Key takeaways
- HPLC purity says how much of a sample elutes as a single peak; mass spectrometry says whether that peak has the molecular weight the sequence predicts.
- Electrospray ionization produces multiply charged peptide ions, so the instrument reports mass-to-charge ratios that must be deconvoluted back to a neutral mass.
- A match between observed and theoretical mass within a few tenths of a dalton is strong evidence of identity, but it cannot distinguish isomers or a scrambled sequence on its own.
- Tandem MS (MS/MS) fragments the peptide along its backbone and reads the sequence from the fragment ladder, which is the most rigorous identity check routinely available.
A certificate of analysis for a research peptide usually carries two headline results. One is a purity figure from high-performance liquid chromatography (HPLC). The other is a mass, reported in daltons, from a mass spectrometer. Readers tend to fixate on the first and skim the second, which is backwards. A chromatogram can show a sample that is 99 percent one compound without saying which compound it is. Mass spectrometry is the measurement that connects the vial to the sequence printed on the label. This note explains what the instrument actually measures, how the number on a COA is derived, what it can and cannot rule out, and why a sequence-level check with tandem mass spectrometry is the stronger standard.
Why a mass is an identity check
Every peptide has a molecular formula fixed by its amino-acid sequence, and every molecular formula has a calculable mass. A synthetic peptide that came out of the reactor with the intended sequence should therefore weigh what the formula predicts, to within the precision of the instrument. If the observed mass is off by 18 daltons, a water molecule has been lost or gained somewhere. Off by 16, an oxidation. Off by the residue mass of a single amino acid, a deletion or an extra coupling. These arithmetic relationships are what make mass such a useful test: the error modes of solid-phase synthesis, described in our note on how research peptides are made, each leave a characteristic fingerprint in the mass.
Mass spectrometry as a tool for large biomolecules only became practical in the late 1980s, when two “soft” ionization techniques were introduced almost simultaneously. Electrospray ionization (ESI), developed in John Fenn’s laboratory, sprays a peptide solution through a charged needle and lets the solvent evaporate until bare, charged molecules remain in the gas phase.1 Matrix-assisted laser desorption/ionization (MALDI), introduced by Karas and Hillenkamp, embeds the analyte in a crystalline matrix that absorbs a laser pulse and gently lifts intact molecules into the instrument.2 Before these methods, the energy needed to ionize a peptide would typically shatter it. Both approaches earned their inventors a share of the 2002 Nobel Prize in Chemistry, and both remain in routine use for peptide identity work.
What the instrument actually reports
A mass spectrometer does not weigh molecules directly. It measures the mass-to-charge ratio (m/z) of ions, which is the mass in daltons divided by the number of elementary charges the ion carries. This distinction matters because ESI, the dominant method for peptide quality control, tends to attach several protons to a single peptide. A 3,000-dalton peptide might appear in the spectrum at m/z 1,501 carrying two protons, at m/z 1,001 carrying three, and at m/z 751 carrying four. The raw spectrum is therefore a family of peaks, not a single one, and the analyst or the software must work backwards from the charge-state envelope to a single neutral mass. This step, called deconvolution, is routine but not trivial, and a COA that lists only an m/z value without stating the charge state is leaving the reader to finish the calculation.
Monoisotopic vs. average mass
The second source of confusion is which mass is being reported. Carbon, nitrogen, oxygen and sulfur each occur in nature as a mixture of isotopes. About 1.1 percent of carbon atoms are carbon-13, one dalton heavier than carbon-12. A peptide with 150 carbon atoms has a meaningful probability that one or more of them is the heavy isotope, so the molecule exists as a cluster of species differing by roughly one dalton each. The monoisotopic mass is the mass of the species built entirely from the lightest isotopes. The average mass is the abundance-weighted mean across the whole cluster. For a small peptide the two differ by a fraction of a dalton; for a 5,000-dalton peptide the gap approaches three daltons. High-resolution instruments resolve the individual isotope peaks and report the monoisotopic value, while lower-resolution instruments see the cluster as a single hump and report something closer to the average. A COA should say which convention it is using, because comparing an observed monoisotopic mass against a theoretical average mass will produce a spurious discrepancy.
