Size Exclusion Chromatography and Peptide Aggregate Detection
A reversed-phase purity trace can report a sample as clean while the vial holds material that is not monomeric at all. That is not a flaw in the trace; it is a consequence of what reversed-phase conditions do to a sample on the way to the detector. Size exclusion exists to look at what those conditions destroy, and understanding the pairing starts with the blind spot rather than with the hardware.
The population reversed-phase never sees
Organic solvent and acid make reversed-phase conditions denaturing. Most non-covalent assemblies come apart under them well before elution, so what emerges as a monomer peak reflects the monomer content of a sample that may have been something else entirely while sitting in buffer.
Size exclusion runs in aqueous buffer close to neutral pH. Non-covalent dimers, oligomers and soluble aggregates therefore survive the trip and register as peaks eluting ahead of the monomer. Covalently linked aggregates, a disulfide-bridged dimer for instance, show up under either method. The non-covalent population appears only here, and it is the population generated by the solution behavior described in peptide aggregation in solution. The two techniques make a real pair in the sense discussed under confirming identity with two independent methods, precisely because neither one carries the other’s blind spot.
A related consequence: species that cannot be pulled apart by a hydrophobic surface can still be separated by size. A dimer co-eluting with its monomer on a reversed-phase column will often resolve on a size exclusion column, since the two differ in dimension even when they behave identically against a hydrophobic packing.
Sorting by how much room a molecule takes up
Porous beads make up the packing. Molecules small enough to diffuse into the pores wander a long path down the column, while molecules too large to enter are confined to the channels between beads and take a short one. Large therefore leaves first and small leaves last, with the entire separation completed inside a single column volume.
There is no retention in the reversed-phase sense, no binding step and no gradient, since mobile phase composition stays fixed from start to finish. Short runs and relatively gentle handling of the sample follow directly from that.
The window, and falling outside it
Two volumes fix the boundaries of anything the column can do.
| Void volume | Where anything too big to enter a pore comes off. Everything above the exclusion limit arrives here at once, unresolved. |
| Total permeation volume | Where anything small enough to access every pore comes off. Everything beneath the lower limit arrives here at once, equally unresolved. |
Only the span between those two points separates anything. Choose a pore size that puts the peptide at one extreme or the other and the column resolves nothing whatsoever, which is far and away the usual explanation for a size exclusion run on a small peptide returning a single uninformative peak.
Apparent weight is not weight
Turning an elution volume into a size means running standards of known molecular weight and fitting a curve to them. What comes out is an apparent molecular weight, and that adjective is doing real work, because the separation answers to hydrodynamic radius rather than to mass.
A chain that is extended or unstructured sweeps out more volume than a compact globular molecule of equal mass, elutes sooner, and reports an apparent weight considerably above the truth. Since plenty of peptides are unstructured in aqueous buffer, any apparent weight read off a globular-protein calibration is biased high by construction and must not be treated as a mass measurement. Measuring mass is the spectrometer’s job, at the accuracy set out in mass accuracy in parts per million.
Charge does not switch off
In the ideal case nothing interacts with the packing at all. Reality intrudes because packing surfaces retain residual charge and peptides carry charge, so ionic interaction runs in competition with the size mechanism. A basic peptide may be held beyond the total permeation volume and look smaller than it is, while an acidic one may be repelled and look larger.
Salt in the mobile phase is the conventional remedy, usually phosphate or acetate at moderate ionic strength, occasionally with a few percent organic solvent added to damp hydrophobic interaction too. Alter that buffer and the answer alters with it, so any size exclusion result is comparable only against another run performed identically. The charge behavior underneath all this is the same one covered in isoelectric point and solubility.
Four ways to get a wrong answer
The sample dilutes several-fold while traveling down the column, and a reversible oligomer whose existence depends on concentration may dissociate en route, ending up under-reported or absent from the result altogether. Aggregates can also adhere to the frit or the packing and simply never emerge; checking mass recovery against an injection onto a column blank catches this, and unaccounted mass should be treated as a finding rather than an annoyance.
Anything insoluble is invisible from the outset, since material filtered out or spun down before injection never enters the answer. A sample prepared by filtration describes the soluble fraction alone, and the report ought to say as much. Finally, resolution here is deliberately modest: separating two species that differ by less than roughly twofold in size is not reliable on one column.
What the detector contributes
Ultraviolet absorbance is the default and brings along the response-factor caveats described in ultraviolet response factors, with an extra wrinkle: an aggregate scatters light and can inflate its own apparent absorbance. Refractive index tracks mass more faithfully but gives up sensitivity to do it.
Putting multi-angle light scattering in line changes what kind of measurement this is. Molar mass comes directly out of the scattering, with no calibration curve anywhere in the process, which disposes of the shape assumption entirely. The tradeoff is that it requires an accurate concentration alongside it, obtained by the approach described in concentration by A280.
Why certificates for short peptides skip it
For a short synthetic peptide, the impurities worth worrying about are deletion sequences, truncation sequences and leftover synthesis reagents. Every one of those sits close in size to the target, making them invisible to size exclusion, while being exactly what a reversed-phase gradient pulls apart. Put a fifteen-residue peptide on a size exclusion column and the usual yield is one peak and nothing learned.
The method earns its keep as chains lengthen, as structure starts to matter, and wherever aggregation state is a stability question rather than a purity question. Alongside a reversed-phase trace it answers something that trace was never designed to address. Not what fraction of the material is the correct molecule, but what fraction of the correct molecule is traveling by itself.
