Why a Single Purity Number on a Peptide Blend Proves Nothing

3 Nov 2026

A single purity percentage on a multi-peptide vial cannot tell you whether every peptide on the label is in the vial. Chromatographic purity is a ratio of peak areas within one injection, so it measures how clean the material is, never how much of each component is present or in what proportion. A vial containing 70 mg of one cheap peptide and none of the two expensive ones can return a higher purity figure than an honestly filled blend. The only certificate that closes that gap reports identity, purity and mass separately for each named component. This page works through one — GLOW Blend lot glow70-0626-01, VialTests report VT-2026-001827-S04 — line by line, with the arithmetic shown so you can check every figure, and sets out what it still does not prove.

The short version

Purity and content are different measurements. Purity is area of the intended peak ÷ total area in one chromatogram. Content is milligrams, read against a calibration curve built from a reference standard. One is a ratio; the other is a quantity.
A blend needs content per component or it needs nothing. Three peptides at three different masses in three different amounts cannot be summarised by one ratio.
Substitution is invisible to a combined purity figure. If a missing peptide is replaced by more of a peptide that is also on the label, every peak in the chromatogram is still an intended peak, so nothing counts as an impurity.
Area percent is not mass percent. On our own CJC-1295 + Ipamorelin lot the chromatogram reads 85:15 and the masses read 53:47. Same vial, same report.
What a good blend certificate contains: for each named peptide — an identity confirmation, a chromatographic purity, and a measured mass against the labelled mass. Plus a lot number that matches the vial and a report number the laboratory will verify.

Why one number cannot describe three peptides

Chromatographic purity is defined for a sample that is supposed to contain one compound. You inject the sample, the detector traces a chromatogram, and the purity figure is the area under the target peak divided by the total area of everything the detector saw:

purity % = (area of the intended peak ÷ total integrated area) × 100

That definition carries an assumption so basic it usually goes unstated: everything that is not the target is an impurity. It is what makes the number meaningful. 99.6% says that for every 1000 units of detector response, 996 came from the molecule you wanted and 4 came from something you did not.

Put three intended peptides in the vial and the assumption collapses. There is no single "intended peak" any more, so there are only two things the laboratory can do with the definition, and both of them destroy the information you actually wanted:

Sum the three intended peaks and report the total as a combined purity. The figure now says "96% of the detector response came from peptides that belong here" — which is a statement about contamination, and says nothing at all about the proportions.
Report one component's purity and let the reader assume it covers the vial. This is the number you most often see on a blend listing with no breakdown underneath it.

Neither answers the question a buyer is actually asking about a blend, which is not "is this clean?" but "is all three of them in there, at the amounts on the label?" That is a question about mass, and purity is not a mass.

The distinction in one sentence

A purity figure describes the quality of what is in the vial. A content figure describes how much of it there is. A blend sold on a purity figure alone has been assayed for the first question and not for the second, and the second is the one that distinguishes it from a vial of the cheapest ingredient.

The substitution a combined figure cannot see

This is the failure mode worth understanding in detail, because it is the one a combined purity figure is structurally incapable of detecting — not unlikely to detect, incapable.

Take the GLOW label: GHK-Cu 50 mg + BPC-157 10 mg + TB-500 10 mg, 70 mg total. The three components are not remotely equal in cost. Using our own single-peptide listings as the reference, because they are public and you can check them:

Component	Our single-peptide price	Per mg	In 70 mg of blend
GHK-Cu 50mg	$49.00	$0.98	50 mg → $49.00
BPC-157 10mg	$59.99	$6.00	10 mg → $59.99
TB-500 10mg	$64.99	$6.50	10 mg → $64.99
Three vials bought separately	$173.98	$2.49	70 mg

Prices as listed on this site. The blend sells for $119.99, which is $1.71/mg — 31.0% below buying the three components separately. That gap is the whole commercial reason a blend exists, and it is also the reason the substitution below is tempting.

GHK-Cu is 6.1 times cheaper per milligram than BPC-157 and 6.6 times cheaper than TB-500. So consider a vial filled with 70 mg of GHK-Cu and nothing else, sold as GLOW. What does a combined purity assay return?

