This annoys me unreasonably, because they use Quebec for ‘Q’, but the correct pronunciation of Quebec is KEH-bek. It’s as dumb as pronouncing Lyon as ‘LIE-un’ instead of ‘lee-OHN’.

You absolutely can not use it to quantify impurities. All you can really do is say whether it starts and finished melting in a range that’s consistent with the claimed substance. It’s a starting point, and that’s all. But if you’re melting fairly far out of range - like 3C or more - and you know that your equipment is correctly calibrated, then you know that there’s some kind of significant issue.

For illicit drugs, there are a few services in the EU that I’m aware of that will run a form of quantitative testing, HPLC and mass spectrometry. IIRC - I know nothing about analytical chemistry - they need a pure sample to compare against; they can’t just run their tests and then search a giant database and tell you what you have. But if you have a few kilos of MDMA, and you’re worried about it being contaminated with fentanil, it’s definitely worth the couple hundred dollars it costs.

I have an actual melting point apparatus, Thermo Scientific, IIRC. You need to use capillary tubes for your samples (very, very small amounts). You also need to check the calibration occasionally against known substances. I do recall something testing very far out of range - like 20C or something - because the sample wasn’t properly dehydrated first. I don’t recall if I ever got around to doing that and re-testing or not.

Point is, this isn’t really a very accurate way of checking your melting point.

And MP alone doesn’t give you a ton of information. Lots of things can melt in the same range.

Melting point is a range, not a single temperatures. Very small amounts of fentanyl aren’t going to change that range enough to really register. You would need to use another test, probably a chemical reagent, to check for fentanyl. Or send your drugs out to be tested by high-pressure liquid chromatography and mass spectrometry.