The Gold Standard of Polymer Science | Detector Response Factors and Mass Fractions in Copolymer SEC

Size-exclusion chromatography, or SEC, separates material according to its behaviour in a column. A concentration-sensitive detector then records a signal. A calibration or another detector supplies information used to assign molar mass.

These are separate steps. Correctly assigning a slice’s molar mass does not guarantee that its integrated detector signal has been converted into the correct polymer mass. Conversely, good concentration weights do not repair an incorrect mass calibration.

Here we isolate detector weighting. Assume two resolved, sufficiently narrow fractions with independently correct slice molar masses M₁ and M₂. Their signal areas are baseline-corrected and measured under the same stable conditions.

We exclude overlapping impurities, concentration-dependent nonlinearity, losses and detector saturation from the initial model. Later chapters explain why those exclusions matter. Keeping one mechanism isolated lets the arithmetic reveal exactly where the apparent distribution changes.

Contents · Next chapter

Let Ai be the integrated signal from fraction i, mi its recovered polymer mass and ki its positive response factor. Define the units so Ai = ki mi. Then mi = Ai/ki.

The mass fraction is wi = (Ai/ki)/Σj(Aj/kj). Dividing signal areas by their response factors must occur before normalising. Normalising raw areas assumes all ki are equal.

For a differential refractive-index concentration detector, refractive-index increment dn/dc is part of the concentration conversion. Its applicable value depends on the material and experimental conditions. The linked NIST work addresses that measurement background.

Our k includes the fixed instrumental and integration factors needed for the chosen units. Do not substitute a literature dn/dc directly for k without checking the instrument’s calibration, flow and signal conventions. The simple equation describes the linear measurement model being tested.

Contents · Previous chapter · Next chapter

Set A₁ = A₂ = 1 signal unit, k₁ = 1 signal unit per milligram and k₂ = 2 signal units per milligram. The recovered masses are m₁ = one milligram and m₂ = half a milligram.

Total recovered polymer is 1.5 milligrams. Thus w₁ = 2/3 and w₂ = 1/3. The second fraction produces the same signal from half as much mass because it has twice the response per milligram.

Raw-area normalisation would give one half to each fraction. It overweights the second fraction and underweights the first. This is a relative-composition error even if another operation later forces the total mass to an apparently correct value.

Check by reconstructing signals from corrected masses: 1×1 = 1 and 2×0.5 = 1. A successful correction must reproduce the observed ledger under the stated response model, rather than merely produce weights summing to one.

Contents · Previous chapter · Next chapter

Let the correctly assigned narrow-fraction masses be M₁ = 50,000 and M₂ = 200,000 grams per mole. With corrected mass fractions, Mw = ΣwiMi = (2/3)×50,000 + (1/3)×200,000 = 100,000 grams per mole.

Using the raw area fractions would instead give 125,000 grams per mole. It overstates Mw by 25% relative to the corrected value, because the higher-molar-mass fraction also happens to have the higher detector response.

That direction is not inevitable. If the lower-molar-mass fraction had the stronger response, uncorrected area weighting could bias Mw downward. The relationship between response and assigned mass determines the effect.

No physical molecules changed when the weights were corrected. The calculation changes the interpretation of the trace. It also shows why a detector calibration problem can affect an apparent molar-mass distribution even when the slice mass axis is independently accurate.

Contents · Previous chapter · Next chapter

For known mass fractions, Mn = 1/Σ(wi/Mi). Corrected weights give Mn = 1/[(2/3)/50,000 + (1/3)/200,000] = 66,666.7 grams per mole.

Using equal raw-area fractions gives Mn = 1/[0.5/50,000 + 0.5/200,000] = 80,000 grams per mole. Corrected dispersity is Mw/Mn = 1.5; the apparent value is 125,000/80,000 = 1.5625.

Both averages and their ratio can change. The error is not generally repaired by multiplying every reported summary by one constant. A constant multiplier would leave dispersity unchanged, whereas this response-weighting correction does not.

The reciprocal expression for Mn reflects molecule counts: a fixed mass of shorter molecules contains more molecules. Do not use an arithmetic average of slice molar masses while calling it a number average unless the weights are actual number fractions.

Contents · Previous chapter · Next chapter

A copolymer contains more than one repeat-unit type. As a simplified teaching model, suppose a fraction with mass fraction x of component B has response k(x) = (1 − x)kA + xkB, at fixed measurement conditions.

This linear mixture relation is an assumption for the exercise, not a universal rule for every copolymer solution. Let kA = 1 and kB = 3 in the same response units. A slice with x = 0 has k = 1; a slice with x = 0.5 has k = 2.

Those two compositions reproduce the response factors in the preceding example. If composition varies with elution position, one bulk k may fail to provide correct slice weights even though it describes the whole recovered sample’s total signal.

The column does not guarantee constant composition across the trace. To apply the correction, establish composition-sensitive information or justified slice-specific response factors rather than assuming every copolymer chain has the same composition.

