Analytical methods
HPLC vs LC-MS for peptides: identity and purity workflow
A Certificate of Analysis usually quotes an HPLC purity and an MS mass. They are different instruments answering different questions. This page shows what each one proves, computes the ions and isotope patterns you should expect, and lays out a workflow.
The short version
Reversed-phase HPLC with UV detection answers how much of the detected material is the main peak (the basics, with a worked example). Mass spectrometry answers whether the main species has the mass of the intended peptide. LC-MS joins the two so that every chromatographic peak can be given a mass. LC-MS/MS goes one step further by fragmenting an ion to test its sequence. None of these on its own proves content (purity is not content).
What each technique can and cannot prove
| Method | What it measures | What it proves | What it cannot prove |
|---|---|---|---|
| HPLC-UV | Absorbance (commonly about 215 to 220 nm) of each peak as it elutes | Share of detected peak area in the main peak; a retention-time fingerprint | Identity; anything that does not absorb or that co-elutes (Karongo et al., 2020 note closely related impurities can co-elute in the generic method) |
| LC-MS (full scan) | Mass-to-charge ratio of ions from each peak | Whether the main peak has the expected mass; masses of neighbouring peaks, which often name deletion, oxidation or protecting-group impurities | Same-mass species; a mass spectrometer's response differs between species, so signal height is not a purity percentage |
| LC-MS/MS | Fragment ions from a selected precursor | Sequence-level evidence; Breindahl et al. (2015) verified melanotan II by retention time plus the relative abundance of five qualifying fragment ions | Content in the vial; stereochemistry in most routine uses |
| High-resolution MS | Mass to several decimal places | Elemental composition of the main species and of impurities; Huo et al. (2022) and Wu et al. (2026) used it to identify and compare impurity profiles | Isomers, unless combined with chromatography or fragmentation |
| Amino-acid analysis or chiral methods | Composition after hydrolysis | Net peptide content; D/L composition | Sequence order (amino-acid analysis explained) |
Why a retention time on an LC-MS run does not match the COA
A purity method and an LC-MS method are frequently different methods. Krokhin and Spicer (2009) describe C18 separations with trifluoroacetic acid or formic acid as the ion-pairing modifier as two of the most popular modes, and propose a hydrophobicity index that depends on the sorbent and on the modifier used. A retention time is therefore only comparable between runs that used the same column, gradient and modifier. Mant et al. (2007) review the HPLC modes used for peptides and their use for structural characterisation.
Reading the mass spectrum: charge states and adducts
In electrospray mass spectrometry, peptide and protein ions are usually observed in multiple charge states, and adduction with other ions such as potassium or sodium often makes the signal pattern for a single analyte complex (Bielow et al., 2010). The table below computes the ions you should expect for three peptides of very different size from their formulas; the sodium and potassium values are offsets of +21.9819 and +37.9559 from [M+H]+.
| Peptide | Formula | Neutral monoisotopic mass | [M+H]+ | [M+NH4]+ | [M+Na]+ | [M+K]+ | [M+2H]2+ | [M+3H]3+ |
|---|---|---|---|---|---|---|---|---|
| Epithalon | C14H22N4O9 | 390.1387 | 391.1460 | 408.1725 | 413.1279 | 429.1018 | 196.0766 | 131.0535 |
| BPC-157 | C62H98N16O22 | 1418.7042 | 1419.7114 | 1436.7380 | 1441.6934 | 1457.6673 | 710.3594 | 473.9087 |
| Semaglutide | C187H291N45O59 | 4111.1154 | 4112.1227 | 4129.1492 | 4134.1046 | 4150.0785 | 2056.5650 | 1371.3791 |
Use the m/z for the charge state your instrument reports. For a multiply charged series, the same neutral mass has to come out of every charge state; the method is worked through in check the mass spectrometry on a COA yourself.