| Quantity | What it means | Where it appears |
|---|---|---|
| Theoretical mass | Calculated from the sequence and any modifications (amidation, acetylation, salt form excluded) | Specification column of the COA |
| Observed m/z | Raw instrument reading for one charge state | Spectrum annotation |
| Deconvoluted neutral mass | Charge states collapsed to a single molecular weight | Result column of the COA |
| Monoisotopic mass | Lightest-isotope species only; requires resolving instrument | High-resolution LC-MS reports |
| Average mass | Abundance-weighted mean over isotope cluster | Lower-resolution or MALDI-TOF reports |
Resolution, accuracy and what “within tolerance” means
Two instrument properties govern how much a mass result can tell you. Resolution is the ability to separate two ions of nearly identical m/z; it determines whether the isotope peaks appear as distinct spikes or blur together. Mass accuracy is how close the measured value is to the true one, usually expressed in parts per million (ppm). A quadrupole instrument might achieve accuracy of a few hundred ppm, meaning a 3,000-dalton peptide is located to within roughly one dalton. Time-of-flight and Orbitrap analyzers routinely reach single-digit ppm, locating the same peptide to within a few hundredths of a dalton.3,4 At that level, most plausible synthesis errors are unambiguous: a deamidation (+0.98 Da) is clearly separable from the natural isotope spacing (+1.003 Da) only on a high-resolution platform.
When a COA states that the observed mass “conforms” to the theoretical value, the acceptance window behind that word depends on the instrument class. Reputable third-party laboratories state the tolerance explicitly. A result reported to one decimal place with a stated tolerance of ±0.5 Da is a legitimate identity check on a mid-resolution instrument. A result reported to four decimal places implies a high-resolution platform and should be accompanied by a ppm error figure. Readers comparing COAs across suppliers should not treat more decimal places as automatically better without knowing what generated them; our guide to reading a certificate of analysis covers how to interpret the method line.
A purity number describes a peak. A mass describes a molecule. Only one of those tells you what is in the vial.
What a single mass cannot rule out
An intact-mass measurement is powerful, but it has a well-defined blind spot: any two molecules with the same molecular formula have the same mass. For peptides this creates three specific gaps. First, sequence scrambling. A peptide with the correct residues coupled in the wrong order weighs exactly what the correct sequence weighs. Second, isomeric residues. Leucine and isoleucine are indistinguishable by mass, as are the D- and L-enantiomers of any amino acid; a racemized residue, which is a known side-reaction in solid-phase synthesis, is invisible to an intact-mass check.5 Third, co-eluting isobaric impurities. If a deletion product happens to be compensated by an unrelated modification of equal mass, the two can overlap.
These gaps are not hypothetical. D’Hondt and colleagues catalogued the impurity classes that arise in synthetic peptide manufacture and noted that several, including diastereomers and certain rearrangement products, require chromatographic or fragmentation-based methods for detection because mass alone will not separate them from the target.5 This is why mass spectrometry and HPLC are complementary rather than redundant. The chromatogram separates species by physical behavior; the spectrometer identifies them by mass. Together they close most of the gaps that either leaves open alone. Our companion note on what HPLC measures covers the other half of that pairing.
Limits of the method
A conforming intact mass is necessary but not sufficient for identity. It confirms the correct molecular formula; it does not confirm the correct sequence or stereochemistry. Laboratories that want sequence-level confidence add tandem MS or amino-acid analysis.
Tandem mass spectrometry: reading the sequence
The method that closes the sequence gap is tandem mass spectrometry, written MS/MS. The instrument isolates one ion from the first spectrum, typically the most abundant charge state of the intact peptide, and collides it with an inert gas. The collision energy breaks the peptide backbone, preferentially at the amide bonds between residues. The fragments are then measured in a second stage of mass analysis. Because the backbone can break at any of the peptide bonds, the result is a ladder of fragments differing in mass by exactly one amino-acid residue each. Reading the mass differences between consecutive rungs recovers the sequence.
The fragment ions are named according to a nomenclature proposed by Roepstorff and Fohlman in 1984 and still in universal use.6 Fragments containing the N-terminus are labeled a, b or c depending on which bond broke; fragments containing the C-terminus are labeled x, y or z. Under the collision conditions most commonly used, b and y ions dominate, and a complete b/y series is considered strong confirmation of sequence.4 Where one residue’s position cannot be resolved, for example two adjacent residues whose order is ambiguous because a fragment is missing, the analyst reports the gap rather than assuming. Steen and Mann’s review remains the standard primer on how fragment spectra are interpreted and where the method’s ambiguities lie.4
In pharmaceutical settings, sequence verification by MS/MS is expected. The FDA’s 2021 guidance on synthetic peptide products that reference recombinant-origin listed drugs asks applicants to characterize the primary sequence and to identify and quantify peptide-related impurities, work that in practice rests on liquid chromatography coupled to tandem mass spectrometry.7 Research-grade material is not held to that regulatory standard, but the analytical logic transfers directly: an intact mass establishes formula, and MS/MS establishes sequence.
