1. Fractionation Range and Exclusion Limit
The fractionation range is the region in which differences in molecular size produce useful differences in pore access and elution. The exclusion limit refers to a size above which molecules are substantially unable to enter the relevant pores and approach the void volume. Definitions, calibration standards, and stated endpoints vary, so values from different media or sources are not always directly comparable.
A molecule near complete exclusion has little internal volume left to distinguish it from a still larger species. A molecule near complete inclusion likewise has little useful pore-volume difference from a smaller neighbor. The best resolving region is generally between these extremes, where the elution response changes meaningfully with hydrodynamic size.
| Reported property | What it suggests | What it does not guarantee |
|---|---|---|
| Protein fractionation range | Approximate mass region for selected protein standards | Behavior of elongated, glycosylated, disordered, or complexed proteins |
| Polymer fractionation range | Calibration for a specified polymer and solvent | Equivalent protein separation in aqueous buffer |
| Exclusion limit | Approximate onset of near-total exclusion under stated calibration | Sharp universal cutoff for every molecule |
| Nominal pore size | Structural indication of pore architecture | Accessible pore distribution under every mobile phase |
2. Select by Hydrodynamic Behavior, Not Sequence Mass Alone
SEC responds to Stokes radius or a related hydrodynamic dimension. A compact globular protein and an elongated protein of the same molecular mass can elute differently. Glycans, intrinsically disordered regions, bound nucleic acid, detergent micelles, lipid, cofactors, and reversible oligomerization can all enlarge apparent size. Shape may also affect access to pore entrances.
Use sequence mass to identify starting candidates, then consider shape and assembly. If possible, obtain a scouting chromatogram or orthogonal size information. For an unknown sample, screen overlapping pore ranges rather than selecting a single medium from a mass table. The objective is not to place the target at a fashionable elution volume; it is to maximize separation from the critical neighbor while retaining recovery.
3. Pore-Size Distribution Controls Selectivity
Real beads contain a distribution of pore sizes and path lengths. Broad distributions can cover a wide range but may provide less local selectivity than a medium optimized for a narrower region. Accessibility can depend on molecular flexibility, surface chemistry, and mobile phase. The same target may occupy a different normalized position on two media with similar stated ranges.
For aggregate removal, both monomer and aggregate should ideally lie within a region that distinguishes their hydrodynamic sizes. If all aggregates are excluded, they may coelute in a single void peak; this can still separate them from monomer but will not resolve aggregate classes. For small-molecule desalting, the desired arrangement is different: product is excluded or nearly excluded, while the small solute is well included.
4. Particle Size, Matrix, and Mechanics Matter
Smaller, more uniform particles can improve efficiency by reducing dispersion and mass-transfer distance, but usually increase pressure. Rigid matrices support higher flow and larger beds; softer matrices may compress. Matrix hydrophilicity and residual charged or hydrophobic groups affect recovery. Particle-size distribution, bead porosity, and packing quality together determine realized performance.
Review pressure–flow data in the actual buffer and temperature, not water alone. High-viscosity formulations reduce allowable velocity. At process scale, column diameter, wall support, distributor design, and compression method matter. A high-resolution particle is not useful if available equipment cannot pack or operate it reproducibly.
5. Place the Critical Pair in a Useful Region
List target, larger impurities, smaller impurities, and expected complexes. Rank which pair controls success. A target centered in a broad range may be less useful than a target positioned where its critical neighbor shows the greatest elution difference. Compare normalized elution positions rather than raw minutes across columns.
If the target elutes near the void, test a larger-pore medium. If it elutes near total permeation, test a smaller-pore medium. If both target and impurity move together across ranges, pore selection may not provide adequate selectivity and an orthogonal mechanism should be considered. Creative BioMart's SEC product category can support candidate review.
6. Compare Candidate Media Fairly
- Use the same representative sample, mobile phase, temperature, and sample history.
- Match bed height and compare at justified linear velocity or residence time.
- Keep sample volume as a controlled fraction of bed volume and avoid overload.
- Measure void and included markers compatible with each medium.
- Calculate normalized elution position, critical-pair resolution, peak width, asymmetry, pressure, and recovery.
- Analyze all relevant peaks and fractions by target-specific methods.
- Repeat the leading medium and challenge a nearby load and flow condition.
Do not rank media from retention time alone. A later target peak is not automatically better resolved. Compare distance between the critical species relative to their widths, plus usable pool recovery and product quality.
7. Match Format to Purpose and Scale
An analytical column prioritizes efficiency, low dispersion, and reproducibility at small loads. A preparative column must add usable volume, recovery, throughput, packing, and pressure. A desalting format prioritizes a large separation between excluded macromolecule and included small solute. Prepacked and self-packed formats differ in flexibility, consistency, hardware limits, and validation burden.