The isotope envelope
A molecule is not a single peak. Natural carbon, hydrogen, nitrogen, oxygen and sulfur have heavier isotopes, so each ion appears as a family of peaks one mass unit apart (for singly charged ions). The computed envelope below uses NIST natural abundances.
| Peptide | M | M+1 | M+2 | M+3 | M+4 | Tallest peak |
|---|---|---|---|---|---|---|
| Epithalon | 1.000 | 0.172 | 0.032 | 0.004 | 0.000 | M (monoisotopic) |
| BPC-157 | 1.000 | 0.749 | 0.322 | 0.101 | 0.026 | M (monoisotopic) |
| Semaglutide | 0.381 | 0.854 | 1.000 | 0.810 | 0.509 | M+2 |
For epithalon the first peak is tallest. For BPC-157 the M+1 peak is about three quarters the height of the first, and for semaglutide the third peak (M+2) is the tallest, so the "monoisotopic" mass is only the first peak, not the strongest. A batch result that quotes the tallest peak of a large peptide as its monoisotopic mass will be off by one or two mass units. At charge z the peaks are spaced 1.0034/z apart in m/z.
Neighbouring peaks: reading mass differences
When an LC-MS run shows a small peak beside the main one, the mass difference is the first clue. A deletion sequence is lighter than the target by one residue mass from the table below; an oxidised methionine is heavier by 15.9949; deamidation adds 0.9840. See common impurities in synthetic peptides and peptide modifications.
| Residue | Name | Residue formula | Mass (Da) |
|---|---|---|---|
| G | Gly | C2H3NO | 57.0215 |
| A | Ala | C3H5NO | 71.0371 |
| S | Ser | C3H5NO2 | 87.0320 |
| P | Pro | C5H7NO | 97.0528 |
| V | Val | C5H9NO | 99.0684 |
| T | Thr | C4H7NO2 | 101.0477 |
| C | Cys | C3H5NOS | 103.0092 |
| L/I | Leu/Ile | C6H11NO | 113.0841 |
| N | Asn | C4H6N2O2 | 114.0429 |
| D | Asp | C4H5NO3 | 115.0269 |
| Q | Gln | C5H8N2O2 | 128.0586 |
| K | Lys | C6H12N2O | 128.0950 |
| E | Glu | C5H7NO3 | 129.0426 |
| M | Met | C5H9NOS | 131.0405 |
| H | His | C6H7N3O | 137.0589 |
| F | Phe | C9H9NO | 147.0684 |
| R | Arg | C6H12N4O | 156.1011 |
| Y | Tyr | C9H9NO2 | 163.0633 |
| W | Trp | C11H10N2O | 186.0793 |
Same-mass impurities need another dimension
Zhang et al. (2023) describe low-level D-amino acid isomers of semaglutide as among the most challenging impurities to characterise. They separated the impurities by reversed-phase UPLC, determined molecular weight by high-resolution MS, collected the isomers, hydrolysed them under conditions that suppress D/L shifting, derivatised with a chiral reagent and analysed by UPLC-MS/MS against standards, and reported D-Ser8, D-His1 and D-Asp9 isomers. The mass of a D-isomer equals that of the target, so a full-scan spectrum cannot see it.
The profile also depends on how the peptide was made. Wu et al. (2026) compared teriparatide reference materials of chemical-synthesis and recombinant origin by LC-HRMS: one isomer and three oxidation impurities were found in all the materials studied, while six amino acid deletion impurities were seen specifically in the chemically synthesised material. Huo et al. (2022) identified one process-related impurity and thirty-two degradation products of the antimicrobial peptide Cbf-14 by HPLC-QTOF-MS/MS, including hydrolysates, isomers and oxidised forms.
An identity workflow for one batch
- Match the label. The lot number on the document is the lot on the vial (how to read a COA).
- Purity trace. HPLC-UV with the column, gradient, modifier and wavelength stated, plus the chromatogram itself.
- Mass of the main peak. Observed m/z, the ion type and charge state, and the neutral mass derived from it, compared with a mass computed from the sequence including every modification.
- Mass of the neighbours. Differences against the residue and modification tables above. A missing residue or +16 points to a named impurity.
- Same-mass question. If stereochemistry or an isomer matters for the experiment, ask for chiral or orthogonal chromatography evidence.
- Retention-time check. If an identity method is built on retention time plus MS/MS, expect it to name the precursor ion. For melanotan II, Breindahl et al. (2015) used the doubly charged precursor at m/z 513; the formula on our melanotan II page gives a computed [M+2H]2+ of 512.78.
