HPLC and mass spectrometry: what each method actually proves about a peptide
A purity figure is the output of a method with a detection limit and a quantifier, not a grade stamped on a substance. How HPLC-UV and mass spectrometry divide the work, what a percentage does and does not mean, and why one measurement on somebody else's sample cannot be read as a standing fact about the vial in your hand.
1 October 2026
Send me the checklist
One email when a new hub goes up. Research use only; nothing else, and one click to leave.
Purity is the number this bench talks about most and understands least. It is a word that sounds like a property of a substance and is actually the output of a specific instrument, on a specific aliquot, on a specific day, at a specific detection limit. Change the detector, the column or the correction practice and you can get a different number out of the same material.
The method-by-method detail is on how peptide purity is measured, and this page does not repeat it. What follows is the narrower question: which method proves what, and why a percentage on its own is not a grade.
The two instruments, dividing the work
HPLC with UV detection separates, then counts. A column separates by retention; a UV detector counts how much light each component absorbs as it comes off. The result you get is a table of peaks with areas, and the percentage is each area over the total. That is a ratio of detector responses, not of masses, which is the single most under-appreciated fact in the whole exchange.
Mass spectrometry identifies. Coupled to the liquid chromatograph, it reads mass-to-charge and confirms identity independently of retention. In the peptide quality-control method published for calcitonin, bivalirudin and exenatide, limits of detection were 0.02, 0.03 and 0.04 µM for the parent peptide and two related impurities, meaning an impurity present at less than 0.1% of a 107 µM API concentration was still detectable; linearity held with r² of 0.995, 0.996 and 0.993, and intra-assay precision was below 10% RSD. That is a well-characterised method, and it is worth noticing what it takes: a named matrix, named impurities, a validation.
They run together, because each covers the other’s gap. Mass spectrometry cannot separate isomers — compounds of the same mass-to-charge ratio are not readily distinguished by it and have to be pushed apart by the column. That is why two-dimensional LC-MS exists as a distinct discipline: one study ranked more than 300 UV and MS chromatograms across 30 column and mobile-phase combinations before it could say which second dimension would resolve peptide isomers. And because UV quantifies against a response rather than a mass, a chromatogram needs a relative response factor to convert area into impurity quantity.
Why a percentage is not a grade
Three things stand between the peak-area table and the word “pure”.
The quantifier. An impurity does not necessarily respond to UV the way the parent does. Same chromophore, same molar absorptivity — then the area percentage approximates the mass percentage. Different chromophore or different molecular weight, and the two numbers come apart in a direction that is not knowable from the chromatogram alone. The relative response factor exists to correct exactly this, and using a default of 1 can overestimate or underestimate the impurity’s true quantity.
The detection limit. Everything below it is not “absent”; it is unseen. A method with a limit at 0.1% of the API concentration cannot tell you what the main peak is contaminated with at 0.02%.
The sample. This is the part most often left out. A purity number is a measurement of the material that was injected. In the 2024 semaglutide study, of 1,080 links collected from search results, 134 (42.3%) led to 59 unique illegal online pharmacies; three vials were delivered, and they measured 7.7% to 14.37% purity against 99% claimed, carried 28.56% to 38.69% more semaglutide than the label declared, and contained endotoxin in every sample between 2.1645 and 8.9511 EU/mg, and none of the three prefilled-pen orders reached the buyer at all.
The Belgian screening came at it from the impurity side instead of the label side, and the spread is wider still: purity from 5% to 75% for cysteine-containing peptides, arsenic and lead several times up to ten times the ICH toxicity limit for parenteral drugs, and speciation confirming the arsenic was entirely in the more toxic inorganic form.
Reading a number that came from somebody else
When the number is not yours, four questions decide what it is worth.
Whose sample? A report describes a lot, and the lot has to be the lot in your parcel. That is a question to ask before ordering, not after.
