Three assays, three questions
RP-HPLC at 220 nm answers: of all the UV-absorbing material the column resolved, what fraction was my main peak?
LC-MS identity answers: does the main peak have the expected molecular mass?
Amino acid analysis (AAA) answers: of the total mass in the vial, how much is actual peptide versus water, salt, and counter-ion?
These are not redundant. The first asks about resolvability, the second about identity, the third about content. A lot can read 99.42% purity by HPLC and 83.1% peptide content by AAA and both numbers are correct — they are measuring different axes.
What AAA is actually doing
In AAA, a small aliquot of the reconstituted peptide is hydrolyzed — heated with acid until the peptide bonds break — and the freed amino acids are quantified. Compare the molar ratios of the released amino acids with the theoretical ratios from the sequence, and you get a direct read on how much of the vial’s mass was peptide and how much was water, acetate, or other non-peptide mass.
This is slow and fiddly. It’s also the only method that will tell you, definitively, that the 99.4% purity number does not include the 15% acetate you’re about to inject.
Our reconciliation
Before a lot is released, we require: HPLC ≥ 99.0%, LC-MS confirmation of the expected [M+H]⁺, and AAA peptide content within spec for the salt form. If all three agree — in that they are telling a consistent story — we release. If one is out, we investigate. If two are out, we destroy and resynthesize.
The most common “disagreement” is HPLC reading high and AAA reading low-in-spec. That’s almost always acetate content creeping up during lyophilization. It’s fixable, usually with a second reverse-phase pass. We have never released a lot with an unresolved assay disagreement.