- Content. A separate measurement (amino-acid analysis, nitrogen or Karl Fischer) tells you what fraction of the vial is peptide.
Our product pages label stated purity as supplier-stated and show the batch COA as not yet provided until a document exists.
Related
- How to read a COA HPLC chromatogram
- N- and C-terminal modifications and mass spectrometry
- How to read an LC-MS trace: a worked peptide example
- Peptide purity: HPLC and mass spectrometry explained
- Solid-phase peptide synthesis explained
- Semaglutide vs tirzepatide: structure and expected m/z
References
- Karongo R, Ikegami T, Stoll DR, et al. (2020). A selective comprehensive reversed-phase×reversed-phase 2D-liquid chromatography approach with multiple complementary detectors as advanced generic method for the quality control of synthetic and therapeutic peptides. J Chromatogr A 1627:461430. PubMed 32823119 · DOI
- Bielow C, Ruzek S, Huber CG, et al. (2010). Optimal decharging and clustering of charge ladders generated in ESI-MS. J Proteome Res 9:2688-95. PubMed 20201597 · DOI
- Zhang B, Xu W, Yin C, et al. (2023). Characterization of low-level D-amino acid isomeric impurities of Semaglutide using liquid chromatography-high resolution tandem mass spectrometry. J Pharm Biomed Anal 224:115164. PubMed 36462248 · DOI
- Wu P, Cheng Y, Li D, et al. (2026). Comparison of Structurally Related Impurity Profiles in Teriparatide From Synthetic and Recombinant DNA Origin Using Liquid Chromatography-High Resolution Mass Spectrometry. Rapid Commun Mass Spectrom 40:e70125. PubMed 42312586 · DOI
- Huo Y, Xu K, Lu Y, et al. (2022). Characterization of structurally related peptide impurities using HPLC-QTOF-MS/MS: application to Cbf-14, a novel antimicrobial peptide. Anal Bioanal Chem 414:6485-6495. PubMed 35840670 · DOI
- Breindahl T, Evans-Brown M, Hindersson P, et al. (2015). Identification and characterization by LC-UV-MS/MS of melanotan II skin-tanning products sold illegally on the Internet. Drug Test Anal 7:164-72. PubMed 24771717 · DOI
- Krokhin OV, Spicer V (2009). Peptide retention standards and hydrophobicity indexes in reversed-phase high-performance liquid chromatography of peptides. Anal Chem 81:9522-30. PubMed 19848410 · DOI
- Mant CT, Chen Y, Yan Z, et al. (2007). HPLC analysis and purification of peptides. Methods Mol Biol 386:3-55. PubMed 18604941 · DOI
- NIST, Atomic Weights and Isotopic Compositions with Relative Atomic Masses, physics.nist.gov: isotope masses and natural abundances used for every computed value on this page.
Frequently asked questions
What is the difference between HPLC and LC-MS for peptides?
HPLC with UV detection separates the sample and reports each peak's area, which gives a purity percentage. LC-MS adds a mass spectrometer after the column so each peak can be assigned a mass, which tests identity and helps name impurities. The two answer different questions.
Is LC-MS a better purity test than HPLC-UV?
Not by itself. Purity percentages are normally taken from UV peak areas, as in the generic quality-control method with a C18 column and 0.1% TFA that Karongo et al. (2020) describe, and mass-spectrometer response varies from species to species. LC-MS is chiefly the tool for assigning peaks.
Why does a peptide show several peaks in a mass spectrum?
Electrospray produces ions in several charge states, and adducts with sodium, potassium or ammonium add further signals (Bielow et al., 2010). The tables on this page compute them for three peptides.
Can mass spectrometry detect every impurity?
No. Impurities with the same mass, such as D-amino acid isomers, need chromatographic separation and extra steps; Zhang et al. (2023) needed separation, high-resolution MS, hydrolysis and chiral derivatisation to characterise D-isomer impurities of semaglutide.
Why is the monoisotopic peak not always the tallest?
Large molecules contain many carbon atoms, so more of the population carries at least one carbon-13. In the table on this page the tallest peak moves from M for epithalon to M+2 for semaglutide.