For scale-up, preserve medium, bed height, packing quality, and the controlling flow parameter. Scale diameter to required bed volume and sample volume. Confirm distributor performance and extra-column volume. If sample supply is scarce, a small column can rank pore ranges, but it may exaggerate the effect of system dispersion.
8. Detect Non-Size Interactions Before Rejecting a Range
Electrostatic interaction often appears as late elution, tailing, concentration-dependent retention, or incomplete recovery. Moderate salt can reduce it if the target remains stable. Hydrophobic interaction may respond to salt identity, organic modifiers compatible with the protein, or a different matrix. Metal-mediated interaction can arise from sample components or hardware. Use no-column controls and complete mass balance.
If changing ionic strength or pH shifts retention substantially without changing target assembly, the column may not be operating as a purely size-based system. A pore range should not be selected to compensate for adsorption; resolve the interaction or choose another matrix. Cleaning history and fouling can create new interaction sites, so include a reference sample over column life.
9. Build a Weighted Selection Matrix
| Criterion | Evidence | Typical decision question |
|---|---|---|
| Critical-pair resolution | Fraction identities, peak centers, widths, and repeatability | Does the medium separate the species that controls acceptance? |
| Recovery and activity | Full mass balance and functional assay | Is apparent resolution created by target loss? |
| Pressure and flow | Pressure–velocity curve in process buffer | Can the desired bed and cycle operate safely? |
| Load tolerance | Volume and concentration challenge | How much usable product is obtained per cycle? |
| Chemical compatibility | Buffer, cleaning, storage, and additive study | Will the medium remain stable through its lifecycle? |
| Implementation | Packing, hardware, supply, documentation, cost | Can the method be reproduced at the required scale? |
10. Resin Specification and Change Control
Record medium identity, matrix, functional surface, stated calibration basis, pore or fractionation information, particle range, pressure limit, chemical compatibility, storage, and packing method. Add the experimental sample, buffer, temperature, column geometry, efficiency, load, flow, normalized elution positions, resolution, recovery, and pressure.
Define acceptable replacement or alternate media through performance, not a shared nominal range. A different medium should be bridged with critical-pair resolution, recovery, product quality, pressure, load tolerance, cleaning, and scale behavior. When calibration data use different standards or endpoints, repeat the sample comparison.
For routine columns, establish an efficiency and asymmetry test, reference-sample window, pressure trend, and maximum cycles or evidence-based replacement criterion. Investigate changes after cleaning, storage, or repacking. A column that still passes a simple plate-count test may have altered selectivity or adsorption for the actual protein, so combine system suitability with product recovery and peak-position trends.
11. Candidate Qualification and Alternate-Resin Strategy
Qualify the leading medium with an independent resin lot and a freshly prepared representative sample. Repeat the critical-pair separation at the target operating condition and at upper load and flow edges. Compare normalized elution position, resolution, asymmetry, recovery, activity, pressure, and pool volume. Lot-to-lot comparison should use the same packed geometry and system whenever possible.
Build a resin specification around demonstrated performance. Include matrix, particle-size range, pore or calibration information, mechanical limit, chemical compatibility, storage, packing method, and required column-efficiency test. Add sample-specific criteria: reference retention window, critical-pair resolution, target recovery, and absence of unacceptable adsorption. Supplier data alone cannot replace these method-specific attributes.
Identify an alternate medium before supply pressure makes the choice urgent. Similar nominal fractionation ranges do not establish equivalence. Bridge the alternate with the actual product and critical impurity, then compare load tolerance, pressure, buffer compatibility, cleaning, and lifecycle. If the alternate shifts the target to a different part of the pore curve, pooling and scale rules may need redevelopment.
For custom or unusual targets, provide the resin team with hydrodynamic information, target and impurity sizes, buffer constraints, sample viscosity, scale, desired bed height, pressure capability, and cleaning needs. Creative BioMart's custom resin development can be evaluated where standard pore distributions or matrices do not provide a useful window. Final selection should still be made with recovery and critical-pair data from the real sample.
Connect this guide to your purification project
Share the protein, sample volume and concentration, current and required buffer, scale, stability limits, target recovery, and analytical requirements. Our team can help evaluate media, dialysis formats, or a fit-for-purpose purification workflow.
References
- Hong P, Koza S, Bouvier ESP. Size-exclusion chromatography for the analysis of protein biotherapeutics and their aggregates. J Liq Chromatogr Relat Technol. 2012;35:2923–2950.
- Erickson HP. Size and shape of protein molecules at the nanometer level determined by sedimentation, gel filtration, and electron microscopy. Biol Proced Online. 2009;11:32–51.
- Carta G, Jungbauer A. Protein Chromatography: Process Development and Scale-Up. 2nd ed. Wiley-VCH; 2020.
