The underlying point is that reading a certificate means checking four things before anything else: traceability to your vial, the test date, the analytical method, and the identity of the testing laboratory.
The test date matters because a certificate six months older than the vial is not evidence about your material, it is evidence about what the material was like half a year ago.
In practice, multiple testing on the same lot reported in a single document suggests internal quality control; identical figures across different lots suggests the same certificate reused.
Research-grade certificates carry no regulatory backing and no standing in quality systems, and that limitation is worth being explicit about.
One qualification: a single result from a single vial at a single time is a point estimate with no error bar, and no amount of decimal places changes that.
If the certificate does not trace to your vial, it is not evidence about your vial.
edited 16 Apr 2025 by thabo_maseko — fixed an arithmetic slip in the third paragraph
8Have you seen anything published on this, or is it inference from the mechanism? – aine_mulcahy 5 months ago Useful. I have added the accept threshold suggestion to my own notes. – s_bhattacharya 7 months ago add a comment