nothing about whether the vial contains bacterial toxins, live microorganisms, toxic metals, leftover solvents from synthesis, or what salt the peptide is bound to. A peptide can be 99% pure by HPLC and still be unsafe to inject.

Approved injectable drugs are tested for all of those things against written standards, and the batch is released only if it passes. Products sold as research chemicals are not required to meet any of them. A 2018 review of falsified peptide and protein drugs by a Belgian national medicines laboratory lists the problems found in seized products: the wrong active ingredient, the wrong dose or none at all, toxic contaminants, and impurities from the product or its manufacture ¹. The rest of this guide takes those hidden contaminants one at a time.

Endotoxin: the contaminant most likely to make you ill

Endotoxin is a component of the outer wall of gram-negative bacteria, released when they die. It survives the filtration that removes live bacteria, it survives freeze-drying, and in the bloodstream it triggers fever, chills and, at high enough doses, shock. Peptides made or handled outside clean manufacturing pick it up easily, which is why injectable drugs are tested for it with the bacterial endotoxins test, usually the Limulus amoebocyte lysate (LAL) assay.

The ceiling is set by body weight, not by the vial. USP chapter 85 recommends a maximum endotoxin exposure of no more than 5 endotoxin units (EU) per kilogram of body weight for most drugs, 0.2 EU/kg for drugs given into the spinal fluid, and 100 EU per square metre for products dosed by body surface area ². Product limits are calculated from that threshold with the formula K/M, where K is 5 EU/kg and M is the maximum dose per kilogram ³. Even the water matters: FDA's inspection guide gives a limit of 0.25 EU/mL for Water for Injection and 0.5 EU/mL for Bacteriostatic Water for Injection ⁴.

What turns up in practice? Researchers in Hungary bought semaglutide from illegal online pharmacies in 2023. Three of six orders for vials arrived; none of the three prefilled pens ordered ever did. All three vials failed most of a packaging inspection, and although no live microorganisms grew, endotoxin was detected in every sample, at 2.16 to 8.95 EU per mg ⁵. Whether that crosses the threshold depends on the dose injected and the person's weight, which is exactly why a certificate needs to report endotoxin per milligram, measured on that batch.

A worked example: why the limit is per kilogram

The per-kilogram ceiling makes the arithmetic personal, and it is worth doing honestly. FDA describes the 5 EU/kg figure as an exposure within one hour, based on an average adult weight of 70 kg ², so a 70 kg adult's threshold is about 350 EU per injection. Take the most contaminated semaglutide vial in the Hungarian study, at 8.95 EU per mg ⁵. A 1 mg injection from it would deliver about 9 EU, roughly 0.13 EU per kg, and even a 2.4 mg dose would deliver about 21 EU, around 0.3 EU per kg. On those numbers, the endotoxin found would not have crossed the fever threshold at semaglutide's small doses.

That is not reassurance; it is the reason a certificate must report the figure. The same contamination per milligram becomes a problem as doses rise, as body weight falls, or when several products and a contaminated diluent are injected together, and nothing about one batch tells you the next. The study's more alarming findings were the others: purity of 7.7% to 14.4% against a 99% label, packaging that failed most quality criteria, and half of the orders never arriving at all ⁵. Endotoxin was simply the contaminant that every vial shared.

Sterility: a separate test from endotoxin

Sterility and endotoxin are often confused. A sterility test asks whether anything in the sample can still grow; an endotoxin test asks whether bacterial toxin is present, alive or dead. A vial can pass one and fail the other, as the semaglutide samples did: no viable microorganisms, yet endotoxin in all of them ⁵. A lyophilised powder that is sterile on the day it was tested can also be contaminated by whoever reconstitutes it, with non-sterile water, a reused needle or a vial stopper that was never swabbed.

For injectable drugs, sterility is built in by manufacturing in controlled conditions and confirmed by testing each batch. A research vial's label promises neither, and a certificate that lists only identity and HPLC purity has not tested for it.

