HPLC peptide purity: how the test works, what it misses, peak purity and aggregation

How HPLC tests peptide purity

High-performance liquid chromatography (HPLC) separates the molecules in a sample and measures each one with a detector. For peptides, labs almost always use reversed-phase HPLC. The column is packed with silica that carries C18 chains. A gradient moves from water to acetonitrile, usually with a small amount of acid such as trifluoroacetic acid or formic acid. Each molecule leaves the column at its own retention time.

A UV detector watches the liquid that leaves the column. Peptide bonds absorb strongly at short wavelengths, so labs read peptides at about 214 or 220 nm. Tyrosine and tryptophan also absorb at 280 nm. In the work that set up USP’s peptide reference standards, peptides were monitored at 220 nm on a C18 column, with triplicate injections from one container.

The data system then integrates each peak. HPLC purity is the main peak’s area as a percentage of the total area of all peaks it counts. That is the whole calculation. Everything else on a purity line, from the wavelength to the peaks left out, changes what that percentage means.

What purity in HPLC means on a COA

Purity is a relative number. It depends on the wavelength, the run length, the integration settings and the rule for which peaks count. ILS, for example, states on its reports that purity is area normalisation at 214 nm and that non-peptide process impurities are excluded when detected. That is a clear rule. Two labs with different rules can print different numbers for the same powder.

One published case shows the size of the effect. A Substack comparison from August 2026 set two labs’ results for one tesamorelin batch side by side. Freedom Diagnostics reported 99.985 % purity and Janoshik reported 97.779 %. A single case does not show which number is right. It shows that the method and integration choices behind a purity figure matter as much as the figure.

Purity is also separate from content. Bachem notes that the net peptide content of research-grade peptides is typically 70 to 90 % of the powder, even when HPLC purity is above 95 %. Its worked example: 5 mg of powder × 85 % net peptide × 98.7 % purity = 4.195 mg of the target peptide. Our guides on peptide purity on a COA and net peptide content go further.

What HPLC purity cannot see

What you want to know	Seen by HPLC-UV purity?	Test that answers it
Is it the right peptide?	No. Similar peptides can elute close together	Identity by LC-MS
How many mg are in the vial?	No	Content per vial against a reference standard
Water in the powder	No	Karl Fischer titration
TFA or acetate salt	Not counted as an impurity	Counter-ion test
Residual solvents	No	Gas chromatography
Endotoxin or live microbes	No	LAL endotoxin test, USP <71> sterility
Impurities that co-elute with the main peak	No, unless the method separates them	A second method or LC-MS across the peak
Aggregates (dimers and larger)	Often not	Size-exclusion chromatography

For medicines, the EMA synthetic peptide guideline applies the Ph. Eur. impurity thresholds, and the purity method must at least meet the reporting threshold. Peptide-related impurities are reported above 0.1 %, identified above 0.5 % and qualified above 1.0 %. ICH Q2(R2) adds that the quantitation limit of an impurity method must be at or below the reporting threshold. A research lab does not have to follow these rules, but they are a useful yardstick for a report.

Peak purity in HPLC is a different check

Peak purity answers a narrower question: does one peak hold one compound? A diode array detector records a full UV spectrum at many points across the peak. The software takes the spectrum at the apex as the reference and compares the others with it. If the spectra change, a second compound is probably co-eluting.

The check has limits. A hidden compound is only flagged if it absorbs UV, has a different UV spectrum and is present at a level the detector can see. Peptide-related impurities, such as a chain missing one amino acid, are structurally close to the main peptide. Their UV spectra can be too similar for the check to separate them.

How to check peak purity in HPLC, then, has three practical answers. Read the diode array purity value as supporting evidence only. Look at mass spectra taken across the peak, which can show a second mass under it. Or run a second method with a different column or pH. TrustPointe prints a co-elution index on its HPLC results, which is one way labs show this check.

Peptide aggregation and HPLC

A reversed-phase run separates molecules by how strongly they stick to the C18 surface, not by size. It is not the usual tool for aggregates. Size-exclusion chromatography (SEC) separates by size and is the common method for dimers and larger species. The EMA guideline lists high-molecular-weight impurities as a specification item where they are relevant.

A 2024 study of insulin powder stored at 60 °C and 75 % relative humidity shows the split. The authors used reversed-phase HPLC for chemical changes such as deamidation, and SEC for covalent aggregation. Some peptide labs sell aggregate testing separately: Janoshik offers a GLP-1 aggregate analysis add-on with its GLP-1 test.

What a good HPLC purity report shows

A purity result you can check prints:

• the wavelength, the column and the gradient;
• the full-run chromatogram, not a view cropped to the main peak;
• a peak table with retention time, area and area % for every integrated peak;
• the integration settings and the reporting threshold;
• a blank injection, so solvent peaks are not counted as impurities;
• system suitability values, such as resolution and peak symmetry (the harmonised USP <621> default symmetry range is 0.8 to 1.8);
• replicate injections and the measurement uncertainty, with digits that match it.

Few reports do this today. In our September 2026 review of public COAs from 25 labs and one rating site, only 4 printed a proper peak table and none printed integration settings. Common problems were charts cropped to the main peak, single-ion traces that cannot show other impurities, censored values such as “> 99 %”, and three decimals with no uncertainty.

Common mistakes when reading HPLC results

• Reading purity as the dose in the vial. Purity says nothing about mg. Check the content result and its basis.
• Treating 99.9 % as better than 98.5 %. Krause Analytical’s director calls area-% purity “really not very useful” as a measure of what is in the vial. A very high figure with a cropped chart deserves more questions, not fewer. See why a 99.9 % result can be a warning sign.
• Comparing purity across labs. Different wavelengths and integration rules give different numbers. Our guide on why labs disagree explains the main causes.
• Taking purity as proof of identity or safety. A clean chromatogram of the wrong peptide is still the wrong peptide, and HPLC does not test for microbes.

HPLC testing near you: sending a sample

You do not need a lab in your city. Peptide labs take samples by courier, so the choice comes down to method, reporting and transit time. Our guides on where to test peptides and how to send peptides for testing cover the practical steps, and peptide purity testing cost covers what labs charge for it.

How HK Analytical reports HPLC purity

Our purity test reports the main peak as a share of all peaks we detect, at the stated wavelength. It also lists the main impurities by retention time. Every number in the report comes with its raw data, such as the chromatogram and the spectra, in a signed record you can check on our verify page. You can see the layout on our example report. Purity sits next to identity by LC-MS and content per vial in one order, and you can add other tests from our list per sample through a quote.