Where synthetic peptide impurities come from: deletion, truncation, and side-chain artifacts in SPPS

A certificate of analysis reports purity as a single number, and that number invites a simple interpretation: the missing few percent is degradation, and better storage would recover it. For synthetic peptides that interpretation is usually wrong. The bulk of the material sitting outside the main HPLC peak was not created in a vial — it was created on a resin bead, one coupling cycle at a time, before the peptide was ever cleaved, purified, or lyophilized. Understanding solid-phase peptide synthesis (SPPS) impurity classes explains why certain sequences are consistently harder to make at high purity, why two lots of the same compound can show different impurity profiles at the same stated purity, and why some impurities are effectively invisible to the analytical method most commonly used to measure them.

This post walks through the main impurity classes generated during synthesis, the chemistry that produces each, and how they present analytically.

The synthesis cycle and where it leaks

Fmoc-based SPPS builds a peptide from the C-terminus toward the N-terminus on an insoluble polymer support. Each residue is added by a repeating two-step cycle: deprotection, in which piperidine removes the Fmoc group from the N-terminal amine of the growing chain, and coupling, in which the next protected amino acid — pre-activated by a coupling reagent — forms an amide bond with that newly exposed amine. The chain elongates one residue per cycle, and everything remains anchored to the resin, so excess reagents and byproducts are simply washed away between steps.

The efficiency of this scheme is also its vulnerability. Because the intermediate is never purified between cycles, any chain that fails a step is not removed. It stays on the resin and continues into subsequent cycles alongside the correct product. A synthesis is therefore not a single reaction with a yield; it is a compounding series of reactions in which every incomplete step permanently enters the final mixture.

The arithmetic is unforgiving. At 99.5% average coupling efficiency, a 10-residue peptide finishes at roughly 96% correct sequence; a 30-residue peptide at roughly 86%; a 40-residue peptide near 82%. At 99% efficiency, the 30-residue case drops to about 74%. This is why crude purity for longer sequences is often modest and why the purification step, rather than the synthesis step, determines what the final material looks like.

Deletion sequences

A deletion sequence is the product of a coupling step that failed on some fraction of the resin-bound chains. The unreacted amine survives the wash steps, gets deprotected in the next cycle along with everything else, and couples the following residue instead. The result is a peptide that is complete at both termini but missing one internal residue.

Deletions are the most analytically troublesome impurity class in SPPS because they are so similar to the target. A peptide missing one small residue — glycine, alanine, serine — differs from the correct sequence by a small mass increment and, frequently, by very little hydrophobicity. In reversed-phase HPLC, retention is governed largely by the hydrophobic surface a molecule presents to the stationary phase. Remove a glycine from a 30-residue chain and that surface barely changes, so the deletion species can co-elute with, or appear only as a shoulder on, the main peak. Standard purity integration will count some of it as product.

Mass spectrometry resolves what chromatography cannot here, since a des-Gly species is 57 Da lighter and a des-Ala species 71 Da lighter than the parent — differences that are trivially distinguished on any modern instrument. This is one of the practical arguments for treating HPLC purity and mass spectrometric identity as complementary rather than redundant analyses. Neither alone characterizes the material.

Certain positions fail more than others. Sterically hindered residues — valine, isoleucine, threonine — couple more slowly. Sequences prone to on-resin aggregation, particularly runs rich in beta-sheet-forming residues, bury the reactive amine inside collapsed chain structure where reagents cannot reach it. Synthesis chemists address these positions with double couplings, elevated temperature, altered solvent systems, or pseudoproline dipeptide building blocks that temporarily disrupt secondary structure. The fact that mitigation is routine is itself informative: difficult positions are predictable from sequence, and a well-developed process anticipates them.

Truncated and capped sequences

A truncation is a chain that stopped growing entirely. Where a deletion skips one residue and resumes, a truncation terminates, leaving a fragment corresponding to the C-terminal portion of the target sequence.

Truncations arise in two ways. In a capped synthesis, unreacted amines are deliberately acetylated after a coupling step — a small amount of acetic anhydride converts the failed chains into a chemically inert acetylated fragment that cannot participate in later cycles. This is a design choice: capping trades a higher count of truncated species for a lower count of deletion species. Since truncations differ substantially in length, and therefore in hydrophobicity, from the target, they are far easier to separate chromatographically than deletions are. Capping deliberately converts a hard purification problem into an easy one.

The second route is incomplete deprotection. If piperidine fails to remove an Fmoc group from a fraction of chains, those chains cannot couple in the next cycle. If the Fmoc group survives to the end of the synthesis, the fragment emerges as a distinctly hydrophobic species — the fluorenyl ring system is large and strongly retained on a C18 column — and generally elutes well after the product, where it is straightforward to detect and remove.

Side-chain artifacts

Amino acid side chains carry protecting groups throughout the synthesis and are unmasked in a final acidic cleavage cocktail, typically trifluoroacetic acid with scavengers. That step is chemically aggressive, and several impurity classes originate there rather than during chain assembly.

Incomplete deprotection leaves protecting groups attached, most often on arginine, whose Pbf group is among the slowest to cleave. The resulting species carries extra mass and considerable extra hydrophobicity. Reattachment is the mirror problem: protecting groups liberated as reactive carbocations can alkylate electron-rich side chains — tryptophan, tyrosine, methionine, cysteine — unless scavengers in the cleavage cocktail intercept them. Scavenger selection is not incidental to the process; it is a determinant of the final impurity profile.

Oxidation is a separate pathway. Methionine oxidizes readily to the sulfoxide, adding 16 Da, and this can occur during cleavage, workup, lyophilization, or storage. Because oxidation is one of the few SPPS-adjacent impurities that also accumulates post-synthesis, a rising +16 Da signal across a stability program is one of the more informative markers for whether material has been handled well. Cysteine-containing peptides add disulfide chemistry: intended intramolecular bridges may form incorrectly, and intermolecular bridges produce dimers and higher oligomers that separate by size-exclusion but may partially co-elute in reversed phase.

Finally, aspartimide formation deserves specific mention because it is sequence-predictable. Aspartic acid followed by glycine, asparagine, or serine can cyclize under the basic conditions of Fmoc removal to form a five-membered aspartimide ring. That ring reopens to give a mixture: the original alpha-peptide, the isomeric beta-peptide, and the corresponding piperidide. The beta-isomer has the same mass as the target — it is a backbone rearrangement, not an addition or loss — so mass spectrometry alone cannot detect it. It requires chromatographic resolution or orthogonal structural methods. Any peptide containing an Asp-Gly motif warrants attention to how the synthesis handled this, typically through backbone protection or modified deprotection conditions.

Reading an impurity profile

Taken together, these classes explain why purity figures need context. A 98% peptide whose remaining 2% is truncated fragments and residual solvent is a different material from a 98% peptide whose remaining 2% is a co-eluting deletion species and an unresolved aspartimide isomer, even though the number on the certificate is identical. The informative document is not the purity value but the chromatogram and the mass spectrum behind it: how many impurity peaks appear, how well resolved they are, whether any masses correspond to predictable synthesis artifacts for that sequence, and whether the method was demonstrated capable of separating the species most likely to be present.

Sequence itself supplies useful priors. Length sets the compounding floor on crude purity. Methionine and tryptophan flag oxidation and alkylation risk. Multiple arginines flag incomplete Pbf removal. Cysteine pairs flag disulfide heterogeneity. Asp-Gly flags aspartimide. None of these makes a compound harder to characterize in principle — they simply indicate which analytical questions are worth asking, and which single-number summaries are least likely to carry the information a researcher actually needs.