How mass spectrometry verifies peptide identity: MALDI-TOF vs ESI-MS

A certificate of analysis that reports 99% purity by HPLC has answered exactly one question: what fraction of the material eluting from the column was a single chromatographic species. It has not established what that species is. Chromatographic retention is a bulk physical property — a function of hydrophobicity, column chemistry, and gradient conditions — and many different molecules can share a retention time. A peptide synthesized with a single incorrect residue, a peptide missing a residue entirely, or a peptide carrying an unintended modification can all elute as a clean, symmetric, apparently pure peak.

Identity is established by mass spectrometry. This is why analytical documentation for a peptide almost always pairs a chromatographic purity figure with a mass determination, and why a COA carrying only one of the two is incomplete. The two dominant techniques for peptide mass determination are matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) and electrospray ionization mass spectrometry (ESI-MS). They differ in how they get the molecule into the gas phase, what the resulting spectrum looks like, and what questions each is best suited to answer.

Why molecular weight is a meaningful identity check

A peptide’s molecular weight is the sum of its residue masses minus one water molecule per peptide bond formed, plus the mass of any modifications. Because the twenty proteinogenic amino acids have distinct and precisely known residue masses, the theoretical monoisotopic mass of a given sequence is a fixed number computable to four decimal places.

This makes mass a sensitive test of sequence fidelity. A deletion — the most common synthesis failure, where a coupling step fails and the chain continues one residue short — shifts the mass by the full residue mass of the omitted amino acid, between 57 Da for glycine and 186 Da for tryptophan. That is an enormous displacement by mass spectrometry standards and is trivially detectable. Incomplete removal of a protecting group leaves a characteristic adduct: a residual tert-butyl group adds 56 Da, a trityl group 242 Da. Oxidation of a methionine residue adds 16 Da. Deamidation of asparagine or glutamine adds approximately 0.984 Da, which is small but resolvable on an instrument with adequate resolution.

The important limitation is that mass alone cannot distinguish sequence isomers. Leucine and isoleucine are exact mass twins, so a substitution between them is invisible. Any rearrangement of the same residue set — a scrambled sequence — has an identical mass. Mass spectrometry confirms composition, not order. Establishing order requires tandem MS, in which a selected precursor ion is fragmented and the sequence is read from the mass differences between fragment ions. Most routine peptide COAs report intact mass only; sequence confirmation by MS/MS is a separate and less commonly commissioned analysis.

MALDI-TOF: soft ionization from a solid matrix

In MALDI, the sample is co-crystallized on a metal target with a large excess of a small organic matrix compound — typically α-cyano-4-hydroxycinnamic acid for peptides, or sinapinic acid for larger species. A pulsed UV laser strikes the crystal. The matrix absorbs the energy, ablates explosively, and carries the embedded analyte into the gas phase while transferring a proton to it. The peptide arrives intact and, characteristically, singly charged.

The ions are then accelerated through a fixed potential into a field-free flight tube. Because all ions receive the same kinetic energy, lighter ions travel faster, and the time taken to reach the detector is proportional to the square root of the mass-to-charge ratio. Measuring flight time yields mass.

The practical consequence of MALDI’s tendency toward single charging is that the spectrum is exceptionally easy to read. A pure peptide of mass M produces essentially one dominant peak at M+1 — the protonated molecular ion, written [M+H]⁺. Interpretation requires no deconvolution: the observed value minus the mass of a proton is the molecular weight. Impurities and adducts appear as additional discrete peaks, and the mass difference between them and the main peak often identifies the defect directly.

MALDI is also comparatively tolerant of buffer salts and involatile additives, which are partly excluded during matrix crystallization. This makes it a robust first-pass identity check on material that has not been extensively cleaned up. Its weaknesses are quantitative: ionization efficiency varies substantially with how the sample crystallized, so peak heights are a poor measure of relative abundance. MALDI answers “is this the right molecule” well and “how much of each species is present” badly.