Honest vial: three clean peptides, weakest component 99.61% pure → combined chromatographic purity ≈ 99.6-99.8% Substituted: 70 mg of GHK-Cu alone our own GHK-Cu lot GHK50-0626-02 assayed at 99.93% → combined chromatographic purity ≈ 99.93% Material cost at the retail prices above: honest $173.98 substituted 70 × $0.98 = $68.60 → 39.4% of the honest cost

The fraudulent vial reports a better purity number than the real one, at 39% of the material cost. Not a similar number — a better one, because GHK-Cu is a 403.9 g/mol tripeptide that is easier to make cleanly than a 43-residue protein. There is no combined-purity threshold that catches this, because the cheat does not introduce an impurity. Every peak in that chromatogram belongs to a peptide printed on the label.

Three other substitutions are equally invisible to the same number: swapping full-length thymosin β4 for the much cheaper Ac-LKKTETQ heptapeptide (both get called "TB-500" on the research market — see our TB-500 guide for why that matters), under-filling one component while over-filling another, and omitting a component entirely while making up the mass with mannitol or another excipient that the detector does not see at 214 nm.

Only one measurement catches all four: a mass, per component, read against a reference standard for that specific molecule. That is what a per-component certificate is for.

Area percent is not composition — measured, on our own lot

There is a tempting shortcut: if the chromatogram separates the three peptides, can you not just read the composition off the relative peak areas? No — and we can demonstrate it from two published figures on the same report rather than from theory.

Our CJC-1295 (no DAC) + Ipamorelin lot cjc10-0626-02 was assayed by Testides (Toronto) by HPLC-UV at 214 nm. That report gives both the named-component peak areas and the measured masses:

Component	Peak area %	Share of intended area	Measured mass	Share by mass
Ipamorelin (711.9 g/mol)	81.08%	84.6%	5.34 mg	53.0%
CJC-1295 (no DAC) (3367.9 g/mol)	14.72%	15.4%	4.74 mg	47.0%
Total	95.80%	100%	10.08 mg (100.8% of label)	100%

By chromatogram that vial looks like 85% ipamorelin. By mass it is 53% ipamorelin. A reader who took the composition off the peak areas would conclude the expensive 29-residue component made up a seventh of the material, when it makes up nearly half of it — and would have judged the vial badly under-filled on CJC-1295 when it is, in fact, filled to 94.8% of its 5 mg label.

The reason is that a UV detector at 214 nm responds to the peptide bond and to aromatic residues, not to mass. Two peptides of different length and different residue composition produce different detector response per milligram, and nothing on the chromatogram tells you the conversion factor. Only a calibration curve built from a reference standard of that particular peptide does, and building one is the extra work that separates a content determination from a purity figure.

The same effect is why the combined 95.80% figure on that lot is not a quality failure. It is a clean blend. The 4.2% of area that is not ipamorelin or CJC-1295 is a small share of a chromatogram dominated by a strongly absorbing short peptide, and the mass figures confirm the vial is filled to label.

Reading a per-component certificate, field by field

A certificate worth the name has nine things on it. If a vendor's blend COA is missing items 3, 4 or 5, it is a purity report wearing a blend's name.

#	Field	What it must say, and why
1	Lot number	Must match the code on the vial you received, not a code "representative of" the batch. Without this the document describes somebody else's vial.
2	Report number, and a way to check it	A number you can enter into the laboratory's own verification form, so the laboratory confirms the document rather than the vendor. A PDF a vendor hosts proves nothing about who wrote it.
3	Identity, per component	Usually mass spectrometry. The measured mass of each species matched to the expected mass of that specific peptide. This is what distinguishes 4963 Da thymosin β4 from the 889 Da fragment.
4	Purity, per component	One figure for each named peptide, not one figure for the vial. A range across the three tells you where the weakest one sits.
5	Content (mass), per component	Milligrams of each peptide, against that peptide's labelled milligrams. The only field that detects substitution, omission and under-fill.
6	Method and instrument	HPLC-UV, HPLC-MS/MS, the wavelength, the instrument. Content figures from different methods are not freely comparable — our own lots show two laboratories differing by up to 16.3% on the same material (the full disagreement).
7	Date of analysis	A certificate is a measurement of a sample on a day, not a permanent property of a product line.
8	What was not tested	Sterility, endotoxin, residual solvents and water content are separate assays. A chromatographic certificate is silent on all of them, and a good one says so.
9	Whether content is net peptide or gross	Net peptide content excludes counterion, residual water and other peptide-related material. Gross powder weight is a different and much less useful quantity.