Which salt form? Peptides are sold as hydrochloride or acetate salts, and the counterion is a separate measurement with its own method — charged aerosol detection and time-of-flight mass spectrometry exist for exactly that, developed at the European Pharmacopoeia’s laboratory department. A purity number for the wrong salt describes a different substance.
Which quantifier, and was it applied? If the report does not say, the number is an area percentage being read as a mass percentage.
What was the detection limit? A figure that only counts what it could see cannot bound what it could not.
None of these questions is answered by the number itself. That is the argument for asking the seller for the lot number, the salt form, the method and the laboratory, and for treating a refusal as data.
What we do not publish
We publish the supplier’s own text verbatim and add no claim of our own. We do not publish a purity figure for any vial, a test certificate, or an assay value, because no supplier on this bench publishes one; an empty field stays empty. We do not repack, we do not relabel, we cannot promise a purity or a result, and nothing here is medical advice. Where a supplier does publish batch paperwork, send the order number and we request it for the batch that shipped.
The document side of the same question is on how to read a peptide certificate of analysis and what third-party testing actually proves.
⚠ Everything sold here is for in-vitro laboratory research only. Not for human or veterinary use.
What each analytical method answers, and what it leaves unsaid
| What | What it determines | Its own limit | The claim it cannot carry | What it means for you |
|---|---|---|---|---|
| HPLC with UV detection | Separation by retention, then area under each peak expressed as a percentage of the total. It resolves what the column can separate and quantifies against a detector response | A percentage of signal, not of mass. Injections differ, UV response differs per chromophore, and the detector sees what absorbs at that wavelength | "99% pure" as a property of the substance. The same peaks, read by a different detector, can give a different number for the same material | Chromatography decides what is separate; the detector decides what counts. The percentage is the detector's arithmetic, not the substance's identity. |
| LC coupled to mass spectrometry | Mass-to-charge, and therefore molecular identity, separately from retention. In a peptide QC method, limits of detection of 0.02, 0.03 and 0.04 µM for a parent peptide and two impurities meant an impurity below 0.1% of a 107 µM API concentration could be detected | Isomers are not separated by it. Compounds with the same mass-to-charge ratio are not readily distinguished by mass spectrometry and must be separated chromatographically | "The peptide is what the label says, therefore the purity is 99%." Mass spectrometry answers identity; the area percentage still answers purity, and the two are different questions | Identity and purity are separate results. A confirmed peptide can still be a mixture, and the confirmation does not change the percentage. |
| Two-dimensional LC with mass spectrometry | Whether the main peak is a single compound. Isomers can be pushed onto a second dimension by column chemistry rather than by mass | Method development, not routine release testing. One study ranked over 300 chromatograms across 30 column and mobile phase combinations to reach that answer | "Peak purity equals batch purity." The main peak can be one compound while a tenth of the material sits in a small, invisible set of impurities | This is the method that tells you whether the big peak is honestly one thing. It is a method-development answer, and it is how a laboratory arrives at a routine test that can be run per batch. |
| Any purity figure, wherever it came from | One sample, on one day, by one method, at one detection limit. In the peer-to-peer semaglutide study three vials measured between 7.7% and 14.37% purity against 99% claimed, and 28.56% to 38.69% more semaglutide than the label declared | Whether it was your vial. The Belgian screening found purity from 5% to 75% for cysteine-containing peptides, arsenic and lead up to ten times the ICH limit, and arsenic entirely in the more toxic inorganic form | A standing fact about a product line. The same vial, re-assayed by a different laboratory with a different method, is a different measurement | Every purity figure has four coordinates: who ran it, on what, by which method, with what limit. A figure quoted without them is a rumour with a decimal point. |
Questions this page answers.
If a purity percentage is printed, what can I conclude?
