Extinction Coefficient Calculator

The coefficient is not a property of the molecule on its own. It is a property of the molecule at one wavelength, which is why every coefficient quoted anywhere carries its wavelength beside it and why a value at 280 nanometres cannot be used to interpret a reading taken at 214 nanometres. The calculation itself is a weighted count: each tryptophan adds 5500, each tyrosine adds 1490, and each disulfide bonded pair of cysteines adds 125, all in inverse molar centimetres. Phenylalanine adds nothing here, because its ring absorbs further into the ultraviolet than the standard wavelength.

The most instructive thing about that list is how little of the sequence it uses. Two peptides of the same length carrying the same number of tryptophans and the same number of cystine pairs return an identical coefficient, however completely different the rest of their residues are. A chain of acidic residues and a chain of basic residues look the same to this model. That is what makes the number usable for a concentration measurement and a poor description of a molecule, and it is worth knowing which of those two you are holding before you use it.

The disulfide term is the one input that cannot be read off a sequence. A cysteine may be free, paired with a second cysteine, or attached to something else entirely, and only the paired form contributes. A one letter sequence cannot distinguish those cases, so the count is supplied by hand rather than inferred, which is why the field exists on this form at all. Entering the number of cysteines assumes every one of them found a partner, and an odd count makes that impossible. The aromaticity page counts the residues that drive the rest of the total.

These coefficients were measured on fully reduced and fully denatured material, and the model is valid to the extent that a measurement reproduces those conditions. Denaturing exposes aromatic side chains that would otherwise sit inside a folded structure, and reducing breaks disulfides so that each cystine term is accounted for once. A sample measured while still folded absorbs less than the model predicts, so the concentration read from it comes out low. That is a systematic bias with a known direction, not random scatter, and knowing the direction in advance is what makes it correctable: denature the sample, or read the result as a lower bound rather than as the concentration.

The second assumption is that the aromatic side chains are the only absorbing species in the cuvette. Anything else that absorbs at 280 nanometres is charged to the peptide, so a buffer with real absorbance, a trace of detergent, or a suspension of fine particles all push the reading up and the derived concentration up with it. The remedy is a blank made from the same solvent without the peptide. Acidic conditions are used for a lot of peptide work for exactly this reason, because a low pH suppresses the absorbance of many buffer components instead of adding to it, and the Beer-Lambert guide sets out the rest of what the measurement needs.

The third assumption is that the chromophores behave independently of one another. The model gives every aromatic ring a fixed coefficient and makes no correction for two of them sitting side by side, for a ring buried in a hydrophobic interior, or for a sample that has associated in the cuvette. Those corrections are small in the cases the method was built for. They are not small at all for a peptide that is aggregating, and aggregation is the failure mode that is hardest to notice from the outside because the spectrum continues to look like a clean absorption band with a plausible shape.

A peptide with no tryptophan and no tyrosine reduces the total to the disulfide term alone. A reduced peptide with no cystine therefore has a coefficient of exactly zero, and the law then says that every concentration whatsoever produces an absorbance of zero. This is not the arithmetic breaking down. It is the correct statement that the measurement carries no information about how much peptide is in the cuvette. One tyrosine is enough to change that completely, and so is a single bonded cystine pair, which is why peptides are often engineered to carry one.

Solving the law for concentration divides the measured absorbance by the product of the coefficient and the path length, so a coefficient that approaches zero produces a very large and very confident looking number assembled entirely out of noise. Nothing about the result looks wrong on the page. The Beer-Lambert concentration calculator refuses a zero or negative coefficient outright for that reason, and the refusal is the useful answer rather than an inconvenience. A small but non zero coefficient is nearly as bad, because it amplifies the same error instead of eliminating it, and the amplification is entirely silent.

For a peptide like that the concentration is not knowable from the sequence. No arithmetic on the letters will produce it, and any specific figure for a specific peptide would have to come from a measurement or from a cited source describing that exact molecule, neither of which this site can supply. What can be said is the shape of the problem instead: the options are a detection method that responds to something else, or a measurement taken at a wavelength where part of the sample does absorb. Saying so plainly is more useful than a plausible figure, because a wrong coefficient propagates directly into a concentration and then into every dilution worked out from that concentration.

The molar coefficient is not the only form this number takes. Divide it by the average molar mass and you get a mass extinction coefficient, conventionally written in mL g^-1 cm^-1, which is the form that pairs with a concentration in grams per litre instead of mol per litre. The arithmetic is straightforward: a coefficient of 12615 for a peptide of average molar mass about 611 gives a mass coefficient of about 20.7. The average molecular weight calculator supplies the second of those two numbers, and converting back is the same division in the opposite direction, so the two forms can never disagree as long as the same molar mass is used for both.

Mixing the two forms is the standing mistake on this subject, and it is wrong by the molar mass rather than by a small factor, which is why it can go unnoticed for so long. The molarity against mass concentration guide sets out the relationship between the two concentration units that sits underneath the conversion, and feeding a mass concentration into the Beer-Lambert absorbance calculator will show you the size of the resulting error directly rather than leaving it for you to reason about. The output is still an absorbance, it is still unitless, and it is simply out by a factor of several hundred.