ESI-MS: soft ionization from solution

Electrospray takes the opposite approach. The sample is introduced as a liquid stream through a fine capillary held at high voltage. The emerging liquid is dispersed into a fine aerosol of charged droplets. As solvent evaporates, the droplets shrink until electrostatic repulsion exceeds surface tension, at which point they fission repeatedly until desolvated analyte ions remain in the gas phase.

Because the analyte is protonated at every accessible basic site — the N-terminus, and the side chains of lysine, arginine, and histidine — electrospray produces multiply charged ions. A single peptide therefore appears not as one peak but as a series, a charge envelope, with each member corresponding to a different number of retained protons. Peaks appear at (M + nH)/n for a range of n.

This is more work to read but carries more information. The charge envelope is internally redundant: each charge state independently reports the molecular weight, and deconvolution software combines them into a single mass value with better precision than any one peak alone. Multiple charging also compresses high-mass species into a lower and more accessible m/z range, which is why electrospray dominates protein work.

The operationally decisive advantage of ESI is that it accepts a flowing liquid, which means it couples directly to liquid chromatography. In an LC-MS run, the column separates the mixture and the mass spectrometer identifies each component as it elutes. A single experiment therefore delivers a purity figure and a mass assignment for every peak — including the impurities. This is a substantially stronger analytical result than a purity number and a separate bulk mass, because it associates each chromatographic species with a specific identity. When a COA reports impurity masses alongside impurity percentages, LC-MS is almost certainly the source.

ESI is correspondingly less tolerant of salts. Involatile buffers suppress ionization and contaminate the source, so samples are typically presented in volatile solvent systems, commonly water and acetonitrile with a small percentage of formic acid.

Reading the mass data on a certificate of analysis

A mass section on a COA generally reports three things: the theoretical molecular weight computed from the claimed sequence, the observed value, and the technique used.

The first check is arithmetic. Compute the expected mass from the sequence independently and confirm it matches the stated theoretical value. A COA whose theoretical mass does not correspond to the sequence it prints has an error at the document level, which raises questions about everything else on the page.

The second check is agreement between theoretical and observed. What counts as acceptable depends on the instrument. A routine MALDI-TOF in linear mode may carry several tenths of a Dalton of uncertainty, and agreement within roughly 1 Da on a peptide of a few thousand Daltons is unremarkable. A reflectron-mode instrument or a modern high-resolution electrospray platform should agree far more closely, and a discrepancy that would be tolerable on the first instrument is a finding on the second. Judging a mass result requires knowing which instrument produced it.

The third check is whether the reported value is monoisotopic or average. The monoisotopic mass uses the lightest isotope of each element; the average mass is weighted across natural isotopic abundance. For a peptide of two to four thousand Daltons the two values differ by one to three Daltons — comfortably large enough to look like a failure if the wrong convention is assumed. High-resolution instruments conventionally report monoisotopic; lower-resolution linear-mode data is usually average. A COA that does not specify which is being reported is ambiguous in a way that matters.

A final consideration is what the technique cannot see. Neither MALDI nor ESI reliably detects counterions. Synthetic peptides purified by reverse-phase chromatography with trifluoroacetic acid in the mobile phase are typically isolated as TFA salts, and the associated counterion mass — which can account for a non-trivial fraction of the weighed material — does not appear in the molecular ion. Residual solvent and water content are likewise invisible. Mass spectrometry establishes what the molecule is; it says nothing about what fraction of the vial contents that molecule represents. That question belongs to net peptide content determination, and it is a separate line on the certificate.

Where the two techniques leave the question

MALDI-TOF and ESI-MS are complementary rather than competing. MALDI is fast, salt-tolerant, and produces a spectrum that can be read at a glance, which suits it to confirming that a synthesis produced the intended molecule. ESI is quantitative, couples to chromatography, and resolves the identity of every species in a mixture rather than the dominant one, which suits it to characterizing what else is present alongside the target.

What neither establishes on its own is the full picture that documentation is often assumed to convey. Intact mass confirms composition but not sequence order. Chromatographic purity describes relative peak area but not identity. Net peptide content describes how much of the weighed solid is peptide at all. These are three separate measurements answering three separate questions, and a certificate reporting only one of them has characterized the material only partially — a distinction worth holding onto when comparing analytical documentation across sources.