Peptide purity, and what the percentage measures

How peptide purity is calculated

A sample is dissolved and pushed through a reversed-phase high-performance liquid chromatography column, RP-HPLC, which separates its components by how strongly each one sticks to the column packing. A detector at the far end registers each component as it comes off, drawing a peak. The area under each peak is proportional to how much of that component passed the detector.

Peptide purity is then the area of the target peak divided by the total area of all peaks, expressed as a percentage. Ninety-nine per cent purity means the main peak accounted for 99% of everything the detector saw.

The detector is almost always a UV absorbance cell set at 214 nanometres, where the peptide bond itself absorbs; 220 nm is the other common setting. Both read the backbone rather than any one side chain, which is what makes them general enough to use across sequences without recalibrating for each. Every Freedom Diagnostics certificate in this library states RP-HPLC at 214 nm, and the second laboratory represented here, pH Solutions, runs its peptide purity assay at 274 nm. It is the same measurement made a different way, and the certificate says which.

Two consequences follow immediately. First, the figure is relative. It compares the target against the other things that were detected, not against the contents of the vial. Second, it depends on the method: the column, the gradient, the wavelength. A well-specified peptide certificate of analysis states those conditions, which is why a chromatogram is worth more than a number typed onto a page.

What a peptide purity percentage leaves out

Most peptide detection uses ultraviolet absorbance at around 214 nanometres, where the peptide bond absorbs. Anything that does not absorb there is not on the chart. Water is invisible. Inorganic salts are invisible. Residual solvent is largely invisible.

Counter-ions are the significant one. Peptides are usually purified by HPLC using trifluoroacetic acid, and they come off that process as a salt: the peptide paired with trifluoroacetate. That counter-ion can account for a meaningful share of the vial's mass, and it does not appear in an area-percent peptide purity figure at all.

How much it accounts for depends on the sequence. A peptide carrying several arginine or lysine residues has more positive charge to balance, so it binds more counter-ion, and the same chemistry makes the dry solid more hygroscopic, quicker to pull water out of the air once the vial is open. Two sequences purified on the same day to the same area percent can therefore differ by several per cent in how much of the powder is actually peptide. Counter-ion can be exchanged for acetate or hydrochloride by a further ion-exchange step, but the exchange is sequence-dependent, is never quite complete, and leaves residual trifluoroacetate that has to be quantified separately if the number is going to mean anything.

So a vial labelled 10mg at 99% peptide purity does not contain 9.9mg of peptide. It contains a fill of lyophilized powder that is 99% pure by chromatography, part of whose mass is water and counter-ion rather than peptide. That is why the certificates here report content as a measured mass against the label strength rather than as a share of the dry solid: across the 42 lots carrying that figure it runs from 93.6% to 131.6% of label, because the fill is set above target so that the labelled mass survives the water and the counter-ion. Nothing dishonest is happening; the two numbers simply answer different questions. The same split between a ratio and a mass is set out from first principles in what is a peptide.

Peptide purity grades, and what each grade is specified for

Suppliers of synthesised peptides sell against specification bands rather than a single standard, and the bands are older than the research-chemical market. They come from the custom-synthesis trade, where the buyer states the measurement they intend to make and the vendor quotes the cheapest grade that will support it. Crude material is the floor: cleaved from the resin, desalted, not chromatographically purified, and sold as such.

The logic of the ladder is worth stating plainly, because it is the opposite of how a retail listing reads. A higher grade is not a better product in the abstract; it is a narrower impurity budget. Antibody production tolerates a wide one, because the target of that work is a sequence and a few per cent of truncated chains does not change which sequence it is. Quantitative work tolerates almost none, because every impurity is mass that was weighed and counted as target.

Two things follow. The first is that a grade describes analytical suitability and nothing else: it is a statement about what a measurement can survive, not a permission or a recommendation for any other use. The second is that once a supplier is quoting 99% and above, the grade has stopped discriminating: the whole catalogue sits in the same band, and the only thing that separates one vial from another is whether the peptide purity figure was measured on that lot and published with the document behind it.

Where the impurities come from

Almost every peptide in this category is made by solid-phase peptide synthesis. A single amino acid is anchored to a resin bead, and the chain is built one residue at a time: deprotect the growing end, couple the next residue, wash, repeat. A fifteen-residue sequence is fourteen coupling cycles, each with its own deprotection and its own wash, and each one an opportunity for a fraction of the chains on the bead to fall out of step.

That is the mechanism behind the impurity profile. A coupling that goes to 99.5% completion leaves half a per cent of the chains one residue short, and because those chains stay on the resin they keep growing, producing a full-length molecule with one amino acid missing from the middle. That is a deletion sequence, and it is the hardest impurity to remove precisely because it is so nearly the target. Capping, which acetylates any chain that failed to couple so it cannot grow further, converts what would have been deletions into truncations: shorter, more polar, and far easier to separate.