How a peptide is run in practice
The dominant configuration for peptide identity work is LC-MS: an HPLC system feeding its eluent directly into an electrospray source. The chromatography separates the target from its impurities in time, and the spectrometer records a spectrum at every point along the run. This arrangement produces, in one experiment, the purity chromatogram, the intact mass of the main peak, and the masses of the minor peaks, so that each impurity can be tentatively assigned (a +16 shoulder as oxidized methionine, for example, or a -71 peak as an alanine deletion). Some laboratories report these assignments on the COA; most report only the main-peak mass. Either is legitimate, but the more complete report is more informative.
MALDI-TOF is the common alternative, particularly for rapid screening. It produces predominantly singly charged ions, so no deconvolution is needed, and it tolerates salts and buffers better than ESI. Its limitations are lower mass accuracy on older instruments and the absence of the chromatographic separation that LC-MS provides, so a MALDI spectrum of an impure sample will show the impurities as extra peaks without telling you their relative abundance as reliably as an HPLC trace would.
Reading the evidence
When a COA reports an LC-MS result, look for four things: the theoretical mass and how it was calculated; the observed neutral mass, not just an m/z; whether monoisotopic or average convention is used; and the stated acceptance tolerance. If any of these is missing, the result is incomplete rather than wrong, but it is harder to evaluate. Wednesday publishes third-party COAs for each lot in the COA library.
Where this fits in a quality picture
Mass spectrometry answers one question with high confidence: is the molecular formula correct? Combined with HPLC it answers a second: how much of the sample is that molecule? Neither answers how much material is actually in the vial by weight, which depends on counter-ions and water content and is the subject of our note on net peptide content versus purity. Nor does either address sterility or endotoxin, which are separate tests. A complete quality picture is assembled from several methods, each with a specific job, and the mass result is the one that anchors everything else to the correct molecule. For an overview of why these tests are best performed by an independent laboratory, see why third-party testing matters.
Frequently asked questions
What does mass spectrometry tell you about a peptide?
It measures the molecular weight of the ions in a sample. Because a peptide’s sequence fixes its molecular formula and therefore its mass, a match between the observed and theoretical mass is evidence that the correct compound is present. Tandem MS goes further and reads the sequence from backbone fragments.
Why does the observed mass on a COA differ slightly from the theoretical mass?
Small differences of a few tenths of a dalton usually reflect instrument accuracy or a mismatch between monoisotopic and average mass conventions. Larger, specific offsets such as +16, +18 or -71 daltons point to chemical differences like oxidation, hydration or a missing residue and should be investigated.
Can mass spectrometry detect a wrong amino-acid sequence?
An intact-mass measurement cannot, because a scrambled sequence has the same formula and mass as the correct one. Tandem mass spectrometry (MS/MS) fragments the peptide and reads the residue order from the fragment ladder, which does detect sequence errors.
What is the difference between LC-MS and MALDI-TOF for peptides?
LC-MS couples chromatographic separation to electrospray ionization, giving purity and mass in one run with multiply charged ions that must be deconvoluted. MALDI-TOF produces mainly singly charged ions and is faster to interpret but lacks the built-in separation of impurities.
Is a mass spec result enough to confirm peptide quality?
No. It confirms identity by formula. Purity requires HPLC, quantity requires net-content methods such as amino-acid analysis, and safety-related attributes such as endotoxin need their own assays. A complete COA reports several methods together.
References & further reading
- Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Electrospray ionization for mass spectrometry of large biomolecules. Science. 1989;246(4926):64-71. doi:10.1126/science.2675315 / PMID 2675315
- Karas M, Hillenkamp F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem. 1988;60(20):2299-2301. doi:10.1021/ac00171a028 / PMID 3239801
- Zubarev RA, Makarov A. Orbitrap mass spectrometry. Anal Chem. 2013;85(11):5288-5296. doi:10.1021/ac4001223 / PMID 23590404
- Steen H, Mann M. The ABC’s (and XYZ’s) of peptide sequencing. Nat Rev Mol Cell Biol. 2004;5(9):699-711. doi:10.1038/nrm1468 / PMID 15340378
- D’Hondt M, Bracke N, Taevernier L, et al. Related impurities in peptide medicines. J Pharm Biomed Anal. 2014;101:2-30. doi:10.1016/j.jpba.2014.06.012 / PMID 25044089
- Roepstorff P, Fohlman J. Proposal for a common nomenclature for sequence ions in mass spectra of peptides. Biomed Mass Spectrom. 1984;11(11):601. PMID 6525415
- US Food and Drug Administration. ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin: Guidance for Industry. May 2021. fda.gov