Our general walkthrough of certificate fields, including the ones specific to single peptides, is how to read a peptide certificate of analysis. Every lot we publish, and the laboratory that issued it, is indexed on the lab results page.

The worked example: GLOW lot glow70-0626-01

Lot glow70-0626-01 has been assayed twice, and the two certificates are a clean illustration of everything above — including, uncomfortably for us, the limits of the first one.

Certificate one: Janoshik Analytical

The release assay reported an overall HPLC purity of 99.469% and confirmed the identity of all three components. It did not determine content. That is a high number from a laboratory that has assayed most of our catalogue, and by the argument on this page it could not have told anyone whether the 10 mg of BPC-157 was in the vial. We published it, labelled it as the only certificate on the lot, and left GLOW out of our vial fill-accuracy study for exactly that reason: it was the one lot with a purity figure and no mass.

Certificate two: VialTests VT-2026-001827-S04

Reported 2 November 2026. Analysed by HPLC-MS/MS on an Agilent 1290 Infinity II coupled to a 6495 QQQ triple quadrupole, each component against its own reference standard:

Component	Label	Measured	% of label	Purity	Share by mass
GHK-Cu copper tripeptide-1, CAS 89030-95-5	50 mg	61.75 mg	123.5%	99.84%	73.6%
BPC-157 pentadecapeptide, CAS 137525-51-0	10 mg	10.55 mg	105.5%	99.61%	12.6%
TB-500 thymosin β4, CAS 77591-33-4	10 mg	11.61 mg	116.1%	99.64%	13.8%
Total net peptide	70 mg	83.91 mg	119.9%	99.61–99.84%	100%

total content 61.75 + 10.55 + 11.61 = 83.91 mg total fill 83.91 ÷ 70 = 1.1987 → 119.9% of label weight shares 61.75 ÷ 83.91 = 73.6% GHK-Cu 11.61 ÷ 83.91 = 13.8% TB-500 10.55 ÷ 83.91 = 12.6% BPC-157

We publish the purity as the range 99.61–99.84%, not as a single average, and that is a deliberate choice. A content-weighted mean works out at 99.78%:

(61.75 × 99.84 + 10.55 × 99.61 + 11.61 × 99.64) ÷ 83.91 = 99.783%

That figure is arithmetically correct and we will not print it as a headline, for two reasons. It is our calculation, not the laboratory's, and presenting it as a certificate value would be presenting our arithmetic as someone else's measurement. And it does what this whole page argues against: it collapses three measurements into one and hides the weakest component, which at 99.61% is the one a reader would actually want to know. A range is the honest summary of a blend. 99.61% is the number that constrains the vial.

Note what the per-component figures establish that nothing else could. All three peptides are present. Each is present at or above its labelled mass. Each is independently clean. The TB-500 component is quantified at 11.61 mg against a 10 mg label and its chromatogram is labelled thymosin β4, so the vial contains the 43-residue protein, not the Ac-LKKTETQ heptapeptide — which agrees with what the same laboratory found in our standalone TB-500 10mg, identified there as CAS 77591-33-4 at 4963.4 g/mol. The substitution scenario in the section above is ruled out by a document rather than by a promise.

The copper problem, and an open question we have not resolved

The GHK-Cu figure deserves scrutiny rather than a victory lap, because 123.5% of label is a large over-fill and there is a plausible accounting explanation for a good part of it.