The final cleavage step brings its own. Side chains are carried through synthesis under protecting groups, and stripping them all off cleanly at the end is a chemistry problem in its own right. A group that survives leaves a molecule heavier than the target and more hydrophobic, while an over-aggressive cleavage cocktail can modify residues it was meant to leave alone. None of this is unusual or a sign of a bad synthesis. It is why purification exists, and why the interesting question on a certificate is not how much impurity there is but which impurity it is.

Degradation is a different class of impurity

Everything above happens during manufacture. A second family appears afterwards, and it is the one a peptide purity figure is least able to describe, because the figure was recorded before any of it had time to occur. A certificate is a measurement of a lot on a date, not a property the material carries permanently.

Water drives most of it. Hydrolysis cleaves the backbone, which is why material is lyophilised in the first place and why a dry cake is stable in a way a solution is not. Oxygen and light do the rest, adding sixteen daltons at a methionine or tryptophan. Deamidation at asparagine or glutamine shifts a mass by one dalton and is easy for a routine method to miss entirely. Aggregation is driven by concentration and by temperature cycling, which is the argument for not thawing and refreezing a vial repeatedly.

This is where purity and storage stop being separate subjects. A 99.9% figure on a lot that has since spent a fortnight at room temperature in a humid room is a historical document. The storage conditions for each state, set out in how to store peptides, are what hold material close to the figure on its certificate. Reconstitution is the moment the clock speeds up, which is why the choice of diluent, bacteriostatic water for peptides or plain sterile water, belongs in the same conversation as the peptide purity percentage.

95, 98 and 99 per cent peptide purity: what actually changes

The difference between 95% and 99% is not four per cent of the material; it is a fivefold difference in impurity. A 95% preparation carries five parts impurity per hundred, a 99% preparation carries one. For work where an impurity can confound a measurement, that ratio is the meaningful comparison.

What those impurities are matters more than how many. A preparation at 98% whose two per cent is a well-separated truncation is a more predictable reagent than one at 99% whose one per cent is a deletion sequence sitting under the shoulder of the main peak, because the second one is both harder to see and closer in behaviour to the thing being measured. This is the argument for the chromatogram over the number: the number cannot distinguish those two cases and the trace does at a glance.

Above about 99% the returns flatten and the cost does not. Each additional decimal costs a further purification pass, a further loss of yield, and a further analytical run to demonstrate it. The practical question stops being how high the number is and becomes whether it was measured, by whom, and on this lot, which is where vetting a research peptide supplier begins.

Why two laboratories report different peptide purity for the same lot

Peptide purity as area percent is not an absolute property of a sample. It is the output of a method, and changing the method changes the number without anything changing in the vial. The same lot can read 98.4% on one system and 99.3% on another, and both results can be correct.

The two laboratories whose certificates are archived here make the point without needing a hypothetical. Freedom Diagnostics reports RP-HPLC at 214 nm and states the result to three decimal places. pH Solutions runs its purity assay at 274 nm and reports it as a pass against a specification of not less than 99%, alongside fill-weight conformance counted vial by vial, 337 of 337 on one lot and 94 of 94 on another. One document gives a measurement, the other gives a conformance statement. Neither is the weaker document; they answer different questions, and reading them as though they were the same scale is the mistake.

The variable that moves a peptide purity number most is the one least often printed. An impurity with no strong chromophore at the chosen wavelength contributes almost no area, so it is not so much under-counted as absent. Integration is the second: where the analyst drew the baseline and the peak boundaries moves a result by tenths of a per cent on the same raw data. This is why 'NLT 99%' and '99.786%' are not comparable statements. The first is a specification the lot had to meet; the second is what was found.

What each analytical method can and cannot bound

No single instrument answers the whole question, and a certificate that runs to one page is usually a certificate that ran one assay. Knowing what each method is blind to is most of what it takes to read one critically.

One thing is true of all 51 lots published here: an RP-HPLC purity result. Identity is confirmed by LC-MS on 48 of the 51, and on the other three by retention-time match against a certified reference standard. Everything else is true of some lots and not others. A measured content figure appears on 42 of the 51, bacterial endotoxin on 17 of the 51, and residual solvents and elemental impurities only on the two pH Solutions certificates. That is why those results sit on the lot record rather than in a catalogue-wide claim. A site that prints an assay panel as universal, when the document linked from the same page runs half of it, has published something a reader can falsify in about ten seconds.

The methods below are the ones worth asking about. Several will not appear on a research-peptide certificate at any supplier, which is a fact about the market rather than a criticism of one seller. It does mean, though, that a supplier claiming them should be able to show the page of the report where they were run.

