Choosing an HPLC Column for Peptides: C18, C8, C4 or Phenyl
Ask a testing lab which HPLC column for peptides it prefers and the answer will usually be C18. It is a sensible starting point, and for many short research sequences it is the right one. It is not a universal answer, though, and the column line printed beside a chromatogram tells a careful reader whether a method was designed around the compound or borrowed from a different one. For QC teams receiving large, multi-vial lots, that one line is worth a minute of attention before the purity figure is entered into the inventory record.
What the stationary phase is doing
Reversed-phase columns are packed with small silica particles whose surfaces carry bonded hydrocarbon chains. Peptides in an aqueous, acidic mobile phase stick to those chains through hydrophobic contact, then release as the proportion of organic solvent (usually acetonitrile) climbs during the gradient. The more hydrophobic a molecule, the later it lets go.
The labels C18, C8 and C4 give the number of carbon atoms in the bonded chain. That number controls how strongly the phase holds on to analytes:
- C18 (octadecyl): the strongest hold of the three and the most widely used phase for small to mid-sized peptides.
- C8 (octyl): a moderate hold, often giving sharper peaks for larger or stickier sequences.
- C4 (butyl): the weakest hold, traditionally chosen for long peptides and small proteins.
Beyond the alkyl phases, a phenyl or phenyl-hexyl phase adds interactions with aromatic rings. Because it separates on a partly different principle, it can pull apart species that a C18 run lumps together.
Choosing an HPLC column for peptides by sequence
There is no single best phase, only a better match for a given molecule. A rough guide based on the sequence itself:
| Sequence character | Typical first choice | Common problem on the wrong column |
|---|---|---|
| Short and polar (a few residues, charged) | C18, sometimes with a polar-embedded or aqueous-stable phase | Elutes near the void, where nothing is resolved |
| Mid-length, mixed hydrophobicity | C18 | Usually behaves well |
| Long or strongly hydrophobic | C8 or C4, wide-pore | Broad, smeared peaks and incomplete recovery on C18 |
| Rich in aromatic residues, or closely related impurities | Phenyl as a second, orthogonal method | Co-elution hidden under a single C18 peak |
A very short, charged peptide may barely interact with any bonded phase and rush out near the solvent front. There, impurities and the main compound come off together and the purity figure means little. At the other extreme, a long hydrophobic sequence can bind C18 so firmly that a portion never elutes within the gradient, so the trace under-represents what is actually in the vial.
Pore size: the parameter people forget
Nearly all of a silica particle’s surface area is inside its pores. If a molecule is too large to diffuse into them, it can interact only with the outside of the particle. Retention weakens, peaks spread and recovery becomes erratic.
Standard practice is a pore diameter of roughly 100 to 120 angstrom for short sequences and small organic analytes, and around 300 angstrom (often called wide-pore) for longer peptides and proteins. The trap is that a long sequence run on a narrow-pore column can still give a tidy-looking chromatogram. The species most likely to be under-counted are the bigger ones, such as aggregates and incompletely deprotected material, which are exactly what an impurity profile is meant to catch.
Other hardware details that shape the trace
- Particle size: smaller particles give narrower peaks and better resolution, at the cost of higher back-pressure.
- Column length: a longer column offers more separating power but longer run times.
- Temperature: a warmer column often sharpens peptide peaks and can change selectivity, so a stated temperature is part of the method, not a footnote.
How the mobile phase interacts with the column
Most peptide purity methods pair water and acetonitrile with an acid modifier. Trifluoroacetic acid, at around 0.1%, remains the classic choice because it pairs with protonated basic side chains and gives narrow, symmetric peaks on silica phases.
The drawback shows up when the column outlet feeds a mass spectrometer. TFA suppresses electrospray ionisation, so LC-MS methods often swap in formic acid. Peak shape usually suffers a little, but mass signal improves considerably. A laboratory confirming identity by LC-MS and purity by UV may therefore run two different mobile phases, and the retention times on the two reports will not line up exactly. That is expected rather than suspicious.
The acid used during purification also tends to end up as the counter-ion in the dried product, which affects net peptide content and the gap between vial mass and peptide mass. It is a separate topic, but it starts at the same bench.
Why column choice matters when comparing lots
For a lab buying the same compound repeatedly, the purity trend across lots is only meaningful if the method underneath it holds still. A switch from C18 to C8, or from a narrow-pore to a wide-pore column, can move the reported figure by more than the real variation between lots. Neither result is wrong. They are measurements taken with different rulers.
Practical steps for anyone logging a bulk or recurring order:
- Record the column phase, pore size and dimensions from each certificate alongside the lot number.
- Note the mobile-phase modifier and detection wavelength at the same time.
- If a new lot reports a noticeably different figure, check these fields first before assuming the material changed.
- Where a single peak is critical to your work, ask whether an orthogonal run (for example phenyl, or a different pH) was performed.
Our catalogue items go out for third-party HPLC and purity analysis. Certificates are published for some products, and cap and crimp colour on each vial ties it back to the matching certificate, which helps when one receipt spans several lots.
Reading the column line on a certificate
A typical method summary might read something like “C18, 4.6 x 250 mm, 5 micrometre, 100 angstrom, gradient of acetonitrile in 0.1% TFA, 214 nm”. Each part answers a question. The phase suggests how strongly the compound was retained. The pore size tells you whether a longer sequence could access the surface. The modifier and wavelength tell you what the detector was set up to see.
When a long, hydrophobic molecule is reported on a narrow-pore C18 column with a generic gradient, the figure is still data, but it is data from a method that was not built for that compound. Weigh it accordingly, and consider asking for the chromatogram if it was not supplied. Column performance on the day is a separate matter, covered by the system suitability checks a laboratory runs before any sample.
These notes concern analytical characterisation only. Bulk Peptides supplies material for in-vitro and laboratory research, not for use in people or animals.
