Figure 1. Size-exclusion chromatography, also called gel filtration, separates proteins according to size. (Wolf, 2015)
1. Separation Mechanism
SEC particles contain a distribution of liquid-filled pores. Molecules too large to enter most pores travel mainly through the interparticle mobile phase and elute early. Smaller molecules explore progressively more internal volume and elute later. The relevant property is hydrodynamic size in the mobile phase, which reflects molecular mass, shape, solvation, conformation, association, and any bound detergent or ligand.
The ideal stationary phase is chemically inert toward the sample. Electrostatic, hydrophobic, metal-mediated, or other adsorption adds retention unrelated to size and may cause tailing or incomplete recovery. Because the mobile phase is usually isocratic, selectivity is established mainly by pores and sample behavior rather than a programmed gradient.
| Term | Meaning | Experimental use |
|---|---|---|
| Void volume | Mobile phase outside pores, estimated with a fully excluded marker | Locates species that cannot access useful pore volume |
| Total permeation volume | Volume approached by a very small, fully included marker | Defines the late boundary of accessible liquid |
| Elution volume | Mobile-phase volume delivered to a peak or centroid | Compares sample behavior under a fixed method |
| Distribution coefficient | Normalized position between void and total permeation limits | Supports calibration and medium comparison |
2. Interpret Column Volumes Carefully
A species at the void volume is not necessarily an aggregate; it may simply exceed the medium's useful pore range. A late species may be small, but it may also interact with the matrix. Determine void and total permeation boundaries with compatible markers, correct for system delay, and avoid treating every peak position as molecular weight.
Peak width contains contributions from the packed bed, injection, tubing, detector cell, connectors, diffusion, and sample heterogeneity. Measure column efficiency with a suitable nonretained or included tracer at a non-overloading concentration. An efficient column cannot restore resolution already lost through a broad injection zone or large extra-column volume.
3. Define the Separation Objective
High-resolution fractionation, aggregate removal, polishing, analytical comparability, and desalting are different objectives. List the target and nearest unwanted species, required purity and recovery, starting volume and concentration, final pool constraints, allowable time, and available pressure. For aggregate removal, characterize whether aggregates are soluble and reversible. For oligomer fractionation, consider exchange between states during the run.
Resolution matters only for a relevant pair. A narrow target peak does not prove separation from an impurity that was not measured. Establish target-specific assays and, when function matters, an activity assay. Retain all fractions during development so adsorption, precipitation, or incomplete collection is not mistaken for separation.
4. Select the Medium and Column
Choose a fractionation range that places the critical species within the region where pore access changes meaningfully. Particle size affects efficiency and pressure; smaller particles often improve efficiency but require tighter control of pressure, packing, and sample cleanliness. Matrix rigidity controls the allowable flow and scale, while hydrophilicity and surface chemistry influence nonspecific interaction.
Bed height increases theoretical plates and can improve separation, but it also increases run time and pressure. Diameter determines cross-sectional area and therefore load volume at a fixed bed height. Evaluate a complete medium—pore distribution, particles, matrix, chemistry, format, and mechanical limits—rather than choosing from an advertised mass range alone. Creative BioMart provides size exclusion chromatography products and column loading support.
5. Prepare and Load the Sample
Clarify the sample without removing or altering species that the method is intended to measure. Match sample and mobile-phase composition where possible. A conductivity, pH, solvent, or viscosity mismatch can distort the band and create baseline artifacts. Inspect recovery after filtration because adsorptive loss can be molecule-specific.
Apply the smallest practical volume for a high-resolution separation. Larger injection volume creates a wider starting zone that must be separated before neighboring species can resolve. Concentrating the sample reduces volume, but excessive concentration can increase viscosity, reversible self-association, precipitation, or nonspecific interaction. Optimize volume and concentration together rather than using a fixed concentration target.
6. Set Flow for Efficiency and Stability
Flow affects diffusion, mass transfer into pores, pressure, run time, and exposure to the mobile phase. Very high flow may reduce pore equilibration and resolution; very low flow can broaden bands through longitudinal diffusion and prolong unstable holds. Run a modest flow study while preserving sample load and collection logic.
Express flow as linear velocity or residence time when comparing columns of different diameter. Monitor pressure in the actual buffer because viscosity changes with composition and temperature. A rising pressure trend can reflect particles, precipitation, fouling, gas, or bed compression. Do not solve an unexplained pressure increase simply by lowering the alarm limit.
