Why Peptide Purity Varies by Batch: 9 Lot Reports (2026)

Why peptide purity varies by batch comes down to four stages, each slightly different every time it runs: the synthesis that builds the chain, the purification that removes by-products, the freeze-drying that turns solution into powder, and the HPLC measurement that produces the number. In nine published lot reports, HPLC purity ran from 99.18% to 99.97%. That band looks narrow, but the share of peak area outside the main peak ran from 0.03% to 0.82%, about a 27-fold spread. Those nine reports cover eight different materials, so most of that spread is between materials rather than between batches, and separating the two takes more than one report.

Sources of batch variation at a glance

Source: Vortex Research Editorial Team, from the literature cited below.

Research materials: Vortex Research publishes its third-party lot reports at /coa/. The values used in this article come from the open lot report dataset, doi:10.5281/zenodo.23578844.

Where batch variation starts: synthesis

Solid-phase synthesis, the route behind most synthetic peptides, grows the chain on resin beads: for every residue added there is a deprotection step and then a coupling step [1, 2]. No step goes to 100% completion. The fraction of chains that reach full length is roughly the per-step completeness multiplied across every coupling, so small differences between runs compound with length.

Source: illustration computed for this article as p⁽ⁿ⁻¹⁾, ignoring deprotection losses and side reactions. Residue counts from the Vortex peptide identity reference (doi:10.5281/zenodo.23578942). These are not measured values for any lot.

The chains that fall short become deletion sequences, missing one or more residues. Other by-products come from side reactions: insertion sequences from excess reagent, diastereomers from racemisation, adducts from incompletely removed side-chain protecting groups, oxidised residues and dimers [2]. Some are sequence-specific. Aspartic acid residues can cyclise to aspartimide under the basic conditions used to remove Fmoc groups, a long-standing problem in Fmoc chemistry [3]; BPC-157 contains two adjacent aspartic acid residues. Methionine and tryptophan side chains oxidise readily; Semax contains one methionine, and MOTS-c contains two methionines and a tryptophan. How much of each by-product forms depends on reagent quality, temperature, mixing and timing, all of which drift slightly between runs. Our overview of how peptides are made describes the synthesis steps in more detail.

Purification decides the final figure

The usual clean-up is preparative reversed-phase HPLC. Crude material goes onto a column, fractions are collected as they elute, each fraction is checked, and the fractions that meet a target are pooled. The pooling decision has a large effect on the final purity. A wide cut keeps more product and more near-eluting impurity. A narrow cut gives higher purity and lower yield.

Column behaviour under heavy loading also varies. In a study of three peptides on an end-capped C18 phase, elution profiles under overloaded conditions depended on trifluoroacetic acid concentration, and at low concentrations showed a sharp front and a long tail [4]. Small differences in loading or mobile phase between purification runs shift where impurities sit relative to the main peak, and so how much of them ends up in the pool.

The dataset shows the effect of this stage. On the stepwise arithmetic above, a 3-residue chain should start purification far cleaner than a 16-residue one. Yet the 100 mg GHK-Cu lot reports 99.34% and the MOTS-c lot reports 99.94%. Chain length alone does not predict the final figure. The comparison is consistent with purification, and the target it was run to, doing most of the work.

Freeze-drying and the non-peptide fraction

After purification, pooled fractions are freeze-dried. The powder that results holds the peptide plus counter-ions, commonly trifluoroacetate from the purification mobile phase, and some residual water. They are not counted as peptide peaks in an area-percent calculation, so they barely touch the HPLC purity figure. They do change how much of the powder is peptide. In one reference-material study, the trifluoroacetate counter-ion alone made up nearly 25% of a synthetic peptide’s mass, and the assigned purity, as a mass fraction, was 691 mg/g [5]. Lot-to-lot differences in drying and salt form therefore show up in net content rather than in purity. Our explainer on how freeze-drying works covers the process.

The measurement varies too

A purity figure is the output of one method on one instrument. Change the method and the same material can report a different number.

• Column chemistry. A comparison of standard, polar-embedded and polar-endcapped reversed-phase columns found large selectivity differences for the same peptide mixtures [6].
• Mobile phase, temperature and gradient. Across 165 peptides, the concentration of trifluoroacetic acid mattered most for peptides carrying several positively charged residues (arginine, lysine, histidine), and column temperature, gradient slope and stationary phase each changed selectivity [7]. Selank, KPV and GHK all contain lysine; Selank and MOTS-c contain arginine.
• Detection and integration. Wavelength, where the baseline is drawn and which small peaks are counted all move an area-percent result. Our explainer on HPLC peptide purity shows how the figure is calculated.

For two lots to be compared fairly, the difference between them has to be larger than the method’s own run-to-run scatter, and both have to be measured the same way.

Worked examples from nine lot reports

The Vortex lot report dataset holds nine reports across eight materials (file generated 22 October 2026). The table adds one column the reports do not print: the share of peak area outside the main peak, which is 100 minus the purity figure.

Source: coa_reports.json, dataset generated 2026-10-22 (doi:10.5281/zenodo.23578844). “Outside main peak” computed for this article. Residue counts from the Vortex peptide identity reference. ILS = ILS Laboratories, San Diego, CA.

Example 1: a narrow purity band hides a wide impurity range

The highest and lowest purities differ by 0.79 percentage points. Expressed as what lies outside the main peak, the lowest row (0.03%) and the highest (0.82%) differ by a factor of about 27. When comparing lots, the outside-main-peak share is the more sensitive number.

