What Is HPLC Peptide Purity Testing?
Every research peptide sold in the United States carries a purity percentage, and almost all of them cluster in the same narrow band: 98%, 99%, 99.4%. The numbers are close enough that they stop discriminating between suppliers, which is precisely the problem. A purity percentage is not a property of the vial. It is the output of one analytical method, run once, on one sample, under conditions the purchaser almost never sees.
This article is about what high-performance liquid chromatography actually measures, what it structurally cannot measure, and what a US laboratory should require on a certificate of analysis before treating a number as comparable across vendors.
The measurement, stated precisely
Reversed-phase HPLC separates a mixture by partitioning it between a non-polar stationary phase and a moving aqueous-organic gradient. Compounds that bind the stationary phase more strongly elute later. A detector — for peptides, almost always a UV absorbance detector — records signal against time, producing the chromatogram reproduced on most batch certificates.
Purity is then computed by area normalization: the area under the main peak divided by the total area of all integrated peaks, expressed as a percentage. That definition carries three consequences that are rarely spelled out on a product page.
It is a ratio of detector response, not of mass. UV detection of peptides at 214 nm works because the peptide bond itself absorbs there, but the individual amino acids do not contribute equally. Kuipers and Gruppen, working in Journal of Agricultural and Food Chemistry (2007, PMID 17539659), measured the molar extinction coefficients of all twenty amino acids and the peptide bond at 214 nm in acetonitrile/formic acid. The peptide bond came in at 923 M⁻¹ cm⁻¹. Tryptophan absorbs roughly thirty times more than a peptide bond; phenylalanine, tyrosine and histidine roughly six times more; proline within a chain about three times more; methionine about the same. Free proline is essentially invisible.
The practical consequence is direct. A synthesis impurity that is missing an aromatic residue absorbs less than the intended peptide per mole, so it is under-represented in the area count. An impurity carrying an extra aromatic residue is over-represented. Area percent and mass percent are related, but they are not the same number, and the gap depends on the specific sequence.
Anything that does not absorb does not exist. This is the failure mode with the cleanest published demonstration. Choules and colleagues, including staff from the United States Pharmacopeial Convention, reported in Journal of Pharmaceutical and Biomedical Analysis (2019, PMID 31671336) that quantitative proton NMR found undeclared mannitol at 20% and 43% w/w in two commercially sourced custom synthetic peptides. Mannitol is highly polar and UV-transparent: it is poorly retained on a reversed-phase column and invisible to the detector even if it did elute. An LC-UV purity assay on those samples would have returned a high number while nearly half the material in the vial was something else. The authors noted that the contaminant was detectable even on a 60 MHz benchtop NMR instrument — this was not an exotic capability problem.
The number moves when the method moves. This is the part most commonly presented as settled and is not. Yeung and colleagues in Analytical Chemistry (2024, PMID 38807522) compared peptide retention across twenty reversed-phase columns with pore sizes from 60 to 300 Å, using both formic acid and trifluoroacetic acid eluents. Wider-pore packing material with a smaller surface area was more retentive — contradicting two assumptions that underpin essentially all peptide method development. Retention peaked near 200 Å with formic acid versus roughly 120–200 Å with TFA, and the effect traced to the physical size of the ion pair formed with each modifier. In the same vein, Gussakovsky and colleagues in Journal of Separation Science (2020, PMID 32818315) showed across roughly 12,000 paired peptides that adding just 0.005% heptafluorobutyric acid to a 0.1% formic acid eluent shifted retention by up to 10% acetonitrile.
Change the column chemistry, the pore size, the modifier, the gradient slope or the detection wavelength and you change which impurities resolve into their own peaks and which hide under the main one. Two vendors reporting 99.2% on different methods have not made the same measurement.
The counterion nobody counts
Peptides purified by reversed-phase HPLC are typically isolated as trifluoroacetate salts, because trifluoroacetic acid is the standard mobile-phase modifier. That counterion is real mass in the vial and it does not appear in a purity percentage.
Cornish and colleagues in American Journal of Physiology (1999, PMID 10567002) found that TFA at 10⁻⁸ to 10⁻⁷ M reduced cell numbers and thymidine incorporation in fetal rat osteoblast cultures, with comparable effects in articular chondrocytes and neonatal mouse calvariae. When they compared TFA and hydrochloride salts of amylin, amylin-(1-8) and calcitonin in osteoblasts, proliferation was consistently lower with the TFA form — producing, in their words, failure to detect a proliferative effect and in one case the wrong attribution of an antiproliferative one. Their conclusion was that this is relevant to any work on purified peptides above roughly 10⁻⁹ M, and that salt exchange should precede biological assessment.
For an in vitro laboratory, that is a design variable sitting entirely outside the number printed on the certificate.
What a certificate should actually show
A number on its own is unfalsifiable. Three fields together are not.