Purity without identity proves nothing

A high peptide purity figure says a sample was clean. It does not say what it was clean of. Both questions have to be answered, and they are answered by different instruments.

Mass spectrometry supplies the second. It measures the molecular mass of the main component and compares it against the theoretical mass calculated from the sequence. Agreement within a small tolerance demonstrates that the substance present is the substance named.

The tolerance matters as much as the match, and it is the part a certificate usually leaves unstated. A modern LC-MS instrument measures mass to within a few parts per million. A report that claims agreement only to within a tenth of a per cent is describing the instrument rather than the sample: for BPC-157, whose molecular weight is 1419.53 daltons, 0.1% is 1.4 daltons either way, a window wide enough to swallow a deamidation, which moves the mass by one. Where a tolerance is quoted in daltons rather than per cent, two daltons is the figure commonly used for a routine confirmation.

A certificate showing a 99% HPLC figure and no mass spectrum has told you half the story, and it is the less important half. Every lot in this catalogue is assayed for both by an independent laboratory, and the report is published against the lot number printed on the vial.

Reading the chromatogram itself

A certificate that attaches its chromatogram is offering the evidence rather than the summary, and the trace can be read without any training in chromatography. Four things are worth looking at, and all four are visible in a few seconds.

What a chromatogram cannot show is a component that never absorbed. Counter-ion, water and inorganic salt leave no mark on a UV trace at 214 nm, so a flat, clean baseline is not by itself evidence that the vial contains nothing but peptide. It is evidence that nothing else absorbing at that wavelength came off the column during the run, which is a narrower and more useful claim.

One lot's peptide purity is not the next lot's

Every peptide purity figure describes one batch. Synthesis is a chemical process with normal variation in it, so the same sequence made twice by the same manufacturer on the same route will not give the same number, and the certificate for a lot bought in March says nothing binding about a lot shipped in September. This is the reason a lot number exists at all.

The spread across the 49 lots here that were reported to three decimal places runs from 99.247% to 99.984%, with a median of 99.786%. Every one of those figures was produced by an outside laboratory on the specific batch, and every certificate is published against its lot number whether or not that batch is still in stock, including the older and lower results, because a library that quietly drops its weakest documents is not a library.

What that makes possible is the only check worth doing. Read the lot number off the vial, find that lot in the library, and confirm that the document names the same compound, the same strength and a date that precedes the release. A supplier that publishes one certificate as a specimen and prints a different lot number on the vials has published a sample, not a record. Looking a lot number up directly takes about as long as reading this paragraph.

Measured peptide purity across this catalogue

Almost everything written about this subject, including most of this article, discusses peptide purity in the abstract. The figures below are not abstract: each one is the HPLC result on a named lot, produced by an outside laboratory, with the signed report published at the lot number.

Alongside purity, these certificates report a measured content figure against the label strength, and it is worth explaining, because it does not read the way people expect. Across the 42 single-compound lots that carry one, measured content runs from 93.6% to 131.6% of the label strength, with a median of 107.9%. Vials are filled above target so that the labelled mass is present after the dry solid's water and counter-ion are accounted for; the certificate is what turns that from an assumption into a number. Nine of the 42 read below the label figure, and those certificates are published too.

The repair panel, the growth panel and the metabolic reference set carry no certificate of their own. They are dispatched as their component vials, each with the certificate for its own lot, which is the only arrangement that keeps a lot number meaningful. Bacteriostatic water is the one line in the catalogue with no peptide purity figure at all: there is no analyte to assay, so the water is specified as USP sterile rather than reported as a percentage.

Who measured it, and what accreditation adds

A peptide purity figure has an author. In-house quality control run by the manufacturer is a real measurement and is not worthless, but it is a party reporting on its own work, and there is no external step between the result and the number on the listing. Independent testing puts a laboratory with no commercial interest in the outcome between the two.

ISO/IEC 17025 is the standard that describes what a testing laboratory has to demonstrate: competence for the specific methods it is accredited for, traceable calibration, documented procedures, and participation in proficiency testing against other laboratories. It is narrower than it sounds. Accreditation is granted for a defined scope, so a laboratory can be accredited and still run an unaccredited method, which is why the scope matters more than the logo. Of the two laboratories behind the certificates here, pH Solutions reports under ISO/IEC 17025 accreditation.

Buyers can commission their own testing, and some do. The costs commonly quoted by contract analytical laboratories put a single RP-HPLC purity run in the region of $50 to $150 a sample, with a comprehensive panel covering identity by LC-MS, amino acid analysis, water and counter-ion content running from several hundred dollars upward; peptide purity testing and what each method costs are set out separately. Independent re-testing of a vial you already hold is the strongest form of verification there is, and it is also the reason a supplier should be comfortable telling you which laboratory produced its own figures and on which lot.