GHK-Cu is a complex, not a plain peptide. The copper(II)-bound form is C₁₄H₂₂CuN₆O₄ at roughly 403.9 g/mol; the metal-free tripeptide Gly-His-Lys is C₁₄H₂₄N₆O₄ at roughly 340.4 g/mol. Those differ by a factor of 1.1865 — 18.7%. So "50 mg of GHK-Cu" is two different masses depending on whether the label means the complex or the peptide inside it, and a quantification can be reported on either convention.

403.9 ÷ 340.4 = 1.1865 → an 18.7% difference between conventions If the measured 61.75 mg is the complex and the label meant metal-free GHK: 61.75 ÷ 1.1865 = 52.04 mg → 104.1% of a 50 mg label total 52.04 + 10.55 + 11.61 = 74.20 mg → 106.0% of label If both are on the same convention: 61.75 mg → 123.5% of label total 83.91 mg → 119.9% of label

So the true fill of this vial is somewhere between 106.0% and 119.9% of label, and which end depends on a convention the certificate does not state. We publish 119.9% because that is what the two documents say when read at face value, and we are flagging the ambiguity rather than quietly choosing the flattering number. We have written to the laboratory to ask which basis the GHK-Cu quantification used. When they answer, this section and the product page will be updated and the change recorded.

Two smaller notes of the same kind. The certificate's report number appears as VT-2026-001827-04 on its first page and VT-2026-001827-S04 on the verification record; we publish the -S04 form because that is what the laboratory's own lookup accepts, and we have asked about the discrepancy. And the 116.1% TB-500 figure sits inside the normal range for our catalogue rather than outside it: across nineteen published lots the median vial holds 113.1% of label, mean 114.9%, range 93.9% to 142.4%, with 17 of 19 at or above label. Over-fill is the rule here, not the exception, and the fill-accuracy study sets out why.

Which of our own blends have per-component data, and which do not

Arguing that blends need per-component certificates obliges us to say where our own catalogue meets that standard. It does on three of four.

Blend	Lot	Per-component content?	What the certificate reports
GLOW Blend GHK-Cu + BPC-157 + TB-500	`glow70-0626-01`	Yes — three masses, three purities	VialTests VT-2026-001827-S04. 61.75 / 10.55 / 11.61 mg; 99.84% / 99.61% / 99.64%. Plus an earlier Janoshik assay at 99.469% overall with no content.
CJC-1295 (no DAC) + Ipamorelin	`cjc10-0626-02`	Yes — two masses, named-component areas	Testides (Toronto), HPLC-UV 214 nm. Ipamorelin 5.34 mg, CJC-1295 4.74 mg, 10.08 mg total (100.8% of label); combined area 95.80%.
Tesamorelin + Ipamorelin Blend	`tesaipa10-0626-01`	Yes — two masses, no purity figure	Janoshik Analytical #235914, per-component content: Tesamorelin 5.42 mg, Ipamorelin 5.89 mg, 11.31 mg total against a 10 mg label. No single purity figure is reported for the blend and we do not print one. Our own confirmation is pending.
Wolverine Blend BPC-157 + TB-500	`HEAL-0626-01`	No — combined only	Testides (Toronto), HPLC-UV 214 nm: 98.64% combined purity, 10.78 mg total content (107.8% of label). The split between the two peptides is not determined on this certificate. By the argument on this page that is the weaker document in our catalogue, and a per-component re-assay is the obvious next thing to commission.

We would rather print that last row than leave the reader to notice it. A combined content figure is a meaningful improvement on a bare purity percentage — it rules out gross under-fill of the vial as a whole — but it does not rule out the two peptides being in the wrong ratio, and BPC-157 and TB-500 differ enough in price that the ratio matters.

Five questions to ask before you accept a blend certificate

Does the certificate give a mass for every peptide on the label? If it gives one total, it cannot detect a wrong ratio. If it gives one percentage, it cannot detect anything about quantity at all.
Does the lot number on the document match the code on the vial? "Representative of this batch" is not the same document.
Can the laboratory confirm the report, independently of the vendor? A report number entered into the laboratory's own form, not a PDF on the seller's server.
Is the content net peptide, and on what basis? Gross powder weight includes counterion and water. For a metal complex such as GHK-Cu, ask whether the figure is the complex or the free peptide — as the section above shows, that one convention moves this vial's stated fill by 14 percentage points.
Which identity method was used? Mass spectrometry distinguishes molecules that a retention time alone does not. On a blend containing TB-500 that is the difference between a 4963 Da protein and an 889 Da fragment sold under the same name.

