Reverse-phase HPLC, at 220 nm, against a reference standard.
Every released lot is profiled on a C18 column with gradient elution, UV detection at 220 nm and quantified as area-percent purity. The chromatogram travels with every shipment.
Six steps from synthesis batch to released lot.
Every peptide goes through the same path. There is no expedited lane, no in-house shortcut, no "this looked fine, ship it." A failure at any step rejects the batch.
-
01
Sample receipt & chain-of-custody
Lyophilized peptide arrives sealed from the synthesis facility. Lot ID, batch, theoretical sequence and MW are logged. The vial is split into two aliquots: one for analytical, one for archive (retained 24 months).
-
02
Reconstitution & dilution
2 mg of peptide reconstituted in 1 mL of mobile phase A (0.1% TFA in water) with light vortexing. Diluted to 1 mg/mL working concentration. Centrifuged 5 min at 13 000 g to remove particulates that would foul the column.
-
03
Column equilibration
250 × 4.6 mm C18 column (5 µm, 100 Å) equilibrated at 95% A / 5% B, 1.0 mL/min, 30 °C until baseline drift is < 0.0005 AU/min. Pressure stable within ± 5 bar across three blank injections.
-
04
Injection & gradient run
20 µL injection loop, full loop fill. Linear gradient from 5% to 95% acetonitrile (0.1% TFA) over 25 min, 5 min hold at 95%, return to initial in 2 min, re-equilibrate 8 min. Total cycle 40 min.
-
05
Detection & integration
Diode-array UV detector, primary trace at 220 nm (peptide bond), secondary at 280 nm (Trp/Tyr). Peaks integrated by valley-to-valley, baseline corrected, area-percent reported relative to the main product peak.
-
06
Release decision
Main peak ≥ 99.0% area: the lot moves to ESI-MS confirmation. Below 99.0%: the lot is rejected, the synthesis batch is investigated, and the QC officer signs the rejection notice. No exceptions, no rounding.
Reading what the trace actually says.
A clean release chromatogram is a flat baseline, a sharp main peak, and very little else. Here is how to read every part of one — including the parts most vendor "COAs" hide by cropping.
Main product peak
Sharp, single, high. Area = <b>99.42%</b> of total integrated signal. Retention 13.4 min — matches the reference standard within ± 0.05 min.
Process-related impurities
Three minor peaks visible at 6.6, 13.0 and 16.5 min. Combined area = <b>0.58%</b>. Likely deletion sequences and oxidation byproducts; identified by ESI-MS in the next assay.
Baseline
Flat, low noise (drift < 0.0005 AU/min), no rolling hump from solvent contamination, no late-eluting "garbage" peak after the main product. The chromatogram returns to baseline cleanly.
The exact specs the lab runs against.
If you can't reproduce the assay from the parameters, the COA is opaque. Here is everything you need to repeat the run on your own instrument.
Stationary phase
| Column | C18 reverse-phase, fully end-capped |
|---|---|
| Dimensions | 250 mm × 4.6 mm i.d. |
| Particle size | 5 µm |
| Pore size | 100 Å |
| Carbon load | ≈ 17% |
| Temperature | 30 °C (column oven) |
Mobile phase & gradient
| Solvent A | H₂O + 0.1% TFA |
|---|---|
| Solvent B | Acetonitrile + 0.1% TFA |
| Flow rate | 1.0 mL/min |
| Gradient | 5% → 95% B over 25 min |
| Hold | 95% B for 5 min |
| Re-equilibration | 5% B for 8 min |
Detection & integration
| Detector | Diode-array UV (DAD) |
|---|---|
| Primary λ | 220 nm (peptide bond) |
| Secondary λ | 280 nm (Trp / Tyr) |
| Sampling rate | 10 Hz |
| Integration | Valley-to-valley, baseline corrected |
| Reporting | Area-% relative to total |
Sample & injection
| Diluent | Mobile phase A |
|---|---|
| Concentration | 1 mg/mL working solution |
| Injection volume | 20 µL (full-loop) |
| Replicates | 3 per lot, ± relative SD reported |
| System suitability | USP <621> criteria, every batch |
| Reference standard | Independent, lot-traceable |
The impurity load doubles between every grade.
A "95% pure" peptide carries ten times the impurity mass of a "99.5% pure" lot. In a 5 mg vial, that's 250 µg of unidentified material vs. 25 µg. In a cell-culture experiment, that difference is the experiment.
Common floor for grey-market peptides. Identity is plausible; reproducibility is not.
The default release threshold for most internet vendors. Better — but truncations still pass undetected.
Hard floor. Anything below is rejected and the synthesis batch is investigated. No rounding.
Where most of our released lots actually land. The 99% line is the floor, not the mode.
What the impurity peaks usually mean.
The minor peaks on a chromatogram are not random noise — each one carries information about how the synthesis went. Here are the patterns we look for and what they signal back to the synthesis facility.
Four conditions that reject the lot, no override.
The release rule is not a guideline. These are the conditions under which the QC officer signs a rejection notice and the lot returns to the synthesis facility for investigation.
Main peak below 99.0% area
The release floor. Even 98.97% is rejected. There is no "almost passed", no rounding to two significant figures, no "we'll release it as the lower-grade SKU" — the batch is sent back.
Single impurity peak above 0.5%
Even when the main peak clears 99.0%, a single related impurity above 0.5% is treated as a synthesis problem worth investigating. Common cause: a truncation that the gradient didn't fully separate.
Retention time off > 0.2 min vs. reference
If the main peak elutes at the wrong time relative to the lot-traceable reference standard, the molecule has likely changed — either a sequence error or a major modification. Sample is held pending MS investigation.
System suitability failure
USP <621> criteria — theoretical plates, tailing factor, capacity factor, signal-to-noise — are checked on a system suitability standard before every batch. If the column or detector is drifting, no sample is run until it's restored.