That somebody measured one sample by one method and got that number on that day. The semaglutide vials bought from sellers without a prescription measured 7.7% to 14.37% against 99% claimed on the labels — so a printed figure and a measured figure can differ by an order of magnitude without anybody having lied, if the printed one was never checked. What the number tells you is what the assay saw, not what is in your hand.
Is mass spectrometry simply the better method?
It answers a different question, and the two are usually run together rather than instead of one another. Mass spectrometry establishes identity from mass-to-charge, and cannot separate isomers — compounds of the same mass have to be separated by the column. Chromatography separates, and UV quantifies against a response. The peptide QC literature puts both in one run because that combination can determine sequence and composition, and quantify impurities even where they co-elute, within a single experiment.
Why does the same peak-area table not give the same purity number in every lab?
Because the percentage is a ratio of detector responses, not of masses. An impurity with a different chromophore, a different molecular weight or a different response factor contributes to the area without contributing the same share of the mass, so the impurity's mass percentage is not its area percentage. The relative response factor corrects for it, and defaulting it to 1 can overestimate or underestimate the impurity in either direction — which is why a single purity figure needs its method, its detector and its correction practice alongside it to mean anything.
What does this shop publish, and what does it not?
We publish the supplier's own product text verbatim and add no claim to it. We do not publish a purity figure, an assay result or a test certificate for any vial, because no supplier on our bench publishes one and we will not invent a number. Where a supplier does publish batch paperwork, send your order number and we request it for the batch that actually shipped.
Where this comes from.
- Zeng K, Geerlof-Vidavisky I, Gucinski A, Jiang X, Boyne MT 2nd. Liquid Chromatography-High Resolution Mass Spectrometry for Peptide Drug Quality Control. AAPS J 2015;17(3):643-651. (PMID 25716148) — limits of detection 0.02, 0.03 and 0.04 µM; an impurity below 0.1% of a 107 µM API concentration was detectable; r² ≥ 0.993; intra-assay precision below 10% RSD.
- Petersson P, Buckenmaier S, Euerby MR, Stoll DR. A strategy for assessing peak purity of pharmaceutical peptides in reversed phase chromatography methods using two-dimensional liquid chromatography coupled to mass spectrometry. Part I. J Chromatogr A 2023;1693:463874. (PMID 36841023) — 30 column/mobile-phase combinations; compounds with the same mass-to-charge ratio are not readily differentiated by mass spectrometry and must be separated chromatographically.
- Kumar Kuril A. The Critical Need for Implementing RRF in the Accurate Assessment of Impurities in Peptide Therapeutics. Anal Chem 2025;97(24):12480-12485. (PMID 40499007) — applying an RRF of 1 by default may over- or underestimate impurities in either direction.
- Ashraf AR, Mackey TK, Vida RG, et al. Multifactor Quality and Safety Analysis of Semaglutide Products Sold by Online Sellers Without a Prescription. J Med Internet Res 2024;26:e65440. (PMID 39509151) — 1,080 links; 134 (42.3%) to 59 unique illegal online pharmacies; measured purity 7.7–14.37% against 99% claimed; content 28.56–38.69% above label; endotoxin in all samples, 2.1645–8.9511 EU/mg.
- Janvier S, Cheyns K, Canfyn M, et al. Impurity profiling of the most frequently encountered falsified polypeptide drugs on the Belgian market. Talanta 2018;188:795-807. (PMID 30029448) — purity 5–75% for cysteine-containing peptides; arsenic and lead up to ten times the ICH toxicity limit for parenteral drugs.
- Ilko D, Nap CJ, Holzgrabe U, Almeling S. Validation and application of an HPLC-CAD-TOF/MS method for identification and quantification of pharmaceutical counterions. Pharmeur Bio Sci Notes 2014;2014:81-91. (PMID 25655245) — how the counterion is identified and quantified, and why the salt form is a separate measurement from the peptide.
- Directive 2001/83/EC on the Community code relating to medicinal products for human use — what a medicinal product is, and therefore what a research compound is not. EUR-Lex.
