A 2015 analysis in PLOS Biology attributed 36 percent of an estimated 28 billion dollars in annual irreproducible preclinical research spending in the United States to reagents and reference materials. The share has been argued over since, but the ordering has not: of the things that make a preclinical result fail to replicate, the material in the tube outranks the statistics applied to it. A cross-sectional international survey of biomedical researchers found 72 percent agreeing their field faces a reproducibility crisis, with 27 percent describing it as significant.
Most of the resulting reform effort has gone into single-entity reagents, where the question is relatively clean. One molecule, one identity check, one purity figure. The research materials catalogue has meanwhile moved in the opposite direction, toward multi-component preparations supplied as a single item. That shift has an analytical consequence the documentation layer has been slow to absorb, because a blend is not one measurement with a higher price. It is several measurements plus a relationship between them.
A blend is n identities plus a ratio
For a single peptide, a certificate of analysis answers two questions. Is this the sequence claimed, usually settled by mass spectrometry, and how much of the material is that sequence rather than something else, usually settled by HPLC or comparable chromatography. Current practice treats a stated purity threshold, confirmed sequence identity, quantified residual trifluoroacetate and a stated bacterial endotoxin limit as the minimum record.
A four-component preparation does not multiply that record by four. It adds a dimension. Each component needs its own identity confirmation and its own purity figure, and the preparation as a whole needs a composition figure: how much of each component is present relative to the others. A blend can be built entirely from components that individually clear their purity threshold and still be the wrong preparation, because purity and proportion are independent properties. Nothing in a per-component purity figure constrains the ratio.
That matters more than it sounds, because the ratio is the part most likely to drift. Purity is set by the synthesis and purification of each component and is reasonably stable for a given route. Proportion is set at the blending and filling step, which is a mass-transfer operation on small quantities of hygroscopic solids, and small-quantity transfers are where loss concentrates.
Three places the record usually thins out
Co-elution hides components rather than flagging them
Chromatographic separation of a mixture is not guaranteed by the fact that the components are different molecules. Peptides of similar length and hydrophobicity can share retention behaviour closely enough that a single peak carries more than one species. Read without a complementary identity method, that peak reports as high purity precisely because two things have been counted as one. Mass spectrometry resolves it, which is why identity confirmation on a blend has to be component-resolved rather than performed on the preparation as a whole.
Counterion load scales with the component count
Residual trifluoroacetate, bound water and counterions travel with every synthetic peptide out of purification, and the amount is a function of the route rather than a constant. In a blend, each component contributes its own non-peptide load, and the preparation inherits the sum. A nominal mass on the vial therefore overstates total peptide content by a margin that is itself a composite, which makes it harder to back-calculate than it would be for a single entity. The number is knowable, but only by measurement.
Stability is not a shared property
Components of a blend do not degrade on the same schedule. Aggregation behaviour, solubility and sensitivity to freeze-thaw cycling differ between sequences, so a preparation that was correctly proportioned at release can shift proportion in storage without any individual component falling below its purity threshold. A certificate describes material as it left the testing laboratory. It does not describe the material on the day of use, which is why a handling log recording lot numbers, storage conditions and reconstitution history is the other half of the same record rather than an optional discipline.
The supply layer is where the gap is widest
A 2026 preprint evaluation of research-grade peptides found that, depending on the criterion applied, between 41.6 and 71.1 percent of grey-market peptides failed fundamental purity requirements. The same analysis found no relationship between purity and endotoxin burden, which rules out using either measurement as a proxy for the other. For a multi-component preparation both findings compound: more components means more independent opportunities for one of them to be the failure, and a single aggregate figure on a document cannot locate which.
Independent testing is the only mechanism that removes production bias from the answer. Per-lot submission to laboratories accredited to ISO/IEC 17025:2017 has accordingly become the reference expectation for chromatographic and mass spectrometric characterisation rather than a premium service.
How this works in practice
A section of the research materials market has started treating the analytical record as part of the product rather than a document issued on request. Bluum Peptides publishes certificates of analysis for every batch, produced by independent third-party laboratories, naming Janoshik Analytical, BioRegen and Freedom Diagnostics as its testing partners. Each certificate carries the lot number, the testing date and the methodology used, with identity confirmed by mass spectrometry or an equivalent peptide-confirmation method and purity assessed by HPLC or comparable chromatography. The catalogue states a minimum purity threshold across products and publishes the measured figure for the current lot of each item, including its multi-component preparations. Material is supplied for research use only, not for human or animal consumption, and is not intended to diagnose, treat, cure or prevent any disease.
The load-bearing detail is the batch granularity. A per-batch document with a date and a method is a record a methods section can cite. A standing claim about the catalogue is not, however accurate it happens to be.
Where blend documentation still runs behind
Even at the better end of the market, the convention is to report purity per component and composition as a nominal formulation. The measurement that would close the loop is a released composition figure: the proportion actually present in the lot, determined rather than specified. That is a harder analysis than a purity check and it is not yet a norm anywhere in the research materials supply chain.
The reporting side has a matching gap. Published work using multi-component preparations tends to name the preparation and omit the lot, which makes the composition unrecoverable after the fact. A reader cannot reconstruct what was in the tube, and neither can the original group once the vial is gone.
Conclusion
Single-entity reagents taught the field a specific lesson, that a label is a claim and a certificate is a measurement, and the two are not interchangeable. Multi-component preparations inherit that lesson and add one of their own. Identity and purity are properties of each component, while composition is a property of the lot, and only the first two are routinely documented. Until the third is measured and reported with the same discipline, a blend will remain the least traceable item in an otherwise well-characterised cupboard.