7. Design the Mobile Phase
The buffer must stabilize every relevant sample species while suppressing unwanted matrix interaction. Moderate ionic strength often reduces electrostatic adsorption, but the appropriate salt and concentration are product-specific. pH should avoid instability and minimize matrix or surface interactions. Cofactors, reducing agents, detergents, or stabilizers may be necessary, yet each can change hydrodynamic size, viscosity, refractive index, detector response, and chemical compatibility.
Degas or manage dissolved gas as appropriate, control temperature, and prepare enough homogeneous mobile phase for equilibration and the full run. Filter only when compatible with the formulation and verify that the filter does not remove a required component. A stable baseline does not establish chemical suitability.
8. A Method-Development Sequence
- Define the critical pair and quantitative recovery target.
- Screen target stability in candidate mobile phases without a column.
- Select two or more pore ranges when hydrodynamic behavior is uncertain.
- Establish void volume, included volume, pressure–flow behavior, and system dispersion.
- Run a low-volume, non-overloaded sample and identify all fractions.
- Optimize flow, load volume, concentration, bed height, and buffer one variable or designed region at a time.
- Confirm with representative sample lots and deliberate condition edges.
- Set pooling rules from composition and quality, not UV alone.
9. Troubleshooting by Pattern
| Observation | Likely causes | Useful checks |
|---|---|---|
| Broad peaks | Large load volume, dispersion, poor packing, diffusion, heterogeneity | Reduce injection volume; test efficiency; inspect connections and sample |
| Late or tailing target | Nonsize interaction, pore trapping, aggregation–dissociation | Change ionic strength or pH; lower load; measure recovery |
| Target at void | Target exceeds pore range, aggregate, or association | Use a larger-pore medium; orthogonally assess assembly |
| Low recovery | Adsorption, precipitation, incomplete collection, filtration loss | Analyze all fractions and surfaces; run no-column controls |
| Run-to-run shift | Buffer, temperature, flow, column condition, or sample change | Use reference sample; verify preparation, calibration, and pressure |
10. Scale-Up and Method Record
Scale preparative SEC by preserving the selected medium, bed height, and justified linear velocity while increasing diameter. Keep sample volume as the same fraction of bed volume only after confirming that concentration and viscosity remain comparable. Account for distributor design, packing quality, system dwell, tubing, detector, and fraction delay. Large equipment can add enough dispersion to erase a small-column advantage.
The method record should include sample history, mobile-phase recipe, temperature, medium and lot, column geometry, packing result, load mass, concentration and volume, flow, pressure, detector settings, all fraction boundaries, recovery, activity, resolution, and pool criteria. Define equilibration, cleaning, storage, maximum pressure, and replacement triggers. Record deviations and unsuccessful conditions; they establish the design boundary and prevent unproductive repetition.
For routine use, add a reference sample or suitable standards that challenge retention and critical-pair resolution. Trend peak position, width, asymmetry, pressure, and recovery. Investigate coordinated drift before system suitability fails. A method is controlled when changes in sample, column, buffer, and instrument can be distinguished, not merely when the chromatogram looks familiar.
11. Implementation Checklist and Acceptance Criteria
Before the first development run, create a one-page protocol that separates fixed conditions from experimental factors. Fix sample history, mobile-phase preparation, temperature, column geometry, equilibration, detector settings, and fraction delay. Deliberately vary only the medium, load, flow, or buffer variable under study. Assign a sample identifier to every fraction and preserve enough material for repeat and orthogonal analysis.
Set numerical or categorical acceptance criteria before optimizing: minimum target recovery and activity, required resolution or impurity reduction, maximum pressure, acceptable peak asymmetry, and pool-volume limit. Add a criterion for column interaction, such as mass balance or recovery after a compatible ionic-strength challenge. A condition that meets resolution only by losing target to the medium should fail.
For a new protein, run a concentration series at the chosen low-volume load. Stable normalized elution position and peak shape support interpretation by size; concentration-dependent shifts suggest association, dissociation, viscosity, or overload. Collect the leading edge, center, and tail separately at least once. Identity and quality data from these fractions reveal whether one apparent peak contains changing molecular populations.
Before routine transfer, challenge small changes in sample volume, concentration, flow, temperature, and buffer preparation. Repeat with an independent sample lot and, when the column will be reused, after representative cleaning. Define the system-suitability sample, allowable position and efficiency windows, pressure baseline, and response to failure. These controls turn a descriptive SEC run into a reproducible method.
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.
- Wolf M. Effective Interactions in Liquid-Liquid Phase Separated Protein Solutions Induced by Multivalent Ions. Universitätsbibliothek Tübingen; 2015. doi:10.15496/PUBLIKATION-4822