Heavy metals and other elemental impurities

Metals reach drugs through catalysts, equipment, reagents and water. Regulators handle them through a risk-based framework, adopted in FDA guidance as ICH Q3D, which sets a permitted daily exposure for each element and route ⁶. For injectable products, USP chapter 232 lists daily limits of 2 micrograms for cadmium, 5 for lead, 15 for inorganic arsenic and 3 for inorganic mercury, lower than the oral limits for cadmium and mercury because injection bypasses the gut ⁷.

A milligram-scale peptide dose makes a gross metal overdose unlikely from the peptide itself, but the framework exists because no one can know without testing, and because some reagents used in peptide chemistry carry metals. A certificate from an accredited laboratory can report an elemental impurity screen. Most research-peptide certificates do not.

TFA: the salt hiding in most synthetic peptides

Almost every synthetic peptide passes through trifluoroacetic acid (TFA). It is used to cut the finished peptide from the resin it was built on and as an ion-pairing reagent in HPLC purification, so peptides usually come out as TFA salts ⁸. That has two consequences that a purity figure hides.

The first is mass. A vial labelled as a given number of milligrams of a TFA salt contains less peptide than that, because part of the weight is counter-ion; how much depends on the sequence, because positively charged amino acids hold more TFA ⁸. The second is biology. In a 1999 study, TFA at concentrations as low as 10^-8 to 10^-7 M reduced the growth of bone-forming cells and cartilage cells in culture, and the TFA salts of three peptides consistently produced less cell proliferation than their hydrochloride salts, enough to hide a real effect or create a false one. The authors recommended converting peptides to the hydrochloride form before testing them ⁹. A 2025 study validated methods for measuring residual TFA and for exchanging it for chloride, and found the salt form changed how readily some peptides crossed a model membrane ⁸.

Approved peptide drugs specify their salt form, usually acetate. A research vial that does not state its counter-ion, or a certificate that does not report TFA content, leaves both the dose and the chemistry uncertain.

Residual solvents and process impurities

Peptide synthesis and purification use organic solvents, and some remain unless they are removed and tested for. Approved drugs follow residual-solvent limits under the same family of international guidelines as elemental impurities; research products follow none. Process impurities also include peptides with a missing or extra amino acid, which HPLC purity does capture, and reagents or protecting groups left behind, which it may not.

Contamination matters in research too. A 2025 laboratory study found that the immune activation attributed to a wheat protein was dominated by bacterial lipopolysaccharide, another name for endotoxin, in the preparation: blocking the endotoxin cut the signal by 92% ¹⁰. If endotoxin can fool a laboratory into crediting a protein with an effect, it can fool a person injecting a gray-market peptide into crediting the peptide with the fever or the inflammation.

What testing of real products finds

Analyses of products from outside the regulated supply keep finding the same problems. A French toxicology laboratory analysed 110 products seized from bodybuilders between 2016 and 2019: of 75 pharmaceuticals, 33% were substandard in dose, 32% counterfeit in content, and only 19% matched their label ¹¹. In the semaglutide test purchases, the vials held 28.6% to 38.7% more semaglutide than labelled, yet measured purity was only 7.7% to 14.4%, nowhere near the 99% printed on the labels ⁵. A falsified product can be over-strength, impure and contaminated at the same time.

What a complete certificate should report

Our COA decoder reads a certificate and flags what is missing. In plain terms, a certificate that supports an injectable product should name the laboratory and its accreditation, the lot number and test date, and then at least: identity by mass spectrometry, HPLC purity with the chromatogram, net peptide content and counter-ion, bacterial endotoxin in EU per mg, sterility, and, ideally, residual solvents and elemental impurities. Most research-peptide certificates stop after the first two.

Even a complete certificate describes only the sample the laboratory received, which may not be the vial you hold. That is the gap our guide to evaluating a peptide source and our page on what research use only means try to explain, and it is why approved products, made and tested under regulation, remain the only ones whose contents are assured.

