Two Peptides at 98 Percent Are Not Equally Pure

Strip the single tryptophan out of a peptide and the resulting fragment all but vanishes from a 280 nm chromatogram while sitting there in plain view at 214 nm. Nothing about the sample changed between those two traces. What differs is how strongly that one molecule absorbs.

Which is the whole difficulty with area percentage. It uses peak area as a stand-in for quantity, and ultraviolet response is not uniform from one molecule to the next.

At 280 nanometers, three residues do all the work

Almost every bit of absorbance at 280 nm traces back to tryptophan. Tyrosine adds a smaller share, cystine a minor one. Molar absorptivities run to roughly 5,500 per molar per centimeter in the tryptophan case and roughly 1,490 for tyrosine.

What follows is stark. Set a peptide carrying one tryptophan beside one carrying two and their 280 nm responses differ by about a factor of two, irrespective of how large either molecule happens to be. Give a peptide neither residue and there is no usable response at all, which rules that wavelength out for a number of compounds.

So the fragment from the opening paragraph is not simply under-counted. A purity figure generated at 280 nm on such a sample errs in a known direction rather than erring safely.

At 214 nanometers, response tracks chain length

Here the chromophore is the peptide bond itself. Every backbone amide adds its share, so a molecule responds roughly in proportion to how many of them it carries.

Count them. Thirty residues means twenty-nine amide bonds; five residues means just four. Inject equal numbers of moles and the two areas come out separated by something in the region of sevenfold. Inject equal masses and the factor shifts again, molecular weights being unequal as well.

Truncated species are therefore under-reported at this wavelength too, systematically. Lose half the chain and roughly half the absorbance goes with it, so 2 percent of such a fragment on a molar basis presents as nearer 1 percent of the total area.

Where the assumption enters

Beer-Lambert explains it. Absorbance, which is what the detector registers, comes out proportional to concentration multiplied by path length multiplied by molar absorptivity. Instrument and injection fix the first two. The third belongs to the molecule and to nothing else.

Peak area accordingly scales with the number of moles crossing the detector times that molecule’s molar absorptivity. When area percentage compares peaks, those products are what it is comparing. Quantities themselves never enter the arithmetic.

Situations where this bites hardest

Three ways out, in rising order of effort

Record at two wavelengths. For any given peak, dividing its 280 nm area by its 214 nm area yields a quantity governed by aromatic content, so a peak whose ratio departs from the main component’s is built differently. The check is qualitative, and it is worth having.

Switch to a detector whose response does not depend on structure. Charged aerosol detection and evaporative light scattering both key on mass instead of chromophore, which removes the problem outright, though calibration then stops being linear.

Or determine the response factors outright, measuring each species of interest against an authentic reference standard. Doing that is what turns area percentage from a ratio into a quantitative statement, and it is what a quantitative method involves.

What the number will and will not support

Two compounds both reported at 98 percent are not thereby equivalent material. Each figure came from some wavelength, applied to a molecule with its own response characteristics, weighed against an impurity population with characteristics of its own. Setting the two side by side assumes a shared scale that does not exist.

The comparison that survives is lot against lot, same compound, same method. Supporting exactly that is the job a purity figure was designed for.

Which makes the wavelength load-bearing information, and a certificate omitting it has withheld what a reader needs. Measured at 214 nm, the figure registers anything with a backbone and gives short species less weight than they deserve. Measured at 280 nm, it registers only what carries an aromatic residue and can overlook entire classes of impurity. Neither answer is incorrect; they are answers to different questions, and the document should say which question it took on. Related reading: reading an HPLC chromatogram and why certificates disagree on purity.