Contents · Previous chapter · Next chapter

The original two-fraction sample has total signal two units and total mass 1.5 milligrams. Its bulk mass-weighted response is therefore kbulk = 2/1.5 = 4/3 signal units per milligram.

Convert each equal-area peak using that bulk factor. Both receive an estimated mass 1/(4/3) = 0.75 milligrams. The total remains correct at 1.5 milligrams, yet the resulting fractions are one half and one half.

The true masses were one and half a milligram. A successful total recovery check therefore does not establish correct allocation across fractions. The same total can conceal offsetting slice errors.

Even a known total mass plus two equal signals does not uniquely determine both response factors. Many positive mass pairs summing to 1.5 can be paired with ki = 1/mi to reproduce the signals. Additional response or composition information is needed to identify the split.

Contents · Previous chapter · Next chapter

A multi-detector SEC experiment may use a refractive-index signal to estimate concentration and light scattering to infer molar mass. The relevant optical response can enter more than one stage, so errors can be coupled.

The present exercise deliberately treats M₁ and M₂ as independently known. It cannot be used unchanged if those slice masses were themselves calculated with an incorrect refractive-index increment.

Likewise, an ultraviolet detector’s response can depend on the amount and environment of absorbing groups. Equal mass concentrations of different compositions need not produce equal absorbance. A detector name alone does not establish a universal response factor.

Trace the measurement chain: observed signal, concentration conversion, molar-mass assignment, weighting and final summary. Mark every place a material-dependent parameter enters. Correcting one stage while holding an already biased later result fixed can leave the final answer inconsistent.

Contents · Previous chapter · Next chapter

Ai = ki mi assumes the area represents only the relevant polymer fraction. A baseline offset, solvent feature or co-eluting additive can add signal without adding that polymer mass. Dividing by ki does not remove contamination.

Incomplete recovery changes the population too. Corrected weights describe recovered material unless recovery of the original sample is established. An extraction or filtration step can preferentially remove part of the distribution.

If ki is close to zero, a small signal uncertainty can create a large relative mass uncertainty. A nearly invisible component cannot be recovered reliably just by using a very large numerical correction.

Record solvent, temperature, detector wavelength, calibration procedure, response-factor uncertainty and recovery evidence. Use another suitable measurement when the response contrast leaves the desired mass fraction poorly identified. A tidy normalised chromatogram does not supply those missing observations.

Contents · Previous chapter · Next chapter

Use A₁ = 3 and A₂ = 2 signal units, with k₁ = 1 and k₂ = 4 signal units per milligram. Corrected masses are three and half a milligram, so the mass fractions are 6/7 and 1/7.

Let the independent slice molar masses again be 50,000 and 200,000 grams per mole. Corrected Mw is (6/7)×50,000 + (1/7)×200,000 = 500,000/7, approximately 71,428.6 grams per mole.

Corrected Mn = 1/[(6/7)/50,000 + (1/7)/200,000] = 56,000 grams per mole. Dispersity is approximately 1.2755. Raw-area weights 0.6 and 0.4 would give Mw = 110,000, Mn approximately 71,428.6 and dispersity 1.54.

Explain the difference without saying the second peak is larger by mass. It supplies 40% of the signal but only about 14.3% of the recovered mass. The detector response is the bridge between those statements.

Contents · Previous chapter · Next chapter

Are equal peak areas ever equal masses? Yes, when the relevant integrated response factors are equal and the linear signal model is valid. Equal factors need evidence, not an assumption hidden by normalisation.

Does this replace SEC calibration? No. It corrects concentration weighting under independently correct slice masses. The calibration guide treats a different route by which the reported distribution can change.

Can bulk composition determine every slice composition? Generally not. It constrains an overall average and leaves many possible distributions across elution position.

Is a universal correction factor available for all copolymers? No. The response must be appropriate to the material, detector and experimental conditions. The invented factors in this guide should never be used as laboratory reference values.

Contents · Previous chapter · Next chapter

A defensible interpretation keeps raw areas, response factors, recovered masses, normalised mass fractions and molar-mass assignments in one ledger. Reconstruct the signal after correction and check units at each step.

The numerical examples here are independent exercises in a stated linear response model. NIST’s refractive-index increment work provides the material-measurement context; it does not certify the invented copolymer mixture relation or supply the example factors.

For teaching, have the learner explain why one sample can pass total-mass recovery while failing slice allocation. Then change which fraction has the stronger response and predict the bias direction before calculating.

For transfer, remember that normalisation removes a common scale factor but not unequal response factors. The same issue appears whenever different constituents produce different signal per unit of the quantity we actually want to compare.

References checked 22 October 2026

André Striegel, NIST: specific refractive-index increment of polymers and measurement conditions

Continue through Polymer Science

Size-exclusion calibration and distorted molar-mass distributions

Extractable additives and incomplete mass recovery

Matching molar-mass averages, different chain populations

Polymer Science root and full subject directory · The Gold Standard of Education · How X Works

Contents · Previous chapter · Return to Polymer Science