Frequently asked questions

Does 99% purity mean a peptide is safe?

No. HPLC purity measures how much of the material is the target peptide rather than related peptides. It does not test for bacterial endotoxin, sterility, heavy metals, residual solvents or the TFA salt.

What is the endotoxin limit for injectable drugs?

USP chapter 85 recommends a maximum endotoxin exposure of 5 EU per kg of body weight for most drugs and 0.2 EU per kg for drugs given into the spinal fluid; product limits are calculated from that with the K/M formula.

What is TFA in peptides?

Trifluoroacetic acid, used in peptide synthesis and HPLC purification, which leaves most synthetic peptides as TFA salts. It adds weight that is not peptide and has affected cells in laboratory studies at low concentrations.

Is a small amount of endotoxin dangerous?

It depends on dose and body weight. The ceiling is 5 EU per kg within an hour, so the same contamination per milligram matters more as the dose rises or body weight falls. Without a measured figure on the certificate, there is no way to know.

Have gray-market peptides been tested for endotoxin?

Yes. In a 2024 study, every semaglutide vial bought from illegal online pharmacies contained endotoxin (2.16 to 8.95 EU per mg), and purity was 7.7% to 14.4% against 99% claimed.

Related on Grey Peptides

• COA decoder
• Verify hub
• How to evaluate a peptide source

Sources

1. Janvier, S., et al. (2018). Falsification of biotechnology drugs: current dangers and/or future disasters? J Pharm Biomed Anal, 161, 175-191. PMID: 30165334
2. FDA. Questions and Answers on Quality-Related Controlled Correspondence: Microbiology (Endotoxin testing). Read October 3, 2026. FDA
3. FDA. Draft guidance: setting endotoxin limits during development of investigational oncology drugs and biological products (K/M formula; K = 5 EU/kg). Read October 3, 2026. FDA
4. FDA. Inspection Technical Guide: Bacterial Endotoxins/Pyrogens (Water for Injection 0.25 EU/mL; Bacteriostatic Water for Injection 0.5 EU/mL). Read October 3, 2026. FDA
5. Ashraf, A. R., et al. (2024). Multifactor Quality and Safety Analysis of Semaglutide Products Sold by Online Sellers Without a Prescription: Market Surveillance, Content Analysis, and Product Purchase Evaluation Study. J Med Internet Res, 26, e65440. PMID: 39509151
6. FDA. Q3D(R2) Elemental Impurities: Guidance for Industry. Read October 3, 2026. FDA
7. United States Pharmacopeia. General Chapter <232> Elemental Impurities: Limits, Table 1 (parenteral daily-dose PDEs: cadmium 2, lead 5, inorganic arsenic 15, inorganic mercury 3 micrograms per day). Text published by USP; read October 3, 2026.
8. Erckes, V., et al. (2025). Towards a Consensus for the Analysis and Exchange of TFA as a Counterion in Synthetic Peptides and Its Influence on Membrane Permeation. Pharmaceuticals (Basel), 18(8). PMID: 40872554
9. Cornish, J., et al. (1999). Trifluoroacetate, a contaminant in purified proteins, inhibits proliferation of osteoblasts and chondrocytes. Am J Physiol, 277(5), E779-83. PMID: 10567002
10. Krouch, D., et al. (2025). Amylase trypsin inhibitors activation of toll-like receptor 4 revisited: The dominance of lipopolysaccharides contamination. Int J Biol Macromol, 310(Pt 4), 143378. PMID: 40288707
11. Fabresse, N., et al. (2021). Analysis of pharmaceutical products and dietary supplements seized from the black market among bodybuilders. Forensic Sci Int, 322, 110771. PMID: 33838562

Educational information, not medical advice. Each dose in this guide names its source, an approved label or a published study. None is a recommendation for you. An unapproved compound has no established safe or effective human dose, and products sold for “research use only” are not made or tested for people. Talk to a doctor before acting on anything on this site, including before you start, stop or change any medicine or dose.