Detecting What Does Not Absorb: ELSD and CAD

Ultraviolet detection assumes the analyte absorbs ultraviolet light. Several compounds do not, and for those the chromatogram from a UV detector is a flat line with the answer somewhere in it. Two detector types solve this by responding to mass instead.

What makes a molecule invisible at 214 and 280 nanometres

Absorbance near 280 nm requires an aromatic ring — tryptophan, tyrosine, or an unnatural residue carrying one. Plenty of peptides have none.

Absorbance near 214 nm requires the peptide bond, which every peptide has. But a molecule that is not a peptide has no backbone amides, and a molecule with very few has very little signal. Bile acids, sugars, lipids, most salts and counter-ions, and many small-molecule research compounds fall below useful response at both wavelengths.

Evaporative light scattering detection

ELSD nebulises the column effluent into a fine spray, evaporates the mobile phase in a heated drift tube, and passes the remaining particles through a light beam. Particles scatter the light and a photodetector measures the scattering.

Anything less volatile than the mobile phase produces a signal, chromophore or not. That is the point. The trade-offs are specific:

• The response is not linear. Signal varies with particle size distribution, so the relationship between amount and response follows a power law rather than a straight line. Quantitation requires a calibration curve, and area percentage across species means less than it does with UV.
• Volatile analytes are lost. Anything that evaporates in the drift tube along with the mobile phase produces no signal.
• Non-volatile mobile phase components produce constant background. Phosphate buffers are incompatible; volatile additives such as formic acid or ammonium formate are required.
• Sensitivity is modest. Typically higher limits of detection than UV for a molecule that does absorb.

Charged aerosol detection

CAD works on a similar principle with a different final step. After nebulisation and evaporation, the particles are charged by a stream of ionised nitrogen, and the resulting charge is measured directly.

Because charge transfer depends less strongly on particle size than light scattering does, CAD gives a more nearly uniform response across different analytes and a wider linear range. In practice it is the more capable of the two and the more expensive.

The same mobile-phase constraint applies: volatile buffers only, and gradient elution causes a baseline change because the proportion of organic solvent affects nebulisation efficiency. Some instruments compensate for this; where they do not, an inverse gradient is plumbed in to hold the composition at the detector constant.

Where each one earns its place

Three situations recur.

An analyte with no chromophore. There is no UV alternative, so the question is only which mass-response detector.

A mixture of species with very different chromophores. UV area percentage distorts the ratio badly when one component has an aromatic residue and another does not. A mass-response detector reports something much closer to composition.

Impurities that might be invisible. A UV method sees only what absorbs. A parallel ELSD or CAD trace can reveal a non-absorbing component — a residual reagent, an excipient, a counter-ion excess — that the UV chromatogram misses entirely.

What they do not do

Neither detector identifies anything. They report that something non-volatile eluted at a given time, with no information about what it was. Identity still comes from mass spectrometry or from comparison against a reference standard.

Neither is a replacement for UV on a peptide with a normal chromophore. UV is more sensitive, linear, cheaper and better understood, and where it works it remains the correct choice.

Reading a certificate that used one

An area percentage from ELSD carries a different meaning from an area percentage from UV, because the response function differs. A 98 percent ELSD figure and a 98 percent UV figure are not the same claim, and neither is comparable to the other.

What matters is that the document names the detector. A purity figure with no detector stated cannot be interpreted at all, and for a compound with no chromophore, a figure attributed to UV detection is a reason to ask what was actually measured. The underlying issue of non-uniform response is covered in why two peptides at 98 percent are not equally pure.

The practical summary

If a compound absorbs, use UV and understand its response factors. If it does not, a mass-response detector is not an alternative to UV — it is the only option, and its non-linear calibration is the price of seeing the analyte at all. Related reading: reading an HPLC chromatogram.