What a per-component certificate still does not prove

A per-component certificate is the right document for a blend. It is not a clean bill of health, and anyone citing this page should cite this section with the rest of it.

It describes a sample, not your vial. A laboratory analysed material from a lot on a date. Fill variation within a lot is real; our nineteen-lot spread of 93.9% to 142.4% is between lots, and nothing here measures within-lot variation.
It says nothing about sterility or endotoxin. Those are separate assays on separate samples. One of our lots carries a bacterial endotoxin certificate; the chromatographic reports do not substitute for it.
It says nothing about residual solvents, water content or counterion identity unless those assays are listed, and on these reports they are not.
It is not a stability statement. Purity and content were what they were on the date of analysis. Neither certificate predicts what the vial contains after a year of warm storage or repeated freeze–thaw.
Accreditation is not established for every laboratory here. Janoshik Analytical and Testides are not ISO/IEC 17025 accredited, and we have not established VialTests' accreditation status. We say so on every page that cites them. Where to get independent testing yourself is covered in our guide to peptide testing in Canada.
It carries no regulatory weight whatsoever. None of these compounds is approved for human use in Canada. Health Canada's public advisory of 9 April 2026 (RA-81874) names GHK-Cu, BPC-157 and TB-500 individually among unauthorised peptides sold online and states explicitly that "for research use only" labelling does not make a product legal to sell for human use. BPC-157 is prohibited in sport under WADA class S0 and thymosin β4 under class S2. A certificate describes a vial; it does not describe a legal position. The legal position is here.

Analytical quality and legality are independent questions, and a vendor who answers the first in order to avoid the second should be read with that in mind.

Questions

What is the difference between purity and content on a peptide certificate?

Purity is a ratio: the area of the intended peak divided by the total integrated area in one chromatogram, expressed as a percentage. It describes how clean the material is. Content is a quantity: milligrams of a specific peptide, derived from the peak area read against a calibration curve built from a reference standard of that molecule. It describes how much there is. A vial can be 99.9% pure and hold half the labelled mass, and a purity figure will not reveal it.

Why can a peptide blend not have a single purity number?

Because chromatographic purity is defined against one intended compound, so that everything else in the chromatogram counts as an impurity. With three intended peptides there is no single target peak, and the only available summary — the three intended peaks added together — reports contamination while saying nothing about proportions. A combined figure cannot distinguish a correctly filled three-peptide vial from a vial of one cheap peptide, because in both cases every peak present belongs to a peptide on the label.

Can you read the composition of a blend off the peak areas?

No. A UV detector at 214 nm responds to peptide bonds and aromatic residues, not to mass, so detector response per milligram differs between peptides of different length and composition. Our CJC-1295 + Ipamorelin lot cjc10-0626-02 demonstrates the size of the error on one report: by peak area the vial reads 85% ipamorelin to 15% CJC-1295, and by measured mass it is 53% to 47%. Converting area to mass requires a calibration curve for each peptide, which is the work a content determination does and a purity figure does not.

What does the GLOW Blend certificate actually report?

VialTests report VT-2026-001827-S04, lot glow70-0626-01, reported 2 November 2026, analysed by HPLC-MS/MS on an Agilent 1290 Infinity II with a 6495 QQQ triple quadrupole. Per component: GHK-Cu 61.75 mg at 99.84% purity (123.5% of its 50 mg label), BPC-157 10.55 mg at 99.61% (105.5% of 10 mg), and TB-500 11.61 mg at 99.64% (116.1% of 10 mg). Total net peptide content 83.91 mg against a 70 mg label, a fill of 119.9%. By weight the vial is 73.6% GHK-Cu, 13.8% TB-500 and 12.6% BPC-157. An earlier Janoshik Analytical assay on the same lot reported 99.469% overall purity and did not determine content.