Consider the certificate linked from the batch COA on our BPC-157 10mg listing, report DBAV-BPC-10-051226, tested 23 May 2026 by Testides Analytical. It records expected content 10.00 mg, measured content mass 10.29 mg, purity 99.41%, and fill accuracy 102.9%, with the chromatogram shown at 214 nm.
Those are three different questions with three different answers:
- Purity (99.41%) — of the material the detector saw, what fraction was the main peak?
- Content mass (10.29 mg) — how much peptide was actually in the vial?
- Fill accuracy (102.9%) — how does that compare with what the label claims?
A certificate carrying only the first is answering the least useful of the three. A vial can be 99% pure and hold 6 mg. It can hold 10 mg of something 92% pure. Purity without content mass cannot distinguish those cases, and content mass is the number that governs how much material a given volume of reconstitution buffer actually contains.
Also worth reading: the detection wavelength, because it determines what was visible at all; the report and lot identifier, because a certificate that cannot be tied to the lot in the box is a marketing asset rather than a control; and the testing laboratory, because in-house and third-party results carry different weight.
The relevant public standards are worth naming even though they are written for pharmaceutical manufacturing rather than research supply: USP General Chapter <621> covers chromatographic system suitability, ICH Q2(R2) covers analytical procedure validation, and ICH Q3A and Q6A cover impurity reporting and specification-setting. A supplier who can say which of these its methods follow is telling you something. One who says “lab tested” is not.
Why this matters more in the US market right now
The domestic research-supply landscape has consolidated. Laboratories that had a settled vendor have been forced to re-source, and re-sourcing means comparing certificates from suppliers whose analytical methods are neither disclosed nor equivalent. A 99.1% from one vendor and a 99.6% from another are not ranked by those digits — they are two different experiments, and the difference between them is smaller than the difference a column change would produce.
Two practical consequences for a US laboratory. First, continuity of method matters more than the last decimal place: a vendor that reports the same fields, from the same laboratory, in the same format, lot after lot, gives you a series you can actually track. Second, holding domestic stock is an analytical argument as well as a logistics one. Maple Research Labs ships same-day from Santa Barbara, California, which keeps material out of prolonged ambient transit — and thermal and aqueous exposure during shipping degrade peptides in ways that no certificate issued weeks earlier can capture.
Maple Research Labs is a Wyoming entity shipping domestically to US laboratories, and every listed compound carries a batch-specific certificate linked from its own product page. We publish the report number so it can be checked against the vial, and we publish content mass and fill accuracy alongside purity because two of those three fields are the ones that change what an experiment actually receives.
The short version
HPLC purity is a real measurement and a useful one. It is also a chromatogram statistic: a ratio of UV detector response, under one set of conditions, blind to anything that does not absorb, and sensitive to method choices the buyer never sees. Read it next to content mass, fill accuracy, wavelength, lot identifier and testing laboratory, and it becomes informative. Read it alone, and it is a number that every supplier in the category can print.
References
- Kuipers BJH, Gruppen H. Molar extinction coefficients of amino acids and the peptide bond at 214 nm for quantitative RP-HPLC-MS. J Agric Food Chem. 2007;55(14):5445-51. PMID 17539659.
- Choules MP, Bisson J, Simmler C, et al. NMR detection of an undeclared constituent in commercially sourced custom synthetic peptides. J Pharm Biomed Anal. 2019;178:112915. PMID 31671336.
- Yeung D, Spicer V, Zahedi RP, Krokhin OV. Ion-pairing-dependent size exclusion effects in reversed-phase peptide separations across twenty columns. Anal Chem. 2024;96(23):9721-9728. PMID 38807522.
- Gussakovsky D, Anderson G, Spicer V, Krokhin OV. Peptide separation selectivity with formic and mixed formic/heptafluorobutyric acid ion-pairing modifiers. J Sep Sci. 2020;43(20):3830-3839. PMID 32818315.
- Cornish J, Callon KE, Lin CQ, et al. Trifluoroacetate as a contaminant in purified proteins and its antiproliferative effect on osteoblasts and chondrocytes. Am J Physiol. 1999;277(5):E779-83. PMID 10567002.
- Aguilar M-I. Reversed-phase high-performance liquid chromatography. Methods Mol Biol. 2004;251:9-22. PMID 14704435.
- United States Pharmacopeia, General Chapter <621> Chromatography.
- ICH Q2(R2), Validation of Analytical Procedures.
- ICH Q3A(R2) and Q6A, impurity reporting and specifications.
Batch data cited from Certificate of Analysis DBAV-BPC-10-051226 (Testides Analytical, tested 23 May 2026), linked from the BPC-157 10mg product listing.
All compounds supplied by Maple Research Labs are for research use only. They are not for human use or consumption, are not drugs, foods or cosmetics, and are not intended to diagnose or affect any condition in people or animals. All work described here refers to in vitro systems and published preclinical animal studies.