Example 2: the one material with two lots

GHK-Cu is the only material in the dataset with two lots: 99.85% (50 mg, VR01-GHK) and 99.34% (100 mg, PS07-GHK100). The outside-main-peak share is 0.15% against 0.66%. It is tempting to read that as batch variation. But the lot is not the only thing that differs. The two reports also differ in lab (Freedom Diagnostics against ILS Laboratories), method label (HPLC-UV against HPLC), test date (20 June against 12 August 2026), labelled fill (50 mg against 100 mg) and requester (commissioned in Vortex’s name against a partner-facility report). With five other differences in play, the dataset cannot attribute the 0.51-point gap to the lot alone.

Example 3: one lab, one week, one panel

Seven reports came from the same lab between 11 and 14 August 2026, with the same seven-assay panel. That is the closest the dataset comes to a controlled comparison, and those seven rows still span 99.18% to 99.97%. But they are seven different materials, so the spread reflects chemistry and purification targets for different molecules, not repeat lots of one.

Example 4: not every row is a peptide

NAD+ and 5-Amino-1MQ are small molecules, not peptides. The stepwise-synthesis explanation does not apply to them, which is another reason to compare lots only within one material.

What a real lot-to-lot comparison needs

One lot report cannot show batch variation. A useful comparison needs:

1. Two or more lots of the same material, at the same labelled fill where possible.
2. The same lab and the same named method, or a stated reason to accept that two methods agree.
3. The outside-main-peak share, alongside the headline purity.
4. Identity confirmed on both or all of the lots being compared, preferably by mass spectrometry, so a change in the main peak is not a change of molecule.
5. The specification the lots were released against, so a difference can be read against the target.

Our dataset does not yet contain a pair that meets the first two conditions. When it does, that comparison will be the useful one.

After the test date: the lyophilised solid

A report describes a lot on its test date. Peptides in the solid state still undergo slow chemical change, including deamidation, oxidation, peptide-bond cleavage and aggregation, at rates that depend on temperature and moisture content [8]. A lot measured again months later, after different storage, can report a different purity without any difference in how it was made. Our explainer on storing lyophilised peptides covers the conditions that slow this down. For reading each field of a report, start with the field-by-field explainer on a peptide COA.

Frequently asked questions

Why does peptide purity differ between lots of the same material?

Each synthesis run completes its coupling steps and side reactions to a slightly different extent, each purification pools fractions to a slightly different cut, and each measurement carries its own method scatter. The final figure reflects all three, plus any change in the solid between manufacture and testing.

Is a 0.5-point purity difference between two lots meaningful?

Only if both lots were measured by the same lab with the same method, and the difference is larger than that method’s repeatability. Read as outside-main-peak share, 99.85% against 99.34% is 0.15% against 0.66%, more than a fourfold difference, but it says little when the lab and method also differ.

Can two labs report different purity for the same lot?

Yes. Column chemistry, mobile phase, temperature, gradient, detection wavelength and integration settings all change how impurities separate from the main peak and how they are counted. A purity figure belongs to the method that produced it.

How many lot reports are needed to judge batch consistency?

At least two lots of the same material on the same method at the same lab, and more is better. With one report, there is nothing to compare against. With reports from different labs or methods, method differences and lot differences cannot be separated.

References

1. Behrendt R, White P, Offer J. Advances in Fmoc solid-phase peptide synthesis. J Pept Sci. 2016;22(1):4-27. PMID 26785684. doi:10.1002/psc.2836
2. D’Hondt M, Bracke N, Taevernier L, et al. Related impurities in peptide medicines. J Pharm Biomed Anal. 2014;101:2-30. PMID 25044089. doi:10.1016/j.jpba.2014.06.012
3. Kong MJW, van den Braak TJHP, Neumann K. Aspartimide formation and its prevention in Fmoc chemistry solid phase peptide synthesis. Chembiochem. 2025;26(18):e202500490. PMID 40857621. doi:10.1002/cbic.202500490
4. Åsberg D, Langborg Weinmann A, Leek T, et al. The importance of ion-pairing in peptide purification by reversed-phase liquid chromatography. J Chromatogr A. 2017;1496:80-91. PMID 28363419. doi:10.1016/j.chroma.2017.03.041
5. Melanson JE, Thibeault MP, Stocks BB, Leek DM, McRae G, Meija J. Purity assignment for peptide certified reference materials by combining qNMR and LC-MS/MS amino acid analysis results: application to angiotensin II. Anal Bioanal Chem. 2018;410(26):6719-6731. PMID 30143839. doi:10.1007/s00216-018-1272-7
6. Mant CT, Cepeniene D, Hodges RS. Reversed-phase HPLC of peptides: assessing column and solvent selectivity on standard, polar-embedded and polar endcapped columns. J Sep Sci. 2010;33(19):3005-21. PMID 21038458. doi:10.1002/jssc.201000518
7. Gilar M, Xie H, Jaworski A. Utility of retention prediction model for investigation of peptide separation selectivity in reversed-phase liquid chromatography: impact of concentration of trifluoroacetic acid, column temperature, gradient slope and type of stationary phase. Anal Chem. 2010;82(1):265-75. PMID 19957962. doi:10.1021/ac901931c
8. Lai MC, Topp EM. Solid-state chemical stability of proteins and peptides. J Pharm Sci. 1999;88(5):489-500. PMID 10229638. doi:10.1021/js980374e

Methods note: written by the Vortex Research Editorial Team. The lot figures come from coa_reports.json (generated 2026-10-22); the outside-main-peak share and the stepwise-coupling illustration were computed for this article and are labelled as such. Residue counts come from the Vortex peptide identity reference. Literature was selected from PubMed for synthesis, chromatography and stability topics, and every PMID was checked against its PubMed record for title, journal and year.

Vortex Research products are sold for in-vitro laboratory research by qualified researchers. Not for human or veterinary use. Not a drug, dietary supplement or cosmetic. Statements have not been evaluated by the FDA.