A chromatogram with a small secondary peak and a light-scattering result dominated by a larger size are not necessarily contradictory. Before comparing them, establish which material each measurement represents and how its reported result is weighted.
The same mixture can produce different kinds of evidence
A manufacturer-hosted study compared SEC with light-scattering detection against batch DLS using mixtures of human serum albumin and alcohol dehydrogenase. The second protein acted as a model for a larger species; it was not an actual albumin dimer.Malvern Panalytical — Comparison of SEC-LS and DLS resolution and detection (opens in a new tab)
SEC partially separated the proteins, whereas DLS did not resolve two populations in that experiment. Nevertheless, the DLS average changed as the mixture composition changed.Malvern Panalytical — Comparison of SEC-LS and DLS resolution and detection (opens in a new tab)
This provides a concrete distinction between detecting a difference and separating its contributors. A shifting average can flag a sample change even when the output does not identify two distinct components.
The example is an instrument application study using proteins. It should not be presented as measured performance for a Novum peptide or as a universal resolution threshold for every DLS instrument.
Detection does not guarantee reliable component quantification
In the same comparison, the light-scattering detector registered the larger protein at a level where the concentration detectors did not support reliable molecular-mass determination. Partial chromatographic overlap also affected the measured mass attributed to albumin.Malvern Panalytical — Comparison of SEC-LS and DLS resolution and detection (opens in a new tab)
A method can therefore reveal that something is present while still lacking the information needed for a robust proportion or molecular-mass assignment. The distinction matters when a result is condensed into an “aggregation percentage”.
| Claim | Evidence to look for |
|---|---|
| Change detected | A reproducible difference from the matched comparison sample. |
| Populations resolved | Separately distinguishable components in the reported output. |
| Amount quantified | A supported relationship between signal and component quantity. |
These levels should remain separate in a report summary. Replacing “a change was detected” with “the sample contains a measured fraction of dimers” adds a conclusion the original observation may not justify.
A size result depends on the measurement physics
A 2011 study compared Taylor dispersion analysis with DLS for peptides and proteins, using size-exclusion chromatography when investigating stressed samples. DLS placed greater weight on aggregates than Taylor dispersion in the stressed formulations.Taylor dispersion analysis compared with DLS for therapeutic peptides and proteins (opens in a new tab)
The authors also reported that their DLS setup could not size the small peptide oxytocin, whereas Taylor dispersion could. That is a finding about the evaluated methods and samples, not a statement that all peptide assemblies are inaccessible to DLS.Taylor dispersion analysis compared with DLS for therapeutic peptides and proteins (opens in a new tab)
For an illustrative reading exercise, suppose two reports show a larger average size and a modest separated secondary fraction. Before calling them inconsistent, ask whether the average weights the contributors in the same way as the fraction calculation.
If the measurement bases differ, agreement should be assessed through a defined comparison rather than by expecting their displayed percentages or averages to match.
Investigate disagreement without choosing a preferred answer
A useful reconciliation record identifies the sample time, preparation, concentration, separation conditions, detection channel and calculation used for each result. It also records whether both measurements came from the same original sample.
Ask what entered the separation, what was recovered and whether all relevant material appears in the displayed chromatogram. For DLS, ask which distribution or average was reported and whether the analysis resolved the proposed populations.
These questions help distinguish a real sample difference from incompatible reporting. They do not diagnose a particular loss mechanism or identify an assembly from size alone.
Sources and further detail
- Malvern Panalytical — Comparison of SEC-LS and DLS resolution and detection (opens in a new tab)
Original 2011 manufacturer-hosted study read, including mixture design, detector limitations and partial resolution. The model mixture is not described as actual peptide aggregation.
- Taylor dispersion analysis compared with DLS for therapeutic peptides and proteins (opens in a new tab)
Original 2011 abstract read. Aggregate weighting and oxytocin sizing are kept specific to the studied methods and materials.
Sources checked 20 September 2026. Worked examples are illustrative unless a supplied report is explicitly identified. This article has not undergone independent scientific peer review.