What a 99% purity figure does not tell you
A purity figure on a certificate of analysis is almost always a chromatographic area percentage: the area under the main peak divided by the total area of all peaks the detector recorded, in a single run, at a single detection wavelength, under a single set of chromatographic conditions. It is a statement about how much of what the detector saw was one thing. It is not a statement about how much peptide is in the vial, not a statement about which molecule that peak is, and not a statement about the material's condition at any date other than the test date. Those are four separate measurements, and a purity figure supplies exactly one of them. Understanding which one is the difference between reading a certificate and merely looking at it.
What does an HPLC area percentage actually measure?
Reversed-phase HPLC separates a mixture by how strongly each component partitions onto a hydrophobic stationary phase. Components emerge at different retention times, a detector — usually ultraviolet absorbance — records a signal, and software integrates the area under each peak. Area percent is a ratio internal to that chromatogram.
The ratio is therefore relative. It compares the sample to itself. Nothing in the calculation references an external standard of known composition, and nothing in it establishes the absolute quantity of anything. A sample reported at 99% area purity and a sample reported at 99% area purity by a different laboratory, on a different column, at a different wavelength, are not necessarily making the same claim.
Detector response is the first reason. UV absorbance depends on what a molecule absorbs at the chosen wavelength, and different species absorb differently. Two compounds present in identical amounts do not produce identical peak areas. Area percent consequently approximates mass percent only when the impurities happen to have a response similar to the main component — which, for closely related synthesis by-products, is often roughly true, and for structurally unrelated contaminants can be badly false.
Why is a peptide's mass not the same as its peptide content?
This is the gap that most often surprises people reading a certificate for the first time. Material purified by preparative reversed-phase HPLC is typically eluted with trifluoroacetic acid, and the peptide is recovered as a TFA salt. The counterion is part of the mass in the vial and is not the peptide. Water is also part of that mass. Neither appears in a chromatographic purity figure, because neither is being asked about.
Erckes and colleagues examined precisely this problem and argued for a consensus approach to reporting and exchanging TFA counterion content in synthetic peptides, validating detection by FT-IR, ¹⁹F-NMR and HPLC with evaporative light-scattering detection, and finding that counterion identity was associated with measurable differences in permeability coefficients depending on sequence and salt form. Their conclusion is the operative one here: counterion content needs to be specified for results to be reproducible between laboratories (Pharmaceuticals (Basel), 2025; PMID 40872554, DOI 10.3390/ph18081163).
The quantity that answers "how much peptide" is peptide content, and it is determined by a different family of methods — amino acid analysis after complete hydrolysis, quantitative NMR, or a mass-balance assay. These do not always agree with one another. A USP-led inter-laboratory study using oxytocin as a case compound surveyed the available quantification approaches and compared their reproducibility, reporting that an HPLC assay run against the same peptide bulk material as the standard showed the lowest inter-laboratory variability, and recommending further exploration of qNMR as a primary method for reference-standard value assignment (J Pharm Biomed Anal, 2019;166:105–112; PMID 30640042, DOI 10.1016/j.jpba.2018.12.028). When trained reference-standard laboratories produce a spread on a well-characterised nonapeptide, a single area percentage from a single run should be read with corresponding modesty.
What can a purity figure hide?
Two things, mainly: species the detector cannot see, and species it sees but cannot separate.
Anything lacking a chromophore at the detection wavelength contributes little or no signal. Residual salts, many solvents, and water fall into this category. They are present in the vial and absent from the area percentage.
Co-elution is the subtler failure. Impurities structurally closest to the target — truncated sequences, deletion sequences, deamidated forms, oxidised forms — are also the ones most likely to share its retention behaviour. A single chromatographic condition can merge them into the main peak. This is why analytical guidance favours orthogonal conditions, and why mass spectrometry is not an optional extra: HPLC establishes that one species dominates, while mass spectrometry establishes what that species is.
The scale of the effect is quantifiable. Stoppacher and colleagues identified and determined impurities in the peptide hormone angiotensin I by LC–high-resolution tandem MS, finding that the dominant impurities were the related fragments angiotensin II and angiotensin III generated by degradation during storage at elevated temperature. They estimated the mass fraction of the major impurities in a candidate reference material at 10.4 mg/g, and noted that failing to correct for them would produce a 1% error in the peptide concentration determined by amino acid analysis (Anal Bioanal Chem, 2013;405(25):8039–51; PMID 23708692, DOI 10.1007/s00216-013-6953-7). The impurities in that case were not exotic contaminants. They were pieces of the target molecule.
Does a purity figure describe the material today?
No. It describes one aliquot on the test date.
Peptides in the solid state continue to degrade, by routes that are well catalogued. Lai and Topp reviewed solid-state chemical stability of proteins and peptides and identified deamidation, peptide bond cleavage, oxidation, the Maillard reaction, beta-elimination and dimerisation/aggregation as the major pathways, with rates governed by temperature, moisture content, excipients and whether the solid is amorphous or crystalline (J Pharm Sci, 1999;88(5):489–500; PMID 10229638). A certificate is a dated snapshot, and the interval between that date and the present is information the certificate does not contain.
This is why the test date is not a formality. It is a load-bearing field.
What makes a purity figure checkable?
Three fields, and the figure is close to meaningless without them: the batch or lot number, the test date, and the named laboratory that signed the report. A number without a batch cannot be tied to any particular material. A number without a date cannot be aged. A number without an attributed laboratory cannot be verified by anyone.
Our own published certificates are structured on that basis, and they are the strongest evidence we can offer because they are measurements rather than claims. Five are currently public: GHK-Cu batch 2026-03 and Retatrutide batch 2026-03, both tested 11 March 2026; MOTS-c batch 2026-03 and NAD+ batch 2026-03, tested 20 March 2026; and BPC-157 batch 2026-03, tested 22 January 2026. Each names Janoshik Analytical as the issuing laboratory, and certificates issued by that laboratory carry a verification key that can be checked directly with the lab rather than taken on our word.
Note what those pages report and what they do not. The Retatrutide certificate carries a specific measured figure of 99.58%. The others are reported as ≥99%, which is a threshold statement rather than a point measurement — a meaningful distinction, and one we would rather state than blur. None of the five constitutes a peptide-content determination, and none should be read as one.
Veridian Research supplies these materials strictly for in-vitro laboratory research. They are not drugs and are not approved for human or veterinary use.
The short version
A 99% purity figure establishes that, under one set of chromatographic conditions on one date, one species accounted for approximately 99% of the detector response. That is a real and useful thing to know. It is not a mass, not an identity, not a shelf life, and not a substitute for the batch number, test date and laboratory attribution that make it possible to check at all.