Impurity reconciliation checker
Paste a purity figure and a list of impurity area percentages; the checker closes the chromatogram, ranks the peaks, and tells you which index rules the certificate would trigger.
| # | % area | Share of impurity total |
|---|---|---|
| no impurity values entered | ||
The default values are a well-behaved certificate: five named peaks summing to 1.40 %, closing against a stated 98.60 % purity to exactly 100.00 %.
What is being checked
closure : purity + Σ(impurity area %) = 100.00 ± 0.02
ordering : largest single impurity ≤ total impurities
consequence : a certificate failing either is internally inconsistent,
regardless of whether the material is good
Chromatographic closure is not a quality test. It is an arithmetic test, and it is the cheapest one available: a certificate whose own numbers do not add up cannot be reconciled against anybody else's, including a re-test of the same lot. The index runs this check on every record it holds and on every supplier certificate, and it publishes the failures under rules AX-R02 and AX-R03.
The two ways a certificate fails this
The sum does not close. Purity 99.2 % with impurities totalling 1.4 % describes 100.6 % of a chromatogram, which does not exist. The commonest innocent explanation is that the two figures were produced on different runs, or that the purity figure was rounded up and the impurity table was not. The commonest guilty explanation is that the purity figure was written first.
The largest single exceeds the total. Arithmetically impossible: a part cannot exceed its whole. Almost always this is a certificate template with the two fields transposed, which is why the index treats it as high severity but not as evidence about the material — it is evidence about the document.
What closure does not tell you
Whether the peaks below the reporting threshold were counted. They were not — 0.05 % area is the conventional threshold and anything under it is neither integrated nor reported, so it is excluded from the impurity total and is silently added to the purity figure. Twenty peaks at 0.04 % is 0.8 % of unreported material, and a certificate carrying that will close perfectly. Closure is a necessary condition and nowhere near a sufficient one.