Why is the GLOW purity published as a range instead of one number?

Because 99.61–99.84% is what was measured, and a single average would be our arithmetic presented as the laboratory's. A content-weighted mean of the three figures works out at 99.783%, but printing that as the headline would hide the weakest component, which at 99.61% is the figure that actually constrains the vial. On a blend a range is the more informative summary.

Is the TB-500 in the GLOW Blend full-length thymosin β4 or the Ac-LKKTETQ fragment?

For lot glow70-0626-01 it is the full-length 43-residue thymosin β4. The third chromatogram on VialTests report VT-2026-001827-S04 is labelled thymosin β4 and the component was quantified at 11.61 mg against a 10 mg label. That agrees with the same laboratory's finding on our standalone TB-500 10mg, where the material is identified as CAS 77591-33-4 at 4963.4 g/mol rather than the roughly 889 Da heptapeptide also sold as "TB-500". If a later lot differs, its certificate will say so.

Why is the GHK-Cu component at 123.5% of its label?

Partly over-fill, which is normal across this catalogue — the median of nineteen published lots is 113.1% of label. Possibly also an accounting convention. GHK-Cu is a copper complex at about 403.9 g/mol enclosing a metal-free tripeptide at about 340.4 g/mol, a difference of 18.7%. If the measured 61.75 mg is the complex while the 50 mg label refers to the free peptide, the figure on a common basis is 52.04 mg, or 104.1% of label, and the whole vial reads 74.20 mg and 106.0% rather than 83.91 mg and 119.9%. The certificate does not state which basis was used. We publish the face-value figures, we are flagging the ambiguity rather than choosing the flattering reading, and we have asked the laboratory which convention applies.

What does a per-component certificate still not prove?

Sterility, bacterial endotoxin, residual solvents, water content and counterion identity are separate assays and are not on these reports. The certificate describes a sample from a lot on a date, so it is not a stability statement and does not measure variation between vials within a lot. Neither Janoshik Analytical nor Testides is ISO/IEC 17025 accredited and VialTests' accreditation status has not been established. And a certificate carries no regulatory weight: none of these compounds is approved for human use in Canada, and Health Canada's 9 April 2026 advisory RA-81874 states that "research use only" labelling does not make a product legal to sell for human use.

Which blends in your own catalogue lack per-component content figures?

One of four. The Wolverine Blend (BPC-157 + TB-500), lot HEAL-0626-01, carries a combined figure from Testides — 98.64% purity and 10.78 mg total content, 107.8% of a 10 mg label — without a split between the two peptides. That rules out gross under-fill of the vial but not a wrong ratio between the two components, and a per-component re-assay is the next thing to commission. GLOW Blend, CJC-1295 + Ipamorelin and the Tesamorelin + Ipamorelin Blend all carry per-component masses.

How to read a peptide certificate of analysis — every field on a single-peptide COA, and the table of all nineteen of our published lots.
Lyophilised peptide vials hold more than the label — nineteen lots measured against label, with the over-fill mechanism.
When two laboratories disagree on content — our own lots differing by up to 16.3% on the same material.
Where to get peptides tested in Canada — commissioning your own assay, and what it costs.
GHK-Cu research guide · BPC-157 research guide · TB-500 research guide
Lab results — every lot, every figure · GLOW Blend

For laboratory and research use only. Not for human consumption, veterinary use, or therapeutic application. None of the compounds discussed is approved by Health Canada and none carries a DIN. We do not provide preparation, dilution or administration guidance for any compound we sell. This page describes analytical chemistry, not a therapeutic position.

Written by the thepeptide.ca research team. First published 3 November 2026. Figures on this page are taken from VialTests report VT-2026-001827-S04, Janoshik Analytical's release assay for lot glow70-0626-01, and Testides reports for lots cjc10-0626-02 and HEAL-0626-01, all indexed on our lab results page. Arithmetic derived from those figures is shown in full above so it can be checked. Where a question is open we say so rather